PARENTING STYLES AND CRIMINAL INVOLVEMENT:

A TEST OF BAUMRIND’S TYPOLOGY

A Dissertation

Submitted to the School of Graduate Studies and Research

in Partial Fulfillment of the

Requirements for the Degree

Doctor of Philosophy

Jason D. Spraitz

Indiana University of Pennsylvania

August 2011

 
 © 2011 by Jason D. Spraitz

All Rights Reserved

ii
 
 Indiana University of Pennsylvania
School of Graduate Studies and Research
Department of Criminology

We hereby approve the dissertation of

Jason D. Spraitz

Candidate for the degree of Doctor of Philosophy

July 5, 2011 Signature on File

_________________________ _________________________________________
Jamie S. Martin, Ph.D.
Professor of Criminology, Chair

July 5, 2011 Signature on File

_________________________ _________________________________________
Kathleen J. Hanrahan, Ph.D.
Professor of Criminology

July 5, 2011 Signature on File

_________________________ _________________________________________
Jennifer J. Roberts, Ph.D.
Associate Professor of Criminology

July 5, 2011 Signature on File

_________________________ _________________________________________
John A. Lewis, Ph.D.
Assistant Professor of Criminology

ACCEPTED

Signature on File
______________________________________ _____________________________
Timothy P. Mack, Ph.D.
Dean
The School of Graduate Studies and Research

iii
 
Title: Parenting Styles and Criminal Involvement: A Test of Baumrind’s Typology

Author: Jason D. Spraitz

Dissertation Chair: Dr. Jamie S. Martin

Dissertation Committee Members: Dr. Kathleen J. Hanrahan

Dr. Jennifer J. Roberts
Dr. John A. Lewis

The current study sought to examine the relationship between perceived parenting styles

and level of criminal involvement. Baumrind’s typology of parenting styles is based on research

conducted approximately 40 years ago (Baumrind, 1967, 1971, 1972; Baumrind & Black, 1967).

Despite its longevity in the social sciences, the typology has enjoyed limited empiricism in the

field of criminology.

Four parenting styles form the crux of Baumrind’s typology. These four styles:

authoritarian, authoritative, neglecting/rejecting, and permissive parenting are defined by the

level of demandingness and responsiveness that parents display. This study provided an in-depth

examination of demandingness, responsiveness, and the four parenting styles that their

convergence creates. Prior empirical research examining the effects that parents’ parenting styles

have on their children is assessed.

A methodology was devised that allowed the researcher to study the perceptions of

parenting styles and the impact that they had on subsequent deviant, delinquent, and criminal

involvement among two different samples – county jails inmates and university students.

University students were randomly selected to participate, were contacted via university email,

and were asked to complete an online survey. A convenience sample of jail inmates was asked to

complete a paper copy of the same survey.

iv
 
 The results from this study suggest that the permissive parenting style is the least

problematic and that those parented by permissive parents are less likely to engage in acts of

deviance, delinquency, and crime. Not surprisingly, respondents parented by neglecting/rejecting

parents were more likely to report engagement in deviant, delinquent, and criminal acts.

Meanwhile, the authoritarian and authoritative parenting styles predicted very few behaviors. No

matter one’s distinction (college student or county jail inmate; female or male; non-white or

white), the style by which one was parented is predictive of behavior.

v
 
 ACKNOWLEDGEMENTS

In this acknowledgment section I would like to thank the individuals who helped me

through the dissertation; I also would like to thank those who have been there for me throughout

this entire four-year process. First, I want to thank my family for their unconditional support. My

parents – Jaci and Curt – and my grandparents – Joanne and Ray – have always encouraged me

to follow my dreams and set my goals high. Words cannot express how much their guidance and

love has meant to me throughout my life. I also want to thank my younger brothers Shaun and

Joshua for all of their brotherly love. Shaun is a writer and Joshua is a musician – both of them

are going to do great things.

Next I would like to thank my dissertation chair, Dr. Jamie Martin. Jamie has been

working with me on this project in one capacity or another for the past three years beginning

when it was simply an idea for an independent study. Her expertise, attention to detail, and time

investment cannot be understated. Without her guidance, not only would this project not be

completed, but it also may never have even come to fruition. Jamie, thank you for your help with

this project and for being a great mentor.

I also would like to thank my dissertation committee. Dr. Kate Hanrahan, Dr. Jen

Roberts, and Dr. John Lewis all provided ideas for improving this project and caught many

mistakes that I made while working on it, and in many ways laid the groundwork for my ability

to complete it. I wrote my first research proposal as a student in Kate’s qualitative class and I

created my first survey instrument as a student in 901 with Jen. The skills that I learned from

each were necessary for the completion of this project. Without John, I may still be testing

countless hypotheses and without all three I may have started an unwanted war of words with a

vi
 
couple of pre-eminent criminological theorists. Kate, Jen, and John: thank you for working with

me on this project and for always having an open door when I needed to chat about something.

There are a couple of other faculty members who I would like to thank. First, I would like

to thank Dr. Dave Myers. Dave was doctoral coordinator when I first arrived and really sold the

program to me when I was deciding what school to attend. He also gave me the opportunity to

work on the editorial staff of Criminal Justice Policy Review, which allowed me to learn how the

peer-review process works. Dave, thank you for all of the opportunities that you have given me

over the past four years and for being a great boss in the process. Next, I would like to thank Dr.

Dan Lee. Dan has been a sounding board for me, both professionally and personally, for four

years; actually, I think I wore a rut in the carpet between our offices with my constant trips back

and forth. Dan, I cannot thank you enough for sharing your time, insight, and expertise.

In addition, I would like to thank every other faculty member for believing in my ability

to teach and giving me the opportunity to do so during the past two years, and for always being

available to answer a question or have a conversation. I would like to thank Christoph Meier,

who helped me in the early stages of my pre-test when he was the director of the Applied

Research Lab. I would be remiss if I did not thank Ms. Dottie Hillard and Mrs. Karen Lemasters

for all of the things that they have helped me with. I always had a mundane question about paper

or office supplies for Dottie, but she always helped me with a smile on her face. Karen was

always tremendous in her role as doctoral secretary, but she also helped CJPR immensely

(thanks for the awesome pens, Karen).

Next, I would like to thank my cohort. I’m not sure how far I would have made it without

James Bowers, Jenni Huck, and Kendra Bowen. On the professional side, thank you for

exhibiting a level of discipline and dedication that I knew I would have to match if I wanted to

vii
 
stick around and for being supportive through qualifiers, TA/Temp applications, real job

applications, and the dissertation. On the personal side, thank you for being my first group of

friends in Indiana, for the game nights at Jenni’s house, flashback Friday’s at James’ house,

Thanksgiving dinners at Kendra’s house, and quirky road trips (be it to the Cheesecake Factory

or home from Cincinnati or waiting for flights in St. Louis). We have been through a lot already

and I hope we can maintain personal and professional ties as we branch out across the country.

I would like to thank a group of individuals who have provided unlimited support during

the past few years; essentially, I consider these three people my best friends (and obviously more

in one instance): Maddie Ursitz (the one instance), Steven Brewer, and Brian Iannacchione.

Words really cannot express how much each of them mean to me and I do not have enough time

to recount all of my favorite moments with them. But, they are all fans of LOST, so I want to

share with them a passage from the closing moments of the series finale that sums up how I feel

about them (and, if you are reading this without having viewed the finale of LOST, consider this

a spoiler alert): “This is a place that you…that you all made together so that you could find one

another. The most important part of your life was the time that you spent with these people.

That’s why all of you are here. Nobody does it alone…You needed all of them, and they needed

you.” Brian, thanks for your friendship. Maddie, thanks for more than that. Steve, thanks for

your friendship and for all of the Qualtrics help early on.

I also would like to thank Phil Stinson. Phil showed me the ropes early on and continues

to be a good friend. Additionally, without Phil’s connections, I am not sure if I would have

gotten into one of the correctional facilities that served as a research site in this study. Phil,

thanks for all of your help over the years. Also, I would like to thank Mari Pierce and Diana

Falco. They were in their third and fourth years when I arrived at IUP, but they took me under

viii
 
their wing almost instantly. They provided guidance and friendship during my early years and

continue to do so today. Thank you, Mari and Diana. Also a word of thanks to Brandon Stroup;

you’ve been a good friend during our time here. Thanks also to my friend and roommate for the

past four years, Jon Rezk, I couldn’t have asked for a better living situation. And thanks to Max

and Pippi, although they will never read this; I’ve probably told those two dogs things that I

haven’t told any people.

I would like to thank the administrators and staff members of the two correctional

facilities that I gained access to, especially Carol, Lesley, Lance, and LaToya. Without their

assistance half of this study would not have been possible. They were the epitome of

professionalism and class. Additionally, I would like to thank everybody who completed the

survey. Without their time and participation NONE of this study would have been possible.

Finally, to the one thing that I have spent the most time with over the past two years, this

dissertation. Thank you, dissertation, for your looming presence and making me a better

academic. When describing you, and I am fairly certain you will agree, I think Bruce Springsteen

sums it up best with his immortal words to Mary in Thunder Road, “you ain’t a beauty, but hey

you’re alright.”

ix
 
 TABLE OF CONTENTS

Chapter Page

I INTRODUCTION ............................................................................................................ 1

The Current Study ............................................................................................................ 3

II LITERATURE REVIEW ................................................................................................. 6

Conceptualization of Baumrind’s Four Parenting Typologies ......................................... 6

Demandingness ................................................................................................... 6

Confrontation .......................................................................................... 7

Monitoring .............................................................................................. 7

Consistent discipline ............................................................................... 8

Responsiveness ................................................................................................... 8

Warmth ................................................................................................... 9

Reciprocity .............................................................................................. 9

Communication and discourse .............................................................. 10

Authoritarian ..................................................................................................... 11

Authoritative ..................................................................................................... 15

Permissive ......................................................................................................... 16

Neglecting/Rejecting......................................................................................... 19

Examining Parenting Style and Child Behavior............................................................. 20

The Link between Parenting Style and Deviant Behavior ................................ 21

Alcohol, tobacco, and substance use..................................................... 22

The Link between Parenting Style and Delinquent/Criminal Behavior ........... 30

Adolescent delinquent behavior ............................................................ 30

x
 
 Criminal behavior ................................................................................. 43

The Current Study .......................................................................................................... 48

Relationship to Prior Literature and Justification for Study ............................. 48

Relevance to the Field of Criminology ............................................................. 50

Pre-Test of Parenting Style Measures ............................................................................ 51

Justification for Pre-Test ................................................................................... 52

Methodology ..................................................................................................... 53

Results ............................................................................................................... 55

Test of internal consistency .................................................................. 55

Test of discriminant validity ................................................................. 56

Conclusion ........................................................................................................ 60

III METHODOLOGY ......................................................................................................... 61

Research Questions and Hypotheses .............................................................................. 61

Site Selection .................................................................................................................. 62

County Jail #1 ................................................................................................... 63

County Jail #2 ................................................................................................... 63

The University .................................................................................................. 64

Sampling and Survey Administration ............................................................................ 64

County Jails....................................................................................................... 64

The University .................................................................................................. 66

Survey Instrument .......................................................................................................... 68

Baumrind Measures .......................................................................................... 69

Other Parenting Measures ................................................................................. 71

xi
 
 Level of Criminal Involvement – Dependent Variable..................................... 72

Demographics ................................................................................................... 74

Reliability and Validity .................................................................................................. 74

Reliability.......................................................................................................... 74

Validity ............................................................................................................. 74

Threats to validity ................................................................................. 75

Human Subject Protections ............................................................................................ 77

Limitations...................................................................................................................... 78

Strengths ......................................................................................................................... 80

Analysis Plan .................................................................................................................. 81

IV ANALYSIS AND RESULTS ........................................................................................ 84

Frequencies and Descriptive Statistics ........................................................................... 85

Demographic Variables .................................................................................... 85

Caretaker Variables ........................................................................................... 89

Maternal Caretaker................................................................................ 90

Paternal Caretaker ................................................................................. 92

Parenting Scales ............................................................................................................. 95

Test of Internal Consistency ............................................................................. 95

Perceptions of Parenting Styles ........................................................................ 96

Dependent Variable Scales ........................................................................................... 100

Internal Consistency of Dependent Variable Scales ....................................... 101

Frequencies of Dependent Variables .............................................................. 102

Total Sample ....................................................................................... 102

xii
 
 University Students ............................................................................. 104

County Jail Inmates............................................................................. 105

White Respondents ............................................................................. 107

Non-White Respondents ..................................................................... 109

Females ............................................................................................... 111

Males ................................................................................................... 112

Bivariate Correlations................................................................................................... 114

Multiple Regression ..................................................................................................... 116

Model One: Effect of Parenting Style on Deviant, Delinquent, and

Criminal Behavior – Full Sample ................................................................... 118

Predatory Property Crimes .................................................................. 118

Predatory Person Crimes..................................................................... 119

Illegal Services Crimes ....................................................................... 120

Public Disorder Offenses .................................................................... 120

Deviant Behaviors ............................................................................... 121

Alcohol/Drugs ..................................................................................... 122

Subscale: Alcohol/Soft Drugs ............................................................. 122

Subscale: Hard Drugs ......................................................................... 123

Summary ............................................................................................. 124

V ANALYSIS AND RESULTS: REGRESSION COMPARISONS .............................. 126

Model Two ................................................................................................................... 126

Effect of Parenting Style on Deviant, Delinquent, and

Criminal Behavior – University Students and County Jail Inmates ............... 126

xiii
 
 Predatory Property Crimes – University Students .............................. 126

Predatory Person Crimes – University Students ................................. 127

Illegal Services Crimes – University Students.................................... 127

Public Disorder Offenses – University Students ................................ 128

Deviant Behaviors – University Students ........................................... 129

Alcohol/Drugs – University Students ................................................. 129

Subscale: Alcohol/Soft Drugs – University Students ......................... 129

Subscale: Hard Drugs – University Students ...................................... 130

Predatory Property Crimes – County Jail Inmates .............................. 130

Predatory Person Crimes – County Jail Inmates................................. 131

Illegal Services Crimes – County Jail Inmates ................................... 132

Public Disorder Offenses – County Jail Inmates ................................ 132

Deviant Behaviors – County Jail Inmates ........................................... 132

Alcohol/Drugs – County Jail Inmates ................................................. 133

Subscale: Alcohol/Soft Drugs – County Jail Inmates ......................... 133

Subscale: Hard Drugs – County Jail Inmates ..................................... 134

Summary ............................................................................................. 135

Model Three: Effect of Parenting Style on Deviant, Delinquent,

and Criminal Behavior – Females and Males ................................................. 136

Predatory Property Crimes – Females ................................................ 136

Predatory Person Crimes – Females ................................................... 137

Illegal Services Crimes – Females ...................................................... 137

Public Disorder Offenses – Females ................................................... 138

xiv
 
 Deviant Behaviors – Females ............................................................. 138

Alcohol/Drugs – Females ................................................................... 139

Subscale: Alcohol/Soft Drugs – Females ........................................... 139

Subscale: Hard Drugs – Females ........................................................ 139

Predatory Property Crimes – Males .................................................... 140

Predatory Person Crimes – Males ....................................................... 140

Illegal Services Crimes – Males ......................................................... 141

Public Disorder Offenses – Males ...................................................... 141

Deviant Behaviors – Males ................................................................. 142

Alcohol/Drugs – Males ....................................................................... 142

Subscale: Alcohol/Soft Drugs – Males ............................................... 142

Subscale: Hard Drugs – Males............................................................ 143

Summary ............................................................................................. 144

Research Hypotheses .................................................................................................... 145

Hypothesis 1: Reported Parenting Styles ........................................................ 145

Hypothesis 2: Effects of Parenting on the Entire Sample ............................... 146

Hypothesis 3: Parenting Effects on University Students

and County Jail Inmates .................................................................................. 147

Hypothesis 4: Parenting Effects on Females and Males ................................. 148

Hypothesis 5: Parenting Effects on Whites and Non-Whites ......................... 148

Summary ...................................................................................................................... 149

VI DISCUSSION AND CONCLUSION .......................................................................... 151

Discussion of Research Findings ................................................................................. 152

xv
 
 Extended Measurement Instrument ................................................................ 153

Impact of Neglecting/Rejecting Parenting ...................................................... 154

Impact of Permissive Parenting ...................................................................... 155

Impact of Authoritative Parenting .................................................................. 156

Summary of the Impact of Parenting Types ....................................... 158

Impact of Race and Sex .................................................................................. 159

Implications for Parents................................................................................................ 162

Strengths and Limitations ............................................................................................. 164

Strengths ......................................................................................................... 164

Limitations ...................................................................................................... 166

Directions for Future Research..................................................................................... 168

Conclusion .................................................................................................................... 171

REFERENCES ............................................................................................................. 173

APPENDICES .............................................................................................................. 184

A. Studies linking parenting style and behavioral problems ......................... 184

B. Studies linking parenting style and substance use .................................... 186

C. Studies linking parenting style and delinquency ....................................... 188

D. Studies linking parenting style and criminality......................................... 191

E. Informed Consent – Jail Inmates ............................................................... 192

F. Informed Consent – University Students................................................... 194

G. Survey Instrument ...................................................................................... 196

xvi
 
 LIST OF TABLES

Table Page

Table 1: Baumrind’s Typology of Parenting Styles ..................................................... 11

Table 2: Criteria for Assignment to PBR Cluster ......................................................... 13

Table 3: PAQ Extension – Items Measuring Maternal Neglecting/Rejecting .............. 52

Table 4: PAQ Extension – Internal Consistency of Scales ........................................... 56

Table 5: Intercorrelations of PAQ Scores ..................................................................... 57

Table 6: PAQ Extension – Intercorrelation Scores ....................................................... 58

Table 7: Comparison of Coefficient Alpha Values: Full Scale vs. Split-Half .............. 70

Table 8: Frequencies and Percentages – Age of Sample .............................................. 85

Table 9: Frequencies and Percentages – Sex of Sample ............................................... 86

Table 10: Frequencies and Percentages – Race of Sample ........................................... 88

Table 11: Frequencies and Percentages for Maternal Caretakers ................................. 90

Table 12: Independent Samples T-Test: Caretaker Variables –

Between Sample, Race, and Sex ................................................................................... 91

Table 13: Frequencies and Percentages for Paternal Caretakers .................................. 92

Table 14: Frequencies and Percentages – Who Respondents Lived With.................... 94

Table 15: Internal Consistency of PAQ-Related Measures .......................................... 96

Table 16: Descriptive Statistics for Measures of Parenting Type................................. 97

Table 17: Independent Samples T-Test – Between Sample, Race, Sex ....................... 99

Table 18: Dependent Variable Scales and Subscales ................................................. 100

Table 19: Internal Consistency of Dependent Variable Scales ................................... 102

Table 20: Frequencies and Percentages for DV Scales – Full Sample ....................... 103

xvii
 
 Table 21: Frequencies and Percentages for DV Subscales – Full Sample ................. 103

Table 22: Frequencies and Percentages for Drug/Alcohol Use – Full Sample........... 104

Table 23: Frequencies and Percentages for DV Scales – University Students ........... 104

Table 24: Frequencies and Percentages for DV Subscales –

University Students ..................................................................................................... 105

Table 25: Frequencies and Percentages for DV Scales – Inmates .............................. 106

Table 26: Frequencies and Percentages for DV Subscales – Inmates ........................ 107

Table 27: Independent Samples T-Test: DV Scales – Sample Groups ...................... 107

Table 28: Frequencies and Percentages for DV Scales – White Respondents ........... 108

Table 29: Frequencies and Percentages for DV Subscales –

White Respondents ..................................................................................................... 109

Table 30: Frequencies and Percentages for DV Scales –

Non-White Respondents ............................................................................................. 109

Table 31: Frequencies and Percentages for DV Subscales –

Non-White Respondents ............................................................................................. 110

Table 32: Independent Samples T-Test: DV Scales – Race ....................................... 110

Table 33: Frequencies and Percentages for DV Scales – Females ............................. 111

Table 34: Frequencies and Percentages for DV Subscales – Females........................ 111

Table 35: Frequencies and Percentages for DV Scales – Males ................................. 112

Table 36: Frequencies and Percentages for DV Subscales – Males ........................... 113

Table 37: Independent Samples T-Test: DV Scales – Sex ......................................... 113

Table 38: Bivariate Correlation Matrix....................................................................... 115

Table 39: Full Sample – Predatory Property............................................................... 118

xviii
 
 Table 40: Full Sample – Predatory Person ................................................................. 119

Table 41: Full Sample – Illegal Services .................................................................... 120

Table 42: Full Sample – Public Disorder .................................................................... 121

Table 43: Full Sample – Deviant Behaviors ............................................................... 121

Table 44: Full Sample – All Alcohol/Drug Use ......................................................... 122

Table 45: Full Sample – Alcohol/Soft Drug Use ........................................................ 123

Table 46: Full Sample – Hard Drug Use .................................................................... 124

Table 47: University Students – Predatory Property .................................................. 127

Table 48: University Students – Predatory Person ..................................................... 127

Table 49: University Students – Illegal Services ........................................................ 128

Table 50: University Students – Public Disorder ....................................................... 128

Table 51: University Students – Alcohol/Soft Drug Use............................................ 130

Table 52: County Inmates – Predatory Property......................................................... 131

Table 53: County Inmates – Predatory Person ........................................................... 131

Table 54: County Inmates – Public Disorder .............................................................. 132

Table 55: County Inmates – Alcohol/Drug Use ......................................................... 133

Table 56: County Inmates – Alcohol/Soft Drug Use .................................................. 134

Table 57: County Inmates – Hard Drug Use .............................................................. 134

Table 58: Female – Predatory Property ...................................................................... 136

Table 59: Female – Predatory Person ......................................................................... 137

Table 60: Female – Illegal Services ............................................................................ 137

Table 61: Female – Public Disorder ........................................................................... 138

Table 62: Female – Deviant Behavior ........................................................................ 139

xix
 
 Table 63: Male – Predatory Property .......................................................................... 140

Table 64: Male – Predatory Person ............................................................................. 140

Table 65: Male – Illegal Services ............................................................................... 141

Table 66: Male – Deviant Behavior ............................................................................ 142

Table 67: Male – Alcohol/Soft Drug Use ................................................................... 143

Table 68: Male – Hard Drug Use ................................................................................ 143

xx
 
 CHAPTER I

INTRODUCTION

There exists a variety of criminological theories that attempt to explain the effects, both

direct and indirect, that the family has on delinquent and criminal behavior. Hirschi (1969)

theorized, among other things, that the bond between parent and child factored into that child’s

level of delinquency and criminality later in life. Building on this, years later Gottfredson and

Hirschi (1990) suggested that poor parental management leads to low self-control in children,

which leads to subsequent delinquent or criminal behavior. Integrated theories that explore the

life-course of criminal behavior also have tried to explain the parent-child crime link. Moffitt

(1993) suggested that juvenile delinquency is caused by neuropsychological vulnerabilities in

children and socialization in a criminogenic environment. Simply, she meant that children with

neuropsychological deficiencies who are born into criminal environments are more likely to

succumb to deviant and delinquent behavior than those who are not born into those types of

environments. Additionally, Moffitt theorized that the most significant predictor of problem

behaviors in children is damaged neuropsychological functions, which are heritable; thus

intimating a direct link from parent to child. Sampson and Laub’s (1993) theory of age-graded

social control, which integrated social control and learning theories, concluded that probability of

delinquency increases when family bonds break down, thus supporting the basic tenets of

Hirschi’s (1969) social bond and Gottfredson and Hirschi’s (1990) self-control theories.

While much has been done in the field of criminology to explore the parent-child crime

link, the fact remains that the discipline is relatively young when compared to other social

sciences and this link also has been the subject of in-depth investigation in other disciplines. As

seen above, the link between parental attachment and support to delinquency and criminality has

1
 
been empirically tested and supported within the field of criminology. However, it is important

to continue to explore, and seek to understand, the connection between parental behavior and

delinquency and criminality in children. There are numerous empirically supported

psychological theories of parenting, such as attribution theory and interaction theory, which have

not received much scrutiny within the field of criminology.

One such psychological typology, Baumrind’s (1966, 1996, 2005; also see Maccoby &

Martin, 1983) typology of parenting styles, has received very little attention from those studying

the effects that parenting style has on the level of delinquent and criminal behavior in

adolescents and young adults (Hoeve et al., 2009). In 1966, Baumrind officially introduced the

conceptual parenting style of “authoritative control” into the child development literature.

Included with this concept were the ideas of authoritarian control and permissive control

(Baumrind, 1966). This early examination of these three controlling behaviors represents the

seminal typology of parenting styles (Buri, 1991; Simons, Simons, & Wallace, 2004). These

styles can be explained best by examining the concepts that they are steeped in – demandingness

and responsiveness (Baumrind, 1996, 2005; Maccoby & Martin, 1983; Simons et al., 2004).

The idea of demandingness is comprised of: direct confrontations between parent and

child, parental monitoring of the child, and consistent parental discipline (Baumrind, 1996). The

idea of responsiveness consists of: warmth, reciprocity, and clear communication and person-

centered discourse (Baumrind, 1996). The convergence of demandingness and responsiveness

results in four different parenting styles: authoritarian, authoritative, permissive, and

neglecting/rejecting. Baumrind (1966) and Simons et al. (2004) discuss three of the different

styles: authoritarian parents display high levels of demandingness and low levels of

responsiveness; authoritative parents also display high levels of demandingness as well as high

2
 
levels of responsiveness; and permissive parents display low levels of demandingness but high

levels of responsiveness. The neglecting/rejecting style, in which parents display low levels of

both demandingness and responsiveness, emerged in later research (Baumrind, 1971, 1991,

2005). The purpose of the proposed study is to examine the effects that parenting style has on

level of criminal involvement in young adults. As the seminal typology of parenting styles (Buri,

1991; Simons et al., 2004), it is unusual that it has not been used more often in the study of

criminal behavior. This deficiency in the empirical literature provided justification for the current

study.

The Current Study

Again, the purpose of this study was to determine what, if any, effects perceived

parenting style has on the level of criminal involvement in adults. Much of the empirical research

on Baumrind’s typology has focused on anti-social behavior in children relative to parenting

style (Barber, Olsen, & Shagle, 1994; Baumrind, 1967, 1971, 1972; Baumrind & Black, 1967;

Lamborn, Mounts, Steinberg, & Dornbusch, 1991; Paulussen-Hoogeboom, Stams, Hermanns,

Peetsma, & van den Wittenboer, 2008; Querido, Warner, & Eyberg, 2002; Vieno, Nation,

Pastore, & Santinello, 2009), adolescent cigarette use (Chassin et al., 2005; Harakeh, Scholte,

Vermulst, de Vries, & Engels, 2004; Huver, Engels, Van Breukelen, & de Vries, 2007; Jackson,

Bee-Gates, & Henriksen, 1994; Jackson, Henriksen, & Foshee, 1998; Mott, Crowe, Richardson,

& Flay, 1999; Simons-Morton, Haynie, Crump, Eitel, & Saylor, 2001), substance abuse

(Baumrind, 1991), and alcohol use (Patock-Peckham & Morgan-Lopez, 2007). Likewise, a

number of studies are devoted to delinquency (Avenevoli, Sessa, & Steinberg, 1999; Chambers,

Power, Loucks, & Swanson, 2000; Hoeve et al., 2007; Jackson et al., 1998; Palmer & Hollin,

2001; Paschall, Ringwalt, & Flewelling, 2003; Simons, Simons, Burt, Brody, & Cutrona, 2005;

3
 
Stattin & Kerr, 2000; Steinberg, Blatt-Eisengart, & Cauffman, 2006; Steinberg, Lamborn,

Darling, Mounts, & Dornbusch, 1994; Steinberg, Mounts, Lamborn, & Dornbusch, 1991;

Walker, Maxson, & Newcomb, 2007). However, very few studies have examined the link

between parenting style and criminal behavior (Haapasalo, 2001; Palmer & Gough, 2007;

Schroeder, Bulanda, Giordano, & Cernkovich, 2010). Based on this lack of research of criminal

outcomes associated with parenting type, it was important that the study was conducted.

There were two units of analysis for the current study. The first unit of analysis was

college students at one university in the northeast, while the second unit of analysis was

incarcerated individuals at two county jails in one northeastern state. The units of analysis were

broken into two sample groups; again, one sample group consisted of college students and the

other consisted of incarcerated men and women. Two sample groups were used because the

research focusing on delinquent and criminal outcomes suggests that higher levels of criminality

and delinquency are associated with certain parenting styles and that lower (and no) levels of

criminality and delinquency are associated with distinctly different parenting styles (Avenevoli

et al., 1999; Haapasalo, 2001; Hoeve et al., 2007; Jackson et al., 1998; Palmer & Hollin, 2001;

Paschall et al., 2003; Simons et al., 2005; Steinberg et al., 1991, 1994, 2006). Despite the fact

that the anticipated outcomes of the current study were expected to mirror the outcomes of prior

investigations, this study built on earlier research in multiple ways.

First, the sample size of many of the studies (Chambers et al., 2000; Haapasalo, 2001;

Palmer & Gough, 2007; Palmer & Hollin, 2001) limits the generalizability of their findings. The

current study built on those small samples. Second, the current study sought to use original data

to assess the link between parenting style and criminal involvement. Hoeve et al. (2007) and

Simons et al. (2005) used secondary data collected from multiple datasets (including, the

4
 
Pittsburgh Youth Study, the Child-rearing and Family in the Netherlands Study, and the Family

and Community Health Study) to analyze the link. This is to take nothing away from secondary

data analysis, but it can be argued that the conceptual definition of “parent” is dynamic and a

current operational definition was necessary in order to gain a clearer understanding of the crime

link under investigation.

With that in mind, Chapter Two provides a discussion of Baumrind’s initial

conceptualization of parenting styles. In addition, the chapter analyzes empirical research that

has tested the link that parenting style has with anti-social, delinquent, and criminal outcomes in

children. The chapter concludes with a discussion of the ever-changing dynamic of the family

and parenting.

Chapter Three provide an explanation of the research methodology that was used in this

study. The overall research design, including sampling frames, formal access, informal access,

and data collection procedures are discussed. In addition, this chapter provides a discussion of

current and past survey design for research of this nature. This chapter also includes a discussion

of data entry and subsequent data storage procedures given the nature of human subject

protections that were implemented. Chapter Three concludes with a discussion of the overall

analysis plan. Chapters Four and Five provide an analysis of the results of the current study.

Chapter Six examines these results and provides a discussion of the results as well as

implications for future research.

5
 
 CHAPTER II

LITERATURE REVIEW

Conceptualization of Baumrind’s Four Parenting Typologies

Baumrind (1966) introduced three distinct types of parental control – authoritarian,

authoritative, and permissive. Over time, a fourth type of parental control – neglecting/rejecting

– was introduced into the literature (Baumrind, 1971; Maccoby & Martin, 1983; Simons et al.,

2004). The formulation of these four typologies is based upon the confluence between the

concepts of demandingness and responsiveness. The following sections discuss the ideas of

demandingness and responsiveness, as well as the four typologies that these ideas create.

Demandingness

Demandingness, according to Baumrind (1996, p. 411), refers to “claims that parents

make on children to become integrated into the family and community by their maturity

expectations, supervision, disciplinary efforts, and willingness to confront a disruptive child.”

Demandingness can be gauged based on the level of direct confrontation, well-defined

monitoring techniques, and patterns of discipline (Baumrind, 1996; Simons et al., 2004) that one

utilizes while parenting. Parents who use high levels of confrontation, monitoring, and consistent

discipline are characterized as demanding; those with low levels of confrontation, monitoring,

and inconsistent discipline are characterized as not demanding. Parents with high demandingness

can be characterized as either authoritarian or authoritative in Baumrind’s typology (Maccoby &

Martin, 1983; Simons et al., 2004). Those with low levels of demandingness can be characterized

as either permissive or neglecting/rejecting (Maccoby & Martin, 1983; Simons et al., 2004). The

following sections provide a more in-depth description of confrontation, monitoring, and

consistent discipline.

6
 
 Confrontation. Baumrind (1996) describes “confronting” parents as those who are

involved in the lives of their children and exercise firm, yet non-coercive, control. She writes (p.

411), “It is not confrontation or the exercise of firm control…but rather the arbitrary, harsh, and

nonfunctional [emphasis added] exercise of firm control that has negative consequences for child

behavior.” The use of arbitrary, harsh, and nonfunctional control creates coercion, which negates

the parents’ attempt to correct child behavior (Baumrind, 1996), thus creating defiance in the

child and hostility in the parent-child relationship (Hoffman, 1960).

In order to eschew the creation of coercion through confrontation, the parents must

explain the reasoning for their demands (Baumrind, 1996). Additional research suggests that if

the confronting parent is supportive, nonpunitive, authentic, and sensitive then parental

confrontation does not negatively impact prosocial behavior in children (Baumrind, 1971, 1983;

Hoffman, 1963). The parents who take these four ideas into consideration while confronting a

child can exhibit more firmness in their commands and expect the child to respond more

positively and prosocially than the parent who confronts a child using coercive techniques.

Monitoring. In this instance, monitoring simply refers to the level of supervision that a

parent exhibits over a child. Close parental monitoring is believed to promote self-regulation and

prosocial behaviors in children (Baumrind, 1996); Patterson (1982, 1986) explicitly suggested

that close parental monitoring deterred antisocial behavior in young males.

Of relevance to this study is the idea that level of parental monitoring is positively related

to socio-economic status (Baumrind, 1996). Simply, as socio-economic status decreases level of

parental supervision also decreases. Baumrind (1996, p. 411) writes that “close supervision, and

the provision of an orderly, consistent regimen require a greater investment of time and energy

than” parents of lower socio-economic standing can provide, since they may work multiple jobs

7
 
in order to make ends meet. Socio-economic status of the family is an important facet to keep in

mind when considering how parenting styles affect child development. As Brooks-Gunn, Britto,

and Brady (1999) report, poverty is detrimental to childhood development. These authors suggest

that impoverished children are at least twice as likely as higher SES children to: repeat a grade in

school, get expelled or suspended from school, and/or drop out of school. In addition,

impoverished children are at a greater risk of having learning disabilities and emotional and/or

behavioral problems.

Consistent discipline. Baumrind (1996, p. 411) suggests that the “contingent use of

positive or negative reinforcers immediately following desired or prohibited child behavior” is

one of the most important factors in child behavioral management. Simply, the parent must

consistently reinforce the child’s behavior, whether good or bad, right away. Snyder and

Patterson (1995) suggest that a parent with inconsistent disciplinary patterns creates an antisocial

and defiant child. On the other hand, consistent parental disciplinary patterns, whether approving

or disapproving, help create prosocial children and are not as detrimental to childhood

development as inconsistent patterns of discipline.

Responsiveness

Responsiveness, according to Baumrind (1996, p. 410), is “the extent to which parents

intentionally foster individuality and self-assertion by being attuned, supportive, and acquiescent

to children’s needs and demands.” This can be gauged based on the level of warmth, reciprocity,

and clear communication and person-centered discourse exhibited by a parent when dealing with

a child (Baumrind, 1996; Simons et al., 2004). Parents who utilize high levels of warmth,

reciprocal behavior, and communication are thought to be highly responsive. Parents who use

low levels of those three facets are thought to be low in responsiveness. Those with high

8
 
responsiveness are thought to be authoritative or permissive in Baumrind’s typology (Maccoby

& Martin, 1983; Simons et al., 2004). On the other hand, parents with low responsiveness can be

characterized as authoritarian or neglecting/rejecting (Maccoby & Martin, 1983; Simons et al.,

2004). The following provides a more in-depth discussion of warmth, reciprocity, and clear

communication and person-centered discourse.

Warmth. As a component of responsiveness, warmth is conceptualized as parents’

expression of love for their child (Baumrind, 1996). Research suggests that prosocial

development occurs in children whose parents utilize affective warmth and empathy while

parenting (Eisenberg, 1992; Hoffman, 1975). On the other hand, a lack of parental warmth is

thought to be associated with aggressive, antisocial, and hostile behaviors in children (Olweus,

1980). Parental dispositions such as anger, disapproval, and discouragement foster antisocial

behavior in children; however, when children are highly antisocial they seek greater levels of

parental approval (Baumrind, 1996). Baumrind also suggests that the concept of parental warmth

is understudied. Thus, warm and supportive parents are likely to exhibit high levels of

responsiveness, while parents who do not show warmth or support are likely to exhibit low levels

of responsiveness.

Reciprocity. The idea of reciprocity in parent-child interactions also is understudied

outside of infant-parent studies (Baumrind, 1996). One study by Parpal and Maccoby (1985)

suggests that children are more likely to comply with the desires of the mother if the mother

complies with the demands of the child. It can be argued that parents who encourage a reciprocal

parent-child relationship will display high levels of responsiveness and those who do not

encourage a reciprocal relationship will not be as responsive. The idea of reciprocity presents a

limitation to the study of parenting styles and behavioral outcomes. Simply, is the parent

9
 
parenting based on reactions to the child’s behavior or is the child’s behavior a reaction to

parenting styles used. This limitation is included in a discussion of the methodology.

Communication and discourse. There are two types of interpersonal discourse that this

section focuses on: position-centered (hierarchy of authority; parent is in charge) and person-

centered (focused on the individual rather than position of authority). Position-centered discourse

is used by parents as a means of legitimizing their authority, while children likely view this type

of interpersonal communication as coercive in favor of the parent (Baumrind, 1996). Person-

centered discourse is not focused on the hierarchical nature of the parent-child relationship,

rather it is intended to create friendly and transformative interactions between the parent and the

child. Instead of relying on coercion, person-centered discourse relies on persuasion (Baumrind,

1996) and is generally considered to be a more effective form of communication than position-

centered discourse (Applegate, Burke, Burleson, Delia, & Kline, 1985). It can be assumed that

parents who use person-centered discourse while interacting with their children are highly

responsive and those who use position-centered discourse have low levels of responsiveness.

As stated, the confluence of demandingness and responsiveness results in four different

parenting styles: authoritarian, authoritative, permissive, and neglecting/rejecting. Table 1

depicts Baumrind’s typology based on high and low levels of parental demandingness and

responsiveness (Maccoby & Martin, 1983; Simons et al., 2004). The following four sections

provide an in-depth discussion of each parenting typology. After the introduction of each

parenting type, an examination of empirical tests related to this typology is provided.

10
 
Table 1: Baumrind’s Typology of Parenting Styles

High Demandingness Low Demandingness

High Responsiveness Authoritative Style Permissive Style

Low Responsiveness Authoritarian Style Neglecting/Rejecting Style

SOURCE.—Adapted from Maccoby and Martin (1983), and Simons et al. (2004)

Authoritarian

As stated, Baumrind (1966) was one of the first to conceptualize the authoritarian

parenting style. According to this first conceptualization (p. 890), and repeated later by

Baumrind (1968, 1971), “the authoritarian parent attempts to shape, control, and evaluate the

behavior and attitudes of the child in accordance with a set standard of conduct, usually an

absolute standard.” In addition to displaying tendencies that are typical of a parent with a high

level of demandingness, the authoritarian parent is punitive, forceful, and believes that a virtuous

child should be obedient and respect the value of having a good work ethic (Baumrind, 1966).

The authoritarian parent attempts to preserve the order of a traditional family structure by

limiting the child’s autonomy, discouraging verbal “give and take” between parent and child, and

demanding that the child accept what the parent says without question (Baumrind, 1966).

According to Baumrind (1971), in order to be classified as “authoritarian” the parent had

to meet a variety of criterion (see Table 2). First, “both parents [had to] have scores above the

median in Firm Enforcement or one parent [had to] score in the top third of the distribution”

(Baumrind, 1971, p. 22). The Firm Enforcement measure of Baumrind’s (1971) Parent Behavior

Rating (PBR) clusters for both mothers and fathers consisted of the following eight items: firm

enforcement, enforcement after initial noncompliance, cannot be coerced by child, requires child

to pay attention, promotes own code of behavior, forces confrontation when child disobeys,

11
 
willingly exercises power to obtain obedience, and child must defer to parental expertise. In

addition, the mother PBR cluster for Firm Enforcement included the item: disapproves of defiant

stance; the father PBR cluster for Firm Enforcement included the following two items: uses

negative sanctions when defied, and has stable/firm views.

Second, “both parents [had to] have scores below the median in Encourages

Independence and Individuality (EII), or one parent [had to] score in the bottom third of the

distribution, or the father [had to] score in the bottom third on Promotes Noncomformity and in

the top third on Authoritarianism” (Baumrind, 1971, p. 22). The EII measure of the PBR clusters

for both mothers and fathers consisted of the following four items: meaningful verbal interaction,

encourages intimate verbal contact, defines child’s individuality clearly, and does not lack

empathic understanding (Baumrind, 1971). Additionally, the mother PBR cluster for EII

included the following 10 items: gives reasons with directives, offers child alternatives, solicits

child’s opinions, listens to critical comments, encourages verbal give and take, promotes

individuality in child, shares decision-making power with child, expresses own individuality,

encourages oppositional behavior, and child must not conform to establishment. The father PBR

cluster for EII included the following seven items: clear ideals for child, clear about parental

role, regards self as competent person, flexible views, can specify aims and methods of

discipline, regards self as potent and knowledgeable, and is secure during home visit.

12
 
Table 2: Criteria for Assignment to PBR Cluster

Authoritarian Authoritative Permissive Neglecting/Rejecting

Firm Enforcement Both parents Both parents Both parents -
above median above median below median
OR one in top
third

Encourages Both parents Both parents - Both parents below

Independence and below median above median median
Individuality OR one in OR one in top
bottom third third
OR

Promotes Non- Father in - - -

conformity bottom third
AND

Authoritarianism Father in top - - -

third

Passive-Acceptant Both parents Both parents One parent in -

below median below median top third
OR one in OR one in
bottom third bottom third

Rejecting - - One parent Both parents above

below median median AND one parent
in top third OR

Expect Participation in - - Both parents -

Chores below median
OR*

Directive - - Both parents -

below median
OR*

Discourages Infantile - - Both in -

Behavior bottom third*

Enrichment of Child’s - - - Both parents in bottom

Environment third AND

Discourage Emotional - - - Both parents in top third

Dependency
*2 of 3 criteria have to be met

13
 
 In addition to meeting criteria for the EII cluster, the father had to meet criteria for both

the Promotes Nonconformity cluster and the Authoritarianism cluster, as stated above. The

Promotes Nonconformity PBR cluster for fathers included the following five items: child must

not conform to establishment, promotes individuality in child, expresses own individuality, sees

child-rearing practices as atypical, and values expressive traits more than instrumental traits

(Baumrind, 1971). The Authoritarianism PBR cluster for fathers included the following 13 items:

does not listen to critical comments, does not solicit child’s opinions, assumes stance of personal

infallibility, does not share decision-making power with child, disobedience does not elicit

further explanation, does not offer child alternatives, becomes inaccessible when displeased, uses

obedience as a salient construct, does not encourage oppositional behavior, does not encourage

verbal give and take, feels that parents’ needs take precedence, does not encourage independent

actions, and does not give reasons with directives (Baumrind, 1971). Again, for the parents to

garner the label of authoritarians, the father had to score in the bottom third of the Promotes

Nonconformity measure and in the top third of the Authoritarianism measure.

Finally, “both parents [had to] score below the median in Passive-Acceptant or one parent

[had to] score in the bottom third” (Baumrind, 1971, p. 22). The Passive-Acceptant PBR cluster

consisted of the following four items for both mothers and fathers: gentle manner, avoids open

confrontation, inhibits annoyance or impatience when child disobeys, and inhibits annoyance or

impatience when child dawdles or is annoying. The Passive-Acceptant measure for mothers

consisted of the additional three items: does not use negative sanctions when defied, does not

discipline harshly, and feels shame when expressing anger. The Passive-Acceptant measure for

fathers consisted of the following two items: does not disapprove of defiant stance and does not

become inaccessible when displeased.

14
 
 The similarities are clear between Baumrind’s (1966) initial conceptual definition of

authoritarian parents to the criterion that Baumrind (1971) demanded parents meet in order to be

defined as authoritarian. Furthermore, the reliability scores for each of Baumrind’s (1971) eight

PBR clusters used to define authoritarian parents range from .83 to .94. DeVellis (2003) suggests

that coefficient alpha scores between .80 and .90 are “very good,” but that “one should consider

shortening the scale” (p. 96) when scores are “much above .90.” One could consider three of

Baumrind’s clusters with coefficient alpha scores of .93, .94, and .94 as being “much above .90.”

The questionable clusters include maternal Firm Enforcement and EII, which consist of 9 and 14

items, and paternal Authoritarianism, which consists of 13 items.

Authoritative

Just as Baumrind (1966) was among the first to conceptualize the authoritarian style of

parenting, she also was one of the first to conceptualize the authoritative style of parenting.

According to this first conceptualization (p. 891), and again repeated by Baumrind (1968, 1971),

“the authoritative parent attempts to direct the child’s activities in a rational, issue-oriented

manner.” The authoritative parent acts with both a high level of demandingness and a high level

of responsiveness. In addition, the authoritative parent welcomes a verbal give and take with the

child, solicits objections from the child, explains the reason for certain rules, and encourages the

child to be autonomous (Baumrind, 1966, 1968, 1971). Despite being quite open-minded, the

authoritative parent does not hesitate to exert firm control and use power, yet one must keep in

mind that the parent does this in order to set standards for the child’s future behavior (Baumrind,

1966, 1968, 1971).

In order to be classified as “authoritative” the parent had to be classified according to a

variety of criterion, as described by Baumrind (1971). First, “both parents [had to] have scores

15
 
above the median in Firm Enforcement, or one parent [had to] score in the top third of the

distribution” (Baumrind, 1971, p. 23). This initial requirement is identical to the first requirement

needed for classification as authoritarian and is discussed in detail above.

Second, in order to be considered authoritative, “both parents [had to] score above the

median in EII or one parent [had to] score in the top third of the distribution” (Baumrind, 1971,

p. 23). When comparing this requirement to the requirement for classification as an authoritarian

one sees that authoritative parents are thought to encourage independence and individuality in

their children at a higher level than authoritarian parents. Again, the items used to create the EII

PBR cluster are detailed above.

Finally, and similar to authoritarian parents, in order to be considered authoritative “both

parents [had to] score below the median in Passive-Acceptant or one parent [had to] score in the

bottom third” (Baumrind, 1971, p. 23). Once again, the Passive-Acceptant scale is discussed

above. It is interesting to note, however, that the Promotes Nonconformity PBR cluster is not

used to determine if a parent is authoritative. While revisiting the items in the Promotes

Nonconformity cluster one realizes that fathers in this cluster do not expect their children to

readily conform, which is typical of parents in Baumrind’s (1966) initial conceptualization of

authoritativeness, in that fathers promote individuality and expressive traits in their children,

which is strikingly similar to the notion of autonomy.

Permissive

In addition to being one of the first to conceptualize the authoritarian and authoritative

typologies, Baumrind (1966) also was among the first to conceptualize the permissive style of

parenting. According to the initial conceptualization (p. 889), and later reiterated by Baumrind

(1968, 1971), “the permissive parent attempts to behave in a nonpunitive, acceptant, and

16
 
affirmative manner toward the child’s impulses, desires, and actions.” As suggested above, the

permissive parent acts with a high level of responsiveness as indicated by some of Baumrind’s

(1966) initial conceptualizations which suggest that the responsive parent is likely to explain

family rules, to encourage children to use the parent as a resource whenever they wish, and to

discourage children from following defined standards of society. The permissive parent also acts

with a low level of demandingness. This notion also is supported by some of Baumrind’s (1966)

early conceptualizations which assert that the permissive parent is likely to make “few demands”

for prosocial behavior, does not use overt power, and avoids implementing control over the

child.

In order to actually be characterized as “permissive,” however, the parent in Baumrind’s

(1971) later studies had to be classified by a variety of criterion, which is similar to the schemes

used to classify authoritarian and authoritative parenting styles. First, “both parents [had to] have

scores below the median on Firm Enforcement” (Baumrind, 1971, p. 23). While revisiting the

items in both the maternal and paternal Firm Enforcement PBR clusters and thinking about the

idea of demandingness, one realizes that it makes sense for the permissive parent to score below

the median on Firm Enforcement while the authoritarian and authoritative parents had to score

above the median; highly demanding parents are likely to be firm enforcers of rules and

regulations while those low in demandingness are not likely to be firm.

Second, “one parent [had to] score in the top third of the distribution on Passive-

Acceptant” (Baumrind, 1971, p. 23). Again, in order to be recognized as authoritarian or

authoritative, both parents had to score below the median or in the bottom third of the

distribution of Passive-Acceptant. Similar to the differences in Firm Enforcement, the

differences between permissive parents and authoritarian or authoritative parents with regards to

17
 
the Passive-Acceptant distribution also make sense theoretically when the idea of demandingness

is taken into consideration.

Third, “at least one parent [had to] have scores below the median on Rejecting”

(Baumrind, 1971, p. 23). The Rejecting PBR cluster consisted of the following three items for

both mothers and fathers: cool towards child, unresponsive towards child, and disciplined

harshly (Baumrind, 1971). In addition to these three items, the maternal PBR cluster for

Rejecting consisted of the following three items: needs of the parent take precedence, assumes a

stance of personal infallibility, and becomes inaccessible when displeased. When thinking about

the concepts associated with responsiveness and the idea that permissive parents are highly

responsive, it is clear that parents must not have high scores on the Rejecting measure in order to

be classified as permissive. It should be noted that the coefficient alpha for the maternal

Rejecting construct was .88 and that it was .82 for the paternal construct; again, DeVellis (2003)

would consider these levels “very good.”

Finally, in order to be classified as permissive two of the following three joint (both

parents) requirements must be met, “Expect Participation in Household Chores [EPHC], below

median score; Directive, below median score; Discourage Infantile Behavior [DIB], low third”

(Baumrind, 1971, p. 24). The joint PBR cluster of EPHC consisted of the following six items:

demand child put toys away, demand child cleans own messes, discourage obstructive behavior,

sets regular tasks, encourages self-help, and demands child dress self (Baumrind, 1971). While

examining Baumrind’s (1966, p. 889) initial notion that the permissive parent “makes few

demands for household responsibility” it is clear that in order to be classified as permissive the

parent would have to score below the median on EPHC. The PBR cluster for Directive consisted

of five items: regimen set for child, fixed bedtime hour, many rules and regulations, and many

18
 
restrictions on both TV and eating (Baumrind, 1971). Baumrind (1966) noted that the permissive

parent did not demand orderly behavior; thus, it also comes as no surprise that permissive parents

also had to score below the median on the Directive cluster. The DIB cluster consisted of five

items: parents initiate toilet training, discourage baby speech and mannerisms, limit bottle and

pacifier usage, demand mature table behavior, and demand mannerly behavior during visits

(Baumrind, 1971). The coefficient alpha for EPHC was .86 (very good); Directive was .78,

which is respectable according to DeVellis (2003); and, DIB was .63, which DeVellis suggests is

undesirable. Behaviors of children with permissive parents are discussed in a later section.

Neglecting/Rejecting

Baumrind’s (1966) initial conceptualization of parenting styles did not include the

“neglecting/rejecting” style. But, based on the confluence of demandingness and responsiveness

(Maccoby & Martin, 1983; Simons et al., 2004), one realizes that a neglecting/rejecting parent is

not warm, reciprocal, confrontational, supervisory, consistent in discipline, or a clear

communicator (Baumrind, 1996). Baumrind’s (1971) introduction of the neglecting/rejecting

parenting style based on PBR clusters supports the assertion that a neglecting/rejecting parent is

neither demanding nor responsive.

Three requirements must be met in order to be classified as neglecting/rejecting

according to Baumrind (1971). First, “both parents [had to] have scores below the median for

EII” (Baumrind, 1971, p. 24). Again, this relates to the level of responsiveness seen in parents

and also is indicative of authoritarian parents who have low levels of responsiveness. Second,

“both parents [had to] have scores above the mean in Rejecting” (Baumrind, 1971, p. 24). By

looking at the items within the Rejecting PBR clusters again, one easily realizes that it makes

sense that scores must be high in this construct in order to be considered neglecting/rejecting.

19
 
 Third, “one parent [had to] score in the top third of the distribution on Rejecting, or…the

family on the Joint clusters [had to] score in the bottom third on Enrichment of Child’s

Environment [ECE], and the top third on Discourage Emotional Dependency [DED]”

(Baumrind, 1971, p. 24). The joint PBR cluster of ECE consisted of the following five items:

invokes cognitive insight, intellectually stimulates environment, sets standards of excellence,

makes educational-based demands on children, and are differentiated and stimulating (Baumrind,

1971); the coefficient alpha score for this construct was .85 (very good). Based on these items,

one can argue that the requirement that parents score in the bottom third of this construct in order

to be labeled neglecting/rejecting makes sense theoretically. The joint PBR cluster of DED

consisted of the following two items: discouraged emotional dependency and not overprotective

(Baumrind, 1971); the coefficient alpha of this construct was .77 (respectable). Again, it makes

sense that a neglecting/rejecting would discourage emotional dependency in the parent-child

relationship. Additionally, a neglectful or rejecting parent is unlikely to be overprotective of their

child. Behaviors of children with neglecting/rejecting parents are discussed in a later section. The

next section discusses how each of these four parenting types have been examined throughout

the social sciences; methodology, measurement, and outcomes of these studies receive much of

the focus.

Examining Parenting Style and Deviant Behavior

A variety of social science literature has focused on the relationship between parenting

style and subsequent child outcomes. Mostly, this includes examinations of antisocial and

general problem behaviors in young children (Baumrind 1967, 1971, 1972; Baumrind & Black,

1967; Paulussen-Hoogeboom et al., 2008) and adolescents (Barber et al., 1994; Lamborn et al.,

1991; Vieno et al., 2009). While these studies have been important in examining the link

20
 
between parenting styles and negative behaviors in young children and adolescents, they are not

as relevant as other research is to the present study.

Additional research has assessed deviant childhood activities like smoking, drinking, and

drug use, while related studies have focused on delinquent behavior. A limited number of studies

have investigated the specific link between parenting style and criminal involvement. The

following section looks at relevant prior literature that has examined the link between

Baumrind’s parenting typologies and deviant behavior.

The Link between Parenting Style and Deviant Behavior

As stated, there is a large body of literature (see Appendix A) that examines the link

between parenting styles and problematic and antisocial behavior in young children (Baumrind

1967, 1971, 1972; Baumrind & Black, 1967; Paulussen-Hoogeboom et al., 2008) and

adolescents (Barber et al., 1994; Lamborn et al., 1991; Vieno et al., 2009). The major findings in

this area suggest that there is a consistent linkage between the authoritative style of parenting and

fewer problem behaviors among children and adolescents. There also exist significant

associations between the authoritarian, permissive, and neglecting/rejecting styles of parenting

and problem behaviors amongst adolescents and children. This initial research of the link

between parenting style and adolescent and childhood behavior, especially the work of Baumrind

(1967, 1971, 1972; also see Baumrind & Black, 1967), provided the catalyst for examinations of

the link between parenting style and deviant, delinquent, and criminal behavior in teenagers and

adults. As such, the remainder of the literature review focuses on studies that examine the

relationship between parenting styles and deviance (i.e., alcohol and tobacco use by minors),

delinquency, and criminality in adolescents and adults.

21
 
 Alcohol, tobacco, and substance use. In addition to examining the relationship between

parenting style and adolescent problem behaviors like lying, cheating, and verbally fighting,

researchers also have looked at the relationship between parenting style and other forms of

deviant (and borderline delinquent) behavior. Some of these behaviors include alcohol use

(Baumrind, 1991; Patock-Peckham & Morgan-Lopez, 2007), cigarette use (Chassin et al., 2005;

Harakeh et al., 2004; Huver et al., 2007; Jackson et al., 1994; Jackson et al., 1998; Mott et al.,

1999; and Simons-Morton et al., 2001), and substance use (Baumrind, 1991). Similar to research

that investigated parenting style and adolescent behavioral problems, this research varies in

methodological procedure and sample characteristics. The following discusses these

methodological differences and the empirical results of these varied studies (also see Appendix

B).

Methodology of alcohol, cigarette, and substance abuse studies. Baumrind (1991)

studied alcohol and substance use patterns relative to parenting style amongst adolescents.

Baumrind utilized data gathered in her earlier studies of parenting styles and childhood problem

behaviors (Baumrind, 1967, 1971, 1972; Baumrind & Black, 1967). For the 1991 study,

Baumrind relied on data gathered during the third phase of data collection when the adolescents

were 15-years-old. The families who participated in Baumrind’s study underwent changes

between the first and third phases of data collection. At phase one all families were considered

intact, meaning that the biological parents were married and living together, but by the third

phase 38% of the original couples had divorced (Baumrind, 1991). The author failed to report the

number of adolescents who were living solely with their mother (or father), thus the results must

be viewed with a small level of skepticism for two reasons. First, Hetherington and Stanley-

Hagan (1997) suggest that juveniles who grow up without a father have problems controlling

22
 
their aggression and psychosocial development – this may lead to increased risk of substance

abuse. Also, while playing the dual role of mother and father, single mothers may be overcome

with stress that undermines their ability to parent effectively, thus creating an environment of

inconsistent parenting (McLanahan &Teitler, 1999). Based on these two ideas, Baumrind (1991)

could have reported descriptive statistics pertaining to the composition of the families of the 139

adolescents observed during the third phase of data collection, but without the benefit of

hindsight this is only a slight omission.

Despite using the same methodology and cohort of participants to study

alcohol/substance abuse and adolescent problem behaviors, Baumrind’s (1991) research team

classified parents into six family types (compared to 15) based on their observations. These six

classifications include (1) authoritative families (authoritative), (2) conscientious/engaged

families (permissive), (3) directive families (authoritarian), (4) good-enough families (mid-level

of demandingness and responsiveness), (5) nondirective families (permissive), and (6)

unengaged families (neglecting/rejecting). Aside from the Baumrind (1991) study, data about

minor deviant behavior relative to parenting style has been gathered using self-report surveys.

Patock-Peckham and Morgan-Lopez (2007) also studied patterns of alcohol usage

relative to parenting styles in a sample of 441 college students who answered self-report surveys.

This sample included 225 men and 216 women who were predominantly Caucasian (75.6%)

with an average age (in years) of 19.48. Limitations of Patock-Peckham and Morgan-Lopez’s

research are (a) participants received class credit for their participation, and (b) non-drinkers

were excluded from participation. To gain a more accurate understanding of the effect that

parenting has on alcohol use researchers must include those who do not consume alcoholic

23
 
beverages otherwise they will not be able to determine which parenting types steer young people

away from (or towards) alcohol use.

In their assessment of the relationship between alcohol consumption and parenting style,

Patock-Peckham and Morgan-Lopez (2007) used Buri’s (1991) PAQ to measure respondents’

perceptions of parenting styles. Consistent with Buri’s original alpha scores, Patock-Peckham

and Morgan-Lopez (2007) reported authoritarian scores of .85 (mother) and .90 (father),

authoritative scores of .82 (mother) and .88 (father), and permissive scores of .77 (mother) and

.81 (father). Similar alpha scores across multiple diverse tests suggest that Buri’s PAQ is a

sufficient measure to use in order to assess perceptions of parenting styles in the proposed study.

The remainder of this section discusses tobacco use relevant to parenting style.

Each of the seven tobacco studies mentioned briefly above also relied upon data gathered

from self-reports. Chassin et al. (2005) sampled 382 adolescents between the ages of 10 and 17

from a midwestern community. Harakeh et al. (2004) relied on data gathered from 1,070 Dutch

adolescents between the ages of 10 and 14. Huver et al. (2007) also reported findings from the

Netherlands; 482 Dutch adolescents between the ages of 12 and 19 completed a 16-page

questionnaire as part of the Study of Medical Information and Lifestyles in Eindhoven (SMILE)

project. Jackson et al. (1994) gathered data from 937 elementary school children in northern

California, while Jackson et al. (1998) sampled 1,236 fourth and sixth graders in North Carolina.

Mott et al. (1999) surveyed approximately 2,300 high school freshmen in southern California.

Simons-Morton et al. (2001) sampled over 4,000 middle school students from a Maryland suburb

near Washington, D.C.

Studies researching juvenile tobacco usage measured parenting styles in a variety of

ways. First, Huver et al. (2007) utilized a 22-item instrument asking about parenting styles that

24
 
was based on the measurement tool used by Lamborn and his colleagues in the behavioral

studies discussed above (Lamborn et al., 1991). Similar to Lamborn et al. (1991), Jackson et al.

(1994) and Mott et al. (1999) conceptualized their measures of parenting style in accordance

with Baumrind’s four typologies. Mott et al. (1999), however, used only two items (how parents

exercised control over their children’s decision-making and how parents reacted to adolescent

smoking) to measure the four parenting types. Somehow, doing so allowed them to discern that

roughly 38% of their sample had authoritative parents, 31% authoritarian, 20% permissive, and

11% neglecting/rejecting. The authors noted that authoritative and authoritarian parents were the

main decision-makers, but that authoritative parents allowed for input from their children,

whereas adolescents with permissive and neglecting/rejecting parents made all of the decisions,

although adolescents in permissive homes consulted with their parents first (Mott et al., 1999). It

should be noted that the characteristics of each parenting style are accurate, but one should be

wary of results based on the limited number of items used to measure each style.

As stated, Jackson et al. (1994) also included measures for Baumrind’s original parenting

styles. Like Mott et al. (1999), there were discrepancies with these measures. Simply, Jackson

and colleagues measured solely for authoritative and non-authoritative parenting styles. In

addition to only measuring for authoritative and non-authoritative (which, is not a style)

parenting, Jackson et al. (1994) used nine test items (six for authoritative and three for non-

authoritative) to examine the relationship. Granted, the nine items capture a wider realm of

behaviors associated with authoritative parenting than Mott’s measures capture, but one should

still be skeptical of the generalizability of results based on a dichotomous measure – especially

results for those who report having non-authoritative parents.

25
 
 Years later, Jackson et al. (1998) developed the Authoritative Parenting Index (API),

which measured demandingness and responsiveness with 20 survey items based exclusively on

Baumrind’s research. Coefficient alpha scores for the API in Jackson et al. (1998) were based on

a numbers of factors including sex, race, and grade-in-school of the respondent. For example,

fourth-grade white males had alpha scores of .82 on the responsiveness scale and .65 on the

demandingness scale. To reiterate, Jackson et al. (1998) used male and female fourth and sixth

graders who were both African American and white. Alpha scores for this group on the

responsiveness measure ranged from .71 to .90, while alpha scores on the demandingness

measure ranged from .65 to .81. While generally recognized as acceptable coefficient alpha

scores they are not as good as Buri (1991), especially the demandingness measure.

Both Chassin et al. (2005) and Harakeh et al. (2004) included items that tested parenting

styles using measures of control (psychological and strict), as well as parental acceptance

(Chassin et al., 2005) and parental knowledge (Harakeh et al., 2004). The control measures

utilized in Chassin et al. (2005) were from the Child Report of Parent Behavior Inventory

(CRPBI). Similar to the PAQ (Buri, 1991), the CRPBI asks respondents their level of agreement

(5-point Likert scale) with a statement about a particular aspect of parenting. Additionally,

acceptance was measured using seven items from the Network of Relationships Inventory (NRI)

that asked about the parent-child relationship. Chassin and colleagues reported healthy

coefficient alpha scores of .91 for each of the two scales. Continuing with the seeming theme of

inconsistent measurements throughout the research, Chassin et al. (2005) used reported scores

from the control and acceptance measures to assign parents to one of Baumrind’s four types.

High levels of control and acceptance were indicative of authoritative parents while low levels of

control and acceptance were indicative of neglecting/rejecting parents (Chassin et al., 2005).

26
 
 Harakeh et al. (2004) also measured aspects of parental control. The first aspect measured

was psychological control, which the authors defined as coercive and manipulative parenting.

Psychological control was measured using nine items that asked respondents their level of

agreement (5-point Likert scale) with statements about the way their parents treated them. Strict

control was similarly measured using four items that assessed adolescents’ perceptions of the

level of control that their parents placed on where they could go and what they could do. Parental

knowledge was measured using four items in a Likert scale format that asked respondents if their

parents knew where they were and what they were doing. Harakeh et al. (2004) report internal

consistency scores ranging from .72 to .87 for the three measures. While reporting healthy

reliability scores, the scales are not arranged in such a way as to let researchers place respondents

into one of four typologies.

Similar to the strict control and parental knowledge measures used by Harakeh et al.

(2004), Simons-Morton et al. (2001) utilized a six item scale measuring parent involvement (how

much does the parent know about the respondent) and a four item scale measuring parent

monitoring (how often does the parent check up on the respondent). Unlike the measure of

seemingly coercive parents used by Harakeh and colleagues, however, Simons-Morton and his

colleagues used a five item measure for parental support (a parent who helps), a seven item

measure for psychological autonomy (a parent who encourages the child to be her/himself), and

a four item measure for parent-child conflict. The sixth measure that Simons-Morton et al.

(2001) used was a six item scale measuring parent expectations (how upset the parent would be

if the respondent engaged in certain behaviors). The following section reports the findings of the

studies discussed above.

27
 
 Empirical results of alcohol, cigarette, and substance abuse studies. The results of

studies that examined the link between parenting styles and adolescent’s use of alcohol, tobacco,

and other substances did not follow the same pattern as studies that examined adolescent

problem behaviors. For the most part, there was little variation in parenting style and tobacco use

(Chassin et al., 2005; Harakeh et al., 2004; Huver et al., 2007; Jackson et al., 1994; Jackson et

al., 1998; Mott et al., 1999; Simons-Morton et al., 2001), alcohol use (Baumrind, 1991; Patock-

Peckham & Morgan-Lopez, 2007), and drug use (Baumrind, 1991; Lamborn et al., 1991). This

section details the results.

Huver et al. (2007, p. 586) suggested, “Specific parenting styles…were not associated

with smoking cognitions or behavior,” which supports findings from both Harakeh et al. (2004)

which found that psychological parental control, strict parental control, and parental knowledge

were not significantly related to adolescent smoking behaviors and Mott et al. (1999), who found

no significant relationship between their two-item measure of parenting style and cigarette use.

The findings from Huver et al. (2007), however, contradict evidence suggesting that children of

neglecting/rejecting parents are more likely to engage in tobacco use than other children

(Chassin et al., 2005) or that children of authoritative parents are likely to exhibit the lowest rates

of intent to and experimentation with, as well as initiation to, smoking (Jackson et al., 1994).

With respect to those who did smoke tobacco, the effect that mothers had on smoking behavior

was not different than the effect that fathers had, and each had similar effects on sons as well as

daughters (Huver et al., 2007).

Continued examination of the relationship between maternal and paternal affects on

deviant behaviors by Patock-Peckham and Morgan-Lopez (2007) reported evidence that

maternal parenting style is not significantly related to alcohol use. Overall, children with an

28
 
authoritative father were less likely than children with an authoritarian father to abuse alcohol.

However, the effect that authoritarian fathers had on their daughters was not seen in the

authoritarian father-son relationship. Simply, the data suggest that authoritarian fathers have a

negative social bond with their daughters, which indirectly leads to alcohol abuse (Patock-

Peckham & Morgan-Lopez, 2007). In addition, college-aged respondents who reported feeling

rejected by their fathers were more likely to succumb to alcohol-related problems than those who

felt rejected by their mothers (Patock-Peckham & Morgan-Lopez, 2007). Similarly, Simons-

Morton et al. (2001) note that adolescents whose parents are highly involved, have high

expectations, and hold their children in high regard are less likely to smoke cigarettes or drink

alcohol than adolescents whose parents display the opposite behaviors. Neither cigarette use nor

alcohol use by the child was significantly related to parenting style (Simons-Morton et al., 2001).

Baumrind (1991) reported on alcohol and drug use. Approximately 25% of the sample

consisted of non-users of drugs and alcohol; these non-users were more likely to have

authoritative parents, as were recreational users (Baumrind, 1991). Parents of both heavy alcohol

users and alcoholics were most likely neglecting/rejecting – this also held true for parents of

heavy drug users and drug addicts (Baumrind, 1991). Incidentally, multiple studies have reported

that children of authoritative and authoritarian parents show little difference in patterns of drug

and alcohol use (Baumrind, 1991; Lamborn et al., 1991), contrary to the results presented by

Patock-Peckham and Morgan-Lopez (2007). Significantly fewer children with authoritative and

authoritarian parents use drugs compared to children with neglecting/rejecting parents (Lamborn

et al., 1991). Additionally, the same can be said of the differences between authoritative and

authoritarian parents and permissive parents. Simply, children from permissive and

neglecting/rejecting families were more likely than children from authoritative and authoritarian

29
 
families to succumb to drug and alcohol use (Baumrind, 1991; Lamborn et al., 1991). Baumrind

(1991) even noted that roughly 30% of the substance use variance can be explained by parenting

type. The contrasting results presented in the studies discussed above suggest that further

research is needed in order to better understand the relationship between parenting style and low-

level deviant behaviors such as alcohol, drug, and tobacco use.

The Link between Parenting Style and Delinquent/Criminal Behavior

In addition to the link between parenting style and deviant behavior, the link between

parenting style and delinquent and/or criminal behavior also has been studied in the social

science literature. The research that has been conducted has utilized a variety of methodologies

and measures, while producing consistent results. In short, respondents who perceive their

parents to be authoritative are less likely to engage in delinquent and criminal behaviors, whereas

those who perceive their parents as neglecting/rejecting are more likely to engage in delinquent

and criminal activities. The following sections discuss the methodological design, including

measurement instruments, and empirical findings of research that has examined the link between

parenting styles and delinquent and criminal behavior.

Adolescent delinquent behavior. The previous section examined literature looking at

the link between parenting style and forms of deviant behavior, including behavior that could be

considered delinquent (cigarette, drug, and alcohol use amongst minors). While acknowledging

that use of those substances by children constitutes delinquent behavior – and that use of illegal

drugs by adults constitutes criminal behavior – this section focuses on non-drug/non-alcohol

offenses. Consistent with the adolescent problem behavior and substance use studies, the

research that is discussed in the following sections employs various methodological designs, yet

reports similar results. The following will discuss the methodology and results from empirical

30
 
analyses of the link between parenting style and adolescent delinquent behavior (also see

Appendix C).

Methodology of delinquent behavior studies. A handful of studies have examined the

explicit link between parenting style and delinquent behavior. The first of these types of research

studies to be addressed (Avenevoli et al., 1999; Steinberg et al., 1994; Steinberg et al., 1991)

relied on the same data that were retrieved in an analysis discussed above (Lamborn et al., 1991),

specifically self-report data from high school students in California and Wisconsin. Sample sizes

varied amongst the four studies: 11,669 (Avenevoli et al., 1999); 4,081 (Lamborn et al., 1991);

7,600 (Steinberg et al., 1991); and, 6,357 (Steinberg et al., 1994).

The majority of the remaining studies relied on participation by adolescents in the United

States (Hoeve et al., 2007; Jackson et al., 1998; Paschall et al., 2003; Simons et al., 2005;

Steinberg et al., 2006; Walker et al., 2007), with additional participation from across the globe

(Chambers et al., 2000; Hoeve et al., 2007; Palmer & Hollin, 2001; Stattin & Kerr, 2000).

Starting with self-report studies in the United States, Jackson et al. (1998) had 1,490 ninth and

tenth grade students in North Carolina complete surveys in which they reported their level of

violent behavior. Paschall et al. (2003) also utilized a self-report questionnaire to examine the

effect parenting style had on delinquent behavior in a sample of 175 male African American

adolescents in a medium-sized city in the southeast; the boys’ female caretakers also were

surveyed. Approximately 90% of the caretakers reported being a biological mother to the boy,

4% were grandmothers, 4% were aunts, and 1% was either stepmothers or foster mothers

(Paschall et al., 2003). In Scandinavia, Stattin and Kerr (2000) had 703 14-year-old Swedish

youths from seven communities in Sweden self-report their levels of delinquency and their

perceptions of their parents. While in the United Kingdom, Palmer and Hollin (2001) asked 94

31
 
students between the ages of 12 and 18 from the West Midlands area of England to fill out self-

report questionnaires. The average age of students in this study was 15.69 years-old;

approximately 70% of this sample was female.

Walker et al. (2007) created their sample of 349 African American (21%) and Hispanic

(71%) boys between the ages of 12 and 17 by randomly selecting homes in eight neighborhoods

in Los Angeles County, California. Researchers conducted structured interviews that lasted up to

90 minutes in duration. Steinberg et al. (2006) also used adolescents between similar ages (14-

17; average age was 16). This study utilized 1,355 adolescents who had been adjudicated for

“felony offenses with the exception of less serious property crimes…misdemeanor weapons

offenses and misdemeanor sexual assault” (p. 48); the authors allowed for a maximum 15% of

the sample to include male drug offenders because they believed drug offenses would otherwise

overtake their sample. Additionally, this study is one of the few to include females (16%) and

most of the respondents lived with their single mother (83%), while only 15% lived with both

biological parents (Steinberg et al., 2006). Steinberg et al. (2006) recruited individuals to

participate in the study based on court-provided information. Researchers met with eligible

participants over the course of two days in sessions that were two hours in duration. The

researcher read the survey off of a computer to the respondent who then self-reported the

information aloud to the interviewer or entered the information into the computer if it was of a

sensitive nature (Steinberg et al., 2006). The final delinquency-parenting study to conduct semi-

structured interviews, used 122 male offenders between the ages of 15 and 22 from the Young

Offender’s Institution in Scotland (Chambers et al., 2000). This sample represented one-fourth of

the population in one of the largest Young Offender’s Institutions in the country (Chambers et

al., 2000).

32
 
 Simons et al. (2005) also used adolescent participants, in addition to their parents.

Contrary to most of the other studies, however, Simons and colleagues performed a secondary

data analysis using data gathered for the Family and Community Health Study (FACHS).

FACHS data were collected from African American families in Georgia and Iowa who had a

child in fifth grade at the time of recruitment for the first wave of data collection. During the first

phase in 1998 this included 867 children (46.1% boys; 46.7% in Georgia) and 738 at wave two

in 2000; the authors noted that they used complete data from 633 children (Simons et al., 2005).

In addition to conducting interviews with the fifth grade child, researchers also interviewed the

child’s primary caregiver and a secondary caregiver when one was available. The child’s primary

caregiver was deemed “a person living in the same household as the target child…who was

responsible for the majority of the child’s care” (Simons et al., 2005, p. 1000). The biological

mother was deemed the primary caregiver in 84% of cases, while the biological father and

grandmother were each primary in 6% of cases. It is interesting to note that 83% of the children

in Steinberg et al. (2006) lived with a single mother and the biological mother was the primary

caregiver in 84% of the cases in Simons et al. (2005). Nationally, nine percent (12.9 million) of

all households are headed by a single parent; 10.4 million (80.6%) single parent families are

headed by a single mother (U.S. Census Bureau, 2009).

Hoeve et al. (2007) also completed a secondary data analysis in their investigation of the

link between parenting style and juvenile delinquency. Data from two studies, the Pittsburgh

Youth Study (PYS) and the Child-Rearing and Family in the Netherlands Study (CFNS), were

used. The authors used PYS data from the years 1995-2000, which included 472 boys. PYS

researchers conducted interviews with the boys, the boys’ parents (or primary caregiver), a

teacher, and collected official juvenile court records (Hoeve et al., 2007). The authors point out

33
 
that 93% of juvenile participants lived with their biological mother, but that only 34.1% of them

lived together with their biological father. Thus, historically it appears that social scientists have

attempted to examine the link between parenting style and adolescent deviance/delinquency by

using children whose parenting situation is very dissimilar from national averages. To complete

their study, Hoeve et al. (2007) examined data collected from 132 male participants in CFNS

who were between the ages of 19 and 27. Respondents completed self-report questionnaires

while researchers conducted interviews with family members.

Measure of parenting styles in studies of delinquent behavior. The authors discussed

above measured the link between parenting style and adolescent delinquency in a number of

ways. First, “parenting style” was subjected to a number of conceptualizations, among them:

acceptance/involvement (Avenevoli et al., 1999; Steinberg et al., 1991, 1994, 2006),

authoritative parenting (Jackson et al., 1998; Simons et al, 2005), bond/attachment (Chambers et

al., 2000; Palmer & Hollin, 2001; Walker et al., 2007) firm control/strictness (Avenevoli et al.,

1999; Palmer & Hollin, 2001; Paschall et al., 2003; Stattin & Kerr, 2000; Steinberg et al., 1991,

1994, 2006), monitoring (Palmer & Hollin, 2001; Paschall et al., 2003; Walker et al., 2007) and

psychological autonomy (Avenevoli et al., 1999; Steinberg et al., 1991). Additionally, Hoeve et

al. (2007) utilized a measure to examine family-related variables, including closeness to

child/affection, consistent discipline, conformity, and responsiveness.

Hoeve et al. (2007) reported that the majority of their measurement items were based on

the work of Maccoby and Martin (1983), which was discussed earlier. The “expression of

affection” measurement was based on nine mother-reported items and five child-reported items

and examined the level of fondness directed towards the child. “Demands for conformity”

studied whether the mother expected the child to abide by fixed rules and was based on the

34
 
mothers’ responses to eight items and children’s responses to four items. “Ignoring” examined

the mothers’ patterns of anger when the child was misbehaving using five items. And,

“responsiveness” used eight items to examine the children’s perceptions of their mothers’ level

of attentiveness. As stated, Hoeve et al. (2007) relied heavily on the work of Maccoby and

Martin (1983) when constructing these scales, thus they are similar conceptually to Buri’s (1991)

PAQ measurement instrument and have a similar range of alpha scores (.70 for ignoring

compared to .75/.74 for mother/father permissive behavior; .87 for expression of affection

compared to .82/.85 for mother/father authoritative behavior). It can be argued that the strength

and utility of the PAQ is reaffirmed every time another measurement tool examines similar

concepts and reports similar internal consistency scores.

The acceptance/involvement measures used in the Avenevoli and Steinberg studies listed

above each used 10 items (Steinberg et al., 1994) to 15 (Avenevoli et al., 1999; Steinberg et al.,

1991). This measure examined the children’s perceptions of their parents’ love for them,

responsiveness to them, and involvement in their lives. The alpha score for this measure across

all three studies was .72. The strictness/firm control measure used in the same three studies

examined the children’s perceptions of their parents’ level of monitoring and rule setting. Each

study utilized nine items in this measure and reported an alpha score of .76 across the board.

Psychological autonomy, used by Avenevoli et al. (1999) and Steinberg et al. (1991, p. 25),

assessed “the extent to which parents employ noncoercive, democratic discipline and encourage

the adolescent to express individuality within the family.” This measure has 12 items in each

study with an alpha score of .72.

Steinberg et al. (2006) used similar measures, parental warmth and parental firmness. The

parental warmth scale utilized nine items that assessed both maternal and paternal levels of

35
 
support and understanding (maternal warmth α = .85; paternal warmth α = .88). Parental

firmness assessed perceptions of parental rules. It was noted that the researchers did not test for

separate levels of maternal and paternal firmness, thus there is only one alpha score (α = .80;

Steinberg et al., 2006).

Jackson et al. (1998) eschewed Steinberg’s authoritative parenting measures because the

authors “judged the item wording either too difficult…or inappropriate for children” (p. 322).

Instead, Jackson and her colleagues developed the API, which, as discussed above, measured

demandingness and responsiveness with 20 survey items based exclusively on Baumrind’s

research. Coefficient alpha scores for the API in Jackson et al. (1998) were based on a number of

factors including sex, race, and grade-in-school of the respondent. For example, ninth-grade

white males had alpha scores of .81 on the responsiveness scale and .81 on the demandingness

scale. Alpha scores for this group on the responsiveness measure ranged from .75 to .87, while

alpha scores on the demandingness measure ranged from .68 to .83. While acceptable coefficient

alpha scores they are not as good as Buri (1991), especially the demandingness measure.

Parental bond was measured by Chambers et al. (2000) using the Parental Bonding

Instrument (PBI) which consists of two scales similar to responsiveness and demandingness –

care and control. Respondents were asked to give their perceptions, using a 4-point Likert scale,

to a variety of statements relating to their parents; each participant was asked to respond

separately for mother and father. Each scale used by Chambers et al. (2000) in the shortened

version of the PBI contained ten statements. The full version of the PBI contains 25 statements

per scale. No internal consistency scores were given. Attachment also was measured by Walker

et al. (2007) by looking at family closeness, parent attachment, and parent monitoring. Closeness

was measured using a three-item scale that asked how often families do things together and how

36
 
much fun they had (α = .72). Attachment was measured using 11 items that asked how well the

adolescent respondent got along with his parent or if he often felt angry towards his parent (α =

.72). Monitoring was assessed through the use of a six-item scale that asked how often the

respondent spoke with his parents and how many of his friends were known to his parents (α =

.63). Walker et al. (2007) decided to combine their scales for “ease of interpretation” (p. 309),

thus dubbing it a scale of parent attachment. Given the combined coefficient score of .56,

unacceptable to DeVellis (2003), they may have been better off leaving the scales separate.

The authoritative parenting measure used by Simons et al. (2005) seemingly incorporated

many of these other measurement items. Researchers who compiled the FACHS dataset, used by

Simons and colleagues, collected caregiver responses to the 21-question authoritative parenting

survey. Five of the questions were about parental monitoring, six were about consistent

discipline, six were about inductive reasoning, two were about problem solving, and two were

about positive reinforcement (Simons et al., 2005). The entire instrument had an alpha score of

.75 during both waves of data collection (Simons et al., 2005).

The other studies examined in this section utilized a variety of different measures in their

attempts to discern the link between parenting style and delinquent behavior. Palmer and Hollin

(2001) looked at parenting through three scales – control mechanisms, guidance mechanism, and

affective bond – each with a variety of subscales, including discipline, punishment, supervision,

communication, attachment, and involvement. Alpha scores are not reported for any of the three

scales or eight subscales (Palmer & Hollin, 2001). Paschall et al. (2003) also used many

measures to study parenting style. Among these measures were a nine-item scale that examined

maternal monitoring of son’s behavior, a six-item scale that looked at perceptions of maternal

control over their sons’ behaviors, an 11-item scale examining mother-son communication, and a

37
 
16-item scaled the examined the relationship between mother and son (do they talk together, joke

around, and have fun with each other).

Stattin and Kerr (2000) introduce a variety of reliable measures of parenting. As part of

their parental monitoring scale, they asked adolescents nine questions relating to their parents’

knowledge of their everyday life including homework and friends (α = .86). Parental solicitation

was a five-item measure that asked if parents would try to solicit information from their children

(α = .77) about their whereabouts and who they socialized with. Parental control used six items

to examine if the adolescent respondents had to gain parental permission for going out and

staying out late (α = .82). Child disclosure measured the adolescents’ willingness to disclose

information about their lives to their parents without being prompted; this scale was created

using five items (α = .81). The alpha scores reported by Stattin and Kerr (2000) are more than

acceptable, but they do not adequately capture the essence of authoritarian, authoritative,

permissive and neglecting/rejecting parenting. Once again, after examining the internal

consistency scores of each of the “parenting style” measures discussed above and how they relate

to Baumrind’s conceptualization of parenting styles it is apparent that the Buri’s (1991) PAQ is

the best tool for measuring Baumrind’s typology of parenting styles.

Second, “delinquency” also was measured in a variety of ways. Simons et al. (2005) used

the conduct disorder section of the Diagnostic Interview Schedule for Children, Version 4

(DISC-IV), which asks the respondent to self-report engagement in 26 acts including shoplifting,

assault, vandalism, and burglary. They note that the alpha score for the data they used was above

.90 at both waves (Simons et al., 2005). Steinberg et al. (2006) used two measures of

delinquency: aggressive offending and income-related offending. The aggressive offending

measure consisted of 11 questions that asked about such things as physically attacking another

38
 
person (α = .76) and the income-related offending measure consisted of 11 questions that asked

about such things as taking property from another person (α = .74). The other studies undertaken

by Steinberg and colleagues (Avenevoli et al., 1999; Steinberg et al., 1991, 1994) measured

delinquency by asking respondents to self-report involvement in six activities, including theft,

vandalism, and carrying a weapon (α = .82).

Hoeve et al. (2007) had different delinquent measures for each of the two studies that

they analyzed. Researchers with the PYS measured robbery (pick-pocketing), gang activity,

assault with a weapon, damaging private property, and petty theft. The CFNS study measured

delinquency by asking respondents about participation in vandalism, theft, arson, fraud, assault,

and sex offenses. Jackson et al. (1998) asked respondents if they carried, used, or threatened to

use a weapon against somebody. Additionally, they asked if the participant had ever been

involved in a physical fight. Similarly, Walker et al. (2007) asked respondents whether they had

committed violent acts, including participating in gang fights, using weapons, and sexually

assaulting another person. Palmer and Hollin (2001) measured delinquency using the 46-item

self-reported delinquency scale developed by Elliott and Ageton (1980). The scale asks about

delinquent acts committed within the past year and has a high test-retest reliability (r = .70 - .95).

Paschall et al. (2003) used a 12-item scale that asked participants if they had committed a

number of different delinquent acts. Included in this measure were questions about brandishing

weapons, using weapons in a fight, selling or using drugs and alcohol, and damaging another

person’s property. Stattin and Kerr (2000) asked about “normbreaking.” Many of the nine items

in this measure were similar to Paschall et al. (2003), including consumption of alcohol and

drugs, damaging property, participating in fights, and stealing. An internal consistency score of

.79 for delinquent behavior measures was reported by Stattin and Kerr (2000).

39
 
 Empirical results of delinquent behavior studies. The data gathered by researchers

investigating the link between parenting style and delinquent behavior are very similar to data

from the problem behavior and deviance studies, but there are some interesting differences. As

expected, the authoritative parenting style is associated with lower levels of reported delinquency

in all cases (Avenevoli et al., 1999; Hoeve et al., 2007; Jackson et al., 1998; Palmer & Hollin,

2001; Paschall et al., 2003; Simons et al., 2005; Steinberg et al., 1991, 1994, 2006). Additionally,

in studies that looked into permissive and neglecting/rejecting (Avenevoli et al., 1991; Steinberg

et al., 1994, 2006), results suggested that those children were most likely to succumb to

delinquent behavior. Surprisingly, Chambers et al. (2000), which sampled incarcerated

offenders, suggested that the majority of respondents perceived their parents to be authoritarian;

this result does not fall completely in line with prior research (discussed above and throughout)

which has suggested that individuals who are most likely to display problem behaviors come

from neglecting/rejecting homes.

Interestingly, however, are results from studies that measured and compared authoritarian

and authoritative parenting style and the outcomes associated with each study (Avenevoli et al.,

1999; Steinberg et al., 1991, 1994, 2006). When measuring delinquency outcomes, data suggest

that there are a limited number of significant differences between authoritative and authoritarian

parenting (Steinberg et al., 1994, 2006). As noted above, these studies measured delinquency

similarly. The only instance of significant differentiation in this set of studies was seen by

Steinberg et al. (2006) in their measure of income-related offenses, which suggests that children

with authoritarian parents are more likely than children with authoritative parents to take money

or belongings from somebody else without that person’s knowledge. Delinquent behavior in

children from both authoritative and authoritarian families saw similar declines in one-year

40
 
follow-ups according to Steinberg et al. (1994). Avenevoli et al. (1999), however, found that

delinquency has a significant negative association with the authoritative parenting style, whereas

there is a positive, albeit non-significant, relationship between delinquency and authoritarian

parenting. Somewhat similarly, in addition to reporting a significant negative relationship

between authoritative parenting and delinquency, Palmer & Hollin (2001) report that

authoritarian parenting has a significant positive relationship with delinquency.

This builds on the Baumrind study (1972), which suggested that the authoritarian style of

parenting is beneficial for African American daughters. Avenevoli et al. (1999) report that

African American adolescents with authoritative families have a higher level of involvement in

delinquent activities than their white counterparts, while African American children with

authoritarian parents have lower levels of delinquent involvement than similarly situated white

children. Seeing similar data, Steinberg et al. (2006, p. 56) opine, “thus it is not that authoritarian

parenting is good for poor, urban, ethnic minority adolescents, but…authoritarian parenting may

not be as bad [emphasis in original] for these adolescents as it has been shown to be for their

middle-class, suburban, white counterparts.”

Walker et al. (2007) reported interesting findings relevant to race. As noted, their

measure of attachment included family closeness and parental monitoring. Their results suggest

that the higher level of attachment that Hispanic youth have with their parents the more likely

they are to revert away from delinquency. When looking at the effects that parental attachment

had on African American adolescents, however, Walker et al. (2007) noted that medium levels of

attachment led to the lowest levels of delinquency, while both low and high levels of attachment

led to high levels of delinquency. Given that parental attachment is essentially a proxy for

41
 
authoritative parenting this result is somewhat surprising, but given the results of Avenevoli et al.

(1999) and Steinberg et al. (2006), not wholly unexpected.

Differences in delinquency levels between children from authoritarian and permissive

families were minimal and not significant (Avenevoli, et al., 1999; Steinberg et al., 1994, 2006).

The greatest differences in delinquent behavior were seen in children from neglecting/rejecting

families. Children from neglecting/rejecting families engaged in significantly more delinquent

behavior than children from authoritative families on all measures (Avenevoli et al., 1999;

Steinberg et al., 1994, 2006). These significant differences occurred between neglecting/rejecting

and the other parenting styles on all other measures other than the aggressive offenses measure in

Steinberg et al. (2006). Thus, children from neglecting/rejecting homes did not differentiate

significantly from children with authoritarian and permissive parents in terms of their levels of

aggressive and assaultive behavior (Steinberg et al., 2006).

Stattin and Kerr (2000) reported one final set of findings. In line with the majority of

empirical results presented in this section, Stattin and Kerr (2000) note that high levels of

parental monitoring, control, solicitation, and child disclosure have a significant negative

relationship with reported delinquent behavior. Interestingly, they also suggest that parental

solicitation of their children’s whereabouts is related to higher levels of delinquency than when

children disclose their whereabouts without being prompted. It can be assumed that child

disclosure is relatively synonymous with the notion of autonomy. Simply, authoritative parents

allow their children to have autonomy, whereas authoritarian parents do not. Thus, it can be

argued that authoritative parents rely more on child disclosure, while authoritarian parents may

be more likely to solicit the information. This argument seems to make the most sense. Sadly,

42
 
Stattin and Kerr (2000) fail to make that argument. The next section examines if these delinquent

tendencies transform into criminal behaviors.

Criminal behavior. The previous sections have examined the link between parenting

styles and adolescent problem behaviors, deviance, and delinquency. This section addresses the

limited empirical data that has been gathered about the parenting style-criminal behavior link. It

is somewhat surprising that a limited number of studies have specifically examined this link

because parenting, in general, is correlated with delinquent and criminal outcomes in children.

But, Baumrind’s typology has been largely ignored in criminological literature. Similar to

previous sections, this section discusses the research methodology and empirical results of

studies that have examined the link between parenting style and criminal behavior (also see

Appendix D).

Methodology of criminal behavior studies. The first study of three that have examined

the link between parenting style and criminal behavior was conducted by Schroeder et al. (2010).

In order to conduct their research, Schroeder and colleagues did a secondary data analysis using

data collected by Cernkovich, Giordano, and Pugh (1985). Data were collected in 1982 (n = 942)

and 1992 (n = 721) from the same modified probability sample of respondents (Schroeder et al.,

2010). The average age of the participants during the second wave of data collection was 25 and

the sample was 45% male and approximately 47% Caucasian (Schroeder et al., 2010).

Additionally, respondents were asked to complete self-reports questionnaires that asked about

their criminal histories and their perceptions of parents’ parenting styles.

In addition to asking participants to self-report their level of criminal involvement,

Schroeder et al. (2010) also utilized two parenting style scales. As discussed above, Baumrind’s

four typologies result from two parenting classifications, demandingness and responsiveness

43
 
(Maccoby & Martin, 1983; Simons et al., 2004). The demandingness scale used by Schroeder

and colleagues included seven items that asked respondents if their parents wanted to know

where they were and who they were with, among other things, while not at home (α= .77). The

responsiveness scale that Schroeder et al. (2010) used included 10 items that asked participants

about their parents’ affection and trust towards them (α = .68). Again, the internal consistency

scores of these measures are not as desirable as those garnered by Buri (1991) when testing the

PAQ.

The second study that examines the parenting style-criminal behavior link (Palmer &

Gough, 2007) has many similarities with the proposed study. Palmer and Gough separated their

respondents into three groups (person offenders, property offenders, and non-offenders). Twenty

young (19.45-years-old) male offenders who had been convicted of assault, rape, and/or

threatening behavior, among other offenses, comprised the group of person offenders, while

twenty young (19.2-years-old) male offenders who had been convicted of burglary, vandalism,

and/or arson, among other offense, comprised the group of property offenders (Palmer & Gough,

2007). The group of non-offenders was comprised of 31 young (19.27-years-old) males. The 71

participants were asked to self-report their criminal histories and their perceptions of their

parents’ parenting styles. Palmer and Gough (2007) note the small sample size and their inability

to generalize their findings to larger populations. As stated, the proposed study is very similar,

yet with a larger sample, thus generalizability should not be as big a concern.

Palmer and Gough (2007) also stray from the PAQ in their measurement of parenting

styles. Instead, they used an abbreviated version of the Egna Minnen Barndorms Uppfostran

(EMBU) scale, which measures rejection, emotional warmth, and overprotection, and provides

separate scores for the mother and the father (Palmer & Gough, 2007). It appears that the

44
 
rejection scale is a proxy for the measurement of the rejecting/neglecting parenting style, the

emotional warmth scale is used to measure the authoritative (or even permissive) parenting style,

and the overprotection scale is a measurement of the authoritarian parenting style. The rejection

scale consisted of six items, the emotional warmth scale consisted of seven items, and the

overprotection scale consisted of ten items; higher scores on each scale suggested higher levels

of perceived rejection, warmth, and parental control (Palmer & Gough, 2007). Internal

consistency scores were not reported, but research (Muris, Meesters, & van Brakel, 2003) using

the EMBU has reported alpha scores ranging from .66 (maternal overprotection) to .81 (paternal

emotional warmth), thus suggesting that the EMBU has internal reliability scores that, again, are

not on par with the PAQ.

The third study, Haapasalo (2001), surveyed 89 male Finnish offenders. These

participants were randomly selected from one of five prisons in Finland. The 89 men averaged

20 years of age and the mean number of convictions amongst the group was eight. The criminal

offenses that these men were convicted of ranged from minor summary-type offenses to serious

and violent crimes (Haapasalo, 2001). Once again, the small sample size presents concerns with

generalizability of the findings.

Haapasalo (2001) conducted research using the Child Report version of the Parent

Behavior Inventory (PBI). This version of the PBI included 108 items (grouped into 18

subscales) that asked participants about their parents’ behaviors and attitudes. The original

version of the PBI included 192 survey items (28 subscales) with an internal consistency score of

.76 (Haapasalo, 2001). Again, based on results reported by Buri (1991), the PAQ appears to be a

stronger measure of distinct parenting styles.

45
 
 Empirical results of criminal behavior studies. Studies examining the link between

parenting style and adult criminal behavior suggest that those who report being parented by

authoritative parents are the least likely to engage in criminal behavior (Schroeder et al., 2010).

Given the pattern of results in research that focused on youth behavioral problems, deviant

behavior, and delinquency, this is not a surprising result. On the other hand, it is somewhat

surprising that the neglecting/rejecting parenting style is not significantly associated with higher

levels of offending (Palmer & Gough, 2007; Schroeder et al., 2010). This section provides

further analysis of these results.

Schroeder et al. (2010) suggests that those who perceive their parents as permissive are

significantly more likely to engage in criminal behaviors at a level higher than those who report

any of the other three parenting styles. Surprisingly, no style of parenting is significantly

associated with criminal offending amongst white respondents, though the authoritarian style is

associated with higher levels of offending by whites (Schroeder et al., 2010); white respondents

with authoritarian parents are more likely than those with permissive or neglecting/rejecting to

be involved in crime. Differences are reported by African American respondents, which suggest

that African Americans with permissive parents are significantly more likely than respondents of

other races to be involved in crime (Schroeder et al., 2010). This is closely followed by African

American respondents with neglecting/rejecting parents who are significantly likely to engage in

criminal activity; respondents with authoritative and authoritarian parents report the fewest

criminal behaviors (Schroeder et al., 2010). These results suggest that parenting style may have a

different effect on African Americans than Caucasians with regard to criminal activity.

Palmer and Gough (2007) report data that both agrees with and deviates from data

reported by Schroeder et al. (2010). As mentioned above, no significant differences exist

46
 
between property, person, and non-offenders who claim that their parents were

neglecting/rejecting. In addition, there are no significant differences between property, person,

and non-offenders who perceived their parents as authoritarian; this is similar to Schroeder et al.

(2010). There are significant differences, however, between the levels of offenders and perceived

maternal or paternal authoritative parenting. Mothers of person offenders were significantly less

likely to display authoritative tendencies than mothers of property and non-offenders; level of

maternal authoritativeness was similarly reported by both property offenders and non-offenders

(Palmer & Gough, 2007). Paternal authoritativeness, though, was significantly higher among

non-offenders than both person and property offenders. The data regarding authoritative mothers

is interesting because it is the first to suggest that some offenders and non-offenders were

subjected to similar levels of authoritative parenting. Again, the sample size in the Palmer and

Gough study limits the generalizability of the findings, thus this may represent an outlier.

The results of Haapasalo (2001) also may have limited generalizability due to sample size

and geographic placement. Haapasalo (p. 115) suggests that “parenting appeared to be

manifested in the dimensions of maternal and paternal rejection.” While not described as having

a lack of demandingness and responsiveness that typifies the neglecting/rejecting parenting style,

rejecting parents in Haapasalo’s study were described as rejectors (selfish, detached, and against

child autonomy) and aggressive (physically and psychologically abusive). Despite its limitations,

this study is relevant because the majority of incarcerated respondents report that their parents

were neglecting/rejecting. When considering the empirical findings of the behavioral, deviant,

and delinquent studies discussed above it makes sense that criminal offenders retrospectively

report that their parents governed according to the neglecting/rejecting style of parenting. Given

the variety of reported perceptions by those who have participated in the limited number of

47
 
studies examining the link between parenting style and criminal behavior, the current study is an

attempt to shed necessary light on an under-studied relationship. The next section: discusses how

the current study is related to the empirical literature, provides justification for the current study,

and assesses the current study’s relevance to the field of criminology.

The Current Study

This study sought to integrate the prior literature in a way that has not been done before.

Simply, the researcher desired to examine the relationship between Baumrind’s four parenting

typologies and level of deviant, delinquent, and criminal involvement. As seen above, multiple

studies have looked at one type of behavior when assessing the impact of parenting on that

behavior, but none have included all of them. The following sections discuss how the current

study related to the prior literature while providing a justification for conducting the research.

This section concludes with an argument that explains why it was necessary to integrate a

typology from the discipline of psychology with the field of criminology.

Relationship to Prior Literature and Justification for Study

Past empirical examinations have assessed the impact that parenting has had on various

problem, deviant, delinquent, and criminal behaviors. It appears that the most favorable and least

problematic behaviors in participants are related to the authoritative style of parenting. At the

other end of the spectrum, it seems that the least favorable and most problematic behaviors are

associated with the neglecting/rejecting style of parenting. While prior literature has been an

important source of information for researchers studying the link between parenting style and

behavior, it also has noticeable gaps. This study sought to fill in some of these gaps.

First, like many of the studies discussed above, the current research asked respondents to

self-report their perceptions of their parents’ parenting styles. Unlike the prior research, however,

48
 
respondents were able to report on the person who fulfilled the role of their caretaker (for

example: biological parent, step parent, foster parent, adoptive parent, or grandparent). As seen

in numerous studies (Paschall et al., 2003; Paulussen-Hoogeboom et al., 2008; Querido et al.,

2002), the main caretaker/parental figure varies between the biological parent, grandparent,

stepparent, foster parent, and aunt/uncle. Despite this variation, no research has sought to explain

how the differences in main caretaker affect the children’s behavioral outcomes. The proposed

study sought to examine this gap, though there was not be enough variation in caretaker type to

perform a complete analysis.

Second, the previous research has rarely sought the opinion of young adults when

examining this relationship. Of the literature being scrutinized above, only five studies

(Chambers et al., 2000; Haapasalo, 2001; Palmer & Gough, 2007; Patock-Peckham & Morgan-

Lopez, 2007; Schroeder et al., 2010) ask young adults their perceptions of their parents’

parenting styles, while the majority relies upon the opinions of adolescents or subjective

observations of researchers. It was the opinion of this researcher that young adults (college-aged

and older) would be able to make better-informed assessments of their parents’ parenting based

on the accumulation of life experiences or of becoming parents themselves.

Third, few studies have examined the link between deviant/delinquent/criminal

involvement and parenting by accessing individuals in correctional settings; this is a somewhat

obvious gap given the lack of adults used in prior research. Of the three studies to directly survey

inmates, two were conducted in Europe (Chambers et al., 2000; Haapasalo, 2001), thus

suggesting a need to examine this link not only using those incarcerated, but also using inmates

in the U.S. Additionally, both studies that took place in Europe relied solely on incarcerated

respondents, whereas the sole U.S.-based study (Palmer & Gough, 2007) compared criminal

49
 
offenders to non-criminal offenders. Given that there is one U.S.-based study examining the link

between parenting style and criminal behavior that utilizes responses from offending and non-

offending samples it is an understatement to say that there is a lack of research about the topic.

Fourth, few differences between male and female respondents have been reported in the

prior research. However, when reported the differences are profound. In addition, a handful of

studies have noted different outcomes for white, Hispanic, and Black respondents. These race-

based differences suggest that children of different races have contrasting reactions to similar

parenting styles. The current study attempted to add to this literature, though limited racial

variation did not allow a complete analysis.

Fifth, a plethora of researchers have attempted to construct measurement instruments to

accurately and consistently measure perceptions of Baumrind’s parenting types, and parenting

styles in general. For the most part, these instruments have been proven as reliable and valid.

But, it can be argued that there exists a need for a universal measure of parenting style. The PAQ

(Buri, 1991) represents a suitable starting point, but pilot research (discussed below) has

extended the PAQ to include neglecting/rejecting parenting behaviors and the current study used

a shortened version of this extension. The implementation of a universal measure that retains

reliability and validity when shortened represents an important addition to the study of the

parenting-child behavior relationship.

Relevance to the Field of Criminology

As a psychological perspective, Baumrind’s typology of parenting styles has yet to

receive its due diligence within the field of criminology. This gap in research is supported by the

lack of empiricism surrounding Baumrind’s typology in criminological circles, while

50
 
Gottfredson and Hirschi’s (1990) self-control theory of crime causation (and parental

management) has received widespread attention within the field (Akers & Sellers, 2009).

Essentially, Gottfredson and Hirschi (1990) argued that parental management consists of

three distinct parts: monitor child behavior, recognize deviant behavior, and apply appropriate

punishment. The authors suggested that children will revert from deviant/delinquent/criminal

behavior if they have parents who meet each of the three criteria. Given the similarity of

Baumrind’s typology to the main tenets of parental management and self control proposed by

Gottfredson and Hirschi, it is argued that the typology deserves empirical examination through a

criminological lens.

Pre-Test of Parenting Style Measures

As stated above, this study used the PAQ (Buri, 1991) in order to determine the parenting

style used by each respondent’s parent(s). Buri’s instrument contained 60 total items – separated

into 30 per parent – that were used to assess authoritarian, authoritative, and permissive

parenting. Missing from the original PAQ, however, were items testing for Baumrind’s fourth

parenting type, neglecting/rejecting. In order to perform a complete test of Baumrind’s parenting

typology it was imperative that all four styles were included in the analysis. Thus, 20 items (10

per parent) were created in an effort to assess the neglecting/rejecting typology. The 10 items

measuring maternal neglecting/rejecting behavior are presented in Table 3; the 10 items

measuring paternal neglecting/rejecting behavior are similar, but the word “father” appears in the

place of “mother” and all gender-based pronouns reflect that change. The following section

outlines the reasons for conducting the pre-test, the methodology used to gather data, and

discusses the findings.

51
 
Justification for Pre-Test

Again, in order to properly assess the relationship between Baumrind’s Typology of

Parenting Styles and level of deviant, delinquent, or criminal involvement, one must measure

each of the four typologies. Although the PAQ has proven to be a reliable and valid measure of

Baumrind’s typology (Buri, 1991), it fails to account for 25% of the proscribed parenting styles.

Thus, it was necessary for the researcher to develop, and subsequently test, questionnaire items

that probe respondents’ perceptions of the neglecting and rejecting aspects of the parenting

methods used by their mothers and fathers. Furthermore, the researcher decided that it would be

unwise to utilize untested survey items for such an important part of the dissertation research.

Based on these very simplistic, yet highly important reasons, a pre-test was conducted in order to

determine the internal consistency and inter-correlations of items included in the maternal

neglecting/rejecting and paternal neglecting/rejecting scales.

Table 3: PAQ Extension – Items Measuring Maternal Neglecting/Rejecting

Statement No. Statement

S4 While I was growing up, my mother did not make any rules and showed no
(MomNeg1) desire to do so

S8 Even when I did something wrong to get my mother’s attention, she didn’t
(MomNeg2) respond to me

S10 As I was growing up, I had no idea what my mother expected of me

(MomNeg3)

S14 My mother never gave me any guidance as I was growing up

(MomNeg4)

S17 I feel like I raised myself more than my mother raised me

(MomNeg5)

S20 I don’t know my mother’s views about discipline because she never
(MomNeg6) communicated them with me

52
 
 S24 As I was growing up, I could do whatever I wanted and my mother didn’t care
(MomNeg7)

S27 I did not learn discipline from my mother

(MomNeg8)

S33 I did not learn effective communication skills from my mother

(MomNeg9)

S39 For the most part, my mother ignored me as I was growing up

(MomNeg10)

Methodology

The entire pre-test survey instrument was loaded into Qualtrics, a web-based software

interface which allows researchers to upload questionnaires and distribute them electronically via

email to potential respondents. The researcher was given a list of 3,500 email addresses for

undergraduate students at one university in the northeast; this list was randomly generated by the

director of the Applied Research Lab (ARL) at the University. Prior to distributing the email to

the 3,500 potential respondents, the researcher copied the informed consent form for the pre-test

and the electronic link to the survey into the Qualtrics email interface.

The random sample of potential respondents were able to access the survey instrument

for approximately five weeks, during which time the researcher sent four reminder emails to

those who had not responded. Of the 3,500 undergraduate students who received an invitation to

complete the survey, only 447 accessed it (12.77%). Out of those 447 participants, 306

responded to each statement relating to the permissive style of mothers, 304 responded to each

statement relating to the neglecting/rejecting maternal style, 301 responded to all maternal

authoritarian items, and 298 responded to the all of the authoritative measures for mothers. As

for the paternal measures, each scale received 195 completed responses.

53
 
 The low response rate is a bit disconcerting and there are likely several factors that

impacted this. First, the survey was initially distributed immediately before the Easter holiday

when the weather was first beginning to turn nice, thus it is understandable why young college

students failed to respond. In addition, one set of reminder emails was sent out with two weeks

remaining in the semester and the final reminder was sent out prior to final exam week. Given

that the timing of the initial invitation and reminders was very close to the end of the semester

when students are preparing for final exams it is not surprising that more people chose to not

participate. Finally, it is important to consider how dynamic response rates are in survey research

and how response rates are influenced by mode of delivery and respondent population (Dillman,

Smyth, & Christian, 2009). Dillman et al. (2009) report a variety of response rates based on the

mode of survey dissemination; this suggests that Internet and e-mail surveys are likely to receive

lower response rates than mail and face-to-face surveys. Additionally, Dillman et al. (2009, p.

10) suggest that the young adult demographic “is probably the most difficult group to get to

respond to surveys by any mode.” In order to improve upon this response rate in the dissertation

project the researcher implemented Dillman and colleagues’ (2009) method for web-based

surveys. Despite, the issues mentioned, it is important to note that the number of completed

surveys is sufficient for completing the planned analysis.

Demographically, the respondents averaged 21.8 years of age, were 75% female, and

85% white. It appears women are overrepresented in the sample, as the University’s student

population is 56% female, but the breakdown between white and non-white respondents is very

similar to the University, whose student body is 87% white (“Facts about IUP,” n.d.).

54
 
Results

In assessing the data, multiple aspects were examined. First, the researcher wanted to

determine if the scale items for neglecting/rejecting parenting were reliable. Second, the

researcher wanted to assess the reliability of Buri’s (1991) original six scales. Finally, the

researcher wanted to assess the intercorrelation scores of the eight extended measures and

determine if they were comparable to Buri’s original six scales given the introduction of the

maternal and paternal neglecting/rejecting scales.

Test of internal consistency. The overriding reason for conducting this initial test of the

survey instrument was to examine the internal consistency of the new measures for the maternal

and paternal neglecting/rejecting parenting styles. The Cronbach alpha scores for Buri’s initial

test are discussed above and included in Table 4 below. As discussed, Buri’s scales can be

characterized as “respectable” and “very good” when using DeVellis’ (2003) criteria for

assessing measurement scales. The Cronbach alpha scores attained in the pre-test were very

similar across all of the original six scales when compared to the outcomes Buri experienced: .85

for Mother’s Authoritarianism, .88 for Mother’s Authoritativeness, .80 for Mother’s

Permissiveness, .88 for Father’s Authoritarianism, .91 for Father’s Authoritativeness, and .79 for

Father’s Permissiveness. As seen, the reliability scores for these six scales are slightly higher

than the alpha scores attained by Buri during his original reliability tests.

It is possible that the slightly higher alpha scores can be explained by minor changes that

this researcher made to the PAQ statements. For example, a statement measuring maternal

authoritativeness in the PAQ reads, “As I was growing up my mother directed the activities and

decision of the children [emphasis added] in the family through reasoning and discipline” (Buri,

1991, p. 113). The same statement was amended for the pre-test to read, “As I was growing up

55
 
my mother directed the activities and decisions that I made through reasoning and discipline.” It

is the opinion of the researcher that this minor difference changed the scope of the statement

from a general parenting behavior applicable to all children in the family to a more specific

behavior geared solely towards the individual evaluating the statement. Thus, the respondents

were able to agree or disagree with the statement based completely on their subjective

interpretations of their parents’ behavior, which may explain the slight increases in reliability

scores.

Aside from similar reliability scores for the six existing measures, the results of the pre-

test suggest that the scales for maternal and paternal neglecting/rejecting behavior also are

respectable and very good. As displayed in Table 4, the coefficient alpha value for Mother’s

Neglect/Rejection was .86 and the alpha value for Father’s Neglect/Rejection was .90. These

alpha values, both for the existing measures and new measures, suggest that the PAQ and the

extension scales are very solid, and the scales were more than acceptable to use in order to

determine perceptions of parenting styles in the current study.

Table 4: PAQ Extension – Internal Consistency of Scales

Buri (1991) α Spraitz Extension α

Mother’s Authoritarianism .85 .85
Mother’s Authoritativeness .82 .88
Mother’s Permissiveness .75 .80
Mother’s Neglect/Rejection - .86

Father’s Authoritarianism .87 .88

Father’s Authoritativeness .85 .91
Father’s Permissiveness .74 .79
Father’s Neglect/Rejection - .90

Test of discriminant validity. As Buri (1991) suggested, if each of the scales were

accurate measures of Baumrind’s typology then participants should provide divergent responses

56
 
to each of the items. For example, items measuring the strict disciplinarian parent (authoritarian)

who is likely to demand a lot, yet communicate little, should diverge from items measuring the

parent who guides a child through life explaining the rules and reasons for the rules

(authoritative). In addition, items measuring permissive parents who are lax disciplinarians, yet

friendly with their children, should diverge from authoritarian and authoritative measures. Each

of these parenting types display different characteristics, thus it is important that these

differences are accounted for in the measurement instrument.

Results from Buri (1991) supported the notion of divergence (see Table 5). Simply,

maternal authoritarianism was inversely and significantly related to maternal authoritativeness

and maternal permissiveness, and paternal authoritarianism was inversely and significantly

related to paternal authoritativeness and paternal permissiveness (Buri, 1991). Additionally,

neither maternal nor paternal permissiveness and authoritativeness were related significantly to

each other (Buri, 1991).

Table 5: Intercorrelations of PAQ Scores

1 2 3
1 Mother’s 1.00
Authoritarianism
2 Mother’s -.48* 1.00
Authoritativeness
3 Mother’s Permissiveness -.38* .07 1.00
4 5 6
4 Father’s Authoritarianism 1.00
5 Father’s Authoritativeness -.52* 1.00
6 Father’s Permissiveness -.50* .12 1.00
SOURCE. – Adapted from Buri (1991)
*p ≤ .0005

As expected, pre-test results of the intercorrelation scales for the original PAQ items

mirrored Buri’s results (see Table 6). As seen in the table, the newly introduced

57
 
neglecting/rejecting scale is intercorrelated with the other scales similarly across both the

maternal and paternal measures. Simply, the neglecting/rejecting scale is positively and

significantly related to the permissive scale, inversely and significantly related to the

authoritative scale, and inversely and not significantly related to the authoritarian scale. These

results present interesting insight into the relationships between parenting types and the

constructs of demandingness and responsiveness.

First, and as expected, the scales for parenting types that do not share similar levels of

both demandingness and responsiveness (authoritarian-permissive; authoritative-

neglecting/rejecting) were inversely and significantly intercorrelated. Because neither of these

two pairs of parenting types have anything in common with each other it is assumed that scales

constructed to measure them should diverge from each other; this assumption is supported by the

findings.

Table 6: PAQ Extension – Intercorrelation Scores

1 2 3 4
1 Mother’s 1.000
Authoritarianism
2 Mother’s -.353* 1.000
Authoritativeness
3 Mother’s -.544* .107 1.000
Permissiveness
4 Mother’s -.084 -.594* .448* 1.000
Neglect/Rejection
5 6 7 8
5 Father’s 1.000
Authoritarianism
6 Father’s -.350* 1.000
Authoritativeness
7 Father’s -.543* .020 1.000

58
 
Permissiveness
8 Father’s -.031 -.602* .538* 1.000
Neglect/Rejection
*p ≤ .01

Second, the scales for parenting types that share similar levels of demandingness

(authoritarian-authoritative; permissive-neglecting/rejecting) were significantly intercorrelated,

but in opposite directions. As discussed earlier, authoritarian and authoritative parents share high

levels of demandingess. The scales for these parenting types were inversely related to each other,

while the scales for permissive and neglecting/rejecting (shared low levels of demandingness)

were positively intercorrelated with each other.

At the same time, the scales for parenting types that share similar levels of

responsiveness (authoritarian-neglecting/rejecting; authoritative-permissive) were not

significantly intercorrelated. Again, authoritarian and neglecting/rejecting parents share low

levels of responsiveness; the scales for these parenting types were inversely intercorrelated.

Authoritative and permissive parents share high levels of permissiveness and the scales for each

are positively intercorrelated.

These results suggest a number of things. First, shared levels of demandingness, whether

high or low, have a greater and more significant impact than shared levels of responsiveness

when assessing intercorrelations. Second, findings suggest that parents who share similar strong

patterns of discipline, but engage in differing levels of communication with their children are

more different from each other than parents who engage in differing levels of communication

and employ similar levels of lax discipline. It can be argued that respondents perceived over-

bearing parents who talked with them differently than over-bearing parents who did not talk with

them, yet perceived little difference between less demanding parents whether or not they

59
 
communicated with them. Third, despite non-significance, results suggest that respondents

perceived communicative parents similarly despite differences in disciplinary techniques,

whereas non-communicative parents with differing methods of discipline were perceived

differently. Thus, it appears that the demandingness construct is more important than

responsiveness in assessing perceptions of parenting style.

Despite the significant positive relationship between the neglecting/rejecting and

permissive scales, it is a generally accepted rule that discriminant validity exists if the correlation

value is less than .85 (John & Benet-Martinez, 2000). The intercorrelation values of .448

between the maternal scales and .538 between the paternal scales fall below this threshold.

Conclusion

Buri’s (1991) original PAQ scales of maternal and paternal authoritarianism,

authoritativeness, and permissiveness continue to be supported, as suggested by the results of the

pre-test. The new scales, maternal and paternal neglecting/rejecting, also appear to be both

reliable and valid measures of Baumrind’s fourth parenting type. The addition of these two

scales proved to be useful to the current study in the short term, and should be useful to the

continued study of the relationship between parenting style and child behavior in the long term.

60
 
 CHAPTER III

METHODOLOGY

This study utilized a cross-sectional design. A quantitative self-report survey instrument

was administered to two distinct sample groups of respondents (incarcerated individuals and

university students) in order to examine the relationship between perceptions of parenting style

and level of deviant, delinquent, and criminal involvement. The following chapter details the

overall research design, site and sample selection (beginning with the incarcerated sample), the

survey methodology, reliability and validity, human subject protections, and limitations and

strengths of the design. First, however, the research questions and hypotheses of the current

study are introduced.

Research Questions and Hypotheses

Using the two sample groups from each site mentioned above, the current study sought to

answer the following research questions:

1. What parenting style is the most likely to be reported by the respondents?

2. When looking at the entire sample, which parenting style will best predict reduced

deviant, delinquent, and criminal involvement; which parenting style will best predict

increased deviant, delinquent, and criminal involvement?

3. How will the impact of different parenting styles on the level of deviant, delinquent, and

criminal behavior compare across sample groups?

4. What impact does the sex of the respondent have on the relationship between parenting

style and level of deviant, delinquent, and criminal involvement?

5. What impact does race have on the relationship between parenting style and level of

deviant, delinquent, and criminal involvement?

61
 
 Based on the empirical research discussed in earlier chapters and the research questions

outlined above, this study analyzed the following hypotheses (with corresponding research

question in parentheses):

H1 (1): The authoritative parenting style will be the most reported style of parenting for

both maternal caretakers and paternal caretakers.

H2 (2): The authoritative parenting style will be the most significant predictor of reduced

involvement in deviance, delinquency, and crime; the neglecting/rejecting parenting style

will be the most significant predictor of increased involvement in deviance, delinquency,

and crime.

H3 (3): Authoritative parenting will have the most significant effect on the behaviors of

the university students; neglecting/rejecting parenting will have the most significant

effect on the behaviors of the county jail inmates.

H4 (4): Female and male respondents will be affected similarly when parented with the

same parenting style.

H5 (5): Non-white and white respondents will be affected similarly when parented with

the same parenting style with the exception of the authoritarian style; non-white

respondents will be less likely than white respondents to engage in deviant, delinquent,

and criminal behaviors when parented by an authoritarian parent.

Site Selection

The current study gathered data from three different sites – two county jails in one

northeastern state and one state university in the northeast. The following sections discuss each

of these different sites.

62
 
County Jail #1

The first county jail is located in a rural area, and is under the purview of its county’s

prison board, which is comprised of the Presiding Judge of the Court of Common Pleas, the

district attorney, the sheriff, the controller, and the county commissioners (61 Pa.C.S.A. § 1731,

2009). This county is home to approximately 87,500 people, roughly 82% of whom are over the

age of 18 (U.S. Census Bureau, 2010). The county jail houses 256 beds for male and female

inmates. According to the jail Warden, the jail may hold up to 40 inmates from any of the State

Correctional Institutions located within the state’s Department of Corrections (DOC); state

inmates were not be eligible to participate in this study because the DOC was not allowing

access to prisoners for research purposes at the time data was gathered.

County Jail #2

The second county jail is located in an urban area, and is under the purview of its

county’s bureau of corrections. This county is home to roughly 1.2 million people,

approximately 80% of whom are over the age of 18 (U.S. Census Bureau, 2010). The jail houses

over 2,000 men and women on any given day (Allegheny County Pennsylvania, n.d.). These

individuals are housed for a variety of reasons, “[A]pprehended fugitives, persons who had

bonds revoked, persons who have been sentenced to Jail at Court, Parole or Probation violators”

(p. 1). Additionally, inmates are delivered to the jail by local police, county sheriffs, state

troopers and constables, as well federal authorities. One can assume that this mixture of reasons

for apprehension and apprehending agency suggests that the level of criminal involvement by

inmates at this jail is extremely varied; likewise, one can assume that the level of criminal

involvement at the rural county jail equally is varied. Thus, it was argued that the two jails were

ideal locations to conduct research.

63
 
The University

The University is a four-year university located in a northeastern state. It is the largest

university affiliated with its state’s higher education system with an enrollment of 14,638

students during the Fall 2009 semester (“Facts about IUP,” 2009). Of the 14,638 students, 44%

are male, 87% are Caucasian, 92% are traditional college-aged, and 12,291 are undergraduates.

In addition, the student body hails from 47 states, 3 U.S. territories, and 72 countries, thus

suggesting the possibility of constructing a very diverse sample for the study.

Sampling and Survey Administration

As stated, the current research intended to sample two groups in an effort to discern the

effects that perceived parenting styles have on levels of deviant, delinquent, and criminal

involvement. The first sample included men and women incarcerated in one of the two jail

facilities discussed above. The second sample consisted of university students currently taking

courses. Due to concerns regarding the accessibility of inmate populations, the researcher

employed multiple sampling procedures. The following discusses the sampling procedures

necessary to establish both an inmate and university student sample.

County Jails

Given the transient nature of jail populations, the researcher had little choice but to use

the non-probability sampling technique of availability sampling for the inmate sample.

According to the urban county jail website (n.d.), the facility oversees roughly 350 in-out/out-in

movements each day. Included in these movements are recently arrested individuals, those who

cannot post bail, and those delivered to the jail by a variety of law enforcement agents, as

discussed above. Additionally, there is daily movement of inmates to and from court

proceedings. Once again, the volume of movement and uncertainty regarding which inmates

64
 
would have been present when the researcher administered the survey suggested that the best

way to procure an inmate sample was to use an availability sample.

Bachman and Schutt (2010) suggest that availability sampling is appropriate when a

researcher is attempting to study perceptions of a particular population. Therefore, availability

sampling was appropriate based on the notion that the researcher was attempting to study the

inmates’ perceptions of their parents’ parenting style. In order to secure this sample, the

researcher met with the Warden and Deputy Warden of the rural county jail, and one of the

Deputy Wardens of the urban county jail. During these meetings, the researcher outlined the

study, allayed any concerns that the jail administration may have had regarding safety and

interference with jail operations, and promised to keep the Deputy Warden updated with the

progress of the research. Jail administrators at each site assured the researcher that access to the

jails would be granted. The researcher completed and submitted institutional clearance

paperwork for access to each jail.

In order to secure participation from inmates at the rural county jail, the researcher

received a list of all county inmates in the facility (state inmates housed in a county facility were

not available for research given the DOC’s moratorium on outside research) from the deputy

warden on each day that research was conducted. The researcher announced the study to inmates

assembled in the common area of each housing unit/pod. County inmates interested in

participating were asked to meet with the researcher in an activity room located within each

housing unit. The Deputy Warden of the rural county jail allowed access to an activity room

within each housing unit as a secure place to administer the surveys. The researcher met with

inmates in the urban county jail on one occasion. The nature and purpose of the study was

explained to inmates in two housing units in the urban jail. The survey was group-administered

65
 
to interested participants in each housing unit. The researcher began all meetings by reading

aloud the informed consent form to the participants (see Appendix E); participants were invited

to keep a copy of this form. After consent was agreed upon, the researcher read each of the

questions or statements to the entire group while allowing adequate time for the participants to

respond. In all instances, the participants did not want the researcher to read the questions aloud.

The researcher also emphasized that the respondents should ask for clarification when a

statement or question was confusing. Participants were able to withdraw at any time by writing

“withdraw” on the survey; most exited the area in which the survey was being administered upon

indicating their desire to withdraw. At the completion of the questionnaire, the researcher

collected all survey instruments (whether complete or incomplete) and exited the housing unit.

The University

As discussed above, the study included a sample of undergraduate students. The survey

administration to this sample differed from the survey administration to the incarcerated sample.

In order to generate the sample of university students, the researcher discussed the scope of the

project with the coordinator of the ARL at the University who then created a randomly generated

list of 2,000 undergraduate email addresses. The email addresses and survey instrument were

uploaded into the Qualtrics, Inc., survey interface, available through the University at

https://iup.qualtrics.com/ControlPanel/.

On one Monday morning in late October 2010 an email was sent to all randomly selected

survey participants. The email message explained the scope of the project, the voluntary nature

of participation in the project, and an active hyper link to the questionnaire. Once the participants

clicked the hyper link they were directed to the survey instrument on the Qualtrics website. To

reinforce the fact that their participation was voluntary, the informed consent form was visible on

66
 
the first page (see Appendix F). Participants were not able to proceed with the survey until they

clicked on an icon indicating that they agreed with the informed consent. Once the participants

completed the survey the results were stored in the researcher’s Qualtrics account.

In order to increase the response rate for the electronic portion of the study, the researcher

employed tactics for web survey implementation outlined by Dillman et al. (2009). First, the

researcher used multiple contacts. As stated, the survey initially was sent to the email accounts of

everybody in the sample on one Monday morning in late October 2010. Dillman et al. (2009)

suggest that there is a greater likelihood of web-based survey completion if the respondents

receive the link on a Monday morning as opposed to any other time of day. Reminder emails

were sent on each of the four successive Mondays to those who had failed to access the survey.

The interface kept track of users who accessed and completed the survey and did not send

reminder emails to those who already had participated; the researcher, however, could not access

this information and did not know the identities of those who had participated, thus ensuring the

anonymity of the respondents.

Second, Dillman et al. (2009) suggest that the researcher send a token of appreciation to

all potential respondents. Because the researcher wanted the respondents to retain anonymity

during this project, complying with this tactic was not feasible. However, the researcher offered

anybody who completed the entire survey an opportunity to be entered in a raffle to win one of

four $25 gift cards to the Co-op Store at the University. In order to do this, the researcher asked

the respondents if they would like an opportunity to win a gift card. If they did, they were asked

to click “yes” before submitting the completed survey to Qualtrics. Respondents who clicked

“yes” were diverted to another screen on which they entered their names and email addresses;

67
 
information from this form was kept separate from survey responses. The identity of the four

raffle winners was known only to the researcher once the raffle was complete.

Another tactic employed to increase participation was to carefully select the subject line

of the email soliciting participation. Dillman et al. (2009) suggest that the subject line should

indicate that the email is about a survey sponsored, in part, by the university. The topic of the

survey also should be apparent by the email title; the email title read “[University] Student

Survey – Seeking Your Opinions About Parenting.” Fourth, instructions for accessing the survey

and a hyper link to the survey were clearly provided within the text of the email. Fifth, and as

briefly mentioned above, once the respondents accessed the survey the first thing they saw was

notification that they were in the right place and a reminder that their participation was voluntary

before they clicked to begin the survey.

The researcher believes that the implementation of these web-based survey techniques

helped increase the number of completed surveys from the sample of university students when

compared to the number who completed the pre-test. Use of some of these techniques may help

solidify some of the web-based parameters of Dillman and colleages’ (2009) tailored design

method.

Survey Instrument

The current study called for the examination of the relationship between perceived

parenting styles and deviant, delinquent, and criminal outcomes. Unlike past research which has

studied the relationship between parenting style and negative behavioral outcomes in young

children using participant observation or adolescents using self-reports (see Literature Review),

this study sought self-reports from adults in two sample groups. The survey questionnaire asked

the respondents to report on their perceptions of parents’ parenting style, other questions related

68
 
to parenting, their own deviant, delinquent, and criminal history, and basic demographic

questions. See Appendix G for a copy of the survey instrument.

Baumrind Measures

Respondents’ perceptions of their parents’ parenting styles were assessed using an

abbreviated version of Buri’s (1991) Parental Authority Questionnaire (PAQ). As discussed in

the literature review, the PAQ was originally created as a means of assessing authoritarian,

authoritative, and permissive parenting behavior. The pre-test, also discussed above, extended

the PAQ to include items for measuring neglecting/rejecting parenting. The proposed study used

an abbreviated version of the extended PAQ in order to measure participants’ perceptions of their

parents’ parenting styles. A shortened version of the PAQ was implemented because a survey

instrument with the full version would exceed 100 items. As discussed above, a limited number

of pre-test participants who accessed the survey actually completed each survey item; it can was

assumed that the length of the original measure contributed to the low completion rate.

Split-half reliability analysis of the pre-test responses was implemented in order to create

the shortened scale. George and Mallery (2010) suggest that applying the split-half technique is

acceptable when a questionnaire contains a large number of items intended to measure similar

concepts. For example, the original PAQ is devised of six scales with ten items each; the

extended version offers two additional ten item scales – a total of 80 items. The researcher

simply created each scale through a trial-and-error process that determined which five items had

the highest alpha score when grouped together. By utilizing split-half reliability analysis of the

pre-test results, the researcher was able to create eight scales with five items each. Simply, the

researcher was able to cut the survey instrument in half while retaining similar Cronbach’s alpha

scores (see Table 7).

69
 
Table 7: Comparison of Coefficient Alpha Values: Full Scale vs. Split-Half

Coefficient Alpha Values

Full Scale Split-Half
Maternal Authoritarian .85 .77
Paternal Authoritarian .88 .83

Maternal Authoritative .88 .87

Paternal Authoritative .91 .90

Maternal Neglecting/Rejecting .86 .82

Paternal Neglecting/Rejecting .90 .90

Maternal Permissive .80 .74

Paternal Permissive .79 .79

The survey items measure Baumrind’s four parenting types; as stated, all eight scales

contain five items. The maternal authoritarianism scale (α = .77) scale asks participants to

indicate their level of agreement with a series of statements such as, “Whenever my mother told

me to do something, she expected me to do it immediately without asking any questions,” and,

“Even if I didn’t agree with her, my mother felt that it was for my own good if I was forced to

conform to what she thought was right.” The paternal authoritarianism scale (α = .83) contains

similar items, including, “My father has always felt that most problems in society would be

solved if we could get parents to strictly and forcibly deal with their children when they don’t do

what they are supposed to do.”

Maternal authoritativeness (α = .87) asks respondents their level of agreement to a series

of statements, including, “As I was growing up, my mother consistently gave me direction and

guidance in rational and objective ways,” as well as, “My mother gave me direction for my

behavior and activities as I was growing up and she expected me to follow her direction, but she

was always willing to listen to my concerns.” The paternal authoritativeness scale (α = .90)

contains those two measures and three others, including, “My father had clear standards of

70
 
behavior for me as I was growing up, but he was willing to adjust those standards to my needs.”

The newly created maternal neglect/rejection scale (α = .82) contains statements such as, “As I

was growing up, I had no idea what my mother expected of me,” and, “My mother never gave

me any guidance as I was growing up.” Paternal neglect/rejection (α = .90) asks participants their

level of agreement to two of the maternal items described above as well as three other items,

including, “As I was growing up, my father did not direct my behaviors, activities, and desires.”

The maternal permissiveness scale (α = .74) asks respondents to indicate their level of

agreement with declarations such as, “Most of the time as I was growing up my mother did what

I wanted her to do when making decisions,” and, “As I was growing up, my mother allowed me

to decide most things for myself without a lot of direction from her.” The same five items are on

the paternal permissiveness scale (α = .79). To reiterate, the researcher believed that an

abbreviated version of the full PAQ scale captured results similar to those that would have been

captured if the longer 80-item version were distributed to participants.

Other Parenting Measures

Because the research discussed in the literature review suggested that “caretaker” is not

synonymous with “biological parent,” the survey instrument directed the participants to click or

mark (depending on electronic of paper survey) the response “indicating which female raised

you the most while you were growing up and respond to the following statements with that

person in mind;” similar directions are given before the respondent begins the paternal portion of

the questionnaire. The answer categories include: Biological Mother, Foster Mother, Adopted

Mother, Step Mother, Grandmother, Sister, Aunt, Female Cousin, Other Female Relative (with a

prompt to indicate who), and None (with a prompt to skip to the next section on the paper

71
 
survey; electronic survey-takers will be re-directed automatically). Again, the response

categories for paternal caretaker are similar, yet reflect gender differences.

The proposed research also sought an answer to the questions of whether respondents

believe their parents’ parenting styles affected their behavior and whether they hold their parents

accountable for the behaviors they engaged in. Participants were asked to indicate their level of

agreement to the following survey items, “My mother’s parenting style had a big effect on my

behavior while growing up,” and, “It is my mother’s fault that I have done the things that I have

done.” Items measuring paternal impact were reworded accordingly.

Level of Criminal Involvement – Dependent Variable

The survey instrument also measured self-reports of the participants’ involvement in

deviant, delinquent, and criminal behavior. This section asked a variety of questions hoping to

elicit the individual respondent’s engagement in a variety of deviant and illegal activities.

Respondents also were asked to estimate the number of times they had taken part in the act. Most

survey items were taken from early research conducted by Elliott and Ageton (1980) and

Thornberry and Farnworth (1982). The self-report tables asked the participants to indicate the

acts that they committed by making an “x” in the box next to the crime/behavior.

Elliott and Ageton (1980) arranged many of the survey items into six subscales:

predatory crimes against persons, predatory crimes against property, illegal service crimes,

public disorder crimes, deviant behaviors, and hard drug use. The latter designation, hard drug

use, was changed to alcohol/drug use in the current study because it asks participants to report on

alcohol use (both underage and of-age use) in addition to the use of nine other substances,

including cocaine, heroin, marijuana, and tobacco. Given the inclusion of tobacco, marijuana,

and alcohol, the use of the term “hard” no longer applied. Items included in the “alcohol/drug

72
 
use” scale did not ask the respondents to estimate the number of times that they had partaken in a

certain substance because it would be nearly impossible for a long-time user to provide an

accurate estimation of frequency of use.

There are a total of eight items measuring crimes against people. Statements indicating

involvement in predatory crimes against persons include, “purposely killed someone,” “used a

weapon to threaten another person,” and, “had (or tried to have) sexual relations with someone

against their will.” Nine items were included in the predatory crimes against property scale, they

include: “purposely damaged or destroyed property belonging to another person,” “stolen money

from another person,” and, “stolen (or tried to steal) something worth more than $50.” A total of

six items were included in the illegal service crime scale. These items include, “sold marijuana”

and “paid to have sexual relations with another person.” There were eight items included in the

public disorder crimes scale, including: “set off a fire alarm for fun,” “been drunk in a public

place,” and “carried a gun without a permit.” Lastly, there were six statements in the deviant

behavior scale including, “cheated on school tests” and “run away from home when younger

than 18.”

Admittedly, these 48 items do not represent an exhaustive measure of involvement in all

types of criminal, delinquent, and deviant activities. However, they are a substantial upgrade

from the limited measures used to examine deviance, delinquency, and criminal behavior in

many of the parenting style studies that were discussed earlier in the literature review.

Furthermore, Palmer and Hollin (2001, p. 89) used a 46-item Elliott and Ageton (1980) scale and

concluded that it displayed “high reliability (test – retest r = 0.70-0.95).” Additionally, Huizinga

and Elliott (1986) noted that the 46-item scale had superior sampling validity than most other

self-report measures; sampling validity is used to determine “whether the items included in a

73
 
scale form an adequate and representative sample of the domain of behavior being investigated”

(p. 311). While not exhaustive of all deviant, delinquent, and criminal behaviors, prior research

has suggested that the survey items that were used to measure level of criminal involvement are

reliable (Palmer & Hollin, 2001) and valid (Huizinga & Elliott, 1986).

Demographics

The questionnaire concluded with questions about the respondents’ demographics.

Specifically, the researcher was interested in the age, sex, and race of the respondents.

Oftentimes, as discussed above, sex is an unused variable when studying parenting typologies.

Additionally, it would have been interesting to examine how the differences in parenting styles

of white and non-white families contributes to deviant, delinquent, and criminal outcomes had

there been enough racial variation in the sample.

Reliability and Validity

Reliability

Simply, reliability is the consistency of a measure (Shadish, Cook, & Campbell, 2002). A

measure is reliable if results are consistent with each other over the course of multiple repeated

tests (Carmines & Zeller, 1979). As mentioned, the reliability scores for the survey items

measuring perceived parenting styles have been found to be “respectable” and “very good”

(DeVellis, 2003) in all of Buri’s tests and in the pre-test that was conducted in preparation for

this current research.

Validity

Carmines and Zeller (1979, p. 12) define validity as the “crucial relationship between

concept and indicator.” In this instance, the main concept is Baumrind’s typology of parenting

styles and the indicator is the survey instrument used to assess perceptions of this typology.

74
 
Thankfully, Buri (1991, p. 118) tested the PAQ and determined that it is a “valid measure of

Baumrind’s parental authority prototypes.” Measurement of the neglecting/rejecting scale in the

pre-test suggested that it also is a valid measure of Baumrind’s typology.

Buri’s (1991) original survey instrument showed significant discriminant validity

between the authoritarian, authoritative, and permissive measures at the .0005 level. The

extended survey instrument displayed discriminant between all four measures of Baumrind’s

parenting types at the .01 level. Shadish et al. (2002, p. 507) explain discriminant validity as the

idea that a “measure of A can be discriminated from a measure of B, when B is thought to be

different from A.” This is important in the current study for two reasons: (a) each parenting style

is different from the other three and (b) in order to limit bias, the researcher must be sure that he

is measuring each parenting style correctly. It appears that these criteria have been met.

Similarly, Buri (1991) examined the criterion validity of his instrument by comparing

measurement outcomes with outcomes from the Parental Nurturance Scale (PNS; Buri,

Misukanis, & Mueller, 1988). Comparisons were made because parental nurturance is an implicit

component of Baumrind’s parenting typology. Buri (1991) believed that the authoritative

measures on the PAQ scale would have a positive significant relationship with the PNS, the

authoritarian measures on the PAQ scale would have a negative significant relationship with the

PNS, and that the permissive measures would not be significant. Measures for both authoritative

and authoritarian parenting were significant at the .0005 level, while the measure for permissive

parenting was not significant at the .10 level. This suggests that, as constructed, the survey

instrument has a high level of criterion validity.

Threats to validity. As is the case with most social science research, there are a variety

of validity threats inherent within the current study. The majority of validity threats revolve

75
 
around one of the two sample groups. Buri (1991) originally tested his survey instrument with

the help of 69-185 college students enrolled in introductory psychology courses and the pre-test

examined a similar instrument with the help of approximately 300 university students, thus the

researcher was less concerned with any validity threats that may have resulted from the use of

college students than with validity threats that may have resulted from using jail inmates as

participants, as will be explained below.

One possible risk was including a survey item (or multiple items) that was not understood

(Fowler, 2009). The researcher assuaged this potential problem by constantly reminding the

inmate participants that they may ask questions if they needed something clarified; a handful of

inmate participants asked questions while completing the questionnaire. Additionally, federal

human subject protections, which are discussed momentarily, mandate that survey questions be

worded in a way that is easily understandable by protected populations. Because the researcher

constantly asked if the respondents needed anything clarified and the majority of respondents did

not request clarification, the researcher anticipates limited validity threats due to failure to

understand what a question was asking.

Memory recall represents another threat to validity (Fowler, 2009). It was anticipated that

that some respondents may not recall specific interactions with their parents, especially if those

interactions occurred one or more decades ago. The researcher felt some assurances knowing that

Buri (1991) had little trouble with his sample of university students, but also felt that this could

have been more problematic for the inmate sample. In instances when respondents were unable

to recall exactly when something happened, the researcher asked them to provide their best

estimate. Obviously, this only limited the potential for error, but failed to eliminate it entirely.

76
 
 Human Subject Protections

One sample of respondents in the current study consisted of men and women who were

awaiting trial, or had been found guilty of or pled guilty to criminal activity in one northeastern

state and are currently detained in one of two county jails, as described above. Thus, the

respondents were categorized as “prisoners” by the United States Department of Health and

Human Services (HHS). “Prisoners” are deemed a vulnerable population and are afforded

additional protections relative to their participation in research.

These additional protections are included in the HHS regulations at 45 CFR 46.302 (c),

“Inasmuch as prisoners may be under constraints because of their incarceration which could

affect their ability to make a truly voluntary and uncoerced decision whether or not to participate

as subjects in research, it is the purpose of this subpart to provide additional safeguards.” Taking

these safeguards into consideration, the current study offered no advantages to prisoners who

decide to participate in the project. This included the provision that any inmates who decided to

participate were not given special parole considerations, as outlined in 45 CFR 46.305 (a).

Further compliance warranted that the survey instrument was worded in language that was

understandable to the respondents and that the participants were not exposed to any risks that

would not be accepted by non-inmates.

Going beyond the specific protections guaranteed to this sample of respondents, this

study ensured that each participant was aware of the voluntary nature of this research. Prior to

survey administration, the researcher reminded each respondent that participation was

completely voluntary. Additionally, participants were made aware that they were free to

discontinue participation at any time while completing the survey questionnaire. The researcher

read an informed consent form (Appendix E, as mentioned above) aloud to the sample of jail

77
 
inmates; their consent was assumed once they began completing the form. All participants in the

sample of jail inmates were allowed to keep a copy of the informed consent form.

As discussed above, the study also had a sample of university students. While human

subject protections are a bit different for non-incarcerated individuals, a few important issues

may have arisen. First, the researcher administered surveys to university students during the Fall

2010 semester, thus it was assumed that some students still may have been younger than 18-

years-old. Students younger than 18-years-old were asked to withdraw from participation.

Second, students also were reminded that their participation was voluntary and that they would

not be subjected to negative sanctions by the researcher, their professors, or the University for

exercising their right to not participate. Again, they also were told that they were free to

discontinue participation at any time while completing the survey.

The researcher maintained the complete anonymity of both sample groups of respondents

by asking that they refrain from placing anything that could identify them on the survey; the

identities of university students who entered the raffle will remain confidential. All completed

paper survey questionnaires will be kept in a secured location accessible solely to the researcher

and dissertation chair, while the electronic questionnaires will be accessible only through the

researcher’s Qualtrics account.

Limitations

As is the case with most of the prior research studying Baumrind’s typology, the current

study has a number of limitations. First, this study employed a cross-sectional research design.

Instead of following adolescents and their parents longitudinally and observing parent-child

interactions, this study sought responses from participants at one point in time. This contributes

to the second limitation of the study – respondent recall. Recall may have been problematic in

78
 
instances when the respondent and his/her parent have been separated for long periods of time or

when the parent-child relationship was shortened or weakened.

These two limitations contribute to the third limitation – all survey responses were self-

reported. As Steinberg et al. (1994, p. 768) suggest, “findings can be interpreted only to show

that adolescents’ adjustment is related to the way in which they subjectively experience their

parents” [emphasis in original]. This problem is minor given the fact that the current research

sought to measure the relationship between respondents’ perceptions of parenting styles and

level of criminality. Had the proposed research sought more than respondents’ perceptions the

self-report nature of the study may have been more problematic. Additionally, as Lamborn et al.

(1991, p. 1062-1063) argue, “self-report measures enable investigators to include substantially

larger and more heterogeneous samples…larger samples may permit the detection of

theoretically important findings that may go unnoticed in smaller-scale research.” Thus, despite

the subjectivity associated with self-report measures, the use of them in this study may allow the

researcher to make important discoveries regarding the typology under investigation.

Finally, reverse causality cannot be ruled out completely. Simply, because of the cross-

sectional design the possibility exists that parenting style may not be the catalyst for criminal

involvement, but rather deviance, delinquency, and anti-social behavior during adolescence may

cause parents to implement different types of parenting strategies. Similar to the findings of

Lamborn et al. (1991), the researcher conducting the study is confident that the findings suggest

that the notion of reserve causality is a minor limitation; despite this, the researcher will be

cognizant about reverse causality when interpreting findings.

79
 
 Strengths

As discussed in the literature review, Hoeve et al. (2009) conducted a meta-analysis that

examined 161 manuscripts in an effort determine the existence and magnitude of the relationship

between parenting and delinquent behavior. Of these 161 studies, the authors noted that “very

few” examined parenting styles; a closer examination revealed that only four studies looked at

parenting style and that one of the four studied each of the four styles. Based on the glaring lack

of empirical data exploring the link between parenting style and criminal involvement, it is

argued that one strength of the current research is the contribution that it will make to the

literature.

Another strength of the study is the use of dual sampling groups. Prior literature, as

discussed earlier, focused almost exclusively on Caucasian, male, adolescents from middle-to-

upper class families. Once again, this study surveyed a heterogeneous mix of young men and

women with different ethnic and racial backgrounds from different socio-economic brackets.

Doing so not only allowed the researcher to explore the differences between Caucasian and

minority outcomes based on historically different parenting styles, but it also allowed for a

comparison between those adjudicated as both delinquent and criminal at some point in their

lives, as well as those who have remained non-criminal/non-delinquent throughout the life-

course. This type of analysis represents another contribution to the literature.

Finally, the current research is cross-disciplinary. Baumrind’s typology originated in the

psychological literature over four decades ago, but has not seen much interest in criminology. To

grow criminology as a discipline it is important that criminologists fully utilize empirical

findings from brethren in other social science disciplines. The integration of a psychological

80
 
typology with a criminological outlook, something that it has seen seldom of, should add to the

growth of both.

Analysis Plan

The researcher used a variety of statistical techniques in order to analyze the data. First,

descriptive statistics were computed in order to allow the researcher to gain a general

understanding of the participants, as well as to allow the researcher to compare both sample

groups to each other. Next, the researcher analyzed the internal consistency values for the

parenting scales. Doing so allowed the researcher to verify the reliability of the abbreviated PAQ

measurement instrument that was used in the study. Because the scales that comprised the

abbreviated PAQ attained Cronbach’s alpha values similar to Buri’s (1991) original values and

the scores obtained in the pre-test it suggested that the shortened version of the PAQ is just as

reliable as the full-length measures and, therefore, will allow for shorter survey instruments and

quicker survey administration in future trials. Third, the researcher verified the validity of the

abbreviated PAQ measure by completing an intercorrelation analysis of the parenting scales.

This ensured that the shortened survey instrument measured what it was intended to measure,

much like the original PAQ and the extended version discussed above.

The researcher also employed multiple regression analysis. Multiple regression uses a

linear combination of multiple independent variables in order to assess their impact on a

dependent variable (Mertler & Vannatta, 2005). In the current research, sex, race, and the four

perceived parenting styles were used to predict level of criminal involvement. The multiple

regression equation used in the current study looked like this:

81
 
Ŷ = B0 + B1X1 + B2X2 + … + BkXk + êi

Where:

Ŷ = the predicted value of the dependent variable, level of criminal involvement

B0 = the y-intercept when X = 0; or, the constant

B = the slope of the regression line

X1 = sex

X2 = race

X3 = maternal authoritarian parenting style

X4 = maternal authoritative parenting style

X5 = maternal neglecting/rejecting parenting style

X6 = maternal permissive parenting style

X7 = paternal authoritarian parenting style

X8 = paternal authoritative parenting style

X9 = paternal neglecting/rejecting parenting style

X10 = paternal permissive parenting style

êi = the predicted error term

Mertler and Vannatta (2005) suggest that researchers address the issue of

multicollinearity prior to performing multiple regression. While the easiest method of

determining multicollinearity is to examine a correlation matrix, Mertler and Vannatta suggest

that researchers obtain tolerance statistics or examine the variance inflation factor (VIF) for each

independent variable. Prior to conducting the multiple regression analysis in the study, the

researcher observed the tolerance statistics and the VIF scores for the independent variables.

Meyers, Gamst, and Guarino (2006, p. 182) suggest, “[T]olerance is the amount of a predictor’s

82
 
variance not accounted for by the other predictors (1 – R2 between predictors).” The lower the

tolerance score, the more likely multicollinearity exists between the independent variables

(Mertler & Vannatta, 2005; Meyers et al., 2006). Typically, tolerance scores lower than .1 are

“problematic” and those around .4 may be concerning (Mertler & Vannatta, 2005; Meyers et al.,

2006). VIF scores are obtained by dividing 1 by the tolerance score (Meyers et al., 2006). VIF

scores around 2.5 are similar to tolerance scores of .4 and VIFs of 10 are essentially the same as

tolerance scores of .1 (Meyers et al., 2006). Finally, the researcher employed a multiple

regression analysis in order to determine the effect that parenting styles, sex, and race had on

delinquent and criminal involvement.

83
 
 CHAPTER IV

ANALYSIS AND RESULTS

The intent of this chapter is to discuss the analyses conducted in this study and present

the findings of those analyses. Frequencies and descriptive statistics for some of the variables

will be presented and discussed first. A discussion about the reliability of the extended parenting

scale measure will be presented second. Third, bivariate correlations between independent and

dependent variables will be examined. A review of the results of the multiple regression analysis

will conclude this chapter.

The analysis presented in this chapter is based on a sample of 409 respondents from a

northeastern state. Of the 409 respondents in the sample, 298 were undergraduate college

students at one mid-sized state university during data collection and the remaining 111 were

incarcerated in one of two county jails. The student sample was generated as follows: using

Qualtrics, a web-based survey tool, the researcher sent the survey questionnaire to a random

sample of 2,000 undergraduate students via their university-issued email address. Of those who

received the email solicitation to complete the survey, 412 (20.6%) viewed at least one question.

However, several students exited the survey prior to reporting on key parenting and behavioral

variables (n = 114). Thus, their incomplete responses were removed from the analyses which left

a remaining sample of 298 students; this represents a response rate of 14.9%. The individuals

who comprised the remainder of the sample were incarcerated in one of two county jails in the

same northeastern state during data collection. Fifty-seven (51.4%) respondents in the inmate

sample were housed in a rural county jail and 54 (48.6%) participants were housed in an urban

county jail. In total, the researcher administered 141 paper questionnaires to the inmates; 111

84
 
inmates chose to participate in the study for an inmate response rate of 78.7%. Again, data are

presented from the combined sample of 409 students and inmates.

Frequencies and Descriptive Statistics

Demographic Variables

Table 8 presents the age of the respondents in the sample. As seen, many of the

respondents in the entire sample were under the age of 21 (46.7%), and the majority were 30 and

younger (79.7%). Approximately 6% (n = 25) of the respondents did not provide their age. When

looking at the university students, one notices that the majority of the student sample was 20-

years-old or younger (63.1%). Additionally, 92% of the university students who responded were

30 and younger. Almost 5% (n = 14) of the student sample did not report their age. The age of

respondents in the county jail inmate sample was more evenly distributed. Approximately 47%

(n = 52) of the inmates reported being 30 or younger, while a little more than 43% (n = 48)

reported being over the age of 30; approximately 10% (n = 11) of the jail inmates did not report

their age.

Table 8: Frequencies and Percentages – Age of Sample

Age Valid n Valid %

18-20 191 46.7
21-30 135 33.0
31-40 33 8.1
41-50 16 3.9
51+ 9 2.2
Unknown 25 6.1
Total Sample 409 100

18-20 188 63.1

21-30 86 28.9
31-40 7 2.3
41-50 3 1.0
Unknown 14 4.7
All Students 298 100

85
 
 18-20 3 2.7
21-30 49 44.1
31-40 26 23.4
41-50 13 11.7
51+ 9 8.1
Unknown 11 9.9
All Inmates 111 99.9

Table 9 contains frequencies and percentages of the demographic variables examined in

this study. Raw numbers and valid percentages are included for sex and race. The overall sample

consisted of 244 females (61.9%) and 150 males (38.1%); fifteen student respondents did not

provide information about their sex, thus there is no demographic information reported in Table

9 for these individuals. Instead, demographic sex information is presented for the remaining 394

respondents. As seen in Table 9, the university sample consisted of 208 females (73.5%) and 75

males (26.5%). According to the University Fact Sheet (Fall 2010), the undergraduate university

population is approximately 57% female and 43% male, thus female students are

overrepresented in the sample. Perhaps this is due to young women being more cooperative or

interested in the subject matter; many female respondents contacted the researcher directly to ask

about the project and their desire to discuss the topic of parenting, while no male participants

were in contact.

Table 9: Frequencies and Percentages – Sex of Sample

Variable Valid n Valid %

Male 150 38.1
Female 244 61.9
Total Sample 394 100

Male 75 26.5
Female 208 73.5
All Students 283 100

Male 75 67.6
Female 36 32.4

86
 
 All Inmates 111 100

Male 42 73.7
Female 15 26.3
CJ1 Inmates 57 100

Male 33 61.1
Female 21 38.9
CJ2 Inmates 54 100

The inmate sample consisted of 36 females (32.4%) and 75 males (67.6%). Again, this

represents an overrepresentation of female inmates when compared to national averages, which

are approximately 12.3% (Minton, 2011); additionally, the number of female respondents from

the rural county jail (n = 15; 26.3%) represents an increase over the typical daily average of 13%

female inmates (L. Simmons, personal communication, December 1, 2010). It is believed that the

percentage of female inmate respondents is high because of survey response rates in the jails.

Simply, female inmates were more willing to respond and engage with the researcher in

conversation about parenting. Additionally, female respondents in the rural county jail were

allowed to leave their housing unit and enter a multi-purpose classroom in order to complete the

survey, whereas male inmates completed the survey in a room attached to their housing units. A

number of women also indicated that had they stayed in the housing unit they would be on

lockdown in their cells, so filling out the questionnaire gave them a brief respite from

punishment.

Respondents were asked to self-report their race. As presented in Table 10, the overall

sample was 82.1% white (n = 321), 9.2% black (n = 36), and 8.6% of the sample (n = 34) self-

reported as Asian/Pacific Islander (n = 7), Hispanic (n = 6), Native American (n = 4), or “other”

(n = 17). The majority of those self-reporting as “other” indicated that they were “bi-racial” and

had one parent who was black and another who was white (n = 12); fourteen student respondents

87
 
and four inmate respondents did not provide information about their race, thus there is no

demographic information reported in Table 10 for these 18 individuals. Instead, demographic

race information is presented for the remaining 391 respondents.

Table 10: Frequencies and Percentages – Race of Sample

Variable Valid n Valid %

African American/Black 36 9.2
Asian/Pacific Islander 7 1.8
Caucasian/White 321 82.1
Hispanic 6 1.5
Native American 4 1.0
Other 17 4.3
Total Sample 391 100

African American/Black 14 4.9

Asian/Pacific Islander 7 2.5
Caucasian/White 252 88.7
Hispanic 4 1.4
Native American 1 0.4
Other 6 2.1
All Students 284 100

African American/Black 22 20.6

Caucasian/White 69 64.5
Hispanic 2 1.9
Native American 3 2.7
Other 11 10.3
All Inmates 107 100

African American/Black 4 7.0

Caucasian/White 45 78.9
Hispanic 2 3.5
Native American 2 3.5
Other 4 7.0
CJ1 Inmates 57 100

African American/Black 18 36.0

Caucasian/White 24 48.0
Native American 1 2.0
Other 7 14.0
CJ2 Inmates 50 100

88
 
 The participants in the university sample were predominantly white (n = 252; 88.7%),

4.9% black (n = 14), and 6.4% were Asian/Pacific Islander, Hispanic, Native American, or

“other” (n = 18). According to the University Fact Sheet (Fall 2010), the undergraduate

population at the time the survey was administered was 87% white and 13% minority, thus the

undergraduate sample is representative of the university population in relation to race.

Additionally, the racial diversity within the inmate sample varied between the rural county jail

and the urban county jail with white respondents comprising 78.9% (n = 45) of the rural sample

and 48% (n = 24) of the urban sample, while black respondents made up only 7% (n = 4) of the

rural sample and 36% (n = 18) of the urban sample. Based on the fact that over 80% of the total

sample is white and no other race accounts for much more than 9% of the total, the race variable

will be dichotomized into “non-white” and “white” categories for the analyses.

Caretaker Variables

The main purpose of the current study is to examine the effects that perceived parenting

styles have on behavioral outcomes, such as deviance, delinquency, and criminality. Before

explaining these effects, one first must examine who respondents identified as their primary

caretakers while growing up. To do so, participants were asked to indicate the woman and the

man who “raised [them] the most while [they were] growing up.” Response categories for the

female caretakers included: biological, adopted, foster, or step mother, as well as grandmother,

sister, aunt, female cousin, other, and none for those who were not raised by a woman. Male

caretaker response categories included: biological, adopted, foster, or step father, in addition to

grandfather, brother, uncle, male cousin, other, and none. The following sections provide

descriptive statistics regarding who the study sample reported living with while they were

growing up.

89
 
 Maternal caretaker. As stated, the researcher was interested in who the participants

identified as their main maternal and paternal caretaker, thus respondents were asked to identify

which male and female “raised [them] the most” while they were growing up; maternal results

are presented in Table 11. With regards to the maternal caretaker, 88.5% (n = 362) of all

respondents indicated that their biological mother raised them, 3.4% (n = 14) noted that their

grandmother raised them, and 1.5% (n = 6) of respondents indicated that they had no female

caretaker while growing up. The remaining 6.6% was scattered amongst adoptive, foster, and

step mothers, as well as sisters and aunts.

Table 11: Frequencies and Percentages for Maternal Caretakers

Variable Valid n Valid %

Biological Mother 362 88.5
Adopted Mother 7 1.7
Grandmother 14 3.4
Aunt 7 1.7
No Maternal Caretaker 6 1.5
Other 13 3.2
Total Sample 409 100

Biological Mother 278 93.3

Adopted Mother 4 1.3
Grandmother 7 2.3
Aunt 3 1.0
No Maternal Caretaker 1 .3
Other 5 1.7
All Students 298 100

Biological Mother 84 75.7

Adopted Mother 3 2.7
Grandmother 7 6.3
Aunt 4 3.6
No Maternal Caretaker 5 4.5
Other 8 7.2
All Inmates 111 100

Table 12 shows that the university student sample and the county inmate sample differed

significantly when reporting maternal caretaker, t(407) = -5.96, p ≤ .001 [university students (M

= 1.24, SD = 1.04) and inmates (M = 2.35, SD = 2.73)]. Biological mother (and biological father)

90
 
were each coded as “1” and each of the other options was coded between “2” and “10,” thus a

mean caretaker score that is closer to “1” indicates that the sample group under investigation has

responded to the survey with a biological parent in mind.

Table 12: Independent Samples T-Test: Caretaker Variables – Between Sample, Race, and Sex

Sample Race Sex

Maternal Caretaker -5.96** 5.65** -2.45*
Paternal Caretaker -3.14* 4.45** -.54
Resided With -1.39 1.38 -1.11
**p ≤ .01, *p ≤ .05

Most university students reported that their main maternal caretaker while growing up

was their biological mother (93.3%; n = 278), whereas just over three-fourths of the inmate

sample named their biological mother as their main maternal caretaker (75.7%; n = 84). A higher

percentage of county jail inmates claimed to have been raised by their grandmother (6.3%; n =

7), an aunt (3.6%; n = 4), or an adopted mother (2.7%; n =3) than the university students who

reported the following: grandmother (2.3%; n = 7); aunt (1.0%; n = 3); and adopted mother

(1.3%; n = 4). Only one university student (0.3%) claimed to have no maternal caretaker, while

4.5% (n = 5) of the inmate sample claimed no maternal caretaker. The individual that

respondents identified as their maternal caretaker also differed significantly based on race (non-

whites vs. whites), t(389) = 5.65, p ≤ .001 [non-whites (M = 2.59, SD = 2.83) and whites (M =

1.32, SD = 1.33)] , and based on sex, t(392) = -2.45, p = .015 [females (M = 1.38, SD = 1.45) and

males (M = 1.82, SD = 2.14)]. Simply, these results indicate that white respondents were

significantly more likely to have been raised by their biological mother than their non-white

counterparts. Additionally, female respondents were significantly more likely to have been raised

by their biological mother compared to male respondents.

91
 
 Paternal caretaker. Results for the identified paternal caretakers are presented in Table

13. Approximately three-fourths of the entire sample (73.3%; n = 297) reported that they were

raised by their biological father. Twenty-six respondents (6.4%) reported being raised by their

stepfather, 3% (n = 12) named an adoptive father as their main paternal caretaker, and more than

10% of the sample (n = 43) reported that they had no male caretaker while growing up. The

remaining 6.7% of responses was scattered amongst foster fathers, grandfathers, brothers, and

uncles; missing data from four respondents is not included in this analysis.

Table 13: Frequencies and Percentages for Paternal Caretakers

Variable Valid n Valid %

Biological Father 297 73.3
Adopted Father 12 3.0
Stepfather 26 6.4
Grandfather 7 1.7
Uncle 7 1.7
No Paternal Caretaker 43 10.6
Other 13 3.2
Total Sample 405 100

Biological Father 234 79.1

Adopted Father 8 2.7
Stepfather 11 3.7
Grandfather 5 1.7
Uncle 4 1.4
No Paternal Caretaker 29 9.8
Other 5 1.6
All Students 296 100

Biological Father 63 57.8

Adopted Father 4 3.7
Stepfather 15 13.8
Grandfather 2 1.8
Uncle 3 2.8
No Paternal Caretaker 14 12.8
Other 8 7.4
All Inmates 109 100

Similar to differences between the university sample and the inmate sample regarding

maternal caretakers, there also were significant differences between the samples in relation to

paternal caretakers, t(403) = -3.14, p = .002 [university students (M = 2.25, SD = 2.83) and
92
 
inmates (M = 3.29, SD = 3.30)] . Nearly 80% (n = 234) of the undergraduate students reported

living with their biological father while growing up. Comparably, less than 58% (n = 63) of the

inmate sample reported living with their biological father. Many inmates reported living with a

stepfather (13.8%; n = 15), while only 3.7% (n = 11) of students considered a stepfather as their

main paternal caretaker. Additionally, 12.8% (n = 14) of the inmate sample reported having no

male caretaker while growing up. Likewise, 9.8% (n = 29) of the university sample grew up

without a male present in the home. The individual that respondents identified as their paternal

caretaker also differed significantly based on race (non-whites vs. whites), t(387) = 4.45, p ≤

.001 [non-whites (M = 3.86, SD = 3.50) and whites (M = 2.18, SD = 2.70)]. Simply, these results

suggest that the university students and county jail inmates reported significantly different

paternal caretakers; university students were much more likely to report growing up with a

biological father, as discussed above, compared to county jail inmates – this is reflected in the t-

test results. There were no significant statistical differences between males and females in the

reporting of paternal caretakers.

In addition to asking respondents to identify their main maternal and paternal caretakers,

the researcher also asked respondents to identify with whom they lived with the most while

growing up. Participants were presented with six combinations including: “Two parents; both

lived there and were involved,” “My mother, but my father was also involved,” “My father, but

my mother was also involved,” “My mother” only, “My father” only, and “Neither…were

involved.” Additionally, the respondents were given the option to self-report any other

combination that the researcher overlooked; many of the respondents who indicated “other”

noted that both caretakers lived together, but that one or both was not involved in the parenting,

93
 
while additional “other” responses suggested that caretakers were involved in a committed

relationship, but one was usually overseas serving in the military.

Table 14 shows that 66.1% (n = 267) of the total sample lived with both caretakers while

growing up. Table 14 also suggests that maternal caretakers had a greater role than paternal

caretakers in raising the respondents. For example, 11.4% (n = 46) reported that they lived with

their maternal guardian although the paternal guardian also was involved, and an additional

11.4% (n = 46) reported that their maternal caretaker was their sole support while growing up.

Paternal caretakers took the lead role in only 2.7% of cases (n = 9). Sadly, three respondents

(0.7%) reported that they lived with neither parent while growing up.

Table 14: Frequencies and Percentages – Who Respondents Lived With

Variable Valid n Valid %

Both Parents 267 66.1
Mother; Father Involved 46 11.4
Father; Mother Involved 6 1.5
Mother Only 46 11.4
Father Only 5 1.2
Neither Parent 3 .7
Other 31 7.7
Total Sample 404 100

Both Parents 204 69.2

Mother; Father Involved 31 10.5
Father; Mother Involved 3 1.0
Mother Only 30 10.2
Father Only 3 1.0
Neither Parent 1 .3
Other 23 7.8
All Students 295 100

Both Parents 63 57.8

Mother; Father Involved 15 13.8
Father; Mother Involved 3 2.8
Mother Only 16 14.7
Father Only 2 1.8
Neither Parent 2 1.8
Other 8 7.3

94
 
 All Inmates 109 100

When comparing the university student sample to the inmate sample, results are

strikingly similar; significant differences do not exist between the university sample and the

inmate sample. The majority of participants in both samples report living with both caretakers

(students: 69.2%, n = 204; inmates: 57.8%, n = 63). Additionally, the maternal caretaker

assumed a lead role in terms of parenting in both samples (students: 20.7%, n = 61; inmates:

28.5%, n = 31). Conversely, paternal caretakers were less often the primary caretakers in both

samples (students: 2%, n = 6; inmates: 4.6%, n = 5). Only one student (0.3%) reported living

with neither caretaker and two inmates (1.8%) also claimed that they did not live with either

caretaker while growing up. Finally, the caretaker groupings that respondents identified as

growing up with did not differ significantly based on race (non-whites vs. whites) or based on

sex.

Parenting Scales

As discussed in the second chapter, the current study utilized Buri’s (1991) Parental

Authority Questionnaire (PAQ) to assess the parenting styles of the respondents’ caretakers. The

original PAQ contains 60 total items – separated into 30 per parent – that are used to assess

Authoritarian, Authoritative, and Permissive parenting. The researcher conducting the current

study created a fourth scale – Neglecting/Rejecting – to measure the fourth style of parenting. As

presented above in Chapter Three, a preliminary analysis of the new scale revealed that it is

reliable and valid across tests for both maternal caretaker and paternal caretaker. The following

section will present additional analyses on the reliability of the Neglecting/Rejecting scale.

Test of Internal Consistency

95
 
 Cronbach alpha scores for Buri’s initial test and the pre-test conducted prior to the current

study are discussed at length in Chapter Two. To reiterate, Buri’s scales have been classified as

“respectable” and “very good” according to DeVellis’ (2003) criteria for scale assessment.

Additionally, according to the pre-test results, the Neglecting/Rejecting scale also is

“respectable” and “very good” for assessing mothers and fathers, respectively. Current analyses

suggest that shortened versions of each scale also are highly reliable measures of Baumrind’s

parenting typologies (Table 15). Using a split-half reliability test, as discussed in an earlier

chapter, the researcher shortened each parenting scale from ten items to five items. By

eliminating 40 items from the questionnaire, the researcher intended to increase the overall

response rate. Alpha values for the shortened scale are highlighted in the column labeled

“Current Study” in Table 15 below.

Table 15: Internal Consistency of PAQ-Related Measures

Buri (1991) α Spraitz Extension Current Study α

(Pre-Test) α
Mother’s Authoritarianism .85 .85 .74
Mother’s Authoritativeness .82 .88 .83
Mother’s Permissiveness .75 .80 .72
Mother’s Neglect/Rejection - .86 .80

Father’s Authoritarianism .87 .88 .76

Father’s Authoritativeness .85 .91 .88
Father’s Permissiveness .74 .79 .77
Father’s Neglect/Rejection - .90 .80

It appears that shortening the parenting measures did not harm the internal consistency of the

scales. Despite these changes, the shortened version of the PAQ remains a “respectable,” “very

good,” and highly reliable measure of Baumrind’s typology of parenting styles.

Perceptions of Parenting Styles

96
 
 As discussed, eight five-item scales were used to assess the perceptions that respondents

had of their caretakers’ parenting styles. Likert-type response categories were used for each item

and included: “Strongly Agree,” “Agree,” “Neither Agree Nor Disagree,” “Disagree,” and

“Strongly Disagree.” Responses were coded on an additive scale with a range of 1 to 5;

“Strongly Agree” responses were scored as a “1” whereas scores of “5” were applied to

responses of “Strongly Disagree.” Thus, the range of scores for each of the scales could be 5 –

25; however, results suggest that no single respondent reported “Strongly Agree” for all items in

the Maternal Permissive, Maternal Neglecting/Rejecting, and Paternal Permissive scales.

Furthermore, no single respondent reported “Strongly Disagree” for all items in the Maternal

Authoritarian scale (Table 16). The reported sample size for each parenting style fluctuates based

on a number of things, including the participants’ willingness to self-report on all items and the

number of respondents who did not have a maternal or paternal caretaker; the sample size for the

paternal parenting styles is much lower because 43 respondents reported not having a male

caretaker while they were growing up.

Table 16: Descriptive Statistics for Measures of Parenting Type

N Mean SD Minimum Maximum

Maternal 395 13.18 3.76 5 24
Authoritarianism

Maternal 387 11.58 4.17 5 25

Authoritativeness

Maternal 392 15.96 3.71 6 25

Permissiveness

Maternal 389 20.01 4.26 6 25

Neglect/Rejection

Paternal 353 13.56 4.07 5 25

Authoritarianism

97
 
Paternal 348 12.98 4.61 5 25
Authoritativeness

Paternal 350 16.35 3.99 6 25

Permissiveness

Paternal 349 18.66 4.49 5 25

Neglect/Rejection

As indicated in Table 16 above, the mean score for Maternal Authoritativeness is 11.58

and represents the lowest mean score amongst all maternal scales. This suggests that respondents

were most likely to “Strongly Agree” with scale items that described an authoritative mother.

Similarly, the mean score for Paternal Authoritativeness is 12.98, which represents the smallest

mean score amongst all paternal scales. Again, this suggests that respondents were most likely to

“Strongly Agree” with scale items that described an authoritative father. Additionally,

respondents were most likely to “Disagree” or “Strongly Disagree” with statements that

suggested their caretakers were Neglecting/Rejecting.

The researcher conducted independent sample t-tests (Table 17) in order to compare the

mean parenting scores based on sample (university students [coded = 0] vs. inmates [coded = 1]),

race (non-white [coded = 0] vs. white [coded = 1]), and sex (female [coded = 0] vs. male [coded

= 1]). First, there were statistically significant differences between sample groups for all

parenting types. Simply, across all parenting types, university students reported different

perceptions of their caretakers than did county jail inmates. Based solely on mean scores,

university students were more likely than inmates to report having authoritative mothers and

fathers as well as permissive mothers and fathers, whereas inmates were more likely to report

having authoritarian and neglecting/rejecting mothers and fathers. These results are consistent

98
 
with research that suggests individuals with authoritative parents are less likely to run afoul of

the law than individuals with neglecting/rejecting parents and authoritarian parents.

Table 17: Independent Samples T-Test – Between Sample, Race, Sex

Sample Race Sex

M. Authoritarian 2.27* -3.27** .78
M. Authoritative -3.15* 1.41 .20
M. Permissive -4.05** 2.38* -.78
M. Neglect/Reject 5.33** -1.59 1.46

P. Authoritarian 1.86† -.49 .37

P. Authoritative -3.06* 1.59 .00
P. Permissive -3.67** -.05 -.82
P. Neglect/Reject 2.22* -.51 .06
**p ≤ .01, *p ≤ .05, †p ≤ .10

When respondents were compared based on race (non-white vs. white), there were only

two significant differences. Non-white respondents were significantly more likely that white

respondents to categorize their maternal caretakers as authoritarian, t(376) = -3.27, p = .001

[non-whites (M = 11.78, SD = 3.21) and whites (M = 13.43, SD = 3.75)]. Meanwhile, white

respondents were significantly more likely than non-white respondents to characterize their

maternal caretakers as permissive, t(374) = 2.38, p = .018 [whites (M = 15.82, SD = 3.70) and

non-whites (M = 17.02, SD = 3.60)]. While mean differences between races were reported for the

other six parenting types none approached levels of significance. Further analysis revealed no

significant differences between white and non-white university students in terms of their

perceptions of their caretakers’ parenting styles.

Additionally, when the entire sample was compared based on the sex of the respondents

there were no significant differences between reported parenting types. Further analyses showed

no significant differences in reported parenting types between male and female inmates and

weak statistical differences in maternal permissiveness [t(276) = 1.72, p = .086] and paternal

99
 
permissiveness [t(249) = 1.65, p =.10] between males and females from the university sample.

The lack of significant findings for between-sex differences suggests that males and females may

have similar perceptions of the way that they are raised.

Dependent Variable Scales

As indicated in the previous chapter, various deviant, delinquent, and criminal behavior

items were taken from Elliott and Ageton (1980) and Thornberry and Farnworth (1982). In the

current study, the dependent variables are organized into six scales (Table 18); the scales are

organized as such: predatory crimes against persons (8 items, including “purposely killed

somebody,” “hit somebody,” and “attacked another person with the intent to injure or kill

them”), predatory crimes against property (9 items, including “purposely destroyed property,”

“stole something valued between $10 and $50,” and “motor vehicle theft”), illegal service crimes

(6 items, including “received money for sex,” “paid money for sex,” and “sold marijuana”),

public disorder crimes (8 items, including “set off a fire alarm,” “drunk in public,” and “made

obscene phone calls”), deviant behaviors (6 items, including “cheated in school,” “lied to

authority figures,” and “skipped class”), and alcohol/drug use (10 items, including substances

such as tobacco, alcohol, marijuana, cocaine, heroin, and methamphetamine). Additionally, the

alcohol/drug use scale can be broken down into two subscales: soft alcohol/drug use (3 items)

and hard drug use (7 items). Elliott and Ageton (1980) arranged the items into six scales, thus

this also was done with the 48 dependent variable items in the current study. The reliability of

the dependent variable scales and descriptive statistics are discussed in the next section.

Table 18: Dependent Variable Scales and Subscales

Predatory - Property Predatory - Person Public Disorder Alcohol/Drug Use

Purposely destroyed Hurt – sought medical Set off fire alarm Tobacco < 18
property attention
Motor vehicle theft Used a weapon Carried gun; no permit Alcohol < 21
Stole something >$50 Killed somebody Thrown objects Marijuana

100
 
Stole something <$10 Hurt – minor way Carried hidden weapon Hallucinogens
Stole something Attacked to injure or kill Disorderly in public Amphetamines
between $10 - $50
Stole money Hit somebody Drunk in public Barbiturates
Held stolen goods Sexual assault Obscene phone calls Heroin
Kept extra change Strong-armed somebody Verbal altercations Cocaine
Trespassing Methamphetamines
Hard Drug Use (sub) Prescription Pain Pills
Illegal Services Deviant Behaviors Hallucinogens
Gave money for sex Ran away < 18 Amphetamines Alc./Soft Drugs (sub)
Get money for sex Lied about age Barbiturates Tobacco < 18
Sold marijuana Cheated in school Heroin Alcohol < 21
Buy alcohol for minor Lied to authority Cocaine Marijuana
Avoid payment Skipped class Methamphetamines
Sold other drugs Suspended from school Prescription Pain Pills

Originally, the alcohol/drug use scale contained 11 items. The additional item asked

respondents to report on their use of alcohol while of legal drinking age; this item was also

included as a fourth item in the alcohol/soft drug use subscale. For a number of reasons, this item

was removed from the analysis. First, approximately 47% (n = 191) of the total sample, and 63%

of the student sample, reported being younger than the current legal drinking age, thus they were

not able to answer this question. Second, when examining all of the behaviors in the alcohol/drug

use scale (and additional subscales) it is noted that all remaining items are unlawful behaviors;

consuming alcohol while of legal drinking age is a lawful behavior, thus its inclusion in the scale

did not make sense conceptually. Based on these two reasons, this item has been removed from

all further analyses in this study.

Internal Consistency of Dependent Variable Scales

Cronbach’s alpha scores for each of the six dependent variable scales and the two

subscales were generated (Table 19). The 8-item predatory crimes against persons scale has an

alpha score of .753, which is considered respectable (DeVellis, 2003). The 6-item illegal services

scale (α = .750) and the 8-item public disorder scale (α = .762) also are respectable. Cronbach’s

alpha scores for the 9-item predatory crimes against property scale (α = .867) and the 10-item

101
 
alcohol/drug use scale (α = .897) are considered very good. The 6-item deviant behavior scale (α

= .656) reports an alpha score that DeVellis considers “minimally acceptable.” Altogether, the

six scales comprised of 48 items represent an acceptable and reliable measure of a variety of

behaviors. Additionally, the two subscales reported differing levels of acceptability. The 3-item

alcohol/soft drug use scale had a respectable alpha score (α = .748), while the 7-item hard drug

use scale had a very good alpha score (α = .912).

Table 19: Internal Consistency of Dependent Variable Scales

Current Study α
Predatory Crimes – Person .753
Predatory Crimes – Property .867
Illegal Services Crimes .750
Public Disorder Offenses .762
Deviant Behaviors .656
Alcohol/Drug Use – All .897
Alcohol/Soft Drug Use .748
Hard Drug Use .912

Frequencies of Dependent Variables

Total sample. Respondents were asked to self-report if they had ever engaged in each of

the 48 deviant, delinquent, or criminal activities listed. Responses were coded “0” for if they had

not and “1” if they had engaged in the behavior. Table 20 displays the frequency with which

respondents reported taking part in each category of deviance and criminality. It should be noted

that there were respondents who did not respond to every question within a dependent variable

scale; non-response was random and not systemic in nature, thus participants were not removed

from the analysis. In most instances, respondents were likely to engage in only a few acts within

each category with the exception of the “General Deviance” scale in which 35.8% of respondents

reported taking part in 4 – 6 of the activities. With regards to the other five scales, over 50% of

respondents reported engaging in as few as two or less of the deviant and criminal behaviors.

102
 
Table 20: Frequencies and Percentages for DV Scales – Full Sample

Predatory Predatory Illegal Public Deviance Alcohol/Drug

Person Property Services Disorder (n = 392) (n = 386)
(n = 388) (n = 388) (n = 386) (n = 388)
0 items 125 (32.2%) 142 (36.6%) 207 (53.6%) 64 (16.5%) 28 (7.1%) 81 (21.0%)
1 item 108 (27.8%) 68 (17.5%) 73 (18.9%) 85 (21.9%) 59 (15.1%) 87 (22.5%)
2 items 65 (16.8%) 45 (11.6%) 39 (10.1%) 74 (19.1%) 64 (16.3%) 53 (13.7%)
3 items 33 (8.5%) 19 (4.9%) 21 (5.4%) 66 (17.0%) 101 (25.8%) 46 (11.9%)
4 items 25 (6.4%) 28 (7.2%) 35 (9.1%) 36 (9.3%) 63 (16.1%) 27 (7.0%)
5 items 15 (3.9%) 23 (5.9%) 7 (1.8%) 21 (5.4%) 56 (14.3%) 20 (5.2%)
6 items 15 (3.9%) 21 (5.4%) 4 (1.0%) 18 (4.6%) 21 (5.4%) 11 (2.8%)
7 items 1 (0.3%) 14 (3.6%) - 18 (4.6%) - 16 (4.1%)
8 items 1 (0.3%) 13 (3.4%) - 6 (1.5%) - 10 (2.6%)
9 items - 15 (3.9%) - - - 15 (3.9%)
10 items - - - - - 20 (5.2%)
Total 388 (100%) 388 (100%) 386 (100%) 388 (100%) 392 (100%) 386 (100%)

When looking at the two subscales for alcohol and drug use, one sees that a slight

majority of respondents have engaged in 2 – 3 behaviors listed under “Alcohol/Soft Drug Use”

while approximately 75% of respondents have partaken in one or fewer forms of hard drugs

(Table 21).

Table 21: Frequencies and Percentages for DV Subscales – Full Sample

Soft Alcohol/Drug Use Hard Drug Use

(n = 388) (n = 386)
0 items 84 (21.6%) 250 (64.8%)
1 item 94 (24.2%) 39 (10.1%)
2 items 68 (17.5%) 19 (4.9%)
3 items 142 (34.7%) 13 (3.4%)
4 items - 19 (4.9%)
5 items - 10 (2.6%)
6 items - 16 (4.1%)
7 items - 20 (5.2%)
Total 388 (100%) 386 (100%)

A more in-depth look at alcohol and drug usage is presented in Table 22. When looking

at overall substance use, a majority of respondents report not using hard drugs whereas alcohol

and marijuana use is high. Over 50% of the sample reported using marijuana, while nearly three-

103
 
fourths of the sample reported using alcohol while younger than the legal drinking age. Of the

three items used in the “Alcohol/Soft Drugs” subscale, only tobacco was used by less than half

of the respondents (45.1%).

Table 22: Frequencies and Percentages for Drug/Alcohol Use – Full Sample

No Yes Total
Tobacco 213 (54.9%) 175 (45.1%) 388 (100%)
Alcohol – Underage 102 (26.3%) 286 (73.7%) 388 (100%)
Marijuana 193 (49.7%) 195 (50.3%) 388 (100%)
Hallucinogens 313 (80.9%) 74 (19.1%) 387 (100%)
Amphetamines 326 (84.5%) 60 (15.5%) 386 (100%)
Barbiturates 337 (87.1%) 50 (12.9%) 387 (100%)
Heroin 329 (85.0%) 58 (15.0%) 387 (100%)
Cocaine 291 (75.2%) 96 (24.8%) 387 (100%)
Methamphetamines 357 (92.2%) 30 (7.8%) 387 (100%)
Pain Killers – No Px 277 (71.6%) 110 (28.4%) 387 (100%)

University students. When looking at the frequency of deviant, delinquent, and criminal

behaviors exhibited by the university student sample one sees similar patterns of behavior in

comparison to the entire sample. Simply, for most dependent variable scales, the majority of

respondents report engaging in 0 – 2 deviant and criminal behaviors per scale (Table 23).

However, there are slight differences. When looking at the “Predatory Crimes Against People”

scale, one sees that no respondents reported taking part in 6 – 8 of the behaviors listed; 4.5% of

the total sample claimed to take part in 6 – 8 of those behaviors. Similarly, when viewing the

“Illegal Services” scale, one sees that no respondents engaged in any more than three of those

behaviors, whereas 11.9% of the total sample reported engaging in 4 – 6 of the listed illegal

service behaviors.

Table 23: Frequencies and Percentages for DV Scales – University Students

Predatory Predatory Illegal Public Deviance Alcohol/Drug

Person Property Services Disorder (n = 287) (n = 282)
(n = 284) (n = 283) (n = 280) (n = 283)
0 items 119 (41.9%) 138 (48.8%) 194 (69.3%) 62 (48.8%) 27 (9.4%) 79 (28.0%)

104
 
1 item 97 (34.2%) 62 (21.9%) 58 (20.7%) 79 (27.9%) 54 (18.8%) 84 (29.8%)
2 items 47 (16.5%) 32 (11.3%) 21 (7.5%) 68 (24.0%) 52 (18.1%) 46 (16.3%)
3 items 13 (4.6%) 14 (4.9%) 7 (2.5%) 51 (18.0%) 90 (31.4%) 36 (12.8%)
4 items 4 (1.4%) 19 (6.7%) 0 (0%) 17 (6.0%) 47 (16.4%) 15 (5.3%)
5 items 4 (1.4%) 11 (3.9%) 0 (0%) 2 (0.7%) 14 (4.9%) 8 (2.8%)
6 items 0 (0%) 6 (2.1%) 0 (0%) 1 (0.4%) 3 (1.0%) 4 (1.4%)
7 items 0 (0%) 0 (0%) - 3 (1.1%) - 5 (1.8%)
8 items 0 (0%) 1 (0.4%) - 0 (0%) - 4 (1.4%)
9 items - 0 (0%) - - - 1 (0.4%)
10 items - - - - - 0 (0.0%)
Total 284 (100%) 283 (100%) 280 (100%) 283 (100%) 287 (100%) 282 (100%)

The same general pattern held for drug and alcohol use as well (Table 24).

Approximately 71% of the student sample reported drinking alcohol or smoking tobacco or

marijuana, but almost 83% reported never using a hard drug; 91.1% reported using one hard drug

or less.

Table 24: Frequencies and Percentages for DV Subscales – University Students

Soft Alcohol/Drug Use Hard Drug Use

(n = 284) (n = 282)
0 items 82 (28.9%) 233 (82.6%)
1 item 87 (30.6%) 24 (8.5%)
2 items 53 (18.7%) 10 (3.5%)
3 items 62 (21.8%) 4 (1.4%)
4 items - 5 (1.8%)
5 items - 5 (1.8%)
6 items - 1 (0.4%)
7 items - 0 (0%)
Total 284 (100%) 282 (100%)

County jail inmates. When looking at the frequency of the self-reported behaviors of the

county jail inmate sample, one sees different frequency of offending when compared to the

university student sample (Table 25). Not surprisingly, members of the county jail inmate sample

reported a higher frequency of engaging in the activities being studied. For example, only 5.8%

of inmates claimed to have taken part in no predatory crimes against people and even fewer

(3.8%) admitted to engaging in zero predatory crimes against property. Thus, approximately

105
 
94% of inmates committed at least one crime against a person and 96% of inmates committed at

least one property crime. Additionally, while nearly 70% of students have not taken part in any

illegal services crimes, almost 88% of participants in the inmate sample reported engaging in at

least one type of illegal service crime. And unlike the university student sample, at least one

county jail participant reported taking part in each deviant, delinquent, or criminal activity within

each scale.

Table 25: Frequencies and Percentages for DV Scales – Inmates

Predatory Predatory Illegal Public Deviance Alcohol/Drug

Person Property Services Disorder (n = 105) (n = 104)
(n = 104) (n = 105) (n = 106) (n = 105)
0 items 6 (5.8%) 4 (3.8%) 13 (12.3%) 2 (1.9%) 1 (1.0%) 2 (1.9%)
1 item 11 (10.6%) 6 (5.7%) 15 (14.2%) 6 (5.7%) 5 (4.8%) 3 (2.9%)
2 items 18 (17.3%) 13 (12.4%) 18 (17.0%) 6 (5.7%) 12 (11.4%) 7 (6.7%)
3 items 20 (19.2%) 5 (4.8%) 14 (13.2%) 15 (14.3%) 11 (10.5%) 10 (9.6%)
4 items 21 (20.2%) 9 (8.6%) 35 (33.0%) 19 (18.1%) 16 (15.2%) 12 (11.5%)
5 items 11 (10.6%) 12 (11.4%) 7 (6.6%) 19 (18.1%) 42 (40.0%) 12 (11.5%)
6 items 15 (14.4%) 15 (14.3%) 4 (3.8%) 17 (16.2%) 18 (17.1%) 7 (6.7%)
7 items 1 (1.0%) 14 (13.3%) - 15 (14.3%) - 11 (10.6%)
8 items 1 (1.0%) 12 (11.4%) - 6 (5.7%) - 6 (5.8%)
9 items - 15 (14.3%) - - - 14 (13.5%)
10 items - - - - - 20 (19.2%)
Total 104 (100%) 105 (100%) 106 (100%) 105 (100%) 105 (100%) 104 (100%)

When examining specific drug and alcohol use tendencies amongst county jail inmates,

the same general pattern emerges, especially in the alcohol/soft drug use subscale (Table 26).

Simply, very few inmates (1.9%) reported never smoking a cigarette, marijuana, or drinking

alcohol. The frequency and percentage of hard drug use shows no discernible pattern with 14.4%

of participants reporting use of only one form of hard drug and nearly 20% reporting use of all

seven forms of hard drugs, while only 4.8% report using five types and 8.7% report using two or

three types.

106
 
Table 26: Frequencies and Percentages for DV Subscales – Inmates

Alcohol/Soft Drug Use Hard Drug Use

(n = 104) (n = 104)
0 items 2 (1.9%) 17 (16.3%)
1 item 7 (6.7%) 15 (14.4%)
2 items 15 (14.4%) 9 (8.7%)
3 items 80 (76.9%) 9 (8.7%)
4 items - 14 (13.5%)
5 items - 5 (4.8%)
6 items - 15 (14.4%)
7 items - 20 (19.2%)
Total 104 (100%) 104 (100%)

An independent samples t-test revealed that there are statistically significant differences

in the deviant, delinquent, and criminal behaviors reported by university students and the county

jail inmates for all six dependent variable scales and both subscales (Table 27). Students reported

committing significantly fewer deviant, delinquent, and criminal acts, which is not a surprising

result given the fact that the more criminally inclined sample group was surveyed while they

were incarcerated.

Table 27: Independent Samples T-Test: DV Scales – Sample Groups

Predatory Person -16.279**

Predatory Property -18.336**
Illegal Services -19.019**
Public Disorder -16.906**
Deviance -10.983**
Alcohol/Drug Use -18.116**

Alcohol/Soft Drug Use -11.384**

Hard Drug Use -17.412**
**p < .001

White respondents. When looking solely at white respondents, self-reported patterns of

offending are comparable to the overall sample (Table 28). The percentage of respondents who

engaged in person-based, property-based, and illegal service crimes decreased as the number of

107
 
behaviors increased, which mirrors the total sample. Additionally, the frequency of white

participants who engaged in public disorder behaviors is similar to the overall sample; the

highest frequency of respondents reported taking part in one public disorder item and then the

frequency percentages decreased. Regarding general deviance of white respondents, the

frequency of self-reported behavior gradually peaked at three items and then decreased; this

same pattern is seen when analyzing the entire sample. This is not surprising given the fact that

white respondents make up approximately 80% of the sample that is being analyzed.

Table 28: Frequencies and Percentages for DV Scales – White Respondents

Predatory Predatory Illegal Public Deviance Alcohol/Drug

Person Property Services Disorder (n = 314) (n = 315)
(n = 313) (n = 312) (n = 313) (n = 313)
0 items 112 (35.8%) 123 (39.4%) 186 (59.4%) 55 (17.6%) 24 (7.6%) 68 (21.6%)
1 item 91 (29.1%) 58 (18.6%) 55 (17.6%) 77 (24.6%) 52 (16.6%) 77 (24.4%)
2 items 52 (16.6%) 33 (10.6%) 32 (10.2%) 64 (20.4%) 56 (17.8%) 42 (13.3%)
3 items 26 (8.3%) 13 (4.2%) 13 (4.2%) 54 (17.3%) 85 (27.1%) 37 (11.7%)
4 items 16 (5.1%) 24 (7.7%) 21 (6.7%) 24 (7.7%) 46 (14.6%) 18 (5.7%)
5 items 9 (2.9%) 19 (6.1%) 3 (1.0%) 16 (5.1%) 39 (12.4%) 12 (3.8%)
6 items 7 (2.2%) 18 (5.8%) 3 (1.0%) 9 (2.9%) 12 (3.8%) 11 (3.5%)
7 items 0 (0%) 6 (1.9%) - 12 (3.8%) - 13 (4.1%)
8 items 0 (0%) 8 (2.6%) - 2 (0.6%) - 8 (2.5%)
9 items - 10 (3.2%) - - - 12 (3.8%)
10 items - - - - - 17 (5.5%)
Total 313 (100%) 312 (100%) 313 (100%) 313 (100%) 314 (100%) 315 (100%)

When looking at the self-reported drug and alcohol use of white respondents (Table 29),

one sees a pattern similar to the overall sample. For example, in the alcohol/soft drug use scale

one sees a gradual movement from 22.5% of participants reporting zero items to approximately

35% of respondents reporting all three items on the subscale. Similar to the overall sample, users

report the same general use frequencies in the hard drug use scale with most white respondents

(66.3%) claiming that they have never used any of the drugs listed. Frequency of use varies

between 2.9% and 9.5% for the other six items.

108
 
Table 29: Frequencies and Percentages for DV Subscales – White Respondents

Alcohol/Soft Drug Use Hard Drug Use

(n = 316) (n = 315)
0 items 71 (22.5%) 209 (66.3%)
1 item 82 (25.9%) 30 (9.5%)
2 items 50 (15.8%) 13 (4.1%)
3 items 113 (35.8%) 10 (3.2%)
4 items - 15 (4.8%)
5 items - 9 (2.9%)
6 items - 12 (3.8%)
7 items - 17 (5.4%)
Total 316 (100%) 315 (100%)

Non-white respondents. The frequency of offending reported by non-white respondents

does not follow patterns similar to those of their counterparts or to the overall sample. When

looking at all six dependent variable scales (Table 30), the frequency of behaviors engaged in

varies up and down from item to item.

Table 30: Frequencies and Percentages for DV Scales – Non-White Respondents

Predatory Predatory Illegal Public Deviance Alcohol/Drug

Person Property Services Disorder (n = 67) (n = 66)
(n = 66) (n = 67) (n = 64) (n = 66)
0 items 11 (16.7%) 16 (23.9%) 17 (26.6%) 9 (13.6%) 4 (6.0%) 13 (19.7%)
1 item 17 (25.8%) 9 (13.4%) 16 (25.0%) 7 (10.6%) 5 (7.5%) 9 (13.6%)
2 items 9 (13.6%) 11 (16.4%) 7 (10.9%) 7 (10.6%) 6 (9.0%) 10 (15.2%)
3 items 6 (9.1%) 6 (9.0%) 7 (10.9%) 8 (12.1%) 14 (20.9%) 8 (12.1%)
4 items 7 (10.6%) 1 (10.5%) 12 (18.8%) 12 (18.2%) 13 (19.4%) 8 (12.1%)
5 items 6 (9.1%) 3 (4.5%) 4 (6.3%) 5 (7.6%) 17 (25.4%) 7 (10.6%)
6 items 8 (12.1%) 3 (4.5%) 1 (1.6%) 9 (13.6%) 8 (11.9%) 0 (0.0%)
7 items 1 (1.5%) 8 (11.9%) - 5 (7.6%) - 3 (4.5%)
8 items 1 (1.5%) 5 (7.5%) - 4 (6.1%) - 2 (3.0%)
9 items - 5 (7.5%) - - - 3 (4.5%)
10 items - - - - - 3 (4.5%)
Total 66 (100%) 67 (100%) 64 (100%) 66 (100%) 67 (100%) 66 (100%)

Despite this fluctuation, non-white respondents report substance use patterns that follow

the same general pattern to that of white respondents (Table 31).

109
 
Table 31: Frequencies and Percentages for DV Subscales – Non-White Respondents

Alcohol/Soft Drug Use Hard Drug Use

(n = 66) (n = 66)
0 items 13 (19.7%) 38 (57.6%)
1 item 11 (16.7%) 9 (13.6%)
2 items 14 (21.2%) 5 (7.6%)
3 items 28 (42.4%) 2 (3.0%)
4 items - 4 (6.1%)
5 items - 1 (1.5%)
6 items - 4 (6.1%)
7 items - 3 (4.5%)
Total 66 (100%) 66 (100%)

An independent samples t-test (Table 32) reveals that the patterns of behavior reported by

white and non-white respondent differed significantly in five of the six dependent variable scales

(predatory person, predatory property, illegal services, public disorder, and general deviance).

Regarding these five scales, the white respondents self-reported significantly fewer acts of

deviance, delinquency, and criminal behavior. With respect to the substance abuse scale (and

both subscales), the patterns of behavior did not differ significantly based on race and white

respondent were less likely to report taking part in any type of drug or alcohol use. It will be

interesting to see if differences in parenting style contribute to any of these differences in

behavior.

Table 32: Independent Samples T-Test: DV Scales – Race

Predatory Person 5.802**

Predatory Property 3.552**
Illegal Services 5.472**
Public Disorder 5.206**
Deviance 4.065**
Alcohol/Drug Use .935

Alcohol/Soft Drug Use 1.345

Hard Drug Use 0.573
**p < .001

110
 
 Females. Female respondents’ self-reports of deviant and criminal behavior is similar in

frequency across all dependent variable scales to the overall sample (Table 33). Simply, the

frequency of offending decreased as the number of negative behaviors increased. It is not

surprising that frequency scores for the female participants correspond to the frequency scores

for the entire sample because females represent almost 62% of the sample; similar patterns were

seen for white respondents and university students.

Table 33: Frequencies and Percentages for DV Scales – Females

Predatory Predatory Illegal Public Deviance Alcohol/Drug

Person Property Services Disorder (n = 237) (n = 239)
(n = 237) (n = 236) (n = 234) (n = 236)
0 items 93 (39.2%) 107 (45.3%) 150 (64.1%) 49 (20.8%) 18 (7.6%) 58 (24.3%)
1 item 77 (32.5%) 49 (20.8%) 48 (20.5%) 68 (28.8%) 42 (17.7%) 67 (28.0%)
2 items 40 (16.9%) 34 (14.4%) 19 (8.1%) 52 (22.0%) 43 (18.1%) 38 (15.9%)
3 items 12 (5.1%) 10 (4.2%) 5 (2.1%) 37 (15.7%) 69 (29.1%) 26 (10.9%)
4 items 10 (4.2%) 13 (5.5%) 7 (3.0%) 17 (7.2%) 40 (16.9%) 11 (4.6%)
5 items 3 (1.3%) 11 (4.7%) 4 (1.7%) 6 (2.5%) 24 (10.1%) 10 (4.2%)
6 items 1 (0.4%) 8 (3.4%) 1 (0.4%) 6 (2.5%) 1 (0.4%) 6 (2.5%)
7 items 0 (0%) 1 (0.4%) - 0 (0%) - 9 (3.8%)
8 items 1 (0.4%) 1 (0.4%) - 1 (0.4%) - 5 (2.1%)
9 items - 2 (0.8%) - - - 5 (2.1%)
10 items - - - - - 4 (1.7%)
Total 237 (100%) 236 (100%) 234 (100%) 236 (100%) 237 (100%) 239 (100%)

In addition, the pattern of alcohol and soft drug use was comparable to the total sample

(Table 34). Regarding the alcohol/soft drug use scale, the frequency of use increased from zero

to one item, then decreased at two items before increasing at three items; this particular result has

been seen throughout the analyses. On the other hand, the frequency of hard drug use by female

respondents decreased as the number of items increased.

Table 34: Frequencies and Percentages for DV Subscales – Females

Alcohol/Soft Drug Use Hard Drug Use

(n = 239) (n = 239)
0 items 59 (24.7%) 182 (76.2%)
1 item 70 (29.3%) 16 (6.7%)

111
 
2 items 44 (18.4%) 8 (3.3%)
3 items 66 (27.6%) 8 (3.3%)
4 items - 11 (4.6%)
5 items - 5 (2.1%)
6 items - 5 (2.1%)
7 items - 4 (1.7%)
Total 239 (100%) 239 (100%)

Males. Frequencies of offending statistics for male respondents follow patterns similar to

female respondents or to the overall sample (Tables 35). More specifically, as the number of

deviant, delinquent, and criminal behaviors increases the frequency of engaging in those

activities appears to decrease.

Table 35: Frequencies and Percentages for DV Scales – Males

Predatory Predatory Illegal Public Deviance Alcohol/Drug

Person Property Services Disorder (n = 146) (n = 145)
(n = 145) (n = 146) (n = 146) (n = 146)
0 items 30 (20.7%) 33 (22.6%) 54 (37.0%) 15 (10.3%) 10 (6.8%) 23 (15.9%)
1 item 31 (21.4%) 17 (11.6%) 23 (15.8%) 16 (11.0%) 15 (10.3%) 19 (13.1%)
2 items 23 (15.9%) 11 (7.5%) 19 (13.0%) 19 (13.0%) 20 (13.7%) 14 (9.7%)
3 items 20 (13.8%) 9 (6.2%) 16 (11.0%) 27 (18.5%) 29 (19.9%) 20 (13.8%)
4 items 14 (9.7%) 13 (8.9%) 28 (19.2%) 19 (13.0%) 21 (14.4%) 16 (11.0%)
5 items 12 (8.3%) 12 (8.2%) 3 (2.1%) 15 (10.3%) 32 (21.9%) 10 (6.9%)
6 items 14 (9.7%) 13 (8.9%) 3 (2.1%) 12 (8.2%) 19 (13.0%) 5 (3.4%)
7 items 1 (0.7%) 13 (8.9%) - 18 (12.3%) - 7 (4.8%)
8 items 0 (0%) 12 (8.2%) - 5 (3.4%) - 5 (3.4%)
9 items - 13 (8.9%) - - - 10 (6.9%)
10 items - - - - - 16 (11.0%)
Total 145 (100%) 146 (100%) 146 (100%) 146 (100%) 146 (100%) 145 (100%)

Similar to the female sample and the overall sample, male respondents were likely to

engage in all three categories of alcohol/soft drug use (Table 36). The pattern of hard drug use

amongst male respondents mirrored hard drug use amongst county jail inmates. A high

percentage of respondents (45.5%) reported using no hard drugs. That percentage gradually

decreased to 3.4% for three items. At four items, the percentage increased to 5.5% before falling

back to 3.4% for five items; the percentage gradually increased over six and seven items.

112
 
Table 36: Frequencies and Percentages for DV Subscales – Males

Alcohol/Soft Drug Use Hard Drug Use

(n = 146) (n = 145)
0 items 25 (17.1%) 66 (45.5%)
1 item 23 (15.8%) 23 (15.9%)
2 items 22 (15.1%) 11 (7.6%)
3 items 76 (52.1%) 5 (3.4%)
4 items - 8 (5.5%)
5 items - 5 (3.4%)
6 items - 11 (7.6%)
7 items - 16 (11.0%)
Total 146 (100%) 145 (100%)

An independent samples t-test was conducted in order to determine the significance of

the mean differences between the behavior of male respondents and the behavior of female

respondents. In all six dependent variable scales and both subscales, males were significantly

more likely than females to engage in deviant, delinquent, and criminal activity (Table 37).

Given these differences, it will be interesting to analyze the effects that perceived parenting

styles have on self-reported behavior.

Table 37: Independent Samples T-Test: DV Scales – Sex

Predatory Person -7.583**

Predatory Property -9.381**
Illegal Services -7.269**
Public Disorder -9.136**
Deviance -4.848**
Alcohol/Drug Use -6.089**

Alcohol/Soft Drug Use -4.387**

Hard Drug Use -5.886**
**p < .01

The results generated by analyzing the frequencies and descriptive statistics of the overall

sample, as well the analyses of the student, inmate, white, non-white, female, and male

respondents specifically, suggest that there are differences in deviant and criminal behavior

113
 
based on race, sex, and institutional affiliation. Given these significant differences, it is

imperative to determine if parenting style contributes to the decision to engage in the behaviors

discussed above. In order to determine the effect that parenting style has on behavior, the

researcher has conducted a multiple regression analysis. Before reporting the results of the

multiple regression analysis, however, the results of a bivariate correlation analysis of the

independent variables are discussed in the next section.

Bivariate Correlations

The purpose of the bivariate correlational analysis was to examine whether significant

relationships existed between the variables. Additionally, this analytical procedure was

undertaken in order to test for multicollinearity. In order to meet the assumptions for multiple

regression analysis, multicollinearity must be absent. The results for the bivariate correlation

matrix are presented in Table 38.

An examination of the bivariate correlation matrix reveals that several independent

variables were significantly correlated with one another. The highest correlations amongst the

independent variables occurred between the parenting variables. For example, maternal

authoritativeness and maternal neglecting/rejecting were the most highly correlated of all the

independent variables (r = -.709, p < .001). Additionally, the relationship between paternal

authoritativeness and paternal neglecting/rejecting also was highly correlated (r = -.625, p <

.001). Given the nature of the authoritative type of parenting and the neglecting/rejecting type of

parenting, the suggestion that these two parenting types are negatively correlated is not

surprising. As discussed earlier, authoritative parents are communicative with their children, yet

demanding at the same time; neglecting/rejecting parents, on the other hand, do not interact with

114
 
Table 38: Bivariate Correlation Matrix

1 2 3 4 5 6 7 8 9 10
1 Mother 1.00
Authoritarian (395)
2 Mother -.285** 1.00
Authoritative (384) (387)
3 Mother -.461** .341** 1.00
Permissive (390) (382) (392)
4 Mother .095 -.709** .003 1.00
Neglect/Reject (387) (379) (384) (389)
5 Father .289** -.044 -.029 .119* 1.00
Authoritarian (348) (341) (346) (342) (353)
6 Father .014 .275** .096 -.226** -.372** 1.00
Authoritative (343) (336) (341) (338) (345) (348)
7 Father -.118* -.009 .404** .068 -.472** .407** 1.00
Permissive (345) (338) (344) (339) (345) (343) (350)
8 Father -.048 -.313** .074 .383** .053 -.625** .067 1.00
Neglect/Reject (344) (337) (342) (338) (344) (341) (344) (349)
9 Race .166** -.073 -.122* .082 .026 -.086 .003 .028 1.00
(378) (371) (376) (373) (342) (337) (339) (338) (391)
10 Sex -.040 -.010 .040 -.075 -.020 .000 .044 -.003 -.114* 1.00
(381) (374) (379) (376) (344) (339) (341) (340) (390) (394)
**p ≤ .01, *p ≤ .05

115
 
their children or hold them to standards of conduct. Simply, one would have a difficult time

finding two parenting types that are more different than authoritative and neglecting/rejecting.

Strong significant negative correlations also exist between authoritarian mothers and

permissive mothers (r = -.461, p < .001) as well as authoritarian fathers and permissive fathers (r

= -.472, p < .001). Again, these strong negative correlations make sense because authoritarian

parents are very demanding of their children, but do not openly communicate with them, while

permissive parents are the complete opposite – they strive to be open and friendly with their

children, but fail to provide proper discipline and standards of conduct.

Not surprisingly, significant positive correlations exist between authoritarian mothers and

fathers, authoritative mothers and fathers, permissive mothers and fathers, and

neglecting/rejecting mothers and fathers. Additionally, significant negative correlations exist

between authoritarian mothers and permissive fathers, authoritative mothers and

neglecting/rejecting fathers, and neglecting/rejecting mothers and authoritative fathers.

Altogether, the results of the correlations suggest that the independent variables that should be

correlated with each other are correlated with each other and that the correlations are in the

correct direction.

Multiple Regression

Many of the analyses discussed above report interesting findings about the relationships

between certain variables, but they are somewhat limited in their ability to draw significant

conclusions about the cause and effect questions that this study seeks to answer. Thus, multiple

regression analysis was used to determine how parenting style affects the reported levels of

deviance, delinquency, and criminality of the respondents. More specifically, the multiple

regression analysis sought to determine how maternal and paternal parenting styles, such as

116 
 
authoritarian, authoritative, permissive, and neglecting/rejecting, along with race and sex

variables, affect offense variables such as crimes against people, crimes against property, illegal

service offenses, public disorder crimes, general deviant behaviors, and drug and alcohol use.

Again, the multiple regression equation used in the study looks like this:

Ŷ = B0 + B1X1 + B2X2 + … + BkXk + êi

Where:

Ŷ = the predicted value of the dependent variable, level of criminal involvement

B0 = the y-intercept when X = 0; or, the constant

B = the slope of the regression line

X1 = sex

X2 = race

X3 = maternal authoritarian parenting style

X4 = maternal authoritative parenting style

X5 = maternal neglecting/rejecting parenting style

X6 = maternal permissive parenting style

X7 = paternal authoritarian parenting style

X8 = paternal authoritative parenting style

X9 = paternal neglecting/rejecting parenting style

X10 = paternal permissive parenting style

êi = the predicted error term

The multiple regression analysis is divided into three models. The first model analyzes

the effects of the parenting variables, race, and sex on each of the six offense scales and two

subscales. Subsequent models examine the effect that parenting style has on offending when:

117 
 
university students are compared to county jail inmates and females are compared to males. The

following discusses each regression model.

Model One: Effect of Parenting Style on Deviant, Delinquent, and Criminal Behavior –
Full Sample

For the initial analysis, a linear regression was conducted using all of the parenting style

variables, race, and sex as independent variables. Each of the dependent variable scales and

subscales were analyzed. Instead of deleting the responses of participants who did not respond to

each item, the researcher replaced all missing values with the mean value (this also was done for

Model Two and Three). The following illustrates the significant findings for the regression

analyses using the entire sample.

Predatory property crimes. In Table 39, the R-square statistic reports the proportion of

variance in the predatory property crimes scale (the dependent variable) that is explained by each

of the independent variables. As an R-square value moves further away from 0 and closer to 1,

theory suggests that an independent variable (or series of independent variables) account for

more of the variation in the dependent variable. The R-square value in this model is .272. This

suggests that the independent variables account for 27.2% of the variance in predatory property

crimes committed by all participants. Additionally, the F in this model is 14.820 (Sig. < .0001),

which suggests that at least one slope in the regression equation does not equal zero. Therefore,

at least one of the independent variables is significant.

Table 39: Full Sample – Predatory Property

B (SE) Beta
Constant 4.621 (1.664) -
Maternal Neglect/Reject -.111 (.042) -.175**
Race -.757 (.309) -.108*
Sex 2.155 (.240) .391**

R-Square = .272

118 
 
F = 14.820
SE = 2.274
**p ≤ .01; *p ≤ .05

When looking at Table 39, one sees that three of the independent variables are

significant: maternal neglecting/rejecting, race, and sex. With regards to the demographic

variables, the results of this model suggest that the level of predatory property offending is likely

to be lower for whites compared to non-whites and that males are more likely than females to be

engaged in property crimes. Only one parenting style variable attained significance in this model.

Results suggest that the likelihood of committing a property offense significantly decreases when

a child is reared by a mother who is not neglecting/rejecting. This suggestion is in agreement

with prior research.

Predatory person crimes. Table 40 reports the effect that the independent variables have

on level of predatory offenses against people. The R-square for this model is .236, which

suggests that the independent variables account for 23.6% of the variance in predatory person

crimes committed by the sample of respondents. The F in this model is 12.294 (Sig. < .0001);

again, this suggests that at least one of the independent variables is significant.

Table 40: Full Sample – Predatory Person

B (SE) Beta
Constant 1.742 (1.070)
Paternal Permissive .079 (.028) .176**
Race -.991 (.199) -.225**
Sex 1.094 (.154) .316**

R-Square = .236
F = 12.294
SE = 1.462
**p ≤ .01

As reported in Table 40, the demographic variables of race and sex are significant once

again. Results suggest that white respondents are significantly less likely than non-white

119 
 
participants to commit crimes against people. Also, males are significantly more likely than

females to commit predatory crimes against people. Results also suggest predatory crimes

against people are likely to increase if one does not have a permissive father. This seems to

contradict prior research, which suggests that permissive parenting leads to increased offending.

Illegal services crimes. The effect that the independent variables have on illegal service

crimes is reported in Table 41. The R-square of the model is .219, which means that 21.9% of the

variance in illegal services offending is accounted for by the independent variables. Additionally,

given the F score of 11.117 (Sig. < .0001) it can be determined that at least one of the

independent variables is significant.

Table 41: Full Sample – Illegal Services

B (SE) Beta
Constant 3.079 (1.123)
Race -.782 (.176) -.203**
Sex .932 (.137) .307**

R-Square = .219
F = 11.117
SE = 1.297
**p ≤ .01

Consistent with the previous models, race and sex are both significant predictors of

offending behavior. White respondents are significantly less likely to commit illegal service

crimes; male respondents are significantly more likely to commit illegal service crimes. No

parenting variables were significant.

Public disorder offenses. As reported in Table 42, the R-square for the model examining

the effect of the independent variables on public disorder offenses is .289. This suggests that the

independent variables account for 28.9% of the variance in public disorder behaviors. Of all R-

square values calculated to this point, the value in this model is the largest. The F-score is also

120 
 
large and significant, 16.169 (Sig. < .0001), suggesting that at least one independent variable also

is significant.

Table 42: Full Sample – Public Disorder

B (SE) Beta
Constant 1.893 (1.240)
Paternal Permissive .091 (.033) .170**
Race -.956 (.231) -.181**
Sex 1.583 (.176) .381**

R-Square = .289
F = 16.169
SE = 1.695
**p ≤ .01

By looking at the table, one can see that both demographic variables (race and sex)

remain significant and continue to predict behavior. Also related to earlier models, is the finding

that public disorder offending is likely to increase when the respondents are not parented by

permissive male caretakers.

Deviant behaviors. The R-square for the full sample deviant behaviors model is .137.

This suggests that the independent variables account for 13.7% of the variance in the deviant

behaviors scale. In addition to being the lowest R-square value up to this point, the F-score also

is small despite the fact that it is significant; 6.305 (Sig. < .0001). Based on the F-score it is

likely that at least one of the independent variables is significant. When looking at Table 43, one

can see that three independent variables are significant.

Table 43: Full Sample – Deviant Behaviors

B (SE) Beta
Constant 3.836 (1.271)
Race -.672 (.203) -.159**
Sex .696 (.157) .210**

R-Square = .137
F = 6.305

121 
 
SE = 1.493
**p ≤ .01

Both demographic variables are significant. Being white remains a significant predictor

of not engaging in negative behaviors – general deviance, in this instance – and being female

also remains a significant predictor of the reduced likelihood of engaging in deviant behaviors.

Again, no parenting styles achieve a level of significance in this model.

Alcohol/Drugs. The full sample alcohol/drug use model produced an R-square of .170,

which suggests that the independent variables account for 17% of the variance in alcohol and

drug use. The F-score in this model is 5.775 (Sig. ≤ .0001) which is significant and suggests that

at least one independent variable in this model is significant. Looking at Table 44, one sees that

two of the independent variables reach significance.

Table 44: Full Sample – All Alcohol/Drug Use

B (SE) Beta
Constant 3.989 (2.413)
Maternal Neglect/Reject -.161 (.066) -.223*
Sex 1.773 (.346) .286**

R-Square = .170
F = 5.775
SE = 2.805
**p ≤ .01; *p ≤ .05

Once again, sex is a significant predictor of the dependent variable; in this instance,

males are significantly more likely to use alcohol and drugs. The presence of neglecting/rejecting

maternal caretakers is significantly predictive of an increased level of alcohol and drug use.

Simply, if one is parented by a mother who is not neglecting/rejecting, they are less likely to use

alcohol and other drugs.

Subscale: Alcohol/Soft Drugs. The model for the first subscale had an R-square of .133

(Table 45). This suggests that the independent variables accounted for 13.3% of the variance in

122 
 
alcohol and soft drug use. The F in this model was significant, 4.348 (Sig. ≤ .0001), which

suggests that at least one of the independent variables in this model was significant. Three

independent variables were a significant predictor of alcohol and soft drug. First, sex is a

significant predictor of alcohol, tobacco, and marijuana use – males are significantly more likely

to use these substances. Similar to the full alcohol and drug use scale, the presence of a

neglecting/rejecting caretaker suggests that respondents will use these types of substances. This

time, however, it is a paternal caretaker who has that effect. Third, the presence of a permissive

paternal caretaker suggests a significantly decreased likelihood of alcohol, tobacco, and

marijuana use.

Table 45: Full Sample – Alcohol/Soft Drug Use

B (SE) Beta
Constant 1.969 (.955)
Paternal Permissive .074 (.024) .248**
Paternal Neglect/Reject -.049 (.022) -.187*
Sex .549 (.137) .228**

R-Square = .133
F = 4.348
SE = 1.115
**p ≤ .01; *p ≤ .05

Subscale: Hard drugs. The final full scale model examined hard drug use. The R-square

for this model is .130, thus suggesting that the independent variables account for 13.0% of the

variance in hard drug usage amongst the respondents. The F-score for the model is 5.957 (Sig. <

.0001), which means it is likely that there is at least one significant independent variable.

Looking at Table 46, one sees two significant independent variables.

123 
 
Table 46: Full Sample – Hard Drug Use

B (SE) Beta
Constant .792 (1.433)
Maternal Neglect/Reject -.092 (.036) -.184*
Sex 1.162 (.206) .268**

R-Square = .130
F = 5.957
SE = 1.958
**p ≤ .01; *p ≤ .05

As with each previous model, sex is a significant predictor of hard drug usage. Simply,

males are more likely than females to engage in this type of criminal behavior. The presence of a

maternal neglecting/rejecting caretaker, once again, significantly predicts an increased likelihood

of engaging in deviant and criminal behavior, as usage is likely to drop if an individual has a

more responsive mother.

Summary. The findings generated throughout this first set of eight models were both

expected and surprising. In all instances, sex was a significant predictor of deviant, delinquent,

and criminal behaviors with males being more likely than females to engage in those behaviors.

This finding was not wholly unexpected given the prevailing notion that men are more criminal

than women. A consensus of the prior literature suggests that children with authoritative parents

are significantly less likely than children of all other parenting types to engage in deviant and

criminal behaviors. Yet, authoritative parenting did not significantly predict anything in the full

sample. The findings that the level of involvement in criminality and delinquency reported by

respondents is linked to the type of mother they had is important. Simply stated, respondents

who were raised by someone other than a neglecting/rejecting mother reported significantly

lower levels of property crimes, overall alcohol/drug use, and hard drug use. Finally, the type of

father one has is related to the level of delinquency and criminal activity reported by

124 
 
respondents. In short, there were increased levels of person-based crimes, public disorder

offenses, and soft drug/alcohol use when one is not raised by a permissive father. The findings

related to permissive male caretaking do not conform to prior research and were completely

unexpected. It is possible that potential reasons for these relationships can be teased out of the

data by examining the differences in institutional setting and sex more closely. The following

chapters will explore these differences.

125 
 
 CHAPTER V

ANALYSIS AND RESULTS – MULTIPLE REGRESSION COMPARISONS

The purpose of this chapter is to continue the discussion of the results of the multiple

regression analyses. The previous chapter ended with an examination of the regression models

for the entire sample. This chapter furthers that discussion with a comparison of the university

student sample and the county jail inmate sample. Additionally, the effect that parenting styles

have on females is compared to the effect that parenting styles have on males. A review of the

hypotheses that guided this study concludes this chapter.

Model Two

Effect of Parenting Style on Deviant, Delinquent, and Criminal Behavior –

University Students and County Jail Inmates

The next set of models examines the influence that the independent variables parenting

style, race, and sex have on the dependent variable scales in the sample of university students.

The university students are examined separately from county jail inmates in an effort to allow the

researcher the ability to determine if parenting style affects behavior differently between the two

groups. Significant results from both samples are discussed and presented below.

Predatory property crimes – university students. The first university student model

examines the commission of a series of predatory property crimes (Table 47). The R-square

value for this model is .091, which means that 9.1% of the variance in predatory property crimes

committed by the students can be explained by the independent variables. A significant F-score,

2.875 (Sig. = .002), also suggests that there is at least one significant independent variable. In

this instance, there is one significant independent variable – sex. The results of this model

suggest that male students are more likely than female students to engage in property crimes.

126 
 
Table 47: University Students – Predatory Property

B (SE) Beta
Constant 2.427 (1.414) -
Sex .816 (.211) .220**

R-Square = .091
F = 2.875
SE = 1.553
**p ≤ .01

The results from the university student sample are similar to results from the full sample.

In both the full sample and the university student sample, sex of the respondent was a significant

predictor of property offending/non-offending. As the analysis continues, it will be interesting to

see if a similar pattern of support emerges for all dependent variable scales.

Predatory person crimes – university students. The R-square value for this

model was .085, which means that 8.5% of the variance in predatory person crimes committed

by university students can be explained by the independent variables (Table 48). One

independent variable in this model was significant: sex. Results suggest that male students are

more likely than female students to commit person-based crimes.

Table 48: University Students – Predatory Person

B (SE) Beta
Constant .330 (.917) -
Sex .438 (.137) .183*

R-Square = .085
F = 2.661
SE = 1.007
*p ≤ .05

Illegal services crimes – university students. Amongst university students, the R-square

for the illegal services model is .053 (Table 49). This means that 5.3% of the variance in the

dependent variable is explained by the independent variables. This is an extremely low R-square

127 
 
value and leaves 94.7% of the variance in illegal services crimes committed by university

students unexplained. Additionally, the F-score is small and not significant, 1.589 (Sig. = .109),

which suggests that no independent variable should be significant. However, one independent

variable, sex, is a significant predictor of illegal service crimes. Consistent with the previous two

models, male university students are significantly more likely than female university students to

engage in illegal service activities. It must be reiterated, however, that given the low R-square

value this model is not a good predictor of illegal service activities amongst college students.

Table 49: University Students – Illegal Services

B (SE) Beta
Constant -.604 (.648) -
Sex .302 (.097) .181**

R-Square = .053
F = 1.589
SE = .711
**p ≤ .01

Public disorder offenses – university students. The R-square in this model was .118,

thus suggesting that the independent variables of parenting style, race, and sex explained 11.8%

of the variance in the dependent variable. In addition, the F-score was significant, 3.827 (Sig. <

.0001), which means that at least one independent variable should be significant. As seen in

Table 50, two independent variables are significant: paternal permissive and sex.

Table 50: University Students – Public Disorder

B (SE) Beta
Constant .450 (1.158) -
Paternal Permissive .065 (.029) .180*
Sex .756 (.173) .245**

R-Square = .118
F = 3.827
SE = 1.272
**p ≤ .01; *p ≤ .05

128 
 
 According to the results, university students are more likely to engage in public disorder

activities when they are not parented by a permissive paternal caretaker. This finding

complements earlier findings for the entire sample both in significance and direction. The second

independent variable that is significant is sex. Again, being a male university student is a

significant predictor of commission of public disorder offenses.

Deviant behaviors – university students. The R-square value in this model is low at

.046. This suggests that the independent variables account for a paltry 4.6% of the variance in

deviant behaviors exhibited by university students, which means that 95.4% of the variance is

unexplained. Additionally, the F-score is low and not significant, 1.369 (Sig. = .194). Consistent

with having an F-score that is not significant no independent variables in this model were

significant. Despite the low R-square value, this model is interesting because it is the first

university student model in which the sex of the respondent is not a significant predictor of

behavior.

Alcohol/drugs – university students. This model had an R-square of .043, which

suggested that the independent variables explained 4.3% of the variance in students’ drug and

alcohol use. The F-score for this model was small and not significant, .906 (Sig. = .529),

suggesting that no independent variables would significantly predict alcohol and drug use; none

were significant.

Subscale: Alcohol/soft drugs – university students. The R-square value of the alcohol

and soft drug use subscale is .045 (Table 51), which is similar to the overall drug use scale for

university students. This means that the independent variables account for 4.5% of the variance

in alcohol, tobacco, and marijuana use amongst university students. The F value for this model is

129 
 
.956 (Sig. = .483). Given the significance level of the F-score it is surprising that one of the

independent variables were significant predictors of students’ alcohol and soft drug use.

Table 51: University Students – Alcohol/Soft Drug Use

B (SE) Beta
Constant 1.115 (1.236) -
Paternal Permissive .066 (.029) .234*

R-Square = .045
F = .956
SE = 1.103
*p ≤ .05

Students with permissive paternal caretakers are significantly less likely to use alcohol

and soft drugs than students who do have some other type of paternal caretakers. Additionally,

this is the second substance use model in which sex is not a significant predictor, which suggests

that tobacco, alcohol, and marijuana use patterns on college campuses cannot be explained the

sex of the student.

Subscale: Hard drugs – university students. The final university student model, hard

drug use, had an R-square value of .032. This suggests that only 3.2% of the variance in

students’ hard drug use can be explained by the independent variables. As with previous models,

an R-square value of .032 is extremely low and suggests that the model is a poor predictor of the

behavior under investigation. Additionally, the F-score is low and not significant, .938 (Sig. =

.499), thus no independent variables are significant. Despite this, results suggest that university

student drug use is predicated largely upon factors other than parents’ parenting style, sex, and

race. The eight models that follow examine the effect that parenting styles had on the sample of

county jail inmates.

Predatory property crimes – county jail inmates. The R-square value for this model

was .126 (Table 52). This suggests that the independent variables defined for the study account

130 
 
for 12.6% of the variance in property crimes committed by county jail inmates in the sample.

The F value for this model is not significant with a score of 1.441 (Sig. = .173). Despite the lack

of significance, which suggest that no independent variable should be significant, the sex of the

inmate is predictive of property offending. Male inmates are significantly more likely than

female inmates to commit predatory property offenses. It is interesting that no parenting

variables are predictive of property offending amongst county jail inmates given the findings of

past literature which suggests that those who exhibit problematic behaviors are likely to have had

neglecting/rejecting parents.

Table 52: County Inmates – Predatory Property

B (SE) Beta
Constant 5.883 (3.219) -
Sex 1.617 (.532) .292**

R-Square = .126
F = 1.441
SE = 2.557
**p ≤ .01

Predatory person crimes – county jail inmates. This model has an R-square value of

.132 (Table 53). That means that over 13% of the variance in the dependent variable, predatory

person crimes committed by county jail inmates, can be explained by the independent variables.

The F value is 1.518 (Sig. = .144) and is not significant. Despite this, one independent variable is

significant. Non-white county jail inmates are significantly more likely to commit these types of

crimes.

Table 53: County Inmates – Predatory Person

B (SE) Beta
Constant 3.472 (2.171) -
Race -.789 (.386) -.211*

R-Square = .132

131 
 
F = 1.518
SE = 1.724
*p ≤ .05

Illegal services crimes – county jail inmates. The model for illegal service crimes

committed by county jail inmates had an R-square value of .130, thus 13% of the variance in

offending can be explained by the independent variables. Despite this, the F value is not

significant, 1.499 (Sig. = .151) and no independent variables are significant predictors of this

type of behavior. Given the results for the past two models, it can be assumed that something

other than parenting style and sex of the offender drives inmates to commit certain offenses.

Public disorder offenses – county jail inmates. The R-square value for this model was

.170 (Table 54). This suggests that the independent variables explain 17% of the variance in

public disorder offenses within the county jail inmate sample. This is a relatively strong R-square

value in social science research. Additionally, the F value is 2.052 (Sig. = .036), which means

that at least one independent variable should be significant. Male county jail inmates are

significantly more likely than female county jail inmates to partake in public disorder offenses.

The finding that males are more likely to engage in deviant or criminal acts is neither surprising

nor unexpected.

Table 54: County Inmates – Public Disorder

B (SE) Beta
Constant 2.958 (2.287) -
Sex 1.098 (.378) .272**

R-Square = .170
F = 2.052
SE = 1.816
**p ≤ .01

Deviant behaviors – county jail inmates. With an R-square value of .072, the

independent variables explain 7.2% of the variance when county jail inmates engage in deviant

132 
 
behaviors. This is a very small R-square value as 92.8% of the variance is explained by other

variables. The F value also is very small and non-significant, .774 (Sig. = .654) and, once again,

no independent variables were predictive of behavior. As a whole, these models are poor

predictors of the commission of crime and deviant behaviors by this sample of county jail

inmates.

Alcohol/drugs – county jail inmates. The R-square value for this model was .262 (Table

55), which suggests that 26.2% of the variance in alcohol and drug use by county jail inmates

can be explained by the independent variables. Additionally, the F value was significant, 2.409

(Sig. = .016), thus leading one to believe that at least one independent variable should be

significant. In this model, one independent variable was significant: race. Race is predictive of

substance abuse. White inmates are significantly more likely than non-white inmates to engage

in drug and alcohol use. The models that immediately follow will discuss the two substance use

subscales relative to the county inmate sample.

Table 55: County Inmates – Alcohol/Drug Use

B (SE) Beta
Constant .918 (3.678) -
Race 3.076 (.778) .494**

R-Square = .262
F = 2.409
SE = 2.684
**p ≤ .01

Subscale: Alcohol/soft drugs – county jail inmates. With an R-square value of .204

(Table 56), over 20% of the variance in county jail inmates use of alcohol, marijuana, and

tobacco is explained by the independent variables. With an F value that barely approaches

significance, 1.739 (Sig. = .090), two independent variables are significant in this model.

133 
 
Table 56: County Inmates – Alcohol/Soft Drug Use

B (SE) Beta
Constant 1.920 (.874) -
Maternal Authoritative .063 (.029) .451*
Race .604 (.185) .424**

R-Square = .204
F = 1.739
SE = .638
**p ≤ .01; *p ≤ .05

County jail inmates were significantly likely to use alcohol, marijuana, and tobacco if

they were parented by women who were not authoritative. This finding complies with prior

research, which suggests that the presence of an authoritative parent should reduce problem

behaviors; surprisingly, this is the first instance of authoritative parenting being a significant

predictor of behavior. Additionally, race is a significant predictor of alcohol and soft drug use.

Once again, white inmates are significantly more likely than non-white inmates to partake in

alcohol and other soft drugs.

Subscale: Hard drugs – county jail inmates. The R-square value for this model was

.161 (Table 57). This suggests that the independent variables being used in the study explain

16.1% of the variance in hard drug use by county jail inmates. The F value is 1.921 (Sig. = .051)

and it is significant, which means that at least one independent variable is significant. Consistent

with findings from the model that examined all drug and alcohol use amongst county jail

inmates, the independent variable race was significant and in the same direction in this model.

Table 57: County Inmates – Hard Drug Use

B (SE) Beta
Constant -1.173 (3.002) -
Race 1.660 (.534) .316**

R-Square = .161
F = 1.921

134 
 
SE = 2.384
**p ≤ .01

Summary. First, it must be pointed out that the R-square values for these 16 models are

not overwhelmingly strong. Overall, the county jail inmate models show more prediction

strength than the models for the university students. When looking specifically at each of the

samples, it is evident that sex is a stronger predictor of behavior amongst university students than

county jail inmates as sex is significant in four of the student models compared to just two of the

inmate models, with an overlap in two models (predatory property crimes and public disorder

offenses).

Parenting variables do not prove to be very strong predictors in either of the two samples.

One paternal variable (permissive) is a significant predictor of behavior in the university student

sample. This result was surprising for two reasons. First, it was anticipated that authoritative

parenting, from either caretaker, would have a greater effect on reducing the likelihood that

university students would involve themselves in deviance and criminality. Second, it was

expected that the neglecting/rejecting parenting style would be responsible for more problematic

behaviors. Perhaps the neglecting/rejecting style of parenting did not have as great of an effect as

anticipated because the frequency of offending amongst university students was not that high and

this type of parenting style was the least reported amongst university students. But, that reason

cannot be applied to county jail inmates because their frequency of offending was much higher

than the university students and only one parenting variable was a significant predictor of the

inmates’ behavior. Not surprisingly, maternal authoritativeness (alcohol/soft drug use) predicted

decreased levels of offending. Possible reasons for these similarities will be examined in the

discussion chapter.

135 
 
Model Three: Effect of Parenting Style on Deviant, Delinquent, and Criminal Behavior –
Females and Males

This section details the effects of parenting style and race on deviant, delinquent, and

criminal behavior. Results of the regression analyses are separated by the sex of the respondents.

The first part of the section will discuss the independent variables that significantly affect the

behavior of females and the second part of the section will examine how the independent

variables significantly affect the behaviors of the male respondents. All significant results are

discussed below.

Predatory property crimes – females. This model had an R-square score of .102 (Table

58), which suggests that the independent variables account for over 10% of the variance in the

female respondents’ likelihood to commit property crimes. One independent variable is

significant, which, given the significance of the F value, 2.950 (Sig. = .002), was somewhat

expected.

Table 58: Female – Predatory Property

B (SE) Beta
Constant 3.554 (1.751) -
Maternal Neglect/Reject -.102 (.045) -.232*

R-Square = .102
F = 2.950
SE = 1.801
*p ≤ .05

The independent variable that reached levels of significance was maternal

neglecting/rejecting. Female respondents were significantly less likely to engage in property

offenses when they were not parented by a neglecting/rejecting mother. As has been the case

throughout this research, the presence of neglecting/rejecting caretakers appears to increase

problem behaviors. This complies with findings from prior research.

136 
 
 Predatory person crimes – females. This regression model had a low R-square score of

.091 (Table 59). Therefore, approximately 9.1% of the variance in predatory person crimes

committed by females can be explained by the independent variables. Additionally, the F score is

significant, 2.597 (Sig. = .007), which suggests that at least one independent variable will be

significant. In this instance, race is a significant predictor of person-based crimes. Simply, non-

white females are significantly more likely than white females to commit these types of crimes.

Table 59: Female – Predatory Person

B (SE) Beta
Constant 1.637 (1.197) -
Race -.818 (.231) -.227**

R-Square = .091
F = 2.597
SE = 1.232
**p ≤ .01

Illegal services crimes – females. The model analyzing illegal service crimes committed

by female respondents had an R-square score of .095 (Table 60). Thus, fewer than 10% of illegal

service activity by females can be explained by the independent variables in the model. One

independent variable, race, is a significant predictor of illegal service activities. Non-white

female respondents are significantly more likely to engage in illegal service crimes.

Table 60: Female – Illegal Services

B (SE) Beta
Constant 1.671 (1.072) -
Race -.582 (.207) -.180**

R-Square = .095
F = 2.716
SE = 1.103
**p ≤ .01

137 
 
 Public disorder offenses – females. The R-square value for this model was .119 (Table

61), which suggests that the independent variables explain nearly 12% of the variance in the

commission of public disorder offenses committed by females. In addition, the F score is

significant, 3.502 (Sig. ≤ .0001), thus indicating that at least one independent variable in the

model is significant. There are two significant independent variables in this model: race and

paternal permissiveness.

Table 61: Female – Public Disorder

B (SE) Beta
Constant 2.069 (1.393) -
Paternal Permissive .081 (.036) .210*
Race -.935 (.269) -.219**

R-Square = .119
F = 3.502
SE = 1.434
**p ≤ .01; *p ≤ .05

As seen in the table above, non-white female respondents are significantly more likely

than their white counterparts to engage in public disorder offenses. Consistent with earlier

findings in this study, the type of father one has plays a role in the child’s behavior. In this

instance, female respondents were significantly more likely to engage in these types of offenses

if they were raised by a father who was not permissive.

Deviant behaviors – females. With an R-square score of .056 (Table 62), the

independent variables in this model explain just over 5% of the variance in the commission of

deviant behaviors by females in this study. Despite the fact that the F value is not significant,

1.551 (Sig. = .131), one independent variable is significant. Once again, non-white female

respondents are significantly more likely than white female respondents to take part in deviant

138 
 
behaviors. This variable in particular has been significant throughout this model; implications

will be examined in the discussion chapter.

Table 62: Female – Deviant Behavior

B (SE) Beta
Constant 2.875 (1.358) -
Race -.792 (.262) -.197**

R-Square = .056
F = 1.551
SE = 1.397
**p ≤ .01

Alcohol/drugs – females. The R-square score for this model is .055. This means that

only 5.5% of the variance in substance use by females can be explained by the independent

variables in the model. The F score does not come close to approaching significance, 1.108 (Sig.

= .360), and no independent variables are significant. This represents a weak model for

predicting alcohol and drug use by females.

Subscale: Alcohol/soft drugs – females. With an R-square value of .063, the

independent variables in this study account for approximately 6% of the variance in alcohol,

tobacco, and marijuana use by females. The F value of .1.278 (Sig. = .252) is not significant.

Again, no independent variables achieve significance in this model.

Subscale: Hard drugs – females. The variance in hard drug use by females that is

explained by the independent variables in this model is a paltry 2.7% (R-square = .027). This

suggests that the model does not adequately measure hard drug use. No independent variables

reach a level of significance, consistent with the presence of an F score that is not significant,

.731 (Sig. = .681). Surprisingly, female drug and alcohol use cannot be explained by the

independent variables in this study. The eight models that follow examine the effect that

parenting styles had on males in this sample.

139 
 
 Predatory property crimes – males. The R-square value in this model was .185 (Table

63), thus suggesting that the independent variables explain 18.5% of the variance in males’

property crimes. In addition, the F score is significant, 3.536 (Sig. = .001). Consistent with the

significant F score, there is one significant independent variable: paternal neglecting/rejecting.

Male respondents were significantly less likely to commit property crimes if they were not

parented by a neglecting/rejecting male caretaker. This finding is consistent with prior research.

Table 63: Male – Predatory Property

B (SE) Beta
Constant 8.342 (3.344) -
Paternal Neglect/Reject -.168 (.084) -.218*

R-Square = .185
F = 3.536
SE = 2.883
*p ≤ .05

Predatory person crimes – males. In this model, the independent variables explain

21.3% of the variance in person-based crimes committed by males (R-square: .213; Table 64).

As expected given the significant F score (4.203; Sig. ≤ .0001), two independent variables

reached levels of significance.

Table 64: Male – Predatory Person

B (SE) Beta
Constant 2.406 (2.066) -
Paternal Permissive .114 (.056) .206*
Race -1.063 (.366) -.232**

R-Square = .213
F = 4.203
SE = 1.781
**p ≤ .01; *p ≤ .05

Once again, the race variable is significant with non-white males being more likely than

white males to commit crimes against people. Additionally, a male without a permissive father is

140 
 
significantly more likely to commit a predatory crime against another person than those raised by

some other type of male caretaker. Given the findings in this study, it appears that the permissive

parenting style has an impact on behavior that was not anticipated by looking at previous

research. Simply, results suggest that permissive fathering is indicative of a lower likelihood of

offending. A more in-depth discussion of this finding will occur in the next chapter.

Illegal services crimes – males. The independent variables for parenting style and race

explain approximately 21% of the variance in illegal service crimes committed by males (R-

square: .214; Table 65). Given the F score (4.245; Sig. ≤ .0001), it is not surprising that there is

one significant independent variable. Yet again, race is a significant predictor of offending in this

sample; non-white males are significantly more likely than white males to commit illegal service

crimes. The R-square score suggests that this is an adequate model to use when predicting the

effects of parenting style and race on minor criminal offending by males.

Table 65: Male – Illegal Services

B (SE) Beta
Constant 2.764 (1.806) -
Race -.924 (.320) -.230**

R-Square = .214
F = 4.245
SE = 1.557
**p ≤ .01

Public disorder offenses – males. Based on an R-square score of .220, one would expect

this model to be adequate to use when predicting the effects of parenting style and race on the

commission of public disorder offenses by males because 22% of the variance in public disorder

offenses can be explained by the independent variables. However, none of the independent

variables were significant predictors of public disorder offending.

141 
 
 Deviant behaviors – males. Once again, this is a fairly strong model to use in order to

predict offending because the R-square score is .204 (Table 66). This suggests that the

independent variables in the model account for 20.4% of the variance in deviant behaviors

exhibited by males. Two independent variables are significant predictors of deviant offending.

The likelihood of engaging in deviant behaviors is significantly reduced when a male is not

parented by an authoritarian male caretaker. This suggests that the presence of “drill sergeant-

type” paternal caretaker does not dissuade males from running away from home, skipping class,

cheating in school, or engaging in other forms of deviant behavior. At the same time, the

likelihood of engaging in deviant activity significantly increases if one is not parented by a

permissive female caretaker. This complements findings pertaining to permissive male

caretakers that are discussed above.

Table 66: Male – Deviant Behavior

B (SE) Beta
Constant 2.912 (1.884) -
Maternal Permissive .124 (.052) .243*
Paternal Authoritarian -.108 (.053) -.195*

R-Square = .204
F = 3.978
SE = 1.624
*p ≤ .05

Alcohol/drugs – males. The R-square value for this model (.167) is smaller than

previous R-square scores in this section. Surprisingly, with an F value that is significant, 2.288

(Sig. = .022), there are no significant independent variables.

Subscale: Alcohol/soft drugs – males. The R-square for this model is .138 (Table 67),

which suggests that nearly 14% of the variance in alcohol, marijuana, and tobacco use by males

can be accounted for by the independent variables. Only one independent variable is significant:

142 
 
permissive parenting by paternal caretakers. In this model, the likelihood that the male

respondent will use alcohol, marijuana, or tobacco significantly increases if he does not have a

permissive father figure. Again, this finding is not consistent with prior research, but it is

consistent with many findings in the current study.

Table 67: Male – Alcohol/Soft Drug Use

B (SE) Beta
Constant 1.564 (1.441) -
Paternal Permissive .099 (.038) .323**

R-Square = .138
F = 1.847
SE = 1.117
**p ≤ .01

Subscale: Hard drugs – males. The R-square value for the final model was .158 (Table

68), which suggests that 15.8% of variance in hardcore drug use by males can be explained by

the independent variables. Three of the independent variables, in particular, are significant.

Table 68: Male – Hard Drug Use

B (SE) Beta
Constant 1.700 (2.765) -
Maternal Authoritative -.166 (.078) -.249*
Maternal Neglect/Reject -.166 (.072) -.267*
Paternal Authoritative .151 (.075) .234*

R-Square = .158
F = 2.927
SE = 2.383
*p ≤ .05

Male respondents are significantly less likely to use hard drugs when they are not

parented by a neglecting/rejecting female caretaker. In an interesting twist, male respondents also

are significantly less likely to use hard drugs when they are not parented by an authoritative

mother figure, while results suggest that male respondents are significantly more likely to use

143 
 
hard drugs if they are not parented by an authoritative male caretaker. This finding suggests that

there may be an inherent difference in a child’s reaction to the use of similar parenting styles by

a male and female parent. A positive reaction to one type of parenting style used by the paternal

caretaker and a negative reaction to the same type of style used by the maternal figure is

essentially unheard of in the literature. Reasons and implications will be discussed shortly.

Summary. In concluding this section, one can easily describe the female and male

models as contrasting. Most R-square scores in the female models were small, with the largest

having a value of .119 (public disorder offense). On the other hand, the R-square values in the

male models were adequate, with the smallest having a value of .138 (alcohol/soft drugs). This

suggests that the combination of race and parenting style may be a better predictor of deviance

and criminality in men than women (three female models and two male models contained zero

significant independent variables).

Upon closer examination, parenting variables reached levels of significance on eight

occasions in the eight male behavioral models. In the female models, parenting variables only

reached significant levels two times. Permissive parenting appears to be the strongest predictor

of good behavior as it was significant in three male models (one maternal; two paternal) and one

female model. Additionally, the neglecting/rejecting parenting conformed to prior research and

continued to predict higher levels of offending in three models (two male; one female). While

neglecting/rejecting was predictive of increased offending, authoritative parenting produced

enigmatic results. In one male model (hard drug use), an authoritative paternal caretaker

predicted decreased drug use, whereas an authoritative maternal caretaker predicted increased

hard drug use by males. The authoritative style of parenting did not achieve significance with

144 
 
females in this sample. These differences will be examined in more detail in the discussion

chapter.

Research Hypotheses

Hypothesis 1: Reported Parenting Styles

The first hypothesis stated that the authoritative style of parenting would be the most

reported style of parenting across responses for both maternal caretakers and paternal caretakers.

As discussed above (Table 16), the mean scores for each of the parenting scales reveal that

respondents were most likely to be parented by authoritative caretakers; thus, the first hypothesis

was supported. Using a 5-point Likert scale, participants were asked to indicate their level of

agreement with a series of statements (“1” = “Strongly Agree;” “5” = “Strongly Disagree). The

mean score for the maternal authoritative scale was 11.58; thus over the five items in the

authoritative maternal scale the average suggests that many participants agreed (or strongly

agreed) that their maternal caretakers was authoritative. Additionally, the mean score for the

paternal authoritative scale was 12.98, which suggests that many participants agreed that their

paternal caretaker was authoritative. On the other hand, respondents were least likely to be

parented by neglecting/rejecting caretakers. The mean score for the maternal neglecting/rejecting

scale was 20.01, suggesting that many respondents disagreed with statements that portrayed their

maternal caretaker as neglecting rejecting. Similarly, the mean score for the paternal

neglecting/rejecting scale was 18.66, which was the highest average score for all paternal scales

and suggests that many participants disagreed with the notion that their paternal caretaker was

neglecting/rejecting.

When comparing reported parenting styles of university students to county jail inmates,

the university students were significantly more likely than county jail inmates to report having

145 
 
authoritative and permissive parents (both maternal and paternal) and county jail inmates were

significantly more likely than university students to report having authoritarian and

neglecting/rejecting parents (both maternal and paternal). Additionally, the maternal parenting

scale with the lowest mean score in the student sample was the authoritative scale; the paternal

scale with the lowest mean also was the authoritative scale. Amongst the county jail inmates, the

maternal parenting scale with the lowest mean was the authoritarian scale; the authoritarian scale

also had the lowest mean score of all paternal scales within the sample of county jail inmates. A

discussion of the implications of these differences is provided in the next chapter.

Hypothesis 2: Effects of Parenting on the Entire Sample

The second hypothesis stated that the authoritative parenting style would significantly

predict lower levels of deviance, delinquency, and crime amongst the entire sample.

Additionally, the hypothesis stated that the neglecting/rejecting parenting style would

significantly predict higher levels of deviant, delinquent, and criminal offender amongst the

entire sample. This hypothesis was supported to a certain extent by the analysis.

Consistent with prior research, neglecting/rejecting parenting predicted an increased

likelihood of offending in two of the dependent variable scales and both sub-scales. The presence

of maternal neglecting/rejecting caretakers predicted higher levels of commission of property

crimes, alcohol/drug use, and hard drug use. The presence of a paternal neglecting/rejecting

caretaker predicted an increased use of alcohol, tobacco, and marijuana throughout the entire

sample.

The parenting type that was most predictive of decreased offending was the permissive

style. Paternal permissiveness was significantly predictive of lower levels of person-based

crimes, as well as public disorder offenses and alcohol, tobacco, and marijuana use. It also

146 
 
should be mentioned that the sex of the respondent was significantly predictive of all six

dependent variable scales and both sub-scales; within this sample, males were more likely than

females to take part in each type of deviance, delinquency, and crime. Additionally, non-whites

were significantly more likely than whites to engage in property and person-based crimes,

commit illegal service and public disorder offenses, and display deviant behaviors. Implications

for these findings are discussed in the next chapter.

Hypothesis 3: Parenting Effects on University Students and County Jail Inmates

The third hypothesis stated that authoritative parenting would have the most significant

effect on the behaviors of university students and that neglecting/rejecting parenting would have

the most significant effect on the behaviors of county jail inmates. Results of the multiple

regression analyses conducted on the sample of university students suggest that authoritative

parenting had no significant impact on that group. After examining the effects that parenting

style had on the university student’s engagement in behaviors included in the six dependent

variable scales and both subscales, it is noticeable that authoritative parenting is not significantly

predictive of involvement, or non-involvement, in these behaviors. This is a surprising finding

given the wealth of research which has suggested that authoritative parenting is the best predictor

of reduced involvement in deviant, delinquent, and criminal behavior. The neglecting/rejecting

hypothesis was not a significant predictor of county jail inmate involvement in any of the

deviant, delinquent, or criminal behaviors. Based on these results, however, as well as those for

the university students, there is mixed support for the third hypothesis and a number of

interesting findings that are discussed in the following chapter.

147 
 
Hypothesis 4: Parenting Effects on Females and Males

The fourth hypothesis stated that female and male participants would experience similar

outcomes when parented with similar parenting styles. For example, participants from both

genders would be less likely to engage in deviant, delinquent, and criminal activities if parented

by an authoritative caretaker. Additionally, those parented by a neglecting/rejecting caretaker

would be more likely to engage in deviance, delinquency, and crime. Again, results were mixed

and there was limited support for this hypothesis.

The two parenting types that produced similar behaviors across both males and females

were neglecting/rejecting and permissive parenting. Amongst the female sample, the

neglecting/rejecting parenting type predicted an increased likelihood of committing predatory

property crimes. In comparison, results from the male sample suggest that the presence of a

neglecting/rejecting paternal caretaker increased respondents’ engagement in property crimes, as

well as hard drug (no parenting types predicted female drug use).

Permissive paternal parenting also had similar effects across the female and male

subsamples, although the effects were not felt on the same types of offenses. This type of

parenting was predictive of decreased involvement in crimes against people, deviance, and

alcohol/soft drug use by males. Permissiveness also predicted decreased involvement in public

disorder offenses by female respondents. There were no other similarities between females and

males, thus the conclusion that limited support was achieved for this hypothesis.

Hypothesis 5: Parenting Effects on Whites and Non-Whites

The fifth hypothesis stated that whites and non-whites would experience similar

outcomes when parented with similar parenting styles with the exception of the authoritarian

parenting style. Based on prior research findings, it was hypothesized that non-white participants

148 
 
would be less likely than white participants to engage in deviant, delinquent, and criminal

behaviors when parented by an authoritarian caretaker. Unfortunately, the sample did not contain

enough racial variance, thus the researcher was not able to directly examine this hypothesis.

However, race is still an important variable in these analyses, as evidenced by its significance in

a number of the models discussed above.

Summary

The results from the analysis provide a number of interesting, as well as unexpected,

findings. In terms of scale development, the results maintain support for the neglecting/rejecting

scale. Thus, the addition to Buri’s (1991) Parental Authority Questionnaire has been reliable and

valid as a ten-question scale and as a five-question scale. This development should aid in future

studies of the effectiveness of certain parenting styles.

Results from the multiple regression analysis were a bit more unexpected, however. For

example, the presence of a permissive caretaker significantly reduced the likelihood that a

member of the sample engaged in: crimes against people, public disorder offenses, and soft

drugs. The presence of a permissive caretaker also reduced the likelihood of public disorder

offenses amongst students and females, soft drug use by students and males, and person-based

crimes and deviant behavior by males. Prior research (Palmer & Hollin, 2001; Querido et al.,

2002; Schroeder et al., 2010; Vieno et al., 2009) suggests that children with permissive parents

should be more likely to engage in deviance, delinquency, and criminality. The effects of the

permissive parenting style proved to be unexpected throughout the entire study. Reasons and

ramifications of this divergence from prior research are discussed below.

For the most part, however, the results were expected. Neglecting/rejecting parenting was

predictive of increased participation in crime and deviance. Authoritative parenting, in the few

149 
 
times that it was significant, was predictive of decreased participation in criminal and deviant

activity. A discussion of all of these findings and their implications, as well as directions for

future research, is presented in the next chapter.

150 
 
 CHAPTER VI

DISCUSSION AND CONCLUSION

Over time, a number of criminologists have attempted to theorize the effects that

parenting has had on delinquent and criminal behavior. As discussed in earlier chapters, leading

theorists such as Hirschi (1969), Gottfredson and Hirschi (1990), and Moffitt (1993) have all

explored different aspects of the link (both direct and indirect) between the behaviors of parents

and the resultant behaviors of their children. For example, Hirschi (1969) suggested that the

parent-child bond influenced the child’s level of delinquency and criminality later in life. Later,

Gottfredson and Hirschi (1990) theorized that low self-control in children is, in part, the result of

poor parental management techniques. Together, these two theories suggest that the level of

discipline exhibited by parents toward their children and the nature of the bond between parent

and child have long-term effects on the behavior – either delinquent and criminal or non-

delinquent and non-criminal – of the child. The focus of the current research project, however,

was on Baumrind’s (1967, 1971, 1972) singular typology that focused on both of these aspects of

parenting (demandingness and responsiveness).

According to Baumrind’s typology, a blend of both demandingness and responsiveness is

critical for successful parenting. A demanding parent is one who uses firm, yet non-coercive,

control, provides effective monitoring and supervision, and applies consistent patterns of

discipline for wrong-doing and praise for good behaviors (Baumrind, 1996). A responsive parent

is one who expresses love, warmth, and has open communication and dialogue with their child

(Baumrind, 1996). As detailed previously, these two concepts have been used to formulate

Baumrind’s typology of parenting styles, which includes: authoritarian, authoritative,

neglecting/rejecting, and permissive.

151 
 
 Limited prior research has focused on the link between Baumrind’s parenting styles and

childhood behaviors. As discussed in Chapter Two, most of these research studies have focused

on three (authoritarian, authoritative, and permissive) of the four parenting types, while

disregarding the importance of the neglecting/rejecting type. Results from these studies suggest

that children with authoritative parents are the least likely to engage in deviant, delinquent, or

criminal acts (see Appendices A, B, C, and D). The few studies that focused on the

neglecting/rejecting parenting style all suggested that children exposed to that type of parenting

were likely to engage in acts of deviance, delinquency, and criminality (Avenevoli et al., 1999;

Baumrind, 1991; Chassin et al., 2005; Lamborn et al., 1991; Patock-Peckham & Morgan-Lopez,

2007; Steinberg et al., 1994, 2006). Additionally, studies focusing solely on authoritarian,

authoritative, and permissive parenting found that the authoritarian and permissive styles of

parenting were positively related to deviant, delinquent, and criminal behavior (Palmer & Hollin,

2001; Querido et al., 2002; Schroeder et al., 2010; Vieno et al., 2009). One last study reported

that children of authoritarian and permissive parents were less likely than children of

neglecting/rejecting parents to use tobacco (Chassin et al., 2005). Given these research findings,

the results of the current analyses are somewhat surprising, yet intriguing. This chapter provides

a discussion of: the research findings (related to parenting styles as well as scale development),

the implications that this research has for families and policy-makers, the strengths and

limitations of the research, and directions for future research.

Discussion of Research Findings

The results from the current study suggest a number of things. First, they suggest that the

extended Parental Authority Questionnaire is a reliable and adequate measure of Baumrind’s

typology of parenting styles. The findings also suggest that the permissive and

152 
 
neglecting/rejecting parenting styles had the most impact on the delinquent and criminal

behavior of the respondents who comprise this sample and further, that the authoritative

parenting style had very little impact. Aside from parenting style, the results also suggest that

race and gender were significant predictors of behavior. The implications of these findings are

discussed below.

Extended Measurement Instrument

As discussed, Buri’s (1991) Parental Authority Questionnaire (PAQ) has effectively

measured Baumrind’s typology since its introduction into the literature two decades ago. To

reiterate, the PAQ contains 60 items and asks respondents to report their level of agreement or

disagreement with those items. Thirty of the 60 items measure maternal styles of parenting and

the remaining 30 items measure paternal styles of parenting. Per each parent, ten of the items

describe the authoritarian style, ten describe the authoritative style, and ten describe the

permissive style. Given the fact that Baumrind’s typology defines four types of parenting, it was

important to devise a measurement instrument that captured each parenting style.

A neglecting/rejecting scale was created to measure maternal parenting and paternal

parenting. After conducting a pre-test and analyzing the results, the internal consistency score for

the maternal neglecting/rejecting scale was .86 and the paternal neglecting/rejecting score was

.90. According to DeVellis (2003), each of these scales was considered very good. Due to

problems with the response rate of the pre-test and informal feedback from some respondents

about the length of the pre-test, the decision was made to shorten the PAQ from 80 items (40

items per parent) to 40 items (20 items per parent). A split-half reliability analysis was performed

in order to determine which five items from each scale provided the best reliability score; the

process that was used is discussed in Chapter Three.

153 
 
 As discussed in Chapter Four, the internal consistency scores for the PAQ measures in

the current study compare favorably with the internal consistency scores generated during the

pre-test. Given the fact that the scales remained “respectable” and “very good” it is important

that researchers consider using these extended measures when researching Baumrind’s typology

for the simple reason that consistent scales need to be used when measuring these concepts; one

of the limitations of prior literature is the fact that measures used to gauge parenting styles are

inconsistent. Granted, the measurement items created for this research project only have been

used in two studies at this point, but they have already proven themselves to be reliable and valid

measures of the neglecting/rejecting parenting style. It is important that they continue to be

utilized, and tweaked as necessary (especially the new neglecting/rejecting scale), in future

research.

Impact of Neglecting/Rejecting Parenting

One of the most expected results that was uncovered in the current study was the

relationship that neglecting/rejecting parenting had with delinquency and criminality. To

reiterate, across the entire sample, maternal neglecting/rejecting was significantly related to an

increased likelihood of property crime and alcohol/drug use, especially hard drug use. Across the

same sample, paternal neglecting/rejecting also was significantly related to an increased

likelihood of alcohol, tobacco, and marijuana use. These results support previous results

discussed earlier in this chapter.

Based on the conceptualization of demandingness and responsiveness provided by

Baumrind (1996), Maccoby and Martin (1983), and Simons et al. (2004), it is argued that Buri’s

(1991) operationalization of Baumrind’s typology and the operationalization of

neglecting/rejecting parenting used in this study are just as appropriate as measurement

154 
 
instruments used in the past. According to Baumrind (1996), neglecting/rejecting parents are

neither warm nor confrontational, do not supervise, nor consistently discipline their children, and

do not communicate expectations that they have for their children. Consequently, children raised

by neglecting/rejecting parents are theorized to exhibit problematic behaviors. The

neglecting/rejecting parental style is the polar opposite of the authoritative type. The

neglecting/rejecting scale developed and used in this study is conceptually sound in the way in

which it is correlated with the other parenting types, and as reported it is also statistically sound.

As expected, the neglecting/rejecting style was the least reported parenting type by the

respondents in this sample, but it had a very significant impact on the results.

Impact of Permissive Parenting

Permissive parenting achieved statistical significance the most frequently in the analyses.

Throughout the analyses (entire sample, university students, county jail inmates, females, and

males) there were 9 instances in which either maternal or paternal permissiveness was a

significant predictor of reduced deviant, delinquent, or criminal behavior; the relationship and

direction was not expected. As discussed in Chapter Five, a plethora of prior research has

suggested that individuals raised by permissive parents are more likely to engage in delinquency

and criminality. Earlier research has suggested that it is not in the best interests of either parent

or child when the child-parent relationship is treated as a friendship by the parent. These earlier

findings suggested that parents must set rules, guidelines, and expectations for their children and

be consistent in disciplining the child.

However, results from the current study suggest that permissive parenting has a positive

impact on behavior. These findings are interesting given the fact that county jail inmates in this

sample were not significantly impacted by permissive parenting, while university students were

155 
 
significantly impacted in regards to two types of crime. Perhaps this suggests that a shift has

occurred in the way that youths respond to permissive parenting. As discussed, past research

suggested that youths respond poorly to the lack of discipline displayed by permissive parents,

but the results from the current study may point to shift in youths’ responses to this type of

parenting. Instead of reacting negatively due to a lack of discipline, it is possible that children are

beginning to respond more positively to the attention bestowed upon them by their parents and

the interactions that they have with their parents. More often than not, children (even college-

aged children) rely on their parents for sustenance, shelter, and money. Current results suggest

that individuals understand their reliance on their parents and may not want to do anything to

disappoint or lose the support of their parents. Still, this is an extremely interesting and

unexpected finding and one that future research must continue to probe.

Impact of Authoritative Parenting

Aside from the findings related to neglecting/rejecting parenting, the lack of significant

findings relating to authoritative parenting is the most surprising result in the current study. A

bevy of prior literature (see Appendices A, B, C, and D) has suggested that authoritative

parenting is the most influential parenting type and that children raised by authoritative mothers

and fathers are the least likely to engage in acts of deviance, delinquency, and crime. There are

only two instances in the current study in which authoritative parenting significantly reduced the

likelihood of respondents engaging in the behavior asked: alcohol/soft drug use by county jail

inmates (authoritative female) and hard drug use by male respondents (authoritative male). More

surprisingly, an authoritative mother was predictive of an increased likelihood that a male

respondent would use hard drugs.

156 
 
 Again, the reason for these findings may be difficult to ascertain, but it appears one of the

primary reasons is the measurement instrument used. The lone study discussed in Chapter Two

to explicitly use Buri’s PAQ was conducted by Patock-Peckham and Morgan-Lopez (2007). The

authors analyzed patterns of alcohol use in a sample of 441 college students at Arizona State

University. According to their results, maternal parenting styles were not significantly related to

alcohol use, whereas the presence of an authoritative father helped decrease alcohol use for

males and females (Patock-Peckham & Morgan-Lopez, 2007). There are some similarities

between this study and the current study: university students are surveyed and maternal

authoritativeness is not a significant predictor of student behavior in either study. Thus, there is

some precedent that cautions researchers against expecting authoritative parenting to reduce

problem behaviors.

It is wholly different, however, to expect authoritative parenting to lead to an increase in

problem behaviors. But, when examining the true nature of authoritative parenting, perhaps one

should not be so quick to assume that it will lead to the best outcomes. To reiterate from Chapter

Two, the authoritative parent welcomes a verbal give and take with the child, solicits objections

from the child, explains the reason for certain rules, and encourages the child to be autonomous

(Baumrind, 1966, 1968, 1971). It is not farfetched to think that a child who has been raised in a

nurturing environment in which warmth and reciprocity abound would be compelled to act upon

the autonomy instilled within oneself (by the parent, no less) and experiment with alcohol and

drugs. As discussed in this sample, males with authoritative mothers are the most likely to use

hard drugs. This does not necessarily mean that authoritative mothers are bad or that male

children are more likely to rebel against their authoritative mother; prior research has stated that

the mere presence of an authoritative parent removes the child’s desire to rebel (Simons &

157 
 
Conger, 2007). Simply, it may mean that the male child’s relationship with an authoritative

mother is one that encourages discussion about drugs and alcohol, experimentation,

consequences, and learning from those consequences. Based on the results, this potential chain of

events is limited to substance use and does not lead to serious or violent crimes against people or

property. Interestingly, the fact that paternal authoritativeness predicts a reduced likelihood of

male respondents using the same types of alcohol and hard drugs may suggest that paternal

discipline carries more clout than maternal discipline and that male children do not want to

disobey their fathers.

Summary of the impact of parenting types. It appears a few things are clear based on

the interesting findings discussed above. First, permissive parenting may have the best

consequences for the child, especially those who participated in this study. Keep in mind, this

does not mean the permissive parenting type is ideal. It simply implies that the respondents in

this sample who reported being raised by permissive parents were not willing to let the lax

discipline that they received from their parents negatively impact them. On the contrary, they

were able to resist the pulls of minor deviance, delinquency, and crime despite the fact that they

did not have parents who were willing to set rules or guidelines for behavior.

Second, children exposed to neglecting/rejecting parents were at as great of a risk as first

thought. Essentially, neglecting/rejecting parents are unconcerned with discipline and

communicating with their children. The failure to instill discipline, provide a solid foundation,

and simply be there for the child is likely to lead alcohol and drug use, general deviance, minor

delinquency and crime, as well as serious and violent criminal behavior. Parents need to be there

for their children. Additionally, this is not to say that the individuals in this study did not have

proper guidance from somebody, they simply indicated that they did not receive it from their

158 
 
neglecting/rejecting parents. Future research should make a better effort to identify who, if

anybody, provided guidance for children in these situations; it could have been provided by a

teacher, a coach, a neighbor, or even a friend.

Finally, the authoritative type of parenting, when measured using the PAQ, may not be as

idyllic as expected. Granted, there were a handful of instances when maternal authoritativeness

predicted a reduced likelihood of alcohol and drug use. But, maternal authoritativeness predicted

an increased likelihood of alcohol and drug use, especially in males, which was surprising.

Recent research (Fearon, Bakermans-Kranenburg, van Ijzendoorn, Lapsley, & Roisman, 2010)

suggests that a strong mother-son attachment reduces the likelihood that the child will exhibit

behavioral problems. While attachment is not synonymous with parenting style, there are

similarities between parent-child attachment and authoritative parenting that suggest a warm and

loving relationship between the parent and the child. Perhaps future research should take an

integrated approach and examine Baumrind’s typology along with the attachment aspect of

Hirschi’s social bond theory. Furthermore, the current research suggested that authoritative

parenting did not reduce the likelihood of committing criminal and delinquent acts, as it was

expected to do. This may be a result of using the PAQ, it may be a result of the particular sample

that was analyzed, or it may be indicative of the true nature of authoritative parenting. Continued

research and further scrutiny of these ideas is absolutely necessary.

Impact of Race and Sex

Throughout much of the current research, sex was a significant predictor of behavior.

When examining the entire sample of respondents, males were significantly more likely than

females to engage in every type of crime and alcohol/drug use category. Given the disparity of

male offenders to female offenders this was hardly surprising. But, when looking solely at the

159 
 
county jail inmates, gender was predictive of only two crime categories – property crimes and

public disorder. This suggests that the gender of the county jail inmates in this sample does not

explain the majority of delinquent and criminal behaviors that they exhibited. On the other hand,

males in the university sample are much more likely than females to commit any number of

offenses, including property crimes, crimes against people, illegal services and public disorder

offenses. Perhaps this simply means that the female university students who participated in this

research are not likely to commit criminal or delinquent acts, although other explanations are

possible and may be uncovered with continued research.

Given the somewhat similar effects that parenting style had on males and females in this

sample, it also could mean that there is an inherent difference – other than parenting – between

female and male university students that causes males to engage in criminogenic behaviors. One

possible explanation may be the different ways that male students and female students cope with

stress and anxiety. Recent research conducted amongst a college sample of 679 students found

that male students were much more likely to utilize negative coping strategies, such as

drug/alcohol use and criminal activity, to deal with stress brought upon by things such as text

anxiety (Bowers, Bowen, Huck, Lee, & Spraitz, 2011). Overall, however, the idea that males are

more likely than females to commit crime is neither new nor novel; the results simply support

prior knowledge and research.

Accurately gauging the impact of race was much more difficult in the current study. Out

of the 391 respondents who identified their race, only 70 were non-white (18.1%). Of the 70

non-white participants, 38 (54.3%) were county jail inmates and 32 (45.7%) were university

students. Given such a small sample of non-white participants, any conclusions must be viewed

carefully, if not skeptically. As reported in Chapter Four, white respondents reported

160 
 
significantly fewer acts of deviance, delinquency, and criminality; being non-white was a

significant predictor of an increased likelihood to engage in five criminal/delinquent behaviors

(race was not a significant predictor of any type of drug or alcohol use). Again, there were more

county jail inmates than university students in the non-white sample, thus it is possible that the

results for the total sample may be skewed.

When examining the sample of university students, race was not predictive of deviance,

delinquency, or criminal behavior. Non-white county jail inmates were more likely to engage in

crimes against people. These findings from the sub-samples provide some support for the results

from the total sample and from recent research. Ghazarian and Roche (2010) found that African

American and Hispanic youth who had engaging mothers (essentially, a proxy for high

responsiveness) were less likely to engage in delinquent behavior. In the current study, to

reiterate, non-white respondents were significantly more likely than white respondents to have

authoritarian mothers (low responsiveness). Thus, there is support for the notion that

authoritarian mothering may not be beneficial for non-whites; meanwhile, the current study was

not able to completely analyze the impact that authoritarian fathering has on non-whites.

White county jail inmates were more likely to use alcohol and all types of drugs. There

may be a number of reasons why white inmates were more likely to engage in substance use.

First, non-white offenders may be more likely to underreport drug use. A plethora of research

has suggested that minority offenders underreport substance use relative to white offenders

(Falck, Siegal, Forney, Wang, & Carlson, 1992; Fendrich & Xu 1994; Katz, Webb, Gartin, &

Marshall, 1997; Kim, Fendrich, & Wislar, 2000; Rosay, Najaka, & Herz, 2007). But, additional

research suggests that whites underreport more than non-whites (Lu, Taylor, & Riley, 2001;

161 
 
McNagny & Parker, 1992). Based on the varied findings, one must keep in mind the self-report

bias associated with drug use amongst samples of offenders.

Second, these results may be caused by the crossover effect. The crossover effect states

that whites are more likely to use drugs and alcohol while younger than 35-years-old and

minorities are more likely to use drugs and alcohol while older than 35-years-old (Watt, 2008).

The average age of the county jail inmates in this sample was 33.16, thus it is possible that if the

same sample of county jail inmates were surveyed two years from now the results may be

different. Recent research supports the idea of a crossover effect by reporting that minority

populations are less likely than white populations to use alcohol and drugs when under the age of

35 (French, Finkbiner, & Duhamel, 2002; Rote & Starks, 2010; Watt, 2004), while reporting that

some minority groups have higher rates of substance use when over the age of 35 (French et al.,

2002). The notion that minority offenders underreport rates of drug and alcohol usage, as well as

the crossover effect, are important ideas that must be kept in mind when examining self-reported

alcohol and drug use in the future.

Implications for Parents

Because this study focused on the effects that parenting had on behavior, this section will

focus on the parenting implications associated with the findings. It is important to keep in mind,

as Steinberg et al. (1994, p. 758) pointed out, that Baumrind’s typology “is a theory about types,

not about specific parenting practices.” Thus, any recommendations for parents that are derived

from the current analysis must go beyond addressing simple parenting behaviors. Instead, parents

must seek to either reinforce or recreate the type of parenting style that they employ while raising

their children.

162 
 
 The authoritative parenting type is discussed extensively in previous literature.

Essentially, this is touted as the best parenting style and research supports that claim (see

Appendices A, B, C, and D). But, the current research has uncovered potential shortcomings of

authoritative maternal parenting relative to the use of drugs. Given the desire of authoritative

parents to be seen as disciplinarians who provide guidance, feedback, and are open to discussion

and debate, it is imperative of them to educate their children in the perils of substance abuse. It is

understandable, and perhaps even acceptable in today’s society, that adolescents and young

adults will experiment with tobacco, alcohol, and marijuana (in addition to other types of drugs);

experimentation allows for autonomous decision-making, which is a hallmark of the

authoritative parenting type. But, those parents who preach discretion and autonomy also must

be prepared to have honest discussions with their children about drugs and alcohol.

In addition to educating their children about substance abuse, the authoritative parent

must be willing to scrutinize their children. It can be argued that increased scrutiny does not

comply with the tenets of the authoritative parenting type. But, an authoritative parent is not just

a sounding board and source of communication. Rather, an authoritative parent also relies on

discipline, direction, positive expectations, and proper guidance in order to achieve goals. In

order to properly guide, the authoritative parent must scrutinize. Simply, discretion and

autonomy can be given to the child, but the parent must know when to limit the autonomy of

their adolescent and teenage children. When substance use borders on substance abuse it is time

for the parent to draw the line.

The permissive parenting type has been discussed extensively in this chapter as a non-

problematic parenting style and one that increases conformity, but the potential negative

ramifications of employing this parenting technique, as suggested by a plethora or prior research,

163 
 
must be discussed once more. Simply, permissive parents seem to believe that they cannot

simultaneously provide discipline for their children and shower them with love and affection. As

seen, the desire to be a friend and not a parent has disastrous results. Parents must be made aware

that a lack of discipline, guidance, ground rules, and expectations has deleterious consequences

for their children. Permissive parents must be told that their children are more likely than

children who are parented with a more balanced approach of discipline and love to commit acts

of deviance, delinquency, and crime in part due to their style of parenting. Once they are made

aware of the potential harm that they are causing their children and the potential gamesmanship

techniques that their children are using, they must be willing to embrace and implement a new

philosophy of parenting.

Given the findings related to neglecting/rejecting parenting, it remains irresponsible to

employ this style of parenting. Additionally, it seems nearly impossible that a

neglecting/rejecting parent would make wholesale changes to their style of parenting. Instead,

attention should be given to children with this type of parent. They must be given the guidance

that they lack. In addition, researchers must continue to examine the effects of this particular

parenting style. This should be more easily accomplished with the newly created

neglecting/rejecting measurement scale. Obviously, further research into this particular type of

parenting only represents one direction for future research on Baumrind’s typology of parenting

styles.

Strengths and Limitations

Strengths

The current study sought to expand the body of knowledge on the effects of parenting

styles in a number of ways. First, it was necessary to develop a standardized measurement

164 
 
instrument in order to completely measure Baumrind’s typology of parenting styles. Granted,

Buri (1991) introduced a respectable, if not very good, Parental Authority Questionnaire (PAQ)

in an effort to gauge the impact that authoritarian, authoritative, and permissive parenting has on

children. In order to examine the impact of neglecting/rejecting parenting, an additional scale

was created, tested, and analyzed to ensure that it measured the fourth parenting style. As

discussed in Chapter Three, a survey instrument containing ten neglecting/rejecting items

measuring maternal and paternal parenting was administered at pre-test; internal consistency

scores were respectable and very good. Each ten-item scale was shortened to five items for the

current study in order to reduce response times and increase response rates. The shortened scales

achieved similar internal consistency scores to the lengthier original scales. With additional

research, this scale has the potential to make an important contribution, and the current study

could not have been completed sufficiently without the creation of the neglecting/rejecting scale.

Second, the creation of the new scale allowed the researcher to test the effects that

neglecting/rejecting parenting had on the participants. As discussed in Chapter Two and above,

very few studies have examined the impact of Baumrind’s fourth parenting type. The importance

of assessing the entire typology cannot be understated; the current study has added to the limited

research on Baumrind’s entire typology.

Third, the current research is only one of a handful of studies to introduce aspects of

criminal behavior into the study of Baumrind’s typologies. Each of the earlier studies that have

examined criminal behavior in this way have been limited either by sample size (Haapasalo,

2001; Palmer & Gough, 2007) or by measurement instrument (Schoeder et al., 2010). Haapasalo

(2001) surveyed 89 male inmates in a Finnish prison and Palmer and Gough (2007) asked 71

young men to complete a questionnaire – 40 were prior offenders and 31 were non-offenders.

165 
 
Schroeder et al. (2010) had a respectable sample (n = 662), but their measurement instrument

only had seven items measuring demandingness and ten items measuring responsiveness. These

efforts have laid the foundation for the current study, which has a sample size of 409 respondents

(298 university students and 111 county jail inmates). While it would be naïve to claim that the

average college student has not engaged in any criminal activities in his or her life it also would

be obvious to claim that the average county jail inmate has a more extensive criminal history

than the average non-inmate, thus the use of both samples was needed in order to assess the

impact of parenting styles on criminal behavior. Additionally, the inclusion of each sample

allowed the researcher to compare effects that parenting has had on college students with the

effects that parenting has had on jail inmates.

Finally, the researcher was able to generate a random sample of university students to

distribute the questionnaire to via Qualtrics, an emerging web-based survey interface. A random

sample of 2,000 undergraduate students was generated from the entire undergraduate population.

Using Qualtrics, the researcher was able to send an email containing the informed consent form

and link to the survey to everybody who was randomly selected to participate. Weekly reminders

were sent to those who had not completed the survey. Instead of randomly selecting classes to

survey, the use of a web-based tool allowed the researcher to randomly select individual

respondents and limit waste generated by the use of paper questionnaires. The use of technology

also allows researchers to draw samples of respondents from multiple locations at once, which

should help enhance future research efforts.

Limitations

There were a handful of limitations associated with the current study. First, the sample

was generated from one mid-sized state university in the northeast and two county jails in the

166 
 
same region. The use of students from one university limits the generalizability of the findings.

In addition, jail administration at both facilities limited the researcher’s access to inmates with

the lowest security risks. Thus, the researcher was not able to survey inmates with more

extensive or violent criminal histories. The ability to replicate this study at other universities and

in other correctional settings throughout the United States undoubtedly would increase the

generalizability of the findings.

A second limitation centers on the sample of inmates who participated. As mentioned, the

interaction with inmates was limited to those who posed the lowest level of risk. In order to be

non-coercive, participation was voluntary, thus the sample of 111 inmate respondents is a

convenience sample. Given the research-related protections afforded to incarcerated individuals

in the United States this was not unexpected. Perhaps in the future more time should be devoted

to meeting with inmates prior to survey distribution in order to explain the project and answer

questions about it. For example, one jail administrator mentioned that many inmates did not want

to participate because they were not comfortable with their reading comprehension skills despite

the administrator’s reassurance that the researcher would go through the entire survey with the

inmate. Situations like that could be resolved beforehand with reassurances from the researcher;

it is definitely something to keep in mind as this research grows.

An additional limitation is the lack of diversity in the sample. Of the 391 respondents

who reported their race, approximately 82% (n = 321) were white. Given the demographics of

the university, which at the time reported a student population that was 87% white (Fact Sheet,

Fall 2010), it was not surprising that nearly 89% (n = 252 out of 284 reported) of the student

sample was white. Nearly 65% (n = 69 out of 107 reported) of the county jail inmate sample was

white, which was higher than the mid-year estimates generated by the Bureau of Justice Statistics

167 
 
that suggested local county jails were 44.3% white, 37.8% black, and 15.8% Hispanic (Minton,

2011). Due to this lack of racial diversity, the researcher was unable to make race-based

comparisons regarding the effects of parenting styles; future research must take this limitation

into consideration and correct it accordingly.

Another limitation of the current study was the overrepresentation of female respondents.

As discussed in Chapter Four, over 70% of the student sample was female despite a university

population that was approximately 57% female (Fact Sheet, Fall 2010). Again, this may be due

to women being more cooperative or attuned to the subject matter. Additionally, females

represented 32.4% of the county jail inmate sample. Minton (2011) noted that at mid-year 2010,

females comprised only 12% of all county jail inmates. This overrepresentation can be attributed

to a number of things, including: the convenient nature of the sample, the willingness of

incarcerated female respondents to discuss ideas related to parenting with the researcher, and the

fact that women in the rural county jail were allowed to leave their housing unit in order to

participate in the research. Although the overrepresentation of female respondents represents a

limitation it also is an opportunity to move the research forward. As discussed in Chapter Two,

very few studies that have examined the link between parenting style and

delinquency/criminality have used female participants; the current study should begin to fill in

some of those gaps.

Directions for Future Research

Future studies on the link between Baumrind’s typology of parenting styles and deviant,

delinquent, and criminal behavior must divert from the current research in a number of ways.

First, future research efforts must be made in other geographical regions. The current research

focused on one state university and two county jails in the northeast. This simply needs to be

168 
 
expanded. Research efforts must be implemented at state and private universities throughout the

country. Access must be sought at a variety of correctional facilities. One thing that limited the

current research was the fact that most of county jail inmate sample were incarcerated for minor

offenses. In order to access a more serious and violent criminal population efforts must be made

to get into state and federal prisons. Obviously, access to other municipal and county jails also

should be sought.

In addition to surveying inmate populations at a variety of correctional institutions, future

research should focus on specific types of offenders. Given the fact that there is little research

pertaining to Baumrind’s typology with criminal offenders, focusing on specific types of

offenders represents a proverbial gold mine of information. It will be interesting to examine how

white collar criminals, murderers, and sex offenders were parented. On the other hand, the

general population should also be included in future research. Surveying specific communities,

cities, and states will allow researchers to determine if the effects that parenting has on criminal

behavior are simply regional or if there are far-ranging implications of parenting styles.

A concerted effort to target more universities and correctional facilities in other areas

should increase the racial diversity of the sample population. According to Ryu (2010),

minorities make up approximately 30% of the population at two-year and four-year colleges and

universities compared to 13% of the population at the university used in the current study.

Additionally, 35% of the current county jail inmate sample was minority, whereas 55% of county

jail inmates nationwide are minority (Minton, 2011). Hopefully, expanding the racial diversity of

the sample will allow the researcher to accomplish things that could not be accomplished given

the racial homogeneity of the current sample.

169 
 
 Early research by Baumrind (1972) claimed that authoritarian parenting benefited African

American females, while more recent research (Avenevoli et al., 1999; Steinberg et al., 2006)

has claimed that authoritarian parenting is not as detrimental to African American children as it

may be to children with other racial backgrounds. Additionally, Schroeder et al. (2010) claimed

that African American children with permissive parents are in the most precarious situation and

are more likely than other African American children to engage in delinquency and crime.

Walker et al. (2007) reported a variety of findings on the impact of parenting styles on Hispanic

and African American youth, most notably that medium levels of parent-child attachment

mitigated delinquency in African American youth while low and high levels predicted

delinquency. Due to the lack of racial diversity in the current sample, the researcher was unable

to substantiate or refute findings from these earlier studies.

Finally, in addition to examining the effects that parenting style has on behavior in

various racial groups, an increasingly diverse sample should allow the researcher to study how

the changing cultural definition of “family” impacts parenting and behavior; data were collected

in order to achieve this goal, but this line of research was outside the scope of the current

analysis. When Baumrind first began her research at Berkeley in the mid-1960s the prototypical

family was made up of a husband and wife and their children. Over the past four decades the

definition of “family” has changed considerably. Currently, married or unmarried couples caring

after their own children, or step children, or adopted children comprise a typical family.

Grandparents raising their grandchildren; aunts and uncles raising their nieces and nephews;

adult brothers and sisters raising their adolescent brothers and sisters comprise a typical family.

Simply, researchers must take these changing definitions into account when examining the

impact that parenting has on criminal behavior.

170 
 
 Conclusion

Seasoned parents and new parents, alike, often joke that parenting does not come with a

manual or a set of rules. Most of the time parents hope to provide the best example for their child

to follow and silently pray that the child does not get into trouble. Although one should not

characterize Baumrind’s typology of parenting styles as a rulebook for parents, it should be

thought of as a guide. Again, Baumrind’s typology does not discuss specific parenting behaviors

nor does it provide a “dos” and “don’ts” list of how to parent. The typology simply defines

parenting types and attempts to lay a foundation of ideal parenting methods.

The current research examined the impact that Baumrind’s four parenting types –

authoritarian, authoritative, permissive, and neglecting/rejecting – had on certain behavioral

outcomes and the mixed results suggest that a parenting manual would do little good.

Neglecting/rejecting parenting was the most harmful type of parenting. For the most part,

neglecting/rejecting parents will eschew disciplining or communicating with their children.

According to the findings of this study and previous research, this leads to an increased

likelihood of property crime as well as alcohol, tobacco, marijuana, and other hard drug use.

Given these results, it is imperative that parents strive to instill discipline in their children while

at the same time creating and maintaining warmth for their children.

Surprisingly, however, the permissive parenting style predicted a reduced likelihood of

offending in most instances. This finding contradicted most of the prior research which has

suggested that parents who fail to instill discipline wind up with children who exhibit a multitude

of behavior problems. Given this surprising finding, more research is needed into the permissive

parenting type because it would be careless to recommend that parents provide zero guidance

and rules for their children.

171 
 
 Finally, the necessity of cross-disciplinary research cannot be understated. Baumrind’s

typology originated in the discipline of psychology and has remained there, save for a limited

number of researchers. As criminologists have slowly begun to introduce the typology into

studies of delinquency and crime, those who study parenting have gained greater insight into

some of the affects that different styles of parenting have on behavioral outcomes. Ideally,

academicians, practitioners, policy-makers, parents, and children should benefit from the

increased integration of parenting styles steeped in psychology and behaviors based in

criminology.

172 
 
 References

Akers, R.L., & Sellers, C.S. (2009). Criminological theories: Introduction, evaluation, and

application. New York, NY: Oxford University Press.

Allegheny County Pennsylvania. (n.d.). County jail: Volume of activity. Retrieved from

http://www.alleghenycounty.us/jail/volume.aspx

Applegate, J., Burke, J., Burleson, B., Delia, J., & Kline, S. (1985). Reflection-enhancing

parental communication. In I.E. Siegel (Ed.), Parent belief systems: The psychological

consequences for children (pp. 107-142). Hillsdale, NJ: Lawrence Erlbaum Associates.

Avenevoli, S., Sessa, F.M., & Steinberg, L. (1999). Family structure, parenting practices, and

adolescent adjustment: An ecological examination. In E.M. Hetherington (Ed.), Coping

with divorce, single parenting, and remarriage: A risk and resiliency perspective (pp. 65-

90). Mahwah, NJ: Lawrence Erlbaum Associates.

Bachman, R., & Schutt, R.K. (2010). The practice of research in criminology and criminal

justice (4th ed.). Los Angeles, CA: Sage.

Barber, B.K., Olsen, J.E., & Shagle, S.C. (1994). Associations between parental psychological

and behavioral control and youth internalized and externalized behaviors. Child

Development, 65, 1120-1136.

Baumrind, D. (1966). Effects of authoritative parental control on child behavior. Child

Development, 37, 887-907.

Baumrind, D. (1967). Child care practices anteceding three patterns of preschool behavior.

Genetic Psychology Monographs, 75, 43-88.

Baumrind, D. (1968). Authoritarian vs. authoritative parental control. Adolescence, 3, 255-272.

Baumrind, D. (1971). Current patterns of parental authority. Developmental Psychology

Monograph, 4(1, Pt. 2), 1-103.

173 
 
Baumrind, D. (1972). An exploratory study of socialization effects on black children: Some

black-white comparisons. Child Development, 43, 261-267.

Baumrind, D. (1983). Rejoinder to Lewis’ reinterpretation of parental firm control effects: Are

authoritative parents really harmonious? Psychological Bulletin, 94, 132-142.

Baumrind, D. (1991). The influence of parenting style on adolescent competence and substance

use. Journal of Early Adolescence, 11, 56-95.

Baumrind, D. (1996). The discipline controversy revisited. Family Relations, 45, 405-414.

Baumrind, D. (2005). Patterns of parental authority and adolescent autonomy. New Directions

for Child and Adolescent Development, 108, 61-69.

Baumrind, D., & Black, A.E. (1967). Socialization practices associated with dimensions of

competence in preschool boys and girls. Child Development, 38, 291-327.

Bowers, J.H., Bowen, K.N., Huck, J.L., Lee, D.R., & Spraitz, J.D. (2011, March). College binge

drinking. Paper presented at the 48th Annual Meeting of the Academy of Criminal Justice

Sciences, Toronto.

Brooks-Gunn, J., Britto, P.R., & Brady, C. (1999). Struggling to make ends meet: Poverty and

child development. In M.E. Lamb (Ed.), Parenting and child development in

nontraditional families (pp. 279-304). Mahweh, NJ: Lawrence Erlbaum Associates.

Buri, J.R. (1991). Parental authority questionnaire. Journal of Personality Assessment, 57, 110-

119.

Buri, J.R., Misukanis, T.M., & Mueller, R.A. (1988). Nothing I ever do seems to please my

parents: Female and male self-esteem as a function of mother’s and father’s nurturance.

St. Paul, MN: University of St. Thomas.

Carmines, E.G., & Zeller, R.A. (1979). Reliability and validity assessment. Newbury Park, CA:

174 
 
 Sage.

Cernkovich, S.A., Giordano, P.C., & Pugh, M.D. (1985). Chronic offenders: The missing cases

in self-report delinquency research. Journal of Criminal Law and Criminology, 76, 705-

732.

Chambers, J.A., Power, K.G., Loucks, N., & Swanson, V. (2000). The quality of perceived

parenting and its associations with peer relationships and psychological distress in a

group of incarcerated young offenders. International Journal of Offender Therapy and

Comparative Criminology, 44, 350-368.

Chassin, L., Presson, C.C., Rose, J., Sherman, S.J., Davis, M.J., & Gonzalex, J.L. (2005).

Parenting style and smoking-specific parenting practices as predictors of adolescent

smoking onset. Journal of Pediatric Psychology, 30, 333-344.

DeVellis, R.F. (2003). Scale development: Theory and application (2nd ed.). Thousand Oaks,

CA: Sage.

Dillman, D.A., Smyth, J.D., & Christian, L.M. (2009). Internet, mail, and mixed-mode surveys:

The tailored design method (3rd ed.). Hoboken, NJ: John Wiley & Sons.

Eisenberg, N. (1992). The caring child. Cambridge, MA: Harvard University Press.

Elliott, D.S., & Ageton, S.S. (1980). Reconciling race and class differences in self-reported and

official estimates of delinquency. American Sociological Review, 45, 95-110.

Falck, R, Siegal, H.A., Forney, M.A., Wang, J., & Carlson, R.G. (1992). The validity of injection

drug users self-reported use of opiates and cocaine. Journal of Drug Issues, 22, 823-832.

Fearon, R.P., Bakermans-Kranenburg, M.J., van Ijzendoorn, M.H., Lapsley, A., & Roisman, G.I.

175 
 
 (2010). The significance of insecure attachment and disorganization in the development

of children’s externalizing behavior: A meta-analytic study. Child Development, 81, 435-

456

Fendrich, M., & Xu, Y. (1994). The validity of drug use reports from juvenile arrestees: A

comparison of self-report, urinalysis and hair assay. Journal of Drug Issues, 24, 99-116

French, K., Finkbiner, R., & Duhamel, L. (2002). Patterns of substance use among minority

youth and adults in the United States: An overview and synthesis of national survey

findings. National Evaluation Data Services. Center for Substance Abuse Treatment.

Caliber Associates: Fairfax, VA.

Fowler, F. J. (2009). Survey research methods (4th ed.). Los Angeles, CA: Sage.

George, D., & Mallery, P. (2010). SPSS for Windows step by step: A simple study guide and

reference 17.0 update (10th ed.). Boston, MA: Allyn & Bacon.

Ghazarian, S.R., & Roche, K.M. (2010). Social support and low-income, urban mothers:

Longitudinal associations with adolescent delinquency. Journal of Youth and

Adolescence, 39, 1097-1108

Gottfredson, M.R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford

University Press.

Haapasalo, J. (2001). How do young offenders describe their parents? Legal and Criminological

Psychology, 6, 103-120.

Harakeh, Z., Scholte, R.H.J., Vermulst, A.A., de Vries, H., & Engels, R.C.M.E. (2004). Parental

factors and adolescents’ smoking behavior: An extension of the theory of planned

behavior. Preventive Medicine, 39, 951-961.

Hetherington, E.M., & Stanley-Hagan, M.M. (1997). The effects of divorce on fathers and their

176 
 
 children. In M.E. Lamb (Ed.), The role of the father in child development (3rd ed., pp.

191-211, 360-369). New York: John Wiley & Sons.

Hirschi, T. (1969). Causes of delinquency. Berkeley, CA: University of California Press.

Hoeve, M., Dubas, J.S., Eichelsheim, V.I., van der Laan, P.H., Smeenk, W., & Gerris, J.R.M.

(2009). The relationship between parenting and delinquency: A meta-analysis. Journal of

Abnormal Child Psychology, 37, 749-775.

Hoeve, M., Smeenk, W., Loeber, R., Stouthamer-Loeber, M., van der Laan, P.H., Gerris, J.R.M.,

& Dubas, J.S. (2007). Long-term effects of parenting and family characteristics on

delinquency of male young adults. European Journal of Criminology, 4, 161-194.

Hoffman, M.L. (1960). Power assertion by the parent and its impact on the child. Child

Development, 31, 129-143.

Hoffman, M.L. (1963). Parent discipline and the child’s consideration for others. Child

Development, 34, 573-588.

Hoffman, M.L. (1975). Moral internalization, parental power, and the nature of parent-child

interaction. Developmental Psychology, 11, 228-239.

Huizinga, D., & Elliott, D.S. (1986). Reassessing the reliability and validity of self-report

delinquency measures. Journal of Quantitative Criminology, 2, 293-327.

Huver, R.M.E., Engels, R.C.M.E., van Breukelen, G., & de Vries, H. (2007). Parenting style and

adolescent smoking cognitions and behaviour. Psychology and Health, 22, 575-593.

Indiana University of Pennsylvania. (n.d.). Facts about IUP. Retrieved from

http://www.iup.edu/WorkArea/linkit.aspx?LinkIdentifier=id&ItemID=44905

Jackson, C., Bee-Gates, D.J., & Henriksen, L. (1994). Authoritative parenting, child

competencies, and initiation of cigarette smoking. Health Education Quarterly, 21, 103-

177 
 
 116.

Jackson, C., Henriksen, L., & Foshee, V.A. (1998). The authoritative parenting index: Predicting

health risk behaviors among children and adolescents. Health Education & Behavior, 25,

319-337.

John, O.P., & Benet-Martinez (2000). Measurement: Reliability, construct validation, and scale

construction. In H.T. Reis & C.M. Judd (Eds.), Handbook of research methods in social

and personality psychology (pp. 339-369). Cambridge, UK: Cambridge University Press.

Katz, C.M., Webb, V.J., Gartin, P.R., & Marshall, C.E. (1997). The validity of self-reported

marijuana and cocaine use. Journal of Criminal Justice, 25, 31-41.

Kim, J.Y.S., Fendrich, M., & Wislar, J.S. (2000). The validity of juvenile arrestees’ drug use

reporting: A gender comparison. Journal of Research in Crime and Delinquency, 37,

419-432.

Lamborn, S.D., Mounts, N.S., Steinberg, L., & Dornbusch, S.M. (1991). Patterns of competence

and adjustment among adolescents from authoritative, authoritarian, indulgent, and

neglectful families. Child Development, 62, 1049-1065.

Lu, N.T., Taylor, B.G., & Riley, K.J. (2001). The validity of adult arrestee self-reports of crack

cocaine use. American Journal of Drug and Alcohol Abuse, 27, 399-419.

Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family: Parent-child

interaction. In E. M. Hetherington (Ed.) & P. H. Mussen (Series Ed.), Handbook of child

psychology: Vol. 4. Socialization, personality, and social development (pp. 1-101). New

York: John Wiley & Sons.

McLanahan, S., & Teitler, J. (1999). The consequences of father absence. In M.E. Lamb (Ed.),

Parenting and child development in nontraditional families (pp.83-102). Mahwah, NJ:

178 
 
 Lawrence Erlbaum Associates.

McNagny, S.E., & Parker, R.M. (1992). High prevalence of recent cocaine use and the

unreliability of patient self-report in an inner-city walk-in clinic. Journal of the American

Medical Association, 267, 1106-1108.

Mertler, C.A., & Vannatta, R.A. (2005). Advanced and multivariate statistical methods (3rd ed.).

Glendale, CA: Pyrczak.

Meyers, L.S., Gamst, G., & Guarino, A.J. (2006). Applied multivariate research: Design and

interpretation. Thousand Oaks, CA: Sage.

Minton, T.D. (2011). Jail inmates at midyear 2010 – Statistical tables. Bureau of Justice

Statistics (NCJ 233431). Washington, DC: U.S. Department of Justice, Office of Justice

Programs.

Moffitt, T.E. (1993). Adolescence limited and life course persistent antisocial behavior: A

developmental taxonomy. Psychological Review, 100, 674-701.

Mott, J.A., Crowe, P.A., Richardson, J., & Flay, B. (1999). After-school supervision and

adolescent cigarette smoking: Contributions of the setting and intensity of after-school

self-care. Journal of Behavioral Medicine, 22, 35-58.

Muris, P., Meesters, C., & van Brakel, A. (2003). Assessment of anxious rearing behaviors with

a modified version of “Egna Minnen Betraffande Uppfostran” questionnaire for children.

Journal of Psychopathology and Behavioral Assessment, 25, 229-237.

Olweus, D. (1980). Familial and temperamental determinants of aggressive behavior in

adolescent boys: A causal analysis. Developmental Psychology, 16, 644-660.

Palmer, E.J., & Gough, K. (2007). Childhood experiences of parenting and causal attributions for

criminal behavior among young offenders and non-offenders. Journal of Applied Social

179 
 
 Psychology, 37, 790-806.

Palmer, E.J., & Hollin, C.R. (2001). Sociomoral reasoning, perceptions of parenting and self-

reported delinquency in adolescents. Applied Cognitive Psychology, 15, 85-100.

Parpal, M., & Maccoby, E.E. (1985). Maternal responsiveness and subsequent child compliance.

Child Development, 56, 1326-1334.

Paschall, M.J., Ringwalt, C.L., & Flewelling, R.L. (2003). Effects of parenting, father absence,

and affiliation with delinquent peers on delinquent behavior among African-American

male adolescents. Adolescence, 38, 15-34.

Patock-Peckham, J.A., & Morgan-Lopez, A.A. (2007). College drinking behaviors: Mediational

links between parenting styles, parental bonds, depression, and alcohol problems.

Psychology of Addictive Behaviors, 21, 297-306.

Patterson, G.R. (1982). Coercive family process: Volume 3. A social learning approach. Eugene,

OR: Castalia.

Patterson, G.R. (1986). Performance models for antisocial boys. American Psychologist, 41, 432-

444.

Paulussen-Hoogeboom, M.C., Stams, G.J.J.M., Hermanns, J.M.A., Peetsma, T.T.D., & van den

Wittenboer, G.L.H. (2008). Parenting style as a mediator between children’s negative

emotionality and problematic behavior in early childhood. The Journal of Genetic

Psychology, 169, 209-226.

Protection of Human Subjects, 45 C.F.R. § 46 (2009). Retrieved from

http://www.hhs.gov/ohrp/documents/OHRPRegulations.pdf

Querido, J.G., Warner, T.D., & Eyberg, S.M. (2002). Parenting styles and child behavior in

African American families of preschool children. Journal of Clinical Child Psychology,

180 
 
 31, 272-277.

Rosay, A.B., Najaka, S.S., & Herz, D.C. (2007). Differences in the validity of self-reported drug

use across five factors: Gender, race, age, type of drug, and offense seriousness. Journal

of Quantitative Criminology, 23, 41-58.

Rote, S.M., & Starks, B. (2010). Racial/ethnic differences in religiosity and drug use. Journal of

Drug Issues, 40, 729-754

Ryu, M. (2010). Minorities in higher education: 24th status report. Washington, DC: American

Council on Education.

Sampson, R.J., & Laub, J.H. (1993). Crime in the making: Pathways and turning points through

life. Cambridge, MA: Harvard University Press.

Schroeder, R.D., Bulanda, R.E., Giordano, P.C., & Cernkovich, S.A. (2010). Parenting and adult

criminality: An examination of direct and indirect effects by race. Journal of Adolescent

Research, 25, 64-98.

Shadish, W.R., Cook, T.D., & Campbell, D.T. (2002). Experimental and quasi-experimental

designs for generalized causal inference. Boston, MA: Houghton Mifflin.

Simons, R.L., Simons, L.G., Burt, C.H., Brody, G.H., & Cutrona, C. (2005). Collective efficacy,

authoritative parenting and delinquency: A longitudinal test of a model integrating

community – and family – level processes. Criminology, 43, 989-1029.

Simons, R.L., Simons, L.G., & Wallace, L.E. (2004). Families, delinquency, and crime: Linking

society’s most basic institution to antisocial behavior. Los Angeles, CA: Roxbury.

Simons-Morton, B., Haynie, D.L., Crump, A.D., Eitel, P., & Saylor, K.E. (2001). Peer and parent

influence on smoking and drinking among early adolescents. Health Education &

Behavior, 28, 95-107.

181 
 
Snyder, J.J., & Patterson, G.R. (1995). Individual differences in social aggression: A test of a

reinforcement model of socialization in the natural environment. Behavior Therapy, 26,

371-391.

Stattin, H., & Kerr, M. (2000). Parental monitoring: A reinterpretation. Child Development, 71,

1072-1085.

Steinberg, L., Blatt-Eisengart, I., & Cauffman, E. (2006). Patterns of competence and adjustment

among adolescents from authoritative, authoritarian, indulgent, and neglectful homes: A

replication in a sample of serious juvenile offenders. Journal of Research on

Adolescence, 16, 47-58.

Steinberg, L., Lamborn, S.D., Darling, N., Mounts, N.S., & Dornbusch, S.M. (1994). Over-time

changes in adjustment and competence among adolescents from authoritative,

authoritarian, indulgent, and neglectful families. Child Development, 65, 754-770.

Steinberg, L., Mounts, N.S., Lamborn, S.D., & Dornbusch, S.M. (1991). Authoritative parenting

and adolescent adjustment across varied ecological niches. Journal of Research on

Adolescence, 1, 19-36.

Thornberry, T.P., & Farnworth, M. (1982). Social correlates of criminal involvement: Further

evidence on the relationship between social status and criminal behavior. American

Sociological Review, 47, 505-518.

U.S. Census Bureau. (2009). Custodial mothers and fathers and their child support: 2007.

Retrieved from http://www.census.gov/prod/2009pubs/p60-237.pdf

U.S. Census Bureau. (2010). Allegheny County, Pennsylvania: State and county quick facts.

Retrieved from http://quickfacts.census.gov/qfd/states/42/42003.html

U.S. Census Bureau. (2010). Indiana County, Pennsylvania: State and county quick facts.

182 
 
 Retrieved from http://quickfacts.census.gov/qfd/states/42/42063.html

Vieno, A., Nation, M., Pastore, M., & Santinello, M. (2009). Parenting and antisocial behavior:

A model of the relationship between adolescent self-disclosure, parental closeness,

parental control, and adolescent antisocial behavior. Developmental Psychology, 45,

1509-1519.

Walker, S.C., Maxson, C., & Newcomb, M.N. (2007). Parenting as a moderator of minority,

adolescent victimization and violent behavior in high-risk neighborhoods. Violence and

Victims, 22, 304-317.

Watt, T.T. (2004). Race/ethnic differences in alcohol abuse among youth: An examination of

risk taking attitudes as a mediating factor. Journal of Ethnicity in Substance Abuse, 3, 33-

47.

Watt, T.T. (2008). The race/ethnic age crossover effect in drug use and heavy drinking. Journal

of Ethnicity in Substance Abuse, 7, 93-114.

61 Pa.C.S.A. § 1731 (2009).

183 
 
 APPENDIX A

Studies Examining the Link between Parenting Style and Adolescent Behavioral Problems

Author Setting Sample Method Measures Findings

Barber, Knox 473 Secondary Parental control Children of

Olsen, & County, TN adolescents data analysis Authoritarian and
Shagle (1994) of self-report Permissive parents
data showed more problem
behaviors than those
of Authoritative

Baumrind Berkeley, 134 Observation Parent Behavior Behavior of children

(1967) CA Caucasian of children in Ratings; see reflected behavior of
families w/ school setting Table X for parents
pre-school and family in description of
age children home setting; the PBR
structured
interviews w/
parents

Baumrind Berkeley, 134 Observation Parent Behavior Fewer problem

(1971) CA Caucasian of children in Ratings; see behaviors in male
families w/ school setting Table X for children of
pre-school and family in description of Authoritative parents
age children home setting; the PBR
structured
interviews w/
parents

Baumrind Berkeley, 16 African Observation Parent Behavior Authoritarian

(1972) CA American of children in Ratings; see parenting appeared
families w/ school setting Table X for beneficial for African
pre-school and family in description of American girls; these
age children home setting; the PBR girls displayed few
structured anti-social behaviors
interviews w/
parents

Baumrind & Berkeley, 95 families Observation Parent Behavior Fewer problem

Black (1967) CA w/ pre- of children in Ratings; see behaviors in male
school age school setting Table X for children of
children and family in description of Authoritative parents
home setting; the PBR
structured
interviews w/
parents

184 
 
Lamborn, Nine high 4,081 high Self-report Parenting style Fewer problem
Mounts, schools in school questionnaire (demandingness behaviors in male
Steinberg, & CA and WI students and children of
Dornbusch responsiveness) Authoritative parents;
(1991) male children with
Neglecting/Rejecting
parents had poorest
outcomes

Paulussen- North 196 pre- Self-report Responsive to Fewer problem

Hoogeboom, Holland, the school age questionnaire child; accepting behaviors in male
Stams, Netherlands children and of child; children of
Hermanns, their consistent Authoritative parents
Peetsma & mothers parenting;
van den discipline;
Wittenboer power
(2008) assertion; love
withdrawal

Querido, Waiting 108 female Self-report Authoritarian, Fewer problem

Warner, & room of a African questionnaire authoritative, behaviors in male
Eyberg pediatric American and permissive children of
(2002) dental clinic caregivers parenting style Authoritative parents;
of pre- children of
school age Authoritarian and
children Permissive parents
showed more problem
behaviors than those
of Authoritative

Vieno, Padua, 840 Italian Self-report Parental control Fewer problem

Nation, Veneto adolescents questionnaire (10 items); behaviors in male
Pastore, & Italy and 657 of parental children of
Santinello their knowledge (4 Authoritative parents;
(2009) mothers items) children of
Authoritarian and
Permissive parents
showed more problem
behaviors than those
of Authoritative

185 
 
 APPENDIX B

Studies Examining the Link between Parenting Style and Alcohol, Cigarette, and Substance Use
by Adolescents

Author Setting Sample Method Measures Findings

Baumrind Berkeley, CA 139 Observation Six family Authoritative

(1991) Caucasian of child for 20 types: parenting likely to
families w/ hours and authoritative, result in non-
high school parents for 30 democratic, drug/alcohol using
age children hours; directive, good- children; heavy drug
structured enough, users and alcoholics
interviews w/ nondirective, likely to have
parents and unengaged neglecting/rejecting
(unengaged) parents

Chassin et al. Midwestern 382 Self-report Parenting style: Those with

(2005) county adolescents questionnaire Behavioral Authoritative,
age 10-17; control (9 items Authoritarian, and
98% from Child Permissive parents
white/non- Report of were significantly
Hispanic Parenting less likely than
Behavior Neglecting/Rejecting
Inventory); to report smoking
Acceptance (7
items from
Network of
Relationships
Inventory)

Harakeh, Six secondary 1,070 Self-report Psychological Psychological

Scholte, schools near adolescents questionnaire control (9 control, strict
Vermulst, de Utrecht, the age 10-14 items); Strict control, and parental
Vries, & Netherlands control (4 knowledge are not
Engels items); Parental significantly related
(2004) knowledge (4 to adolescent
items) smoking behavior

Huver, Eindhoven, 482 Dutch Self-report Parenting style Parenting style not
Engels, van the adolescents questionnaire (22 items) associated with
Breukelen, & Netherlands adolescent smoking
de Vries
(2007)

186 
 
Jackson, Six schools in 937 Self-report Authoritative (6 Lowest rates of
Bee-Gates, & Northern adolescents questionnaire items) and non- intention, initiation,
Henriksen California in grades 3-8 authoriative (3 and experimentation
(1994) items) parenting with smoking are
related to
authoritative
parenting

Jackson, North 1,236 fourth Self-report Authoritative Children of

Henriksen, & Carolina and sixth questionnaire Parenting Index authoritative parents
Foshee graders (demandingess significantly less
(1998) and likely to report
responsiveness) tobacco/alcohol use

Mott, Crowe, Los Angeles 2,352 ninth- Self-report Parenting style No significant
Richardson, and San graders questionnaire (2 items) findings related to
& Flay Diego parenting style and
(1999) counties (CA) smoking

Patock- Arizona State 441 college Self-report Authoritarian, Maternal styles not
Peckham & University; students questionnaire authoritative, significantly related
Morgan- Tempe, AZ and permissive to alcohol use;
Lopez (2007) parenting style paternal
authoritativeness
less likely to result
in alcohol abuse

Simons- Maryland 4,263 middle Self-report Involvement (6 Children of parents

Morton, suburb; near school questionnaire items); who displayed
Haynie, Washington, students expectations (6 positive behaviors
Crump, Eitel, D.C. items); were less likely to
& Saylor monitoring (4 smoke and drink
(2001) items); support than those whose
(5 items); parents displayed
psychological negative behaviors
autonomy (7
items); parent-
child conflict (4
items)

187 
 
 APPENDIX C

Studies Examining the Link between Parenting Style and Delinquency

Author Setting Sample Method Measures Findings

Avenevoli, Nine high 11,669 high Self-report Parents’ love, Authoritative

Sessa, & schools in CA school questionnaire responsiveness, parenting
Steinberg and WI students and involvement associated with
(1999) (15 items); lower levels of
strictness (9 delinquency;
items); children of
psychological permissive and
autonomy (15 neg/rej more
items) likely to be
delinquent;
authoritarian
parenting not as
bad for African
American
children

Chambers, Young 122 male Semi-structured Parental Offenders

Power, Offender’s offenders, age interviews Bonding perceived
Loucks, & Institution in 15-22 Instrument (care parents to be
Swanson Scotland and control) controlling and
(2000) not caring
(authoritarian)

Hoeve, Pittsburgh, PA 472 boys in Secondary data Parental Authoritative

Smeenk, and the first cohort of analysis of self- affection (5 child parenting
Loeber, Netherlands Pittsburgh report data items/9 adult); associated with
Stouthamer- Youth Study; conformity (4/8 lower levels of
Loeber, van 132 male items); ignoring delinquency
der Laan, participants (5 adult items);
Gerris, & from Child- responsiveness
Dubas (2007) rearing and (8child items)
Family in the
Netherlands
Study

Jackson, North Carolina 1,490 ninth Self-report Authoritative Children of

Henriksen, & and tenth questionnaire Parenting Index authoritative
Foshee (1998) graders (demandingness parents
and significantly
responsiveness) less likely to
report violent
behavior

188 
 
Palmer & West Midlands 94 students, Self-report Perceptions of Authoritative
Hollin (2001) area of age 12-18 questionnaires parenting significant
England (control, negative
guidance, and relationship
affective bond) with
delinquency;
more
authoritarian
style significant
positive
relationship
with
delinquency

Paschall, Medium-size 175 Self-report Monitoring (9 Monitoring and

Ringwalt, & southeastern adolescent questionnaire items); Control control
Flewelling city African (6 items); significant
(2003) American Communication negative
boys and their (11 items); P-C association
mothers Relationship (16 with
items) delinquency;
control had
higher impact
in father-absent
families

Stattin & Kerr Seven mid- 703 14-year- Self-report Monitoring (9 All measures
(2000) Sweden old Swedish questionnaire items); significant
communities youths Disclosure (5 negative
items); relationship
Solicitation (5 with delinquent
items); Control behavior;
(6 items) solicitation
related to
higher
delinquency
than disclosure

Steinberg, Philadelphia, 1,355 juvenile Interview with Parental warmth Authoritative

Blatt- PA, and offenders juvenile and firmness parenting
Eisengart, & Phoenix, AZ associated with
Cauffman lower levels of
(2006) delinquency;
children of
permissive and
neg/rej more
likely to be
delinquent;
authoritarian
parenting not as
bad for African

189 
 
 American
children
Steinberg, Nine high 6,357 high Self-report Parents’ love, Authoritative
Lamborn, schools in CA school questionnaire responsiveness, parenting
Darling, and WI students and involvement associated with
Mounts, & (10 items); lower levels of
Dornbusch strictness (9 delinquency;
(1994) items) children of
permissive and
neg/rej more
likely to be
delinquent
Steinberg, Nine high 7,600 high Self-report Parents’ love, Authoritative
Mounts, schools in CA school questionnaire; responsiveness, parenting
Lamborn, & and WI students students and involvement associated with
Dornbusch assigned to one (15 items); lower levels of
(1991) of 16 niches strictness (9 delinquency
items);
psychological
autonomy (15
items)
Simons, Multiple sites 633 fifth Secondary data 21-item Authoritative
Simons, Burt, in GA and IA graders and analysis of authoritative parenting
Brody, & primary FACHS dataset parenting survey associated with
Cutrona caretaker; (interview data) lower levels of
(2005) secondary delinquency
caretaker
when
available
Walker, Los Angeles 349 African Structured Attachment Higher levels
Maxson, & County American and interviews (includes family of attachment
Newcomb Hispanic closeness and led to lower
(2007) boys, age 12- parental levels of
17 monitoring) delinquency in
Hispanic youth;
Medium
attachment led
to low
delinquency in
African
American
youth with high
and low
attachment
leading to high
delinquency

190 
 
 APPENDIX D

Studies Examining the Link between Parenting Style and Criminality

Author Setting Sample Method Measures Findings

Haapasalo Five Finnish 89 Finnish Self-report Parent Behavior Specific style did
(2001) prisons male inmates questionnaire Inventory (108 not emerge
items; 18 significantly;
subscales) most self-reports
suggested
rejection

Palmer & Not specified 40 young Self-report EMBU scale Maternal

Gough (2007) male questionnaire authoritativeness
offenders higher for non-
(property and offenders and
person); 31 property
young male offenders;
non- paternal
offenders authoritativeness
higher for non-
offenders

Schroeder, Large 662 Self-report 7-item Permissive

Bulanda, metropolitan Caucasian questionnaire demandingness parenting more
Giordano, & area in OH and African scale; 10-item likely to result in
Cernkovich American responsiveness criminal
(2010) adults scale behavior.
Amongst whites,
authoritarianism
more likely to
lead to crime;
permissiveness
amongst African
Americans

191 
 
 APPENDIX E
Informed Consent/Invitation to Participate – Jail Inmates 

192 
 
 Informed Consent Form

You are invited to participate in a research study. The following information is provided in order
to help you to make an informed decision whether or not to participate. If you have any questions
please do not hesitate to ask. You are eligible to participate because you are currently housed in a
county jail in the Commonwealth of Pennsylvania. If you are under the age of 18, however, you
are not permitted by law to complete this survey.

My name is Jason Spraitz and I am asking for your participation to help me complete my
dissertation research. I am a doctoral student at Indiana University of Pennsylvania; neither I
nor my research is connected with any law enforcement agency, any court system, any
local/county jail, or the Pennsylvania Department of Corrections. The purpose of this study
is to examine the effects that perceived parenting style has on behavior. Participation will require
approximately 45-60 minutes of your time. Participants in this study will not be subject to risk
beyond a minimal level.

Your participation in this study is voluntary. You are free to decide not to participate in this
study or to withdraw at any time. If you choose to not participate simply sit quietly or return to
the main housing unit. If you choose to withdraw while completing the survey simply write the
word “withdraw” on the survey in front of you, your request to withdraw will be respected and
the incomplete survey will be shredded. If you choose to participate your identity will remain
anonymous; please do not place any identifying information (such as your name or ID number)
on the survey as I do not want to able to identify which response came from a particular
respondent who completes the survey. Your responses will be considered only in combination
with those from other participants. The information obtained in the study may be published in
peer-reviewed journals or presented at professional meetings but your identity will remain
anonymous.

Your participation in this study is completely voluntary. Thank you in advance for your
assistance with this project. If you have any questions or comments please feel free to contact
me or my dissertation chairperson, Dr. Jamie Martin.

Jason D. Spraitz, M.S. Jamie Martin, Ph.D.

Doctoral Candidate Professor
Department of Criminology Department of Criminology
G-13 Wilson Hall, 411 North Walk G-18 Wilson Hall, 411 North Walk
Indiana, PA 15705-1002 Indiana, PA 15705-1002
Email: j.spraitz@iup.edu Email: jmartin@iup.edu

The Indiana University of Pennsylvania Institutional Review Board has approved this project for
the Protection of Human Subjects (Phone: 724-357-7730).

I have read and understand the information on the form and I consent to volunteer to be a subject
in this study. I understand that my responses are completely anonymous and that I have the right
to withdraw at any time. Providing responses implies my consent to participate.

193 
 
 APPENDIX F
Informed Consent/Invitation to Participate – University Students

194 
 
Greetings,

You are receiving this email message because you are invited to participate in a research study. The
following information is provided in order to help you to make an informed decision whether or not to
participate. If you have any questions please do not hesitate to ask; you can use the phone number or
email address listed below to contact me. You are eligible to participate because you are an undergraduate
student at IUP who was randomly chosen to participate in the study. Students under the age of 18,
however, are not permitted by law to complete this survey. Although the opinions of those under the age
of 18 are important it would be appreciated if those under 18 would not click on the survey link below.

##LINK HERE##
My name is Jason Spraitz and I am asking for your participation to help me complete my dissertation
research. The purpose of this study is to examine the effects that perceived parenting style has on
behavior. Participation will require approximately twenty minutes of your time. Again, your participation
would be most helpful and I would be very appreciative. Once you complete the survey you will be able
to enter yourself in a raffle to win one of four $25 gift cards to the Co-op Store. Participants in this
study will not be subject to risk beyond a minimal level.

Your participation in this study is voluntary. You are free to decide not to participate in this study or to
withdraw at any time. If you choose not to participate you simply do not have to click on the link to the
survey. If you choose to withdraw while completing the survey, your request to withdraw will be
respected and no information pertaining to you will be collected. If you choose to participate your identity
will remain anonymous; there is no way for you to place any identifying information on the survey. Your
responses will be considered only in combination with those from other participants. The information
obtained in the study may be published in peer-reviewed journals or presented at professional meetings
but your identity will remain anonymous. The researcher will be unable to identify which response came
from a particular student who completes the survey.

Your participation in this study is completely voluntary. Thank you in advance for your assistance with
this project. If you have any questions or comments please feel free to contact me or my dissertation
chair, Dr. Jamie Martin.

Jason D. Spraitz, M.S. Jamie Martin, Ph.D.

Doctoral Candidate Professor
Department of Criminology Department of Criminology
G-13 Wilson Hall, 411 North Walk G-18 Wilson Hall, 411 North Walk
Indiana, PA 15705-1002 Indiana, PA 15705-1002
Phone: 724-357-1247 Phone: 724-357-5975
Email: j.spraitz@iup.edu Email: jmartin@iup.edu

The Indiana University of Pennsylvania Institutional Review Board has approved this project for the
Protection of Human Subjects (Phone: 724-357-7730).

I have read and understand the information on the form and I consent to volunteer to be a subject in this
study. I understand that my responses are completely anonymous and that I have the right to withdraw at
any time. Clicking on the link implies my consent to participate.

195 
 
 APPENDIX G

Survey Instrument

196 
 
Section I:

This section contains statements about your mother and/or mother figure. Please indicate your
level of agreement or disagreement with the statement by circling the letter next to it.
Before you begin, please circle the letter indicating which female raised you the most while you
were growing up and respond to the following statements with that person in mind.

1. The woman who raised me the most while I was growing up was:

A. Biological Mother F. Sister

B. Adopted Mother G. Aunt
C. Foster Mother H. Female Cousin
D. Step Mother I. Other Female Relative: __________________
E. Grandmother J. None ÆSkip to #22 if selected

2. Even if I didn’t agree with her, my mother felt that it was for my own good if I was forced to conform
to what she thought was right:

A. Strongly Agree
B. Agree
C. Neither Agree nor Disagree
D. Disagree
E. Strongly Disagree

3. As I was growing up, I knew what my mother expected of me, but I also felt free to discuss those
expectations with my mother if I felt that they were unreasonable:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

4. As I was growing up, I had no idea what my mother expected of me:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

5. While I was growing up my mother felt that children should have their way in the family as often as
the parents do:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

197 
 
 6. Whenever my mother told me to do something, she expected me to do it immediately without asking
any questions:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

7. As I was growing up, my mother consistently gave me direction and guidance in rational and objective
ways:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

8. My mother never gave me any guidance as I was growing up:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

9. My mother felt that children need to be free to make up their own minds and do what they want to do,
even if parents might not agree:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

10. My mother always felt that more force should be used by parents in order to get their children to
behave the way they are supposed to:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

198 
 
 11. My mother gave me direction for my behavior and activities as I was growing up and she expected me
to follow her direction, but she was always willing to listen to my concerns:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree

12. I feel like I raised myself more than my mother raised me:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

13. Most of the time as I was growing up my mother did what I wanted her to do when making decisions:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

14. As I was growing up, my mother let me know what behavior she expected of me, and if I didn’t meet
those expectations, she would punish me:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

15. As I was growing up, my mother gave me clear direction for my behaviors and activities, but she was
also understanding when I disagreed with her:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

199 
 
 16. I don’t know my mother’s views about discipline because she never communicated them with me:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

17. As I was growing up, my mother allowed me to decide most things for myself without a lot of
direction from her:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

18. As I was growing up, I knew what my mother expected of me and she insisted that I conform to those
expectations simply out of respect for her authority:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

19. As I was growing up, if my mother made a decision in the family that hurt me, she was willing to
discuss that decision with me and admit if she had made a mistake:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

20. I did not learn effective communication skills from my mother:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

200 
 
 21. As I was growing up, my mother allowed me to form my own point of view on family matters and she
generally allowed me to decide for myself what I was going to do:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

Section II:

This section contains statements about your father and/or father figure. Please indicate your level
of agreement or disagreement with the statement by circling the letter next to it.
Before you begin, please circle the letter indicating which male raised you the most while you
were growing up and respond to the following statements with that person in mind.

22. The man who raised me the most while I was growing up was:

A. Biological Father F. Brother

B. Adopted Father G. Uncle
C. Foster Father H. Male Cousin
D. Step Father I. Other Male Relative: __________________
E. Grandfather J. None ÆSkip to Section III if selected

23. Even if I didn’t agree with him, my father felt that it was for my own good if I was forced to conform
to what he thought was right:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

24. As I was growing up, I knew what my father expected of me, but I also felt free to discuss those
expectations with my father if I felt that they were unreasonable:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

201 
 
 25. As I was growing up, I had no idea what my father expected of me:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

26. While I was growing up, my father felt that children should have their way in the family as often as
the parents do:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

27. My father always felt that more force should be used by parents in order to get their children to behave
the way they are supposed to:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

28. As I was growing up, my father consistently gave me direction and guidance in rational and objective
ways:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

29. My father never gave me any guidance as I was growing up:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

202 
 
 30. My father felt that children need to be free to make up their own minds and do what they want to do,
even if parents might not agree:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

31. My father felt that wise parents should teach their children who the boss is in the family:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

32. My father had clear standards of behavior for me as I was growing up, but he was willing to adjust
those standards to my needs:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

33. I feel like I raised myself more than my father raised me:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

34. Most of the time as I was growing up my father did what I wanted her to do when making decisions:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

203 
 
 35. My father has always felt that most problems in society would be solved if we could get parents to
strictly and forcibly deal with their children when they don’t do what they are supposed to do:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

36. My father gave me direction for my behavior and activities as I was growing up and he expected me to
follow his direction, but he was always willing to listen to my concerns:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

37. I did not learn discipline from my father:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

38. As I was growing up, my father allowed me to decide most things for myself without a lot of direction
from him:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

39. As I was growing up, I knew what my father expected of me and he insisted that I conform to those
expectations simply out of respect for his authority:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

204 
 
 40. As I was growing up, my father gave me clear direction for my behaviors and activities, but he was
also understanding when I disagreed with him:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

41. As I was growing up, my father did not direct my behaviors, activities, and desires:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

42. As I was growing up, my father allowed me to form my own point of view on family matters and he
generally allowed me to decide for myself what I was going to do:

A. Strongly Agree
B. Agree
C. Neither Agree Nor Disagree
D. Disagree
E. Strongly Disagree

I am interested in who was responsible for raising you while you were growing up. In the
following question, “parents” refers to the individuals who you consider to be your mother and
father.

43. When I was growing up, I lived with:

A. Two parents; both parents lived there and were involved in raising me
B. My mother, but my father was also involved in raising me
C. My father, but my mother was also involved in raising me
D. My mother, and she was the only one who raised me
E. My father, and he was the only one who raised me
F. Neither of my parents were involved in raising me
G. Other (please explain) ______________________________________

205 
 
 Section III:

This section contains questions asking about behaviors that you may or may not have engaged in.
Please answer questions as honestly as possible.

Read the following events and mark the “Have you ever?” box with an “X” if you have engaged
in that activity. Then estimate the number of times that you have engaged in the activity and
place the NUMBER in the box.

Action  Have you ever?  How many times? 

Hurt someone badly enough that they needed medical treatment    

Purposely damaged or destroyed property belonging to another person    

Paid to have sexual relations with another person    

Set off a fire alarm for fun    

Ran away from home while under the age of 18    

Used a weapon to threaten another person    

Stolen (or tried to steal) a motor vehicle, such as a car or motorcycle    

Been paid for having sexual relations with someone    

Carried a gun without a permit    

Lied about your age to gain entrance or to purchase something (e.g., lying    
about your age to buy liquor or get into a bar)
Purposely killed someone    

Stolen (or tried to steal) something worth more than $50    

Sold marijuana    

Thrown objects (such as rocks, snowballs, or bottles) at cars or people    

Cheated in school (e.g., copying another student’s answers or cheating on a    

test)
Hurt someone in a minor way (e.g., slap in face, push to ground)    

Lied to an authority figure (such as a teacher, boss, or parent)    

206 
 
 Read the following events and mark the “Have you ever?” box with an “X” if you have engaged
in that activity. Then estimate the number of times that you have engaged in the activity and
place the NUMBER in the box.

Action  Have you ever?  How many times? 

Carried a hidden weapon other than a plain pocket knife    

Skipped classes without an excuse    

Attacked someone with the idea of seriously hurting or killing them    

Stolen (or tried to steal) things worth less than $10    

Bought alcohol for a minor or provided alcohol for a minor    

Been loud, rowdy, or unruly in a public place (e.g., disorderly conduct)    

Been suspended from school    

Hit (or threatened to hit) another person    

Stolen money or other things from another person    

Failed to return extra change that a cashier accidently gave you    

Been drunk in a public place    

Had (or tried to have) sexual relations with someone against their will    

Avoided paying for such things as movies, bus or subway rides, and food    

Made obscene telephone calls, such as calling someone and saying dirty things    

Used force (strong-arm methods) to get money or things from another person    

Stolen (or tried to steal) things worth between $10 and $50    

Broken into a building or vehicle (or tried to break in) to steal something or    
just to look around
Knowingly bought, sold or held stolen goods (or tried to do any of these    
things)
Sold drugs other than marijuana    

Gotten into verbal altercations (shouting/screaming) with another person    

207 
 
 Read the following list of alcohol and drugs and mark the “Have you ever?” box with an “X” if
you have used that substance.

Substance  Have you ever? 

Tobacco (e.g., Cigarettes, Chew) while under the age of 18

Alcohol (e.g., beer, wine, liquor) while younger than legal drinking age

Alcohol (e.g., beer, wine, liquor) while of – and older than – legal drinking age

Marijuana

Hallucinogens (e.g., LSD, Acid, Mescaline, Peyote)

Amphetamines (e.g., Uppers, Speed)

Barbiturates (e.g., Downers, Reds)

Heroin

Cocaine (e.g., Powder, Crack)

Methamphetamines

Prescription Pain Killers without valid prescription (e.g., Vicodin, Percocet)

92. Did the way that your parents raised you contribute to your taking part in the behaviors
asked about above? Yes No

93. Do you blame your parents for the behaviors that you took part in? Yes No

94. Are you more likely to keep your feelings inside you or let them out? Inside Outside

Section IV:

95. How old are you? _____

96. What is your sex? Male Female

208 
 
97. With what race do you most identify:

African American/Black Asian/Pacific Islander

Caucasian Hispanic
Native American Other _______________ 

209