Reconceptualizing the Nature and Consequences of Part-Time Work

Author(s): Daniel C. Feldman

Source: The Academy of Management Review, Vol. 15, No. 1 (Jan., 1990), pp. 103-112
Published by: Academy of Management
Stable URL: http://www.jstor.org/stable/258108 .
Accessed: 12/06/2014 17:05

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The Academy
of Management Review.

http://www.jstor.org

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions
C Academy of Management Review, 1990, Vol. 15, No. 1, 103-112.

Reconceptualizing the Nature

and Consequences
of Part-Time Work
DANIELC. FELDMAN
University of South Carolina

This article presents a theoretical framework for understanding the

impact of part-time work on employees' attitudes and behaviors. A
series of hypotheses also are presented to explain the varying con-
sequences that different types of part-time employment arrange-
ments, work-context factors, and demographic variables have on the
experiences of part-time workers. Future issues for theory develop-
ment and research methodology are discussed as well.

Part-time workers are a growing segment of in accommodating the particular needs and de-
the U.S. labor force. With the exception of Swe- mands of part-time workers, especially in terms
den, no major industrialized country has nearly of retaining, scheduling, and rewarding them
as high a percentage of part-time workers as the (Mahlin & Charles, 1984; Gannon, 1975; Wer-
United States (Kahne, 1985). Defined by the U.S. ther, 1976).
government as employees who work fewer than Second, part-time work has become an im-
35 hours a week, part-time workers now number portant employment opportunity for three major
about 19 million, which is just over 20 percent of demographic groups in our society: younger
the work force (Nardone, 1986). Since 1955, the workers (ages 16 to 24), older workers (65 and
proportion of part-time workers to the total work over), and female workers. Women represent
force has doubled (Kahne, 1985). Despite their two thirds of the total of part-time workers, while
importance, however, part-time workers are nearly two thirds of the male part-time workers
still, in many ways, "missing persons" in orga- are under 24 or over 65 (Nardone, 1986). How
nizational research (Rotchford & Roberts, 1982). part-time work is designed and managed will
Research on part-time labor has become have a great impact on the quality of life of these
more critical for at least two reasons, other than three groups of workers.
their sheer volume. First, part-time labor has For example, although many people assume
emerged as an important labor supply for entire that part-time employment builds character and
industries. As an example, service and retail responsibility in young students, Greenberger
trade industries rely most heavily on part-time and Steinberg (1986) present some disturbing
workers. Indeed, almost 75 percent of part-time empirical evidence to the contrary. They noted
employees in America work in these industries; that
between 1979 and 1985, over 2.4 million part-
time workers were added to these sectors alone. extensive part-time employment may have a
deleterious impact on youngsters' schooling;
The literature suggests that service and retail that [part-time] working appears to promote,
companies will need to be especially innovative rather than deter, some forms of delinquent be-

103

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions
 havior; that [part-time]working, especially in largely determine the motivations and satisfac-
the high-stress jobs held by many teenagers, tions of part-time workers.
leads to higher rates of alcohol and drug abuse;
and that for many youngsters, [part-time]work- Second, although some research has been
ing fosters the development of negative atti- conducted on differences among demographic
tudes toward work itself. (p. 4) groups in their reactions to part-time work, little
theoretical work has explained those differ-
Greenberger and Steinberg are particularly ences. This paper argues that different demo-
concerned that the insufficient adult supervision graphic groups may use systematically different
young adults receive results in situations in frames of reference to evaluate their part-time
which students learn much more about goofing work experiences, and they may focus on differ-
off than about work skills. ent context factors when they determine
In contrast, the effective use of older workers whether their partial inclusion in the work force
has very real benefits, both for the elderly work- is attractive, irritating, or irrelevant.
ers themselves and for their younger col- Third, research on part-time work typically
leagues. As McDonald's McMasters program has assumed that the very nature of part-time
has demonstrated, older workers provide a work directly and systematically influences
valuable source of part-time employees. More- workers' attitudes and behaviors on the job.
over, programs at corporations such as Merck However, it is the thesis of this paper that there
Pharmaceuticals, Coopers and Lybrand, and also is some reciprocal causality; that is, em-
Kaiser Permanente suggest that the availability ployees who have certain job attitudes and work
of part-time work for those over 62 might induce behaviors may gravitate toward part-time work
them to leave the full-time work force sooner instead of full-time work. Other research-
(opening up slots for promotion for younger method issues, such as the mediating and/or
workers) while providing a stable, longer- moderating roles that work-context factors play
lasting part-time work force among postretirees in the experiences of part-time workers, are also
(Bosworth & Holden, 1983; Pfeffer & Baron, 1988; discussed.
Rothberg & Cook, 1985). Managers will have to In each section, series of hypotheses are pre-
learn how to utilize these three groups of part- sented that can be used to direct future re-
time workers productively, in ways that are nei- search. Although not exhaustive due to page
ther harmful to the employees nor to the compa- constraints, these hypotheses were chosen for
nies. their relevance to key dependent variables
This article attempts to reconceptualize the re- (e.g., job satisfaction and commitment, produc-
search on the nature and consequences of part- tivity, and withdrawal behavior), for their rele-
time workers in three ways. First, most of the vance to the largest sets of part-time workers
research on part-time workers has focused on (e.g., part-time student workers, part-time fe-
the differences between full-time and part-time male workers with children at home, etc.), and
workers (e.g., Jackofsky & Peters, 1987; Wake- for their relevance to managerial practice (e.g.,
field, Curry, Mueller, & Price, 1987). It is the the- compensation and scheduling issues).
sis of this paper that there are major theoretical
differences among types of part-time work ar-
Types of Part-Time
rangements (e.g., temporary part-time versus
permanent part-time, seasonal versus year-
Work Arrangements
round, part-time work as one's main job versus There are at least five dimensions on which
part-time work as a second job, etc.) that not part-time work arrangements can be arrayed.
only attract different groups of workers but also Permanent/Temporary. Permanent part-time

104

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions
workers are employed fewer than 35 hours per lower paying, less interesting jobs (e.g., clerk,
week on a continuing basis, whereas temporary household help) that have noncomparable
workers are hired for limited periods of time to fringe benefits and less supervisory responsibil-
deal with fluctuating workloads or short-term ities (Deuterman & Brown, 1978; Howe, 1986;
personnel shortages (Howe, 1986; Moberly, Nardone, 1986; Owen, 1978). Hypothesis 1 sug-
1986; Nollen, 1978). gests that part-time workers whose jobs are
Organization-hired/Agency-hired. Part-time somewhat regular (they are employed year-
workers may be recruited and hired by one round as their main employment) will be the
company by whom they are paid, or they may most likely to obtain the better compensated,
work for a personnel agency that places them in more interesting jobs within their job category.
a variety of corporations and pays them directly Also, part-time workers who are working part-
(Howe, 1986; Moberly, 1986). time voluntarily are more likely to be satisfied
Year-round/Seasonal. Some part-time work- with their jobs than employees who have been
ers hold part-time employment year-round; forced to reduce their hours or who are unable
other part-time workers hold employment only to locate full-time jobs.
during certain seasons of the year (e.g., Christ- Hypothesis 2: Part-time workers will be more
mas, summer) (Hom, 1979). dissatisfied with pay, fringe benefits, and the
Main-job/Second-job ("Moonlighting"). Some work itself when their employment arrange-
part-time workers are employed fewer than 35 ments are (a) temporary and (b) seasonal.
hours per week in one job, which is their only Research on the impact of part-time work on
source of salaried income; others ("moon- job attitudes suggests there are not systematic
lighters") work part-time jobs as second jobs to differences in overall levels of satisfaction be-
supplement their income (Best, 1981; Owen, tween part-time and full-time employees (Rotch-
1978; Ronen, 1984). ford & Roberts, 1982). However, it is quite likely
Voluntary/Involuntary. Traditionally, the Bu- that part-time workers may be more dissatisfied
reau of Labor Statistics (Nardone, 1986) has dis- with certain aspects of their jobs (e.g., pay and
tinguished between voluntary part-time workers fringe benefits) than full-time workers (Eber-
(those who work fewer than 35 hours per week hardt & Shani, 1984; Hall & Gordon, 1973; Hom,
by choice) and involuntarily part-time workers 1979; Miller & Terborg, 1979). Moreover, Logan,
(those who work fewer than 35 hours per week O'Reilly, and Roberts (1973) and Rotchford and
because of the unavailability of full-time jobs, Roberts (1982) suggested that the perceptual
because of cutbacks of hours assigned during maps of job satisfaction may be different for
business downturns, etc.). part-timers than for full-time employees; part-
Major differences in job attitudes and job be- time workers may weight factors such as sched-
haviors among part-time workers are expected, uling flexibility higher in their overall assess-
depending on the types of part-time employ- ment of job satisfaction. Hypothesis 2 suggests
ment arrangements they have. Several hypoth- that part-time workers' dissatisfaction with pay,
eses on these relationships are presented and fringe benefits, and the work itself will be
discussed below. greater the more peripheral or tangential they
Hypothesis 1: Part-time workers will be more are to an organization. Temporary and seasonal
satisfied with theirjobs when their employment workers are hypothesized to be the part-time
arrangement is (a) year-round, (b) their main workers most likely to be placed in routine, low-
job, and (c) voluntary. paying jobs without fringe benefits (e.g., clean-
Substantial evidence at present points to no- up crews and inventory counters) (Howe, 1986;
ticeable job segregation of part-time workers in Owen, 1978).

105

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions
 Hypothesis 3: Part-time workers will be less in- ductive time in terms of start-up each day, there
volved in their jobs and committed to their or- is some evidence to suggest that this can be off-
ganizations when their employment arrange-
set by higher per-hour productivity on the job
ments are (a) temporary, (b) agency-hired, (c)
seasonal, and (d) second jobs. and less fatigue (Mahlin & Charles, 1984; Wer-
ther, 1976). It is suggested in Hypothesis 5 that
Hypothesis 3 proposes that the more tangen- part-time workers are more likely to be produc-
tial part-time employees are to an organization,
tive when their work arrangements are perma-
the less likely they will feel involved in their jobs
nent, year-round, organization-hired main jobs.
and committed to their organizations (Logan et
These workers are the most likely to receive ad-
al., 1973). Temporary work, employment by per-
equate training for their jobs, to become most
sonnel agencies, seasonal work, and moon-
fully integrated into the work group, and to stay
lighting are hypothesized to lead to decreased
with the organization long enough to perform at
commitment to the focal part-time employer be-
peak efficiency for extended periods of time.
cause such arrangements lead to split loyalty
among other organizations (school, one's main
job, the personnel agency, etc.) and make more Role of Demographic Variables
salient the partial nature of the work contract.
Although the type of part-time work arrange-
Hypothesis 4: Part-time workers are more likely ment certainly has an impact on the attitudes
to turn over when their employment arrange- and behaviors of employees, it also is clear that
ment is (a) a moonlighting job or (b) involuntar-
ily part-time.
individual differences, particularly demo-
graphic variables, play a major role. The vari-
The research on the impact of part-time work ables that have been most closely explored in
on job behaviors indicates that part-time work- the context of part-time work are age (Bosworth
ers have higher levels of turnover (Gannon & & Holden, 1983; Soumerai & Avorn, 1983), gen-
Nothern, 1971; Jackofsky, Salter, & Peters, 1986; der (Haring, Okun, & Stock, 1984; Rothberg &
Owen, 1978). Researchers have found that part- Cook, 1985), and marital status and number of
timers are more likely to leave their jobs than children (Hart, 1987; Nakamura & Nakamura,
full-timers because they are more dissatisfied 1983; Presser, 1986). In general, consistent with
with the pay and are less likely to receive the the organizational literature, older part-time
fringe benefits tied to longevity (Kahne, 1985). workers are more satisfied than younger work-
Moonlighters and involuntarily part-time work- ers; female part-time workers are more satisfied
ers are hypothesized to leave their jobs more than males; and part-time workers, who feel ex-
frequently because they will be most likely to tensive conflict between roles (either from school
search for positions that pay better; for the invol- duties or from family/home responsibilities) are
untarily part-time worker this may be at another more dissatisfied than those who do not have
firm and for the moonlighter this may be either such conflicts (Greenberger & Steinberg, 1986;
at the primary job or at an altogether different Hall & Gordon, 1973; Presser, 1986). Similarly,
company. research has suggested that younger, part-time
Hypothesis 5: Part-time workers are more likely students and married women who have school-
to be productive when their work arrangements aged children are more likely to have higher
are (a) permanent, (b) year-round, (c) organiza- rates of absenteeism and turnover due to the
tion-hired, and (d) their main job.
greater demands on their time by their other
No systematic differences have been found roles (Best, 1981; Hart, 1987; Kahne, 1985; Naka-
between full-time and part-time workers in mura & Nakamura, 1983).
terms of productivity (Greenwald & Liss, 1973). There are at least two distinct reasons why
Although part-time workers may lose some pro- demographic variables might influence the job

106

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions
satisfaction and job behaviors of part-time work- worked before retirement (Kahne, 1985). Be-
ers. First, different demographic groups might cause in such situations pension and social se-
systematically gravitate toward different types curity benefits usually provide disincentives for
of part-time work arrangements, and these ar- more than 20 hours of work per week, they also
rangements are, in turn, differentially reinforc- are likely to be voluntarily part-time and are
ing. For example, students frequently might end likely to hold only one part-time job with one
up working at summer jobs, which historically organization.
have been low paying. Second, demographic
Hypothesis 8: Females who have children at
groups might weigh facets of part-time work dif- home are more likely to gravitate toward work
ferently when they assess their overall job satis- arrangements that are (a) temporary, (b)
faction or motivation. For instance, having con- agency-hired, and (c) voluntarily part-time.
genial co-workers may be more important for Research suggests that women are more
postretirees than for students, and having op- likely to be voluntarily and temporarily part-
portunities to develop new skills may be more time workers, largely because of the responsi-
important to homemakers who are entering the bilities involved in taking care of a home and
work force than to postretirees. Illustrative hy- children (Deuterman & Brown, 1978; Kahne,
potheses on both sets of demographic issues are 1985; Nardone, 1986). In addition, as Hypothesis
presented next. 8 predicts, they are more likely to use personnel
Hypothesis 6: Part-time student workers are agencies than their male counterparts. Howe
more likely to gravitate to work arrangements (1986) suggested that women who have family
that are (a) temporary, (b) organization-hired, responsibilities seek employment through tem-
(c) seasonal, (d) voluntarily part-time, and (e) porary agencies in order to test the market by
their main jobs.
sampling a range of work environments; they
The majority of students who work part-time also use this opportunity to acquire needed skills
do so to earn extra income while they attend and experiences while building their confidence
classes; they usually do not convert their part- and marketability.
time jobs into full-time work opportunities Hypothesis 9: Flexibility in scheduling will be
(Greenberger & Steinberg, 1986). Consequently, most important to (a) students under age 24, (b)
as Hypothesis 6 suggests, students under age 24 married workers who have children, and (c)
are more likely to be voluntarily employed in moonlighters.
either part-time temporary jobs during the As Ronen (1984) and Levine (1987) suggested,
school year or seasonal jobs during the sum- flexibility in scheduling their work plays an im-
mer. Young workers also use employment portant role in the experiences of part-time work-
agencies less (Howe, 1986), perhaps because so ers. Part-time workers who are given flexibility
many of the jobs offered by such agencies are in scheduling their work will have more positive
clerical in nature or perhaps because the varied job attitudes and will be less likely to change
job opportunities offered by them may require jobs. Hypothesis 9 suggests that for these three
more transportation to firms widespread across groups of workers who have systematic con-
their metropolitan area. straints on their time, flexibility in scheduling
Hypothesis 7: Workers over age 62 are more their work is especially important. This flexibility
likely to gravitate to work arrangements that are could be determined by the number and sched-
(a) organization-hired, (b) main job, (c) volun- ule of hours weekly or by their obtaining the
tarily part-time. desired number and schedule of hours on a con-
Workers over age 62 often gravitate to part- tinuous basis.
time employment in the firms where they Hypothesis 10: Part-time workers under age 24

107

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions
 and over 62 will be more satisfied with fewer bivariate comparisons between groups of part-
hours of work per week (i.e., fewer than 20 time workers to more complex comparisons. For
hours); part-time workers between 25 and 62
will be more satisfied with more hours of work instance, cluster and discriminant analysis can
per week (i.e., more than 20 hours). be used to explore the impact of groups of de-
mographic variables across a wide array of at-
Hypothesis 10 suggests that younger and titudinal and behavioral outcomes. Demo-
older workers may react differently to part-time graphically, these clusters might include part-
work experiences than employees between time students under age 24, women who have
ages 24 and 62. For student workers, more than school-aged children, part-time retirees over
20 hours of work per week is likely to interfere age 62, and males aged 25-62. It might be pre-
with studying (Greenberger & Steinberg, 1986); dicted, for example, that part-time retirees over
for older workers, more than 20 hours work is age 62 have the highest rates of job satisfaction,
likely to interfere with their leisure time or it that women who have school-aged children
could produce marginally less income for them have the highest rates of absenteeism, and that
because of reduced Social Security benefits part-time students under age 24 have the high-
(Harriman, 1982). In contrast, adult employees est rates of turnover. Simple bivariate compari-
who want to supplement their incomes through sons of age and gender might be masking more
part-time work will, in general, prefer to work interesting phenomena.
more hours per week.
In order to better understand the satisfactions Frames of Reference
and frustrations of different demographic Besides types of work and demographic sta-
groups, it is important to know more about what tus, research suggests that perceptions of equity
expectations they bring with them to the job, also influence part-time workers' feelings of sat-
what needs they are trying to fulfill, and what isfaction. As Rotchford and Roberts (1982) and
goals they are trying to accomplish. For exam- Miller and Terborg (1979) pointed out, a major
ple, little behavioral research has been con- theoretical issue which needs to be addressed
ducted about either what motivates different de- vis-&-vis part-time workers is the frame of refer-
mographic groups to enter the work force on a ence they use to evaluate their experiences. Tra-
part-time basis, or what (besides job dissatisfac- ditionally, that frame of reference was full-time
tion) motivates them to leave. It could be spec- workers who held the same job, and for some
ulated that part-time students enter the work groups of part-time workers that, indeed, may
force for the money (Greenberger & Steinberg, be the case. However, for others, full-time work-
1986), that "late-entering" homemakers take ers may not be the frame of reference. Much
part-time jobs to develop new work skills, and more research is needed on the different frames
that older workers enter, in large part, for stim- of reference part-time workers use and the ex-
ulation and diversion (Kahne, 1985; Soumerai & tent to which their partial inclusion influences
Avorn, 1983); however, these assumptions have their perceptions of equity and feelings of job
not been tested empirically. Since it might be satisfaction. Several hypotheses, such as those
expected that these motivations for working presented next, might be able to guide these
part-time influence employees' experiences on future research efforts.
the job and their tenure in the organization,
Hypothesis 11: Full-time workers in the same or-
such research is clearly warranted. ganization are more likely to be taken as the
With few exceptions (e.g., Hom, 1979; Jackof- frame reference by part-time workers who are (a)
sky & Peters, 1987), patterns of demographic permanent, (b) organization-hired, (c) year-round,
variables also have not been sufficiently inves- and (d) working solely at one part-time job.
tigated. Researchers should move away from Hypothesis 11 suggests that the more perma-

108

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions
nent and regular the part-time job, the more are identified, both the extent and the conse-
likely the part-time workers will be to use their quences of that inequity will be largely un-
full-time counterparts as a frame of reference. known.
Part-time employees who are permanently em- In addition, the frame of reference issue
ployed solely by one firm on a year-round basis should be considered in terms other than those
will have the most contact with comparable full- of the immediate work situation. For some indi-
time workers, and they will have the most viduals, part-time work plays a very minor role
knowledge about the relative rewards and per- in their lives; it is more a diversion than an in-
formance-outcome contingencies of part-time come-generating proposition, or it clearly takes
versus full-time work. a lower priority than family, children, or leisure
Hypothesis 12: Part-time student workers are
pursuits. For others, part-time work plays a very
more likely to use as their comparative others major role in their lives; the pay is essential to
part-time student workers at other organiza- meeting living expenses, and their performance
tions. on the job may be critical for obtaining longer
Most part-time student workers are only tem- hours, better job duties, or full-time jobs. In con-
porarily employed by an organization. They are trast, it might be speculated that part-time work
less likely to be familiar with the organization's is much less central in the lives of students than
pay and benefits schedules for full-time workers, it is to regularly employed adult workers, and it
and they are less likely to consider themselves is less central to moonlighters who hold other
as entitled to comparable treatment. Hypothesis jobs than to retired persons who have no other
12 suggests that part-time student workers are employment. Understanding the role of part-
much more likely to be attuned to differences in time work within a person's overall life as well
their jobs as compared to the jobs of their stu- as understanding it in contrast to the work expe-
dent friends (part-time workers at other compa- riences of others are major avenues for future
nies) rather than to those of their full-time col- researchers to pursue.
leagues. A third issue on frames of reference that needs
additional research is the distinction between
Hypothesis 13: Moonlighters are more likely to part-time workers and peripheral workers. In
use as their frame of reference their full-time
some research studies, the term part-time work-
colleagues on their main jobs.
ers is used synonymously with the term periph-
For moonlighting employees, the rate of pay eral workers, an aggregate of employees who
for a second job is likely to be compared to the have low career involvement, low organiza-
rate of pay for their primary jobs. Because many tional commitment, and low job attachment.
moonlighters work extra hours to make up for However, part-time work does not necessarily
insufficient income from one job, a second job, imply feelings of being peripheral, and full-time
which pays substantially less per hour than their workers in some organizations may feel just as
primary job, might be highly undesirable. Thus, peripheral as their part-time counterparts. It is
Hypothesis 13 suggests that moonlighters are important to discover which factors lead part-
likely to use as their frame of reference the full- time workers to feel peripheral to their organi-
time colleagues on their main jobs. zations and which factors lead to feelings of high
Gaining more knowledge about these frame involvement. Some variables that might be in-
of reference issues is vital for researchers who vestigated in this light are socioeconomic status,
study part-time workers because implicit in the involvement in family activities and/or commu-
study of job attitudes and job performance is the nity affairs, status or skills required by the occu-
notion of perceived inequities with full-time pation, and the social climate on the job. For
workers. However, until the comparative others example, part-time professional workers (such

109

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions
 as nurses or teachers) might be much more in- Work-context
volved in their jobs than part-time menial labor- factors
ers are; older workers might be much more com-
mitted to their organizations if those organiza-
Types of part-time _ _ , Job attitudes
tions provide a desired pleasant social work arrangements s and behaviors
atmosphere.

Causal Models of Part-Time Work Demographic

Traditionally, researchers of part-time work factors
have assumed that part-time work has a direct Figure 2. Proposed framework for research.
impact on job attitudes and behaviors, whereas
some work-context factors (e.g., wages) and de-
mographic variables (e.g., gender) are the mod- 1987; Ronen, 1984), and union activity (Moberly,
erating variables. This framework for research 1986), moderate the relationships between type
is displayed in Figure 1. of employment arrangement and job attitudes/
However, past research as well as the hypoth- job behaviors. Most of the research on part-time
eses presented here suggest that the causal re- work has assumed that work-context factors
lationships may be more complex, and a more moderate these relationships (James & Brett,
complex model of these causal relationships ap- 1984), that is, researchers have assumed implic-
pears in Figure 2. itly that the relationships between part-time
Although many types of part-time work ar- work and job attitudes are a function of the level
rangements can influence job attitudes and be- of a third variable (e.g., wages, fringe benefits,
haviors, there may be some reciprocal causality or flexibility in scheduling). However, it is also
as well; for example, individuals who enjoy and possible that these work-context variables medi-
place greater value on work may be more likely ate the relationship between part-time work and
to be employed full-time (or for relatively long job attitudes, that is, the impact of part-time
hours part-time). Similarly, workers who have work on job attitudes may be indirect through
many years of professional training and educa- these work-context variables. Part-time work, for
tion may be more motivated to seek full-time em- example, may lead to lower pay and fringe ben-
ployment, despite family responsibilities, be- efits, which in turn leads to lower job satisfaction
cause of their professional socialization. and higher turnover. Future research should
Research on part-time workers has previously address whether these work-context variables
noted that certain work-context factors, espe- are moderator variables, mediating variables,
cially wages (Owen, 1978), fringe benefits or in some cases, as James and Brett (1984) sug-
(Moberly, 1986), flexibility of scheduling (Levine, gest, both.
Similarly, as discussed earlier, demographic
factors may not only moderate the relationships
Work-context between types of part-time work arrangements
moderators
and job attitudes/job behaviors, but also may
have direct effects on the types of part-time work
Part-time Jr to which employees gravitate and on important
-- Job attitudes
employment and behaviors attitudinal and behavioral outcomes.
A At this stage in the research on part-time
workers, there has been little attention paid to
Demographic causal models, and even less attention paid to
moderators issues of reciprocal causation. Through confir-
Figure 1. Traditional framework for research. matory statistical techniques and more rigor-
110

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions
ously stated research problems, a greater un- job behaviors, and on the ways they moderate
derstanding of the causal relationships among the relationships between part-time work ar-
variables in the area of part-time work can be rangements and important job outcomes. Al-
gained, and, consequently, a greater under- though a variety of work-context factors have
standing of the nature and consequences of been identified as relevant to the experiences of
part-time work can be gained as well. part-time workers (e.g., scheduling flexibility
and availability of fringe benefits), further re-
search is needed to discover whether these
Conclusion work-context factors serve as mediating or mod-
erating variables between part-time work ar-
As the number of part-time workers continues rangements and job attitudes and behaviors.
to increase, and as they become more critical to There are other avenues for improving the
the labor market, gaining a better understand- methodological rigor of research on part-time
ing of the nature and consequences of part-time employment, and these include a focus on cau-
employment becomes more important as well. sal models and reciprocal causality. Research-
There are several directions for future research ers should sample not only across heteroge-
and investigation. neous job types, but also across important work-
In terms of theory development, researchers context factors, demographic variables, and
should pay careful attention to the types of part- types of work arrangements.
time work arrangements employees face and In terms of management practice, compensa-
the differing consequences these arrangements tion and scheduling are issues that seem most
have for workers' attitudes and performance. In important to pursue. As the work force ages, re-
addition, researchers should be more con- search on ways of attracting and retaining older
cerned with studying the comparative others workers will be important. The notion of using
that part-time employees use to evaluate their part-time workers to buffer full-time employees
jobs, and looking more carefully at the role that from unpleasant jobs or layoffs also should be
part-time work plays within an individual's investigated further.
overall approach to his or her life. Across all areas of future research, however,
The roles of demographic factors and work- one common theme emerges. It is time to move
context factors have implications for both theory research on part-time workers away from bi-
and research methods. More research is needed variate comparisons with full-time workers and
on the ways demographic factors influence toward more theoretically and methodologically
workers to gravitate to part-time work, or the sophisticated examinations of complex sets of re-
ways they directly influence job attitudes and lationships.

References
Best, F. (1981) Work sharing. Kalamazoo, MI: Upjohn Insti- specific organizational characteristics and overall satis-
tute. faction. Academy of Management Journal, 27, 893-900.
Bosworth, T. W., & Holden, K. C. (1983) The role part-time
Gannon, M. J. (1975) Management of peripheral employees.
job options play in the retirement timing of older Wiscon-
Personnel Journal, 54, 482-486.
sin state employees. Aging and Work, 6(1), 31-36.
Deuterman, W. V., Jr., & Brown, S. C. (1978) Voluntary part- Gannon, M. J., & Nothern, J. C. (1971) A comparison of short-
time workers: A growing part of the labor force. Monthly term and long-term employees. Personnel Psychology, 24,
Labor Review, 101(1), 3-10. 687-696.
Eberhardt, B. J., & Shani, A. B. (1984) The effects of full-time Greenberger, E., & Steinberg, L. (1986) When teenagers
versus part-time employment status on attitudes toward work. New York: Basic Books.
1ll

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions
Greenwald, C. S., & Liss, J. (1973) Part-time workers can Moberly, R. B. (1986) Temporary, part-time, and other atyp-
bring higher productivity. Harvard Business Review, ical employment relationships in the United States. Labor
51(1), 20-21. Law Journal, 37, 689-696.
Hall, D. T., & Gordon, F. E. (1973) Career choices of married Nakamura, A., & Nakamura, M. (1983) Part-time and full-
women: Effects on conflict, role behavior, and satisfaction. time work behavior of married women: A model with a
Journal of Applied Psychology, 58, 42-48. doubly truncated dependent variable. Canadian Journal
Haring, M. J., Okun, M. A., & Stock, W. A. (1984) A quanti- of Economics, 16, 229-257.
tative synthesis of literature on work status and subjective
Nardone, T. T. (1986) Part-time workers: Who are they?
well being. Journal of Vocational Behavior, 25, 316-324.
Monthly Labor Review, 109(1), 13-19.
Harriman, A. (1982) The work/leisure trade-off. New York:
Praeger. Nollen, S. D. (1978) Permanent part-time employment. New
York: Praeger.
Hart, R. A. (1987) Working time and employment. Boston:
Allen & Unwin. Owen, J. D. (1978) Why part-time workers tend to be in low-
wage jobs. Monthly Labor Review, 101(1), 11- 14.
Hom, P. (1979) Effects of job peripherality and personal char-
acteristics on the job satisfaction of part-time workers. Pfeffer, J., & Baron, J. N. (1988) Taking the workers back out:
Academy of Management Journal, 22, 551-565. Recent trends in the structuring of employment. In B. M.
Howe, W. J. (1986) Temporary help workers: Who are they, Staw & L. L. Cummings (Eds.), Research in organizational
what jobs they hold. Monthly Labor Review, 109(1), 47-49. behavior (Vol. 10, pp. 257-304). Greenwich, CT: JAIPress.

Jackofsky, E. F., & Peters, L. H. (1987) Part-time versus full- Presser, H. B. (1986) Shift work among American women
time employment status differences: A replication and ex- and child care. Journal of Marriage and the Family, 48,
tension. Journal of Occupational Behavior, 8, 1-9. 551-563.
Jackofsky, E. F., Salter, J., & Peters, L. H. (1986) Reducing Ronen, S. (1984) Alternative work schedules. Homewood, IL:
turnover among part-time employees. Personnel, 63(1), Dow Jones-Irwin.
41-43.
Rotchford, N. L., & Roberts, K. H. (1982) Part-time workers as
James, L. R., & Brett, J. M. (1984) Mediators, moderators, and
missing persons in organizational research. Academy of
tests for mediation. Journal of Applied Psychology, 69,
Management Review, 7, 228-234.
307-321.
Kahne, H. (1985) Reconceiving part-time work. Totowa, NJ: Rothberg, D. S., & Cook, B. E. (1985) Part-time professional.
Rowman and Allanheld. Washington, DC: Acropolis Books.

Levine, H. Z. (1987) Alternative work schedules: Do they Soumerai, S. B., & Avon, J. (1983) Perceived health, life sat-
meet workforce needs? Personnel, 64(1), 57-62. isfaction, and activity in urban elderly: A controlled study
of the impact of part-time work. Journal of Gerontology,
Logan, N., O'Reilly, C., & Roberts, K. H. (1973) Job satisfac-
38, 356-362.
tion among part-time and full-time employees. Journal of
Vocational Behavior, 3, 33-41. Wakefield, D. S., Curry, J. P., Mueller, C. W., & Price, J. L.
Mahlin, S. J., & Charles, J. (1984) Peak-time pay for part-time (1987) Differences in the importance of work outcomes be-
work. Personnel Journal, 63, 60-65. tween full-time and part-time hospital employees. Journal
of Occupational Behavior, 8, 25-35.
Miller, H. E., & Terborg, J. R. (1979) Job attitudes of part-time
and full-time employees. Journal of Applied Psychology, Werther, W. B. (1976) Mini-shifts: An alternative to overtime.
64, 380-386. Personnel Journal, 55, 130-133.

Daniel C. Feldman (Ph.D., Yale University) is Profes-

sor of Management and Business Partnership Foun-
dation Fellow at the University of South Carolina Col-
lege of Business Administration. Correspondence re-
garding this article can be sent to him at the
University of South Carolina, College of Business Ad-
ministration, Department of Management, Columbia,
SC 29208.
The author wishes to thank Barton A. Weitz and
Nancy R. Klich for their comments on the manuscript.

112

This content downloaded from 185.2.32.21 on Thu, 12 Jun 2014 17:05:58 PM

All use subject to JSTOR Terms and Conditions