Available online at www.sciencedirect.

com

Learning and Individual Differences 18 (2008) 166 – 175

Beliefs and perceptions about the future: A measurement

Abstract

Human's ability to consider the future, willingness to make sacrifices in the present to obtain something better in the future has been a
significant part of our success as a species (Suddendorf, T., & Corballis, M. C. (1997). Mental time travel and the evolution of the human mind.
Genetic, social, and general psychology monographs 123, 133–167.). Although the importance of thinking about the future is apparent to many
educators it has only begun to gain prominence within educational research (Kauffman, D. & Husman, J. (2004). Effects of time perspective on
student motivation: Introduction to a special issue. Educational Psychology Review, 16, 1–7.). So that educational researchers may continue to
consider some of the many dimensions of “future thinking” more fully, this article discusses Future Time Perspective Theory and presents some
evidence for the validity of four constructs within FTPT. Over the course of three studies four subscales consistently emerged: Extension, Speed,
Connectedness, and Value.
© 2008 Elsevier Inc. All rights reserved.

Keywords: Motivation; Time perspective; Post-secondary education

1. Introduction Lacante, 2004). FTP looks at time not as a physical thing we all
share, but rather as an individual psychological phenomenon.
The study of humans' perceptions of the future has been Research on the development of students' mental represen-
examined under the heading of Future Time Perspective (FTP). tations of the future has only just begun (Suddendorf & Busby,
FTP relates to the perception of time rather than actual physical 2005). Researchers across psychological domains seem to agree
time as it passes on the calendar or as recorded by the clock. The that adolescence is a critical period for development of students'
temporal context (i.e., how far individuals plan into the future), present and future identity (Kerpelman & Mosher, 2004; Simons
the clarity with which individuals perceive future needs, and the et al., 2004). Researchers also agree that by late adolescence
degree to which the present is connected to the past and the (if not before) students' mental representations of the future do
future, describe the time space that individuals consider when influence their academic motivation (Kerpelman & Mosher,
making decisions about their present. The needs of the indi- 2004; Malka & Covington, 2005; Oyserman, Gant, & Ager,
vidual become the objects (goals) within this temporal context. 1995; Pizzolato, 2006). For these reasons, research on time per-
The further INTO the future an individual's time perspective is spective across psychological domains has focused on college
extended, the greater the number of goals and plans to reach students' time perspective (Bembenutty & Karabenick, 2004;
those goals an individual has (Simons, Vansteenkiste, Lens, & Levy & Earleywine, 2004; Malka & Covington, 2005; Sheldon,
Vansteenkiste, Strathman, & Joireman, 2005; Vansteenkiste,
Simons, Soenens, & Lens, 2004; Wilson & Ross, 2003).
⁎ Corresponding author. Box 870611 College of Education, Arizona State
Although evidence exists that college students' thoughts
University, Tempe AZ 85287, United States. Tel.: +1 480 965 3993; fax: +1 480 about their futures do have an impact on their academic achieve-
965 0300. ment (Malka & Covington, 2005; Shell & Husman, 2001;
E-mail address: Jenefer.Husman@asu.edu (J. Husman). Simons, et al., 2004) and study persistence (Shell & Husman,
1041-6080/$ - see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.lindif.2007.08.001
 J. Husman, D.F. Shell / Learning and Individual Differences 18 (2008) 166–175 167

2001) and temporal perspective has been included in the be related to behavior that might reduce exposure to the
Expectancy × Value model of motivation (Wigfield & Eccles, HIV virus (Rothspan & Read, 1996). A similar construct delay
2002), future time perspective has not been as widely incor- discounting (willingness to forgo a small present reward for the
porated into conceptions of self-regulated and strategic learning promise of a larger future reward) has been shown to contribute
(e.g., Graham & Weiner, 1996; Pintrich, 2003; Zimmerman, to smoking cessation (Audrain-McGovern et al., 2004).
1998). Educational psychologists for the most part have not De Volder and Lens (1982) research was the basis for the
examined students' beliefs about time and their own abilities to initial development of the instrument described in the current set
influence or manage the passage of time. We believe that this of studies. In their study, De Volder and Lens measured valence
lack of examination is partly due to a lack of suitable instruments by asking participants the importance of various goals that had
for assessing FTP among student populations. Although ins- different valences. They argued that a participant, who valued
truments have been created that measure aspects of FTP (Shell & proportionately more distant goals, would also have a stronger
Husman, 2001; Zimbardo & Boyd, 1999), a comprehensive future time perspective than a participant who valued fewer
measure of FTP constructs has not yet been available. Also, distant goals. This measure was intended to be both context
many existing FTP instruments have unsatisfactory reliability specific and to assess general individual differences in partici-
and construct validity. In the research presented here, we attempt pants valuing of the future. Results of this study indicated that
to clarify the constructs in future time perspective research that both valence and instrumentality impact student achievement.
are most useful for post-secondary educational research and to
introduce a psychometrically sound instrument that measures 1.3. Connectedness
these constructs.
A cognitive aspect of FTP is the ability to make connections
1.1. Relevant dimensions of time perspective between present activities and future goals.
The cognitive aspect of FTP makes it possible to anticipate the
FTP has a long history, particularly in European psychology,
more distant future; to dispose of longer time intervals in which
of being studied as acquired and rather stable sets of beliefs and
one can situate motivational goals, plans, and projects; and to
expectations about the future (Lens & Rand, 1997; Tromms-
direct present actions toward goals in the more distant future.
dorff, 1994; Zimbardo & Boyd, 1999). The dimensions we
As a consequence, those actions acquire a higher utility value
focused on in the development of our instrument – valence,
(Eccles & Wigfield, 2002), and the present activities are
extension, speed and connectedness – are constructs that re-
perceived as more instrumental (Miller et al., 1999; Simons,
searchers have traditionally thought of as part of the relatively
2001). (Simons et al., 2004, p. 123)
stable FTP set of beliefs. Although viewed as stable, in the short
term, evidence indicates these aspects of time perspective can We wish to differentiate between the perception of utility
change and are rooted in social context and culture (Jones, 1994; value that comes from the view that a specific activity is
Seginer & Halabi, 1991; Bond & Smith, 1996). instrumental for choosing a future goal and the general tendency
to make connections between present activities and future goals.
1.2. Valence As described by many authors (e.g., Brown & Jones, 2004;
Simons, et al., 2004; Shell & Husman, 2001) individuals with
De Volder and Lens (1982) have operationalized valence strong FTP will, all things being equal, be more likely to make
as the importance individuals place on goals attainable in connections between their present activities and future goals.
the future. The conception of valence has also been operatio- Many researchers have tried to capture the cognitive as-
nalized as delay discounting (Holt, Green, & Myerson, 2003; pects of FTP with varying degrees of success. Daltrey and
Schoenfelder & Hantula, 2003; Simpson & Vuchinich, 2000; Langer (1984) developed a scale called the Daltrey Future Time
Trope & Liberman, 2000). Future time perspective theorists Perspective Test that contained 5 subscales: extension, coher-
have argued that all things being equal, goals that are more ence, directionality, density, and attitude. These subscales were
distant in time are going to be perceived as less valuable than moderately-to-highly correlated with each other (0.42 to 0.86).
more immediate goals. For those with a strong FTP, however, Daltrey and Langer concluded that, because of the moderate to
the value of goals that are within an individual's time per- high correlational relationships, the 5 subscales were actually
spective will be less affected by the lack of proximity (Husman representative of an underlying unitary construct. They per-
& Lens, 1999; Raynor & Entin, 1983). Valuing long-term future formed an exploratory factor analysis (Varimax rotation) on
goals is one indicator of a strong FTP. the scores for the 5 subscales. This secondary principal factor
Valuing of the future, or valence, has been shown to be analysis produced only 1 principal component, indicating that 1
associated with adaptive behavior and positive motivation in secondary factor made up of the 5 subscales did in fact exist.
post-secondary educational settings (Shell & Husman, 2001; However, other researchers could not successfully replicate
Turner & Schallert, 2001) and secondary educational setting Daltrey and Langer's results. A factor analysis of the Daltrey
(De Volder & Lens, 1982). Research on the willingness to Future Time Perspective Test revealed 13 factors, not 4, or 1
sacrifice the present for future goals, although often not labeled (Regeth, 1996). Regeth did not find meaningful patterns in any
Valence or Value, has focused on adult populations, predomi- of the 13 factors, and the items that were intended to create the 5
nantly in health psychology. For example it has been shown to original subscales did not load together. The primary factor did
168 J. Husman, D.F. Shell / Learning and Individual Differences 18 (2008) 166–175

contain the majority of the items, however. After examining the between wishes, dreams, and goals. Individuals may have dreams
primary factor, Regeth retained 10 of the original 80 items. for 10 or 20 years into the future, but when making decisions, they
Those ten items proved to be reasonably internally consistent will generally not take those dreams into account. Another aspect
(alpha coefficient was 0.78) and have reasonable construct, of extension not measured by the existing inventories is the
predictive, and concurrent validity. differential perception in the temporal closeness of goals. Our
We believe that a general concern for the future and plan inventory attempts to examine students' extension using a survey
fullness about the future links these scales. Some example items method referencing points common to undergraduate students.
will help to illustrate this point. “I have been thinking a lot about
what I am going to do in the future.” (Gjesme, 1979, pp. 179); 1.5. Speed
“Since I feel that the future is something beyond my control,
planning for it becomes a waste of time.” (Regeth, 1996); and, “I The ability to anticipate and plan for the future has been
consider how things might be in the future, and try to influence attributed to strong cognitive FTP. When he developed a scale to
those things with my day to day behavior.” (Strathman, Gleicher, measure students' orientation toward the future Gjesme (1983)
Boninger, & Edwards, 1994, pp. 752). Each of these items deals included a subscale that focused on the lack of plan fullness
with either a concern for or lack of concern for planning for the for the future, needing external regulation to manage upcoming
future. We have termed this general feeling of connectedness to events, and a sense of being overwhelmed by approaching dead-
and plan fullness about the future connectedness. Our measure lines. This subscale represented the speed at which individuals
of connectedness has predicted student achievement in post- feel time is passing. Speed has also recently been included
secondary educational settings (Malka & Covington, 2005; as a dimension of temporal integration (Drakulic, Tenjovic, &
Shell & Husman, 2001). Lecic-Tosevski, 2003).
The purpose of the series of studies presented here is to report
1.4. Extension the development of a scale that reliably assesses post-secondary
students' connectedness, valence, extension, and speed. Most of
Daltrey and Langer (1984) defined extension as “how far the current research on FTP, instrumentality, and time orientation
ahead a person projects one's thoughts” (pg. 719). Goals that has focused on post-secondary students; therefore, in these
exist within an individual's “time horizon” generally seem studies we also focused on post-secondary students.
closer, are more distinct, and hold greater importance for the
individual than goals outside the habitual time space. The more 2. Study 1
extended one's time horizon, the nearer and more important
long-term goals will seem (Tucker, Vuchinich, & Rippens, The purpose of Study 1 was to extend the Future Time
2002). Perspective Scale (FTPS; Shell, 1985; Shell & Husman, 2001)
Extension has been measured using many different techni- to measure four constructs within FTP: Extension, Connected-
ques. One technique, the Rappaport Time Line (Rappaport, ness, Speed and Valence. These instruments were examined
Enrich, & Wilson, 1985) has been used in clinical and coun- both at the construct level and at the item level.
seling psychology with adolescents and adults (Lennings, 2002; The instrument used was a product of scales first developed
Rappaport, 1991; Rappaport et al., 1985; Rappaport & Fossler, by Shell (1985; see also Shell & Husman, 2001) and Gjesme
1993). Participants are given a blank 24-in. strip of paper, and (1979). Shell's instrument was based on the concepts of
told to think of it as representing their entire life. They are asked instrumentality and valence as proposed by De Volder and Lens
to place significant life experiences from the past, present, and (1982). The primary difference between Shell's and De Volder
future on the line, and indicate an age for each experience. and Lens' instruments was that Shell's instrument attempted to
Extension is measured through computing the total number of measure a more general conception of both instrumentality and
years that the experiences listed extend into the past or future. valence. By not being specific to academic life or to specific
The most common self-report measure is the Life Events classes he hoped to avoid some of the problems discussed
Inventory (Nurmi, 1991). Using this technique, participants are earlier. Shell reported generally good psychometric properties
asked to list events or future states of being that they feel will for these scales, with Cronbach's alpha coefficients of 0.78 for
occur. The participants are then asked to indicate the age they the valence scale and 0.83 for the instrumentality scale. Full
think they will be when they reach a given event. The mean age details of factor analysis and item analysis can be found in Shell
the participants project themselves to be when each event (1985) and Shell and Husman (2001). Both Shell's scale and
is reached is then used as an index of FTP extension (e.g., Gjesme's scale were designed to examine future time per-
Trommsdorff & Lamm, 1980). Listing techniques have been used spective as a more general trait-like construct. We expected
to examine post-secondary age students' extension (Lennings, many of Gjesme's items to merge with Shell's to produce a
1994). This technique, however, has been criticized because it more accurate measure of connectedness. Speed was measured
encourages subjects to think about the future in ways they might using a subscale from Gjesme's scale. Given that extension
not have on their own (Lens, 1988). Because differentiating has been considered an important part of FTP for many years
between wishes, dreams, and goals would significantly affect the (Lennings & Burns, 1998; Stouthard & Peetsma, 1999; Vazquez
outcome on these measures, it is important to target only goals. & Rapetti, 2006 Daltrey & Langer, 1984), we used Daltrey
The Life Events Inventory, however, has no way to differentiate and Langer's definition of extension “how far ahead a person
 J. Husman, D.F. Shell / Learning and Individual Differences 18 (2008) 166–175 169

projects one's thoughts.”(p. 719) as a starting point and con- the item component loadings and the scree plot. Based on the
structed questions and incorporated them into the composite examination of the scree plot (Cattell, 1966), we retained four
questionnaire (see Table 4 for example items). principal components. Component loadings that exceeded 0.30
Our final instrument contained 47 items: 14 items from were considered to be salient; if an item failed to load on any
Gjesme's scale (6 Involvement items; 3 Anticipation items; 3 component, it was removed. Two items that had both low
Occupation items; and 2 Speed items), 25 items from Shell's loadings (b0.30) and loaded on more than one component were
scale (16 instrumentality items and nine valence items) and 6 removed (see Table 1).
original items to measure extension. Each item contained a stem Although some of the items from Gjesme's (1979) scale did
statement that the participants were asked to rate using a five produce one independent component, many of the items from
point Likert scale 1 (Strongly Disagree) to 5 (Strongly Agree). that scale loaded on the same factors as did those from Shell's
We expected that four factors would be found: valence, con- (1985) scale. Gjesme's involvement items also loaded with
nectedness, extension, and speed. items from Shell's instrumentality factor to create the current
connectedness scale. Items from both speed and anticipation
3. Method loaded together to create what we are now referring to simply as

3.1. Participants
Table 1
Over the course of two semesters, a total of 398 under- Study 1 item component loadings
graduate student volunteers (245 women; 148 men) from upper- Item Component Component Component Component
division classes in Educational Psychology completed the
1 2 3 4
questionnaire. The students were part of the Educational Psy-
Connectedness 1 0.68 − 0.04 0.22 − 0.18
chology subject pool and were required to participate in a
2 0.62 − 0.02 0.23 − 0.06
research to complete the class; students who did not wish to 3 0.62 − 0.14 0.03 0.21
complete the research study were given an alternate assignment. 4 0.60 − 0.20 − 0.15 0.25
The students in these courses are typically students from across 5 0.59 − 0.09 0.21 − 0.07
the university, as the course meets requirements for general 6 0.59 − 0.09 − 0.07 0.19
7 0.56 − 0.16 − 0.14 0.17
studies. Although the students ranged in age from 18 to 53 the
8 0.56 − 0.04 0.19 0.02
modal age of the students was 22. Each of the participants 9 0.55 − 0.05 0.18 0.06
received credit as part of a research participation requirement in 10 0.51 − 0.16 0.17 − 0.04
their course. 11 0.51 0.14 0.24 − 0.14
12 − 0.48 0.28 − 0.05 0.31
13 0.48 − 0.07 − 0.04 − 0.03
3.2. Procedure
14 − 0.48 0.17 − 0.20 0.28
15 − 0.47 0.22 − 0.27 0.27
The Future Time Perspective Scale (FTPS) was administered 16 − 0.47 0.25 − 0.06 0.16
in paper and pencil form by the researchers and graduate student 17 0.46 0.09 0.11 0.07
assistants to groups of 20–50 participants as one of several 18 − 0.45 0.04 − 0.13 0.33
19 0.42 0.06 0.19 0.01
questionnaires dealing with students' beliefs and attitudes
20 − 0.38 0.25 − 0.20 0.30
toward school. The participants came to the survey room at their 21 0.37 − 0.16 0.19 0.14
convenience once during a three-week period in the middle of Value 1 − 0.16 0.75 0.03 − 0.03
the semester. The FTPS was the first scale the students were 2 0.03 0.74 − 0.04 0.01
asked to complete. The students were told that they were being 3 − 0.07 0.72 − 0.04 0.07
4 − 0.11 0.69 − 0.13 0.11
asked to complete these questionnaires because we were at-
5 − 0.10 0.65 0.03 0.00
tempting to develop questionnaires to measure students' beliefs. 6 − 0.07 0.61 0.00 − 0.01
Students were allowed as much time as they needed to complete 7 − 0.07 0.59 − 0.02 − 0.05
the questionnaires. 8 − 0.11 0.52 − 0.04 − 0.07
9 − 0.33 0.34 − 0.15 0.24
Extension 1 − 0.07 0.20 − 0.69 − 0.12
4. Results
2 − 0.07 0.12 − 0.68 0.00
3 0.11 − 0.11 0.59 0.25
Principal component analysis with Varimax rotation was 4 − 0.12 0.12 − 0.57 − 0.03
used to establish the construct validity of the four subscales of 5 0.22 0.16 0.43 0.00
the FTPS. Although Exploratory Factor Analysis (EFA) with an 6 0.11 0.06 0.43 − 0.09
7 0.23 0.05 0.36 0.06
oblique rotation is often employed to factor analyze pilot data,
Speed 1 0.08 0.14 0.23 0.68
our goal with this initial examination of the scale was to analyze 2 0.16 0.10 0.24 0.66
the orthogonal structure of the adapted scale and to determine if 3 0.19 0.02 0.29 0.62
clean components could be extracted. In order to determine 4 − 0.21 0.00 − 0.17 0.43
whether the instrument contained the expected four compo- 5 − 0.05 − 0.11 − 0.09 0.35
6 0.01 − 0.04 0.05 0.35
nents, or whether some other solution was better, we examined
170 J. Husman, D.F. Shell / Learning and Individual Differences 18 (2008) 166–175

speed. Two of the items Gjesme had labeled “occupation” the items that comprise this subscale emphasize valuing the future
loaded with our extension items to create the extension factor. over the present.
The third component was comprised of the items from
4.1. Reliability Gjesme's scale. These items were a combination of the speed
and anticipation items from Gjesme's scale. The concept which
After the elimination of four items that had low loadings on these items seemed to have in common were that they dealt with
their primary component, the reliability of the scale created by a lack of ability to structure time, and a feeling that time is
using the remaining items in the scale was calculated. Nine moving quickly, too quickly perhaps. We thought this may have
items remained in the valence component with a Cronbach something to do with a lack of ability to manage or control time.
alpha of 0.83; twenty one items remained in the connected- In hopes of both understanding this component better and
ness component with a Cronbach alpha of 0.88; seven items increasing its reliability we added two items that concerned
remained in the extension component with a Cronbach alpha of planning for and anticipating the future. This lack of ability
0.70; and six items remained in the speed component with a to control time has been a central theme in many studies of
Cronbach alpha of 0.64. FTP (Bjorgvinsson & Wilde, 1996; Zimbardo & Boyd, 1999;
Regeth, 1996).
5. Discussion We labeled the fourth component Extension. Four of the
seven items that had been created to measure extension loaded
The first component combined items from Gjesme's scale onto this component. Occupation items from Gjesme's scale
and Shell's scale. The high loading of Gjesme's involvement also loaded onto this component. After reading the items which
items on a component dominated by items from Shell's instru- loaded onto this component it was clear that each of the items
mentality seemed consistent with our conceptualization of discussed individuals' perception of the distance into the future;
connectedness. The content of the items which represent this passage of time was a major theme.
component do appear to focus on an individual's connectedness After completing the exploratory principal component anal-
to and involvement in the future. The content of the items from ysis, colleagues examined the items for coherence and clarity
Gjesme's scale which loaded with Shell's instrumentality scale within each of the components. The examination led us to
also seem to focus on an individuals' connectedness to and change the wording of some of the items [e.g. “The future is too
involvement in the future (see Table 2). From previous research uncertain to be used as a guide to action today” was changed to
(Regeth, 1996), it was expected that some of Gjesme's items “The future is too uncertain to help me decide what to do today.”
would reflect connectedness. We feel that the combination of and “It is important to go without now, to reach long range
items from Gjesme's and Shell's scale strengthened the scales goals.” was changed to “I will go without things now to reach
ability to measure this aspect of connectedness, which we be- long range goals.”].
lieve to be unique to FTP research; the belief that present de- Overall, the results of this study indicated that there are four
cisions and actions affect future outcomes. separate constructs that comprise our measure of FTP. These
The second component was valence. Shell's scale remained constructs are consistent with the constructs present in the
intact, and none of the items from Gjesme's scale had high literature, and they all refer to students' perceptions and beliefs
loadings on this component. This was expected because Gjesme's about the future. We recognized that our next step was to refine
scale does not claim to measure any aspect of valence. All of the the instrument itself.
items that Shell had labeled as “valence” in his original study had
strong loadings on this component. Although Shell had labeled 6. Study 2
this subscale “Valence” in his original study, after further
consideration, we decided to rename this subscale “Value” as Using the information gathered in Study 1, the revised scale
was constructed using the items that we had retained from Study
1. This scale also incorporated the changes in wording as
described above. So that the instrument could be applicable to a
Table 2 broader range of educational settings, we decided to eliminate
Example items from the subscale Connectedness
the two questions that dealt specifically with exams.
Example items from Shell's scale Item-scale r We added two items relating to structuring the future and two
One should be taking steps today to help realize future goals. 0.59 extension items, both of which asked the students if six months
What one does today will have little impact on what 0.41 seemed like a long period of time. Studies have indicated that at
happens 10 years from now.
six months individual differences in extension become apparent
Planning for the future is a waste of time.a 0.59
(Lens & Moreas, 1994). We also changed the wording of one of
Example items from Gjesmes' scale the extension items so that it would measure a similar time period.
I have been thinking a lot about what I am going to do in the 0.45 We changed “Sometimes it seems like the day will never end”
future. which had originally loaded with other extension items, to
It's really no use worrying about the future.a 0.51 “Sometimes it seems like the semester will never end”. At the time
I think about the future only to a very small extent.a 0.56 the students were given the questionnaire, the end of the semester
Note: a = items are negatively scored. was five months in the future. The final scale had 46 items. Each
 J. Husman, D.F. Shell / Learning and Individual Differences 18 (2008) 166–175 171

item contained a stem statement that the participants were asked very few items. Therefore, for the final study, we added items to
to rate using a standard Likert scale of 1—Strongly Disagree to the Speed and Extension scales. These changes resulted in the
5—Strongly Agree. We expected that the four components would final 33-item questionnaire that was used for Study 3.
be found again: value, connectedness, extension, and speed. Certain changes were noted in the reliability of the scale
following the addition (two for connectedness and two for
6.1. Participants extension) and elimination (two items dealing with exams) of
specific items, as well as alteration of the wording of several
A total of 380 undergraduate student volunteers (gender others. Specifically, reliability dropped from 0.83 in the first
unknown) from 16 sections of a multi-section lower-division study to 0.76 for valence and 0.89 to 0.80 for connectedness
class in Educational Psychology were asked to complete the between the first and second studies. In addition, reliability for
questionnaire. The participants were all enrolled in a course extension (0.60 to 0.74) and speed (0.64 to 0.66) was increased.
focused on providing learning strategies instruction. Students This suggests that item changes were relatively successful for
take this course because they have heard it was not difficult, improving the reliability of the extension and speed subscales,
they are required to as part of their academic probation, or be- while also hindering the reliability of the valence and connec-
cause someone told them it would help them be more successful. tedness scales. Although item changes may have been a signif-
Each of the participants received credit as part of the homework icant factor in the observed differences between Study 1 and
requirement in their course. Study 2, another factor may have also influenced the results. As
noted earlier, samples for both studies were relatively dissimilar
6.2. Procedure in nature. Study 1 subjects were drawn from upper-division
educational psychology courses, whereas those in Study 2 were
Participants were recruited at the beginning of the semester. drawn from a lower-division course (first and second year
Each student was given the FTPS to take home as part of a subjects). Therefore, it is possible that subjects from each study
larger course evaluation. Each instructor handed out the ques- may have differed both in terms of age-related differences and
tionnaire to their students and requested that they return the
questionnaire on the following Monday. This gave them one
Table 3
weekend to complete the questionnaire. Students were given Study 2 item component loadings
course credit for returning the FTPS by the due date.
Item Component Component Component Component

7. Results 1 2 3 4
Connectedness 1 0.66 0.01 0.09 − 0.02
This instrument was analyzed using the Varimax rotation 2 0.64 − 0.02 0.09 − 0.14
3 0.61 0.08 0.01 0.04
principal component analysis. Using both scree plot and eigen-
4 0.61 − 0.06 0.05 − 0.01
values, we again verified that the four-component solution was 5 0.53 0.23 0.07 0.03
best. These components were again identified as value, connec- 6 0.52 0.14 − 0.04 − 0.02
tedness, extension, and speed (see Table 3). Nine items that 7 0.52 0.08 0.04 − 0.05
were found to have a low loading (b.30) and/or multiple loadings 8 0.50 0.24 0.24 0.04
9 0.48 0.02 0.10 0.01
were eliminated.
10 0.47 0.22 0.14 0.01
11 0.46 0.23 0.10 − 0.10
7.1. Reliability 12 0.46 0.00 0.09 − 0.29
13 0.46 0.18 0.19 − 0.15
After the elimination of items that had low loadings on their 14 0.42 0.15 − 0.06 0.01
15 0.33 0.01 0.10 − 0.11
primary component, the reliability of the scale was estimated.
Value 1 0.03 0.66 0.01 0.08
Seven items remained in the value component with a Cronbach 2 − 0.12 0.66 0.06 − 0.03
alpha of 0.76; 15 items remained in the connectedness com- 3 0.27 0.63 0.09 − 0.01
ponent with a Cronbach alpha of 0.80; six items remained in the 4 0.10 0.63 − 0.01 0.14
extension component with a Cronbach alpha of 0.74; and five 5 0.17 0.62 0.15 − 0.08
6 0.26 0.56 0.06 0.11
items remained in the speed component with a Cronbach alpha
7 0.17 0.54 0.00 − 0.13
of 0.66. Extension 1 − 0.15 − 0.08 − 0.79 0.18
2 0.04 0.13 0.78 0.02
8. Discussion 3 − 0.04 0.02 − 0.78 0.10
4 0.20 0.21 0.59 − 0.17
5 − 0.06 0.09 − 0.58 0.03
As with Study 1, survey items again loaded on the four com-
6 0.15 0.17 0.36 0.24
ponents identified as components of Future Time Perspective in Speed 1 − 0.09 − 0.06 − 0.09 0.79
the literature: connectedness, valence, speed, and extension. 2 − 0.04 − 0.11 − 0.06 0.75
Some items did not load on any of the components and were 3 − 0.14 0.11 − 0.12 0.74
eliminated. This occurred for several items that were originally 4 0.21 0.05 0.33 0.47
5 − 0.22 0.09 − 0.15 0.40
intended for the speed and extension scales leaving them with
172 J. Husman, D.F. Shell / Learning and Individual Differences 18 (2008) 166–175

cohort-effects. If this was indeed the situation, then changes solution was estimated by using the maximum likelihood pro-
made to the questionnaire following Study 1 may have held less cedure. The FTPS measurement model was estimated by the
relevance for subjects in Study 2. It is interesting to note, LISREL 8.01 statistical program (Joreskog & Sorbom, 1993). It
however, that although the populations from Study 1 were has been suggested that more than one goodness-of-fit statistic
dissimilar from those in Study 2 the component analysis pro- needs to be reported (Bollen, 1989). We have chosen six mea-
duced an almost identical component structure for the scale, sures to determine adequate fit of the proposed model to the data:
indicating that although not yet perfect the scale may be useful the chi-square fit statistic (χ2), the chi-square ratio (χ2/df), the
for all college populations. As a final check of the structure goodness-of-fit statistic, non-normed fit index(NNFI), Normed
of the 33 item questionnaire a final confirmatory analysis was Fit Index (NFI), and the comparative fit index (CFI). Because of
conducted. In study three the structure of the scale was tested the high sensitivity of the χ2 procedure to sample size and
using Structural Equation Modeling, with a sample drawn from because models involving many variables tend to have high χ2
a course similar to that of Study 1. values, it has been suggested that adjusted fit indexes be reported
(Hoyle, 1995). If the χ2 is nonsignificant, it is assumed the
9. Study 3 proposed model is consistent with the data. The GFI, NNFI, CFI
and NFI all need to be above 0.9 for the model to be considered a
9.1. Participants good fit with the data (Bollen, 1989). Bollen suggests that a χ2
ratio below 2.5 is also an indicator of model good fit.
For the third study a total of 473 undergraduate student
volunteers (257 women; 212 men, and 4 of unknown gender)
from upper-division classes in Educational Psychology were
asked to complete the questionnaire. Although the students were
primarily upper-division students some freshman and sopho- Table 4
Future Time Perspective Scale
mores did volunteer (18 = Freshman; 25 = Sophomores; 125 =
Juniors; 301 = Seniors). These students were recruited from the Subscale Item
course used in Study 1, but were recruited during a subsequent Speed: I find it hard to get things done without a deadline.
semester. All students were part of the educational psychology I need to feel rushed before I can really get going.
I always seem to be doing things at the last moment.
subject pool and as such were required to participate in a research
Extension August seems like a long way off.
study or complete an alternate assignment. It often seems like the semester will never end.
Half a year seems like a long time to me.
9.2. Procedure In general, six months seems like a very short period of time.
September seems very near.
Value Given the choice, it is better to get something you want in the
The FTPS was administered to the participants as the first of
future than something you want today.
several questionnaires dealing with students' beliefs and Immediate pleasure is more important than what might happen
attitudes toward school. The authors and graduate student re- in the future.
search assistants administered the pencil and paper question- It is better to be considered a success at the end of one's life
naires to groups of 20–50 participants. The participants came to than to be considered a success today.
The most important thing in life is how one feels in the long
the survey room at their convenience once during a three-week
run.
period in the middle of the semester. Each student was allowed It is more important to save for the future than to buy what one
1 h to complete the questionnaires. Most students completed the wants today.
questionnaires within the allotted time. All students completed Long range goals are more important than short range goals.
the FTPS. What happens in the long run is more important than how one
feels right now.
Connectedness I don't think much about the future.
10. Results I have been thinking a lot about what I am going to do in the
future.
10.1. Reliability It's really no use worrying about the future.
What one does today will have little impact on what happens
ten years from now.
Cronbach's (1951) alpha coefficients were estimated for the
What will happen in the future is an important consideration in
four subscales. The Cronbach's alpha coefficients were greater deciding what action to take now.
than 0.70 for all four subscales indicating that the scale has good I don't like to plan for the future.
internal reliability. The specific alphas were 0.72 for the valence It's not really important to have future goals for where one
subscale, 0.82 for the connection subscale, 0.72 for the speed wants to be in five or ten years.
One shouldn't think too much about the future.
subscale, and 0.74 for the extension subscale.
Planning for the future is a waste of time.
It is important to have goals for where one wants to be in five or
10.2. Construct validity ten years.
One should be taking steps today to help realize future goals.
To examine the construct validity of the final scale, we con- What might happen in the long run should not be a big
consideration in making decisions now.
ducted a confirmatory component analysis. A four-component
 J. Husman, D.F. Shell / Learning and Individual Differences 18 (2008) 166–175 173

In the component loading matrix, the patterns of parameters Our scale focuses on this perception as the operationalization of
were restricted so that each item was estimated only under the Speed.
latent component (i.e., the four subscales) that it was hy- Many instruments have been created in order to operationa-
pothesized to represent. The other loadings were fixed at 0. lize FTP. Our intent was to distill the constructs of future time
After the initial CFA, the modification indices of the model perspective and time orientation down to the fundamental com-
were examined. Six items were eliminated based on their poor ponents that are uniquely representative of FTP. We believe that
fit and their lack of conceptual fit. All of the items eliminated in our connectedness, extension, speed, and valence compo-
had been weak in the two previous studies, having low or mul- nents, we have accomplished this task. We have found a stable
tiple loading. These items could conceptually represent more than set of constructs that reflect a persons' FTP that can be reliably
one construct. The adjusted model was then estimated. The fit measured. This series of studies has provided evidence of the
indexes of the adjusted model indicated a good fit χ2(318, construct validity of the FTPS, further research examining di-
N = 453) 465.45, p b 0.000, χ2/df = 1.46, GFI = 0.93, NNFI = 0.94, vergent validity between the FTPS and other measures of future
NFI = 0.84, CFI = 0.94. time perspective (e.g., the Rappaport Time line, Zimbardo Time
Perspective Scale). Additional research is needed to examine
11. General discussion the predictive validity of the FTPS, specifically examining the
expected positive relationship between the FTPS and educa-
The final version of the FTPS has 27 items (see Table 4). We tional and motivational beliefs such as self-efficacy and goal
have shown that the dimensions represented by the four sub- orientation.
scales in the instrument were distinct and stable across two Much of the research on FTP has been conducted with
exploratory and one confirmatory component analyses in three populations outside of the United States. The population that
studies. The strong and replicated four-component structure of our participants were recruited from is made up of citizens of
this instrument indicates that FTP, as we have conceptualized it, the United States and has only a small minority representation.
is not a unitary construct as has been argued by some research- Many studies have shown that time perspective is strongly af-
ers (Daltrey & Langer, 1984; Gjesme, 1979, 1983). Rather, fected by culture (e.g. Jones, 1988, 1994). Replication of these
consistent with other future time perspective researchers (Lens, results with college students from other cultures, nationalities, and
1988; Nuttin & Lens, 1985; Nurmi, 1991), FTP appears to be a ethnic backgrounds will be important. Future research should also
multi-dimensional phenomenon which has both dynamic and include test–retest reliability and convergent validity with other
cognitive components. measures of FTP.
The importance individuals place on goals attainable in
the future has been referred to as Valence (De Volder & Lens, References
1982). In our scale we use the word “Value” to represent the
valuing of the future, particularly the willingness to sacrifice in Audrain-McGovern, J., Rodriguez, D., Tercyak, K. P., Epstein, L. H., Goldman,
the present to support the future. This aspect of future time P., & Wileyto, E. P. (2004). Applying a behavioral economic framework to
understanding adolescent smoking. Psychology of Addictive Behaviors, 18,
perspective is present in other measures of future orientation/
64−73.
perspective (Brown & Jones, 2004; Zimbardo & Boyd, 1999). Bembenutty, H., & Karabenick, S. A. (2004). Inherent association between
The cognitive aspect of FTP that incorporates plan fullness for academic delay of gratification, future time perspective, and self-regulated
the future, the tendency to make connections between present learning. Educational Psychology Review, 16, 35−57.
activities and future goals and outcomes, as well as a general Bjorgvinsson, T., & Wilde, G. J. S. (1996). Risky health and safety habits related
to perceived value of the future. Safety Science, 22(1–3), 27−33.
concern for future consequences is represented in our scale by
Bollen, K. A. (1989). Structural equations with latent variables. Oxford, England:
Connectedness. The items in this scale are a combination of John Wiley & Sons.
items from Gjesme's Future Orientation Scale and Shell's Time Bond, M. H., & Smith, P. B. (1996). Cross-cultural social and organizational
Perspective scale (Shell, 1985; Shell & Husman, 2001). One psychology. Annual Review of Psychology, 47, 205−235.
aspect of FTP that has been examined across many sub- Brown, W. T., & Jones, J. M. (2004). The substance of things hoped for: A study
of the future orientation, minority status perceptions, academic engagement,
disciplines of psychology is Extension (Okhuysen, Galinsky, &
and academic performance of black high school students. Journal of Black
Uptigrove 2003; Rappaport, 1991). Extension is the amount Psychology, 30(2), 248−273.
time that is contained within an individual's habitual time space. Cattell, R. B. (1966). The meaning and strategic use of factor analysis. In R. B.
The FTPS asks students to indicate if activities outside of a six Cattell (Ed.), Handbook of multivariate experimental psychology (pp. 243−274).
month time frame are perceived as “far away”. FTP theory and Chicago: Rand McNally.
Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests.
research indicates that activities that occur outside an indi-
Psychometrika, 16, 297−334.
vidual's habitual time space or time perspective will seem Daltrey, M. H., & Langer, P. (1984). Development and evaluation of a measure
farther away than those things that occur within the time space. of future time perspective. Perceptual and Motor Skills, 58(3), 719−725.
The concept of Speed, first articulated by Gjesme (1983) is De Volder, M., & Lens, W. (1982). Academic achievement and future time
focused on the speed at which time seems to move. Within an perspective as a cognitive motivational concept. Journal of Personality and
Social Psychology, 42, 566−571.
individual's personal time space, activities seem to be closer,
Drakulic, B., Tenjovic, L., & Lecic-Tosevski, D. I. (2003). Time integration
but time is also manageable, if an individual is not future ori- questionnaire. Construction and empirical validation of a new instrument
ented they will not be as able to organize their future activities for the assessment of subjective time experience. European Journal of
and will therefore perceive the future as ‘rushing toward’ them. Psychological Assessment, 19(2), 101−116.
174 J. Husman, D.F. Shell / Learning and Individual Differences 18 (2008) 166–175

Eccles, J. S., & Wigfield, A. (2002). Motivational beliefs, values, and goals. Pizzolato, J. E. (2006). Achieving college student possible selves: Navigating
Annual Review of Psychology, 53, 109−132. the space between commitment and achievement of long-term identity goals.
Gjesme, T. (1979). Future time orientation as a function of achievement motives, Cultural Diversity and Ethnic Minority Psychology, 12(1), 57−69.
ability, delay of gratification, and sex. The Journal of Psychology, 101, Rappaport, H. (1991). Measuring defensiveness against future anxiety: Telepres-
173−188. sion.Current Psychology, 10 (1/2).
Gjesme, T. (1983). On the concept of future time orientation: Considerations of Rappaport, H., Enrich, K., & Wilson, A. (1985). Relation between ego identity and
some functions and measurements' implications. International Journal of temporal perspective. Journal of Personality and Social Psychology, 48(6),
Psychology, 18, 443−461. 1609−1620.
Graham, S., & Weiner, B. (1996). Theories and principles of motivation. In D. C. Rappaport, H., & Fossler, R. J. (1993). Future time, death anxiety, and life
Berliner, & R. C. Calfee (Eds.), Handbook of educational psychology purpose among older adults. Death Studies, 17(4), 369−379.
(pp. 63−84). New York: Macmillan. Raynor, J., & Entin, E. (1983). The function of future orientation as determinant
Holt, D. D., Green, L., & Myerson, J. (2003). Is discounting impulsive? of human behavior in step-path theory of action. International Journal of
Evidence from temporal and probability discounting in gambling and non- Psychology, 18, 463−487.
gambling college students. Behavioural Processes, 64(3), 355−367. Regeth, R. (1996). A measurement of future time perspective. International
Hoyle, R. H. (1995). Structural equation modeling: Concepts, issues, and Journal of Psychology, 31(3–4), 4495.
applications. Thousand Oaks, CA, US: Sage Publications. Rothspan, S., & Read, S. J. (1996). Present versus future time perspective and HIV
Husman, J., & Lens, W. (1999). The role of the future in student motivation. risk among heterosexual college students. Health Psychology, 15, 131−134.
Educational Psychologist, 34(2), 113−125. Schoenfelder, T. E., & Hantula, D. A. (2003). A job with a future? Delay
Jones, J. M. (1988). Cultural differences in temporal perspectives: Instrumental discounting, magnitude effects, and domain independence of utility for
and expressive. In J. E. McGrath (Ed.), The social psychology of time: New career decisions. Journal of Vocational Behavior, 62(1), 43−55.
perspectives (pp. 21−38). Newbury Park, CA: Sage. Seginer, R., & Halabi, H. (1991). Cross-cultural variations of adolescents' future
Jones, J. M. (1994). An exploration of temporality in human behavior. In R. C. Schank orientation: The case of Israeli Druze versus Israeli Arab and Jewish male.
& M. R. Lepper (Eds.), Beliefs, reasoning and decision making (pp. 389−411). Journal of Cross-Cultural Psychology, 22, 224−237.
Hillsdale, NJ: Lawrence Erlbaum Associates Publishing. Sheldon, K. M., Vansteenkiste, M., Strathman, A., & Joireman, J. (2005).
Jöreskog, K. G., & Sörbom, D. (1993). LISREL 8: Structural equation modeling with Personal goals and time travel: How are future places visited, and is it worth
the SIMPLIS Command Language. Chicago: Scientific Software International. it? Understanding behavior in the context of time: Theory, research, and
Kerpelman, J. L., & Mosher, L. S. (2004). Rural African American adolescents' application (pp. 143−163). Mahwah, NJ, US: Lawrence Erlbaum Associates
future orientation: The importance of self-efficacy, control and responsi- Publishers.
bility, and identity development. Identity, 4, 187−208. Shell, D. (1985). Achievement motivation: Interactive effects of locus of
Lennings, C. J. (1994). An investigation of the effects of agency and time control, expectancy attribution, self-efficacy, goal-setting and future time
perspective variables on career maturity. Journal of Psychology: Inter- perspective on academic performance. Unpublished master's thesis,
disciplinary and Applied, 128(3), 243−253. University of Nebraska, Lincoln, NE.
Lennings, C. J. (2002). Perceptions of time in adolescents and young adults: The Shell, D. F., & Husman, J. (2001). The multivariate dimensionality of per-
use of the time line. Journal of Applied Health Behavior, 4(1–2), 46−51. sonal control and future time perspective in achievement and studying.
Lennings, C. J., & Burns, A. M. (1998). Time perspective: Temporal extension, Contemporary Educational Psychology, 26, 481−506.
time estimation, and impulsivity. Journal of Psychology: Interdisciplinary Simons, J. (2001). De betekenis van verschillende types instrumentaliteit voor
and Applied, 132(4), 367−380. de motivatie en de doelgerichtheid bij studenten. Unpublished doctoral thesis,
Lens, W. (1988). The motivational significance of future time perspective: The KULeuven, Belgium.
homecoming of a concept. Psychologica, 1, 27−46. Simons, J., Vansteenkiste, M., Lens, W., & Lacante, M. (2004). Placing
Lens, W., & Moreas, M. A. (1994). Future time perspective: An individual and motivation and future time perspective theory in a temporal perspective.
a societal approach. In Z. Zaleski (Ed.), Psychology of future orientation Educational Psychology Review, 16, 121−139.
(pp. 23−28). Lublin: Towarzystwo Naukowe KUL. Simpson, C. A., & Vuchinich, R. E. (2000). Reliability of a measure of temporal
Lens, W., & Rand, P. (1997). Combining intrinsic goal orientations with discounting. Psychological Record, 50(1), 3−16.
professional instrumentality/utility in student motivation. Polish Psychological Stouthard, M. E. A., & Peetsma, T. T. D. (1999). Future-time perspective: Analy-
Bulletin, 28(2), 103−123. sis of a facet-designed questionnaire. European Journal of Psychological
Levy, B., & Earleywine, M. (2004). Discriminating reinforcement expectancies Assessment, 15(2), 99−105.
for studying from future time perspective in the prediction of drinking Strathman, A., Gleicher, F., Boninger, D. S., & Edwards, C. S. (1994). The
problems. Addictive Behaviors, 29, 181−190. consideration of future consequences: Weighing immediate and distant
Malka, A., & Covington, M. V. (2005). Perceiving school performance as outcomes of behavior. Journal of Personality and Social Psychology, 66(4),
instrumental to future goal attainment: Effects on graded performance. 742−752.
Contemporary Educational Psychology, 30(1), 60−80. Suddendorf, T., & Busby, J. (2005). Making decisions with the future in mind:
Miller, R. B., DeBacker, T. K., & Greene, B. A. (1999). Perceived Developmental and comparative identification of mental time travel.
instrumentality and academics: The link to task valuing. Journal of Learning and Motivation, 36(2), 110−125.
Instructional Psychology, 26(4), 250−260. Trommsdorff, G. (1994). Future time perspective and control orientation: Social
Nurmi, J. (1991). How do adolescents see their future? A review of the devel- conditions and consequences. In Z. Zaleski (Ed.), Psychology of future
opment of future orientation and planning. Developmental Review, 11, 1−59. orientation (pp. 39−62). Lublin: Towarzystwo Naukowe KUL.
Nuttin, J., & Lens, W. (1985). Future time perspective and motivation: Theory Trommsdorff, G., & Lamm, H. (1980). Future orientation of institutionalized and
and research method. Hillsdale, N.J.: Erlbaum. non institutionalized delinquents and non delinquents. European Journal of
Okhuysen, G., Galinsky, A. D., & Uptigrove, T. A. (2003). Saving the worst for Social Psychology, 10, 247−278.
last: The effect of time horizon on the efficiency of negotiating benefits Trope, Y., & Liberman, N. (2000). Temporal construal and time-dependent
and burdens. Organizational Behavior and Human Decision Processes, 91, changes in preference. Journal of Personality and Social Psychology, 79(6),
269−279. 876−889.
Oyserman, D., Gant, L., & Ager, J. (1995). A socially contextualized model of Tucker, J. A., Vuchinich, R. E., & Rippens, P. D. (2002). Predicting natural
African American identity: Possible selves and school persistence. Journal resolution of alcohol-related problems: A prospective behavioral economic
of Personality and Social Psychology, 69, 1216−1232. analysis. Experimental and Clinical Psychopharmacology, 10, 248−257.
Pintrich, P. R. (2003). A motivational science perspective on the role of stu- Turner, J. E., & Schallert, D. L. (2001). Expectancy-value relationships of shame
dent motivation in learning and teaching contexts. Journal of Educational reactions and shame resiliency. Journal of Educational Psychology, 93(2),
Psychology, 95, 667−686. 320−329.
 J. Husman, D.F. Shell / Learning and Individual Differences 18 (2008) 166–175 175

Vansteenkiste, M., Simons, J., Soenens, B., & Lens, W. (2004). How to become Wilson, A. E., & Ross, M. (2003). The identity function of autobiographical
a persevering exerciser? Providing a clear, future intrinsic goal in an autonomy- memory: Time is on our side. Memory, 11(2), 137−149.
supportive way. Journal of Sport and Exercise Psychology, 26(2), 232−249. Zimbardo, P. G., & Boyd, J. N. (1999). Putting time in perspective: A valid,
Vazquez, S. M., & Rapetti, M. V. (2006). Future time perspective and motivational reliable individual-differences metric. Journal of Personality and Social
categories in Argentinean adolescents. Adolescence, 41(163), 511−532. Psychology, 77(6), 1271−1288.
Wigfield, A., & Eccles, J. S. (2002). The development of competence beliefs, Zimmerman, B. J. (1998). Developing self-fulfilling cycles of academic regu-
expectancies for success, and achievement values from childhood through lation: An analysis of exemplary instructional models. In D. H. Schunk &
adolescence. In A. E. Wigfield & J. S. (Eds.), Development of achievement B. J. Zimmerman (Eds.), Self regulated learning: From teaching to self-
motivation. A volume in the educational psychology series (pp. 91−120). reflective practice (pp. 1−19). New York: Guilford Press.
San Diego, CA: Academic Press.