Citation
Role Models, Possible Selves, Perceived Self-Efficacy, and Perceived Self-Control as Predictors of GPA in College Students

Material Information

Title:
Role Models, Possible Selves, Perceived Self-Efficacy, and Perceived Self-Control as Predictors of GPA in College Students
Creator:
Marshik, Tesia T.
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (57 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.A.E.)
Degree Grantor:
University of Florida
Degree Disciplines:
Educational Psychology
Committee Chair:
Ashton, Patricia T.
Committee Members:
Algina, James J.
Graduation Date:
5/1/2008

Subjects

Subjects / Keywords:
Academic achievement ( jstor )
Cognitive models ( jstor )
College students ( jstor )
Gratification ( jstor )
Learning ( jstor )
Modeling ( jstor )
Role models ( jstor )
Self control ( jstor )
Social psychology ( jstor )
Willpower ( jstor )
Educational Psychology -- Dissertations, Academic -- UF
achievement, college, control, education, efficacy, goals, gpa, models, self, students
Genre:
Electronic Thesis or Dissertation
bibliography ( marcgt )
theses ( marcgt )
Educational Psychology thesis, M.A.E.

Notes

Abstract:
Self-control predicts many important developmental outcomes including achievement, emotional stability, and successful social relationships. The potential for self-control to predict academic achievement in college students is particularly important because few good predictors of academic achievement in college students exist and college retention and graduation rates are decreasing. The purpose of the present study was to identify factors that predict self-control and GPA in college students. Participants included 163 undergraduate students at the University of Florida who completed take-home self-report questionnaires. A recursive path analysis was used to test a model relating students' role models, possible selves, perceived self-efficacy, perceived self-control skills, and GPA. Partial support for the model was found, as total role models predicted the number of balanced possible selves; the number of balanced possible selves predicted Delay of Gratification; perceived self-efficacy predicted perceived self-control; and Planful Thinking predicted GPA. These results are consistent with social cognitive theory, as role models influenced students' possible selves, which in turn influenced students' self-control skills and academic achievement. However, the lack of relationships among some of the variables indicates that more research is needed to examine the relationships among role models, possible selves, and self-control. In particular, researchers should use domain-specific measures of academic possible selves and academic self-control to predict students' academic achievement. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.A.E.)--University of Florida, 2008.
Local:
Adviser: Ashton, Patricia T.
Statement of Responsibility:
by Tesia T. Marshik

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright by Tesia T. Marshik. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
7/11/2008
Classification:
LD1780 2008 ( lcc )

Downloads

This item has the following downloads:


Full Text






control. Specifically, the model depicts the following predictions regarding mediation effects: (a)

the number of balanced possible selves will mediate the relationship between the total number of

role models and the perceived self-efficacy variables, (b) the self-efficacy variables will mediate

the relationship between the number of balanced possible selves and the perceived self-control

variables, and (c) the perceived self-control variables will mediate the relationship between the

perceived self-efficacy variables and GPA. In the conceptual model depicted in Figure 1-1,

ellipses indicate latent variables and rectangles indicate observed variables.

Perceived Self-
Efficacy for Delay of Perceived Delay of
SGratification Gratification


Balanced Perceived Self-
Role Models o Efficacy for Perceived Emotion GPA
(total number) Possible Emotion Regulation
Selves Regulation

Perceived
Self-Efficacy Perceived Planful
for Planful I Thinking
Thinking


Figure 1-1. Theoretical model of the relationships among role models, possible selves, perceived
self-efficacy, perceived self-control, and GPA.

Theoretical Significance

The results of this study will serve to integrate several research areas and will help

psychologists better understand the links between individuals' role models, possible selves, self-

control skills, and academic achievement. Markus and Nurius (1986) have suggested that

possible selves are shaped by and greatly influenced by social, cultural, and environmental

factors and can thus be facilitated or hindered according to the quality of these factors. However,

few studies have examined the connections between individuals' social relationships and their

possible selves, and none have examined the associations between role models and possible









p < .05). There were no significant indirect effects on Perceived Emotion Regulation or

Perceived Delay of Gratification. However, there was a significant indirect effect of the number

of balanced possible selves on Perceived Delay of Gratification (/ = .08p < .05), which was

mediated by Perceived Self-Efficacy for Delay of Gratification. Finally, the only significant

direct effect on GPA was Perceived Planful Thinking (f = .22, p < .05). Support was also found

for a significant indirect effect of Perceived Self-Efficacy for Planful Thinking on GPA (f = .10,

p < .05), which was mediated by Perceived Planful Thinking.









relationships with others. Several researchers (Anthis, Dunkel, & Anderson, 2003; Day,

Borkowski, Punzo, & Howsepian, 1994; Kao, 2000; Knox, Funk, Elliott, & Bush, 2000) have

demonstrated the influence of sociocultural factors on children's possible selves, reporting that

prevalent gender, racial, and ethnic stereotypes were reflected in the types of possible selves

children and adolescents envisioned. Thus, given that possible selves are greatly influenced by

social and cultural factors, it is likely that individuals' positive and negative role models

influence the types of possible selves they construct. However, no studies have specifically

examined the relationship between individuals' role models and their possible selves. In this

study, I examined the relationship between the total number of students' role models (including

positive and negative role models) and their balanced possible selves. Balance means having

both a positive and negative aspect of a future goal, or having a hoped-for self and a

corresponding feared self in the same domain (Oyserman & Markus, 1990). The balance

measure of possible selves incorporates both positive and negative possible selves and should

therefore capture the effects of both positive and negative role models. On the basis of previous

studies, I hypothesized that individuals with more role models have more balanced possible

selves because they are able to envision the positive and negative possibilities associated with

their future goals. The next link in the proposed model is from balanced possible selves to

perceived self-efficacy for self-control.

Balanced Possible Selves and Perceived Self-Efficacy for Self-Control

Research into the likely roles of possible selves suggests that (similar to role models) they

function as incentives for future behavior, and they provide an evaluative and interpretive

context for the current self-concept. Markus and Nurius (1986) suggested, "The efficient

performance of almost any task, whether relatively mundane, or more complex, requires the

construction of the possible self that carries out the action, completes the task, or masters the









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts in Education

ROLE MODELS, POSSIBLE SELVES, PERCEIVED SELF-EFFICACY, AND PERCEIVED
SELF-CONTROL AS PREDICTORS OF GPA IN COLLEGE STUDENTS

By

Tesia T. Marshik

May 2008

Chair: Patricia Ashton
Major: Educational Psychology

Self-control predicts many important developmental outcomes including achievement,

emotional stability, and successful social relationships. The potential for self-control to predict

academic achievement in college students is particularly important because few good predictors

of academic achievement in college students exist and college retention and graduation rates are

decreasing. The purpose of the present study was to identify factors that predict self-control and

GPA in college students. Participants included 163 undergraduate students at the University of

Florida who completed take-home self-report questionnaires. A recursive path analysis was used

to test a model relating students' role models, possible selves, perceived self-efficacy, perceived

self-control skills, and GPA. Partial support for the model was found, as total role models

predicted the number of balanced possible selves; the number of balanced possible selves

predicted Delay of Gratification; perceived self-efficacy predicted perceived self-control; and

Planful Thinking predicted GPA. These results are consistent with social cognitive theory, as

role models influenced students' possible selves, which in turn influenced students' self-control

skills and academic achievement. However, the lack of relationships among some of the

variables indicates that more research is needed to examine the relationships among role models,

possible selves, and self-control. In particular, researchers should use domain-specific measures









Turner, S. (1996). Big brothers: Impact on little brothers' self-concepts and behaviors.
Adolescence, 31, 875-882.

U.S. Census Bureau. (2000). Educational attainment 2000. Retrieved October 10, 2007, from
http://www.census.gov/prod/2003pubs/c2kbr-24.pdf

Weinberg, R., Grove, R., & Jackson, A. (1992). Strategies for building self-efficacy in tennis
players: A comparative analysis of Australian and American coaches. Sport Psychologist,
6, 3-13.

Wills, T. A., & Stoolmiller, M. (2002). The role of self-control in early escalation of substance
use: A time-varying analysis. Journal of Consulting and Clinical Psychology, 70, 986-
997.

Wolfe, R. N., & Johnson, S. D. (1995). Personality as a predictor of college performance.
Educational & Psychological Measurement, 55, 177-185.

Woodward, W. R. (1982). The "discovery" of social behaviorism and social learning theory,
1870-1980. American Psychologist, 37, 396-410.









APPENDIX A
ROLE MODELS QUESTIONNAIRE

Who influences the way you act and the type of person you try to be? We all probably observe
other people who inspire us to behave in certain ways. These people whom we desire to be like,
or to avoid being like, can be thought of as our "role models." Role models may range from close
relatives and friends, to coworkers, superstars, historical figures, and even fictional characters.
Positive role models are individuals who possess desirable qualities and who inspire others to
emulate these qualities. Negative role models are individuals who possess undesirable qualities
and who motivate others to avoid these qualities. Think about your current role models. Who do
you want to be like? Who do you want to avoid being like?

In the spaces below, please list and describe your current positive role models (people you desire
to be like). If you need more room, use the back of this paper. You do not have to provide names,
but please consider the following questions:


What do you admire about this individual?
What positive qualities does he or she possess?
What is his or her relation to you?
In what ways does this person motivate you?


My current positive role models
What I am doing to become like this person:
(People who I want to be like and the desirable
qualities they possess):
(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)









that, from 1983 to 2007, approximately one third of the students who attended college dropped

out before their second year. In addition, the report revealed that non-return rates were increasing

and that little has changed in the last two decades regarding 5-year graduation rates, still

hovering around 50%. The National Center for Education Statistics (2005) reported similar

results for the years between 1989 and 1995 and noted that 5-year graduation rates have not

changed despite increased access to colleges. In a more recent study, the U.S. Census Bureau

(2000) reported that one in three Americans drops out of college, and that this number appears to

be steadily increasing. Clearly, the numbers of students who drop out of college demonstrate the

need to identify factors that contribute to this problem. Perceived self-efficacy has been shown to

have a strong link to academic achievement (Bandura, 1997), hence, it is important for

researchers to identify potential predictors of perceived self-efficacy and academic achievement

at the college level.

It is likely that individuals who have confidence in their ability to engage in self-control

have higher levels of self-control than individuals with lower perceptions of self-efficacy for

self-control. Furthermore, it is likely that individuals with higher self-control achieve at higher

rates because they are better able to organize their resources, plan ahead, engage in more

effective problem solving strategies, and control factors that potentially interfere with successful

performance (such as unwanted thoughts and emotional distractions). These skills may be

particularly important and useful in academic settings, where students are regularly required to

complete assignments and tasks within time constraints, and some evidence suggests self-control

is a significant predictor of academic achievement and school performance. For example,

Mischel and colleagues (Mischel, Shoda, & Peake, 1988; Shoda, Mischel, & Peake, 1990)

conducted a series of studies and reported that children's delay of gratification at age 4









APPENDIX C
PERCEIVED SELF-EFFICACY MEASURE

For each of the following items, indicate how certain you are that you could perform the
following tasks by choosing the appropriate number (1 7) on the following scale and marking it
on the Scantron sheet.

ANSWER SCALE: 0-6
0 1 2 3 4 5 6
Cannot do Moderately Certain
certain can do can do


1. When I am feeling down, I make myself feel better by thinking positive thoughts.
2. When I feel pain, I keep myself from thinking about it by thinking of other things.
3. When I fail, I stop worrying about it by thinking of how I can be successful in the future.
4. When I am faced with a difficult problem, I solve it by taking a step-by-step approach.
5. When I have a lot of work to do, I create a plan to complete it effectively.
6. When I have a bad habit, I overcome it by first identifying everything that supports the habit.
7. When I have to complete an unpleasant task I do it right away.
8. When I can choose a small reward immediately or a larger reward later, I choose to wait for
the larger reward.
9. When I have a difficult job to do, I do it right away even though I would rather be doing
something else.









LIST OF TABLES


Table page

3-1 Intercorrelations, means, and standard deviations of variables in path analysis (N=
1 6 3 ) .......................................................................................... . 4 3

3-2 Total, direct, and indirect effects in the proposed model (N= 163).............................44









Finally, the only perceived self-control factor that predicted GPA was planful thinking.

This finding suggests that students who report that they are better able to plan ahead and envision

how they will approach and solve problems achieve at higher levels than individuals who report

that they lack this self-control skill. It is not clear why the other factors (delay of gratification

and emotion regulation) were not significantly related to GPA, but upon closer inspection of the

perceived self-efficacy items, the items for planful thinking all deal with problem-solving ability

(a skill that is especially relevant to the academic setting). The items for emotion regulation and

delay of gratification are more general and context-free. Future studies should utilize a more

specific, multidimensional measure of academic self-control in order to better predict GPA.

In sum, this study identified some predictors of self-control skills and academic

achievement. These findings are relevant to researchers attempting to identify college students

who are likely to succeed and those who are likely to struggle academically. In particular, this

study identified potential targets of intervention for individuals who are struggling academically

or who lack certain self-control skills. The results of this study suggest that an intervention aimed

at improving the balance of students' possible selves may improve some self-control skills (i.e.,

planful thinking), which may improve GPA. Oyserman et al. (2006) conducted an intervention

aimed at improving the quality of students' possible selves. The intervention was successful in

that students' academic initiative, grades, and standardized test scores improved, while absences

and school misconduct declined. These results suggest that parents, teachers, and other role

models may influence students' academic achievement. Specifically, parents, teachers, and other

significant social influences may influence students' possible selves, which in turn may affect

their ability to regulate their behaviors and responses, which may ultimately affect achievement.









the score for each item could range from -3 to +3, with no neutral response at 0, and total scores

ranging from -108 to +108. However, the scaling method was modified in this study so that

items were scored from 0 to 5, and the total score potentially ranged from 0 to 180. A

preliminary exploratory factor analysis of 176 undergraduate students' responses to the items on

the Self-Control Schedule yielded seven factors, three of which were examined in the present

study because they had the highest reliability estimates and were well represented by the items:

Emotion Regulation, Planful Thinking, and Delay of Gratification. In the sample obtained for

this study, coefficient alpha for these variables were .72, .62, and .59, respectively.

Perceived Self-Efficacy

A nine-item measure of perceived self-efficacy created for this study was used to assess

participants' confidence in their ability to control their thinking, emotions, and behavior (see

Appendix C). These questions have Likert-type response options ranging from 0 (cannot do) to 6

(certain can do). A high score indicates high perceived self-efficacy for self-control. An

exploratory factor analysis of 176 undergraduate students' responses to the items on the measure

yielded three factors: Perceived Self-Efficacy for Delay of Gratification, Perceived Self-Efficacy

for Emotion Regulation, and Perceived Self-Efficacy for Planful Thinking. In this study, internal

consistency estimates using coefficient alpha were .77, .68, and .79, respectively

GPA

As part of a demographic questionnaire, students were asked to report their current overall

GPA (see Appendix D). Students' GPAs ranged from 2.3 to 4.0, with a mean of 3.32 and a

standard deviation of .42.

Procedures

With the permission of course instructors, participants were recruited from courses in

educational psychology. Volunteers were asked to complete the questionnaires either during









ACKNOWLEDGMENTS

It is a pleasure to thank the many people who helped make this thesis possible. The

support, encouragement, and wisdom of the professors in the Educational Psychology and

Psychology Departments have inspired me to do my best. It is difficult to overstate my gratitude

to my advisor and chair, Dr. Patricia Ashton, for her instruction, guidance, patience, and

encouragement throughout this process. I would have been lost without her. I am also greatly

indebted to Dr. James Algina for his excellent instruction and advice regarding statistical

procedures and interpretation, and for spending hours responding to my questions and reviewing

my computer programs. I would also like to thank my professor and friend, Dr. John Bengston,

for his stimulating conversation, challenging questions, ready ear, and incisive mind. The

aforementioned individuals have contributed most to my development as a student, and as a

person. They practice what they teach and are exemplars for their students. Furthermore, I would

like to thank my student colleagues for their stimulating contributions to class discussions and

for their support and encouragement. I am also grateful to Elaine Green and Linda Parsons for

helping the department run smoothly, for always taking care of things, and for assisting me in

many different ways, especially with reminders about deadlines and requirements. Lastly, and

most importantly, I acknowledge with gratitude the enduring support of my family. My husband,

Jesse, has listened with patience to the ramblings of my ideas and endured countless nights of me

working frantically on the computer. He has always encouraged me to pursue my academic

career to its fullest potential and he continually lifts my spirit. I also wish to thank my siblings,

especially my sister, for being excellent role models and teachers throughout my life and for their

perpetual support and encouragement. Finally, I am grateful to my parents and grandparents for

raising me, supporting me, teaching me, loving me, and encouraging me to strive towards

excellence.












Table 3-2. Total, direct, and indirect effects in the proposed model (N = 163)
Variable Effect 1 2 3
1. Total role models Total -- -- --
Direct


2. Balanced possible selves




3. Self-Efficacy for Delay of Gratification




4. Self-Efficacy for Emotion Regulation




5. Self-Efficacy for Planful Thinking




6. Perceived Delay of Gratification




7. Perceived Emotion Regulation




8. Perceived Planful Thinking




9. GPA




Note. -- means the effect is not in the model.
*p<.05.


Indirect -- -
Total .39* --
Direct .39* -
Indirect -- -
Total -.06 -.16
Direct -- -.16
Indirect -.06 --
Total -.02 -.06
Direct -- --
Indirect -.02 --
Total .05 .14
Direct -- --
Indirect .05 --
Total .03 .08* .55*
Direct .03 -- .55*
Indirect -- .08* --
Total .02 .04
Direct -- --
Indirect .02 .04
Total -.02 -.06
Direct -- --
Indirect -.02 -.06
Total .00 .00 .09
Direct -- -- --
Indirect .00 .00 .09


.60*
.60*







-.03


.35*
.35*


.10*


-.03 .10*


--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--

.22*
.22*









academic self-control and found a significant relationship between them (such that higher

balance predicted higher levels of perceived self-control). Furthermore, Oyserman et al. reported

that balance in academic possible selves significantly predicted GPA in high school students.

Future studies should use measures of these variables that refer specifically to academic

achievement.

As hypothesized, the perceived self-efficacy measures predicted their corresponding

perceived self-control factors. Specifically, students' perceived self-efficacy for delay of

gratification predicted their perceptions of their ability to delay gratification; students' perceived

self-efficacy for emotion regulation predicted their perceptions of their ability to regulate their

emotions, and students' perceived self-efficacy for planful thinking predicted their perceptions of

their ability to engage in planful thinking. Thus, the extent to which students believed that they

were capable of regulating their behavioral responses predicted their self-reports of their ability

to do so. These results are consistent with social cognitive theory (Bandura, 1997), which posits

that human behavior is learned observationally through modeling the behaviors, attitudes, and

emotional reactions of others, and that perceived self-efficacy can be affected by social

influences. These results also support Bandura's conception of perceived self-efficacy as being

domain-specific, although all of the self-efficacy factors were significantly, positively correlated

(presumably because they all dealt with self-control in general). Furthermore, perceived self-

efficacy for planful thinking significantly predicted GPA (through an indirect effect mediated by

planful thinking). This result is consistent with an extensive amount of literature indicating that

perceived self-efficacy has a positive impact on individuals' success and achievement (Bandura,

1997; Schunk, 1984, 2003).









Possible Selves

An open-ended questionnaire modeled after Oyserman and Markus (1990) was used to

obtain information about students' possible selves (see Appendix B). On the measure,

individuals are asked to generate their hoped-for and feared possible selves and list their

strategies for obtaining or avoiding them The measure consists of a description of possible selves

and instructs students to think about their own possible selves and then to list all of their negative

and positive possible selves. Cross and Markus (1991) found that college students' responses to

open-ended measures of possible selves ranged from simple one-word descriptions to elaborate

and vivid descriptions of both hoped-for and feared possible selves. The results of additional

studies have shown that open-ended measures of possible selves elicit unique and diverse sets of

individual responses (Knox, Funk, Elliott, & Bush, 1998; Knox et al., 2000; Leonardi et al.,

1998; Oyserman & Markus, 1990). Following the procedures used by Oyserman and Markus

(1990), Oyserman, Terry, and Bybee (2002), and Dunkel and Anthis (2001), participants'

positive and negative possible selves were coded into six categories (achievement, interpersonal

relationships, personality traits, material and lifestyles, physical and health-related, and

negative) and balance was assessed by tallying the number of connections between students'

positive and negative possible selves in the same domain. Possible selves were double coded and

interrater agreement was 94% (all disagreements were discussed to agreement).

Perceived Self-Control

The Self-Control Schedule (Rosenbaum, 1980a) was used to assess participants' self-

control. The Self-Control Schedule is composed of 36 items, and each item is scored on a 6-point

Likert-type scale, with responses that range from very uncharacteristic of me to very

characteristic of me. The Self-Control Schedule score is the sum of the individual items after

reversing the scores of some items. A high score indicates a high level of self-control. Originally,









significantly less depressed, coped better with their disability (epilepsy), and maintained a

stronger belief in their control over their health and their seizures compared to individuals low in

self-control. Simons, Lustman, Wetzel, and Murphy (1985) examined self-control and depressed

patients' response to cognitive behavioral therapy and reported that patients entering cognitive

therapy with relatively high Self-Control Schedule scores responded more favorably to cognitive

therapy than patients with low scores. In contrast, Lewinsohn and Alexander (1990) reported that

adolescents low in self-control exhibited an increased probability of becoming depressed.

Finally, some researchers have reported relationships between self-control (as measured

by the Self-Control Schedule) and self-reported patterns of alcohol consumption and tobacco use

in young adults (Carey, Carey, Carnrike, & Meisler, 1990; Katz & Singh, 1986). Carey et al.

compared undergraduates' self-reported patterns of alcohol consumption and found that the

heaviest drinkers had the lowest scores and that individuals who drank infrequently or not at all

had the highest scores on the Self-Control Schedule. Several studies also suggest that smokers

have significantly lower scores on the Self-Control Schedule than non-smokers (Carey et al.,

1990; Kennett, Morris, & Bangs, 2006). Similarly, Katz and Singh (1986) reported that ex-

smokers scored significantly higher on the Self-Control Schedule than did smokers who had

attempted to quit but failed. On the basis of their results, the authors suggested that ability of

smokers to quit smoking may have been attributable to the better coping skills associated with

high self-control. Kennett et al. replicated these results and reported that individuals who quit

smoking exhibited just as much self-control as individuals who never smoked, even after

controlling for age differences. Taken together, these studies suggest that self-control skills may

protect against substance abuse. Furthermore, these results support the notion that self-control

skills may serve as a protective factor against depression, given that substance use (and abuse) is










of academic possible selves and academic self-control to predict students' academic

achievement.









effects of aversive stimuli compared to individuals with low resourcefulness. From these studies,

it is clear that learned resourcefulness involves a variety of important self-regulatory processes.

Yet despite the promising results associated with this construct, research provides inconsistent

conclusions regarding the factorial structure of the Self-Control Schedule. For example, Gruber

and Wildman (1987) conducted an exploratory factor analysis and subsequently reported that

only three significant factors emerged: problem-focused coping, mood and pain control, and

externality (which the authors concluded is the reverse of self-efficacy). Other researchers

conducted factor analyses across groups. Edwards and Riordan (1994) performed separate

varimax rotations for Black and White students and reported 14 and 12 factors, respectively, that

were difficult to interpret. Redden, Tucker, and Young (1983) also used varimax rotation and

obtained six factors. The authors, however, cautioned that interpretation of the factors should

"proceed tentatively" because of "a lack of a clear, strong factor structure" (pp. 84-85).

Thus, the factor structure of the Self-Control Schedule is still unclear, though research

suggests that the items on the Self-Control Schedule load on more than the four factors that

Rosenbaum proposed as composing the construct of learned resourcefulness (albeit the factors

that emerged in previous studies do reflect similar themes). From the information available, it

appears that all previously published factor analyses were conducted using varimax rotation.

Constraining the factors of the Self-Control Schedule to an orthogonal solution when theory and

the nature of the items suggest they should be correlated likely led to erroneous conclusions. To

address these concerns, I conducted an exploratory factor analysis of the Self-Control Schedule

using promax rotation, which allowed the factors to correlate (Marshik, 2007). Prior to data

collection, two items were eliminated due to their sensitive content. These items ("If I would

smoke two packages of cigarettes a day, I probably would need outside help to stop smoking,"









CHAPTER 3
RESULTS

Preliminary Analyses and Descriptive Statistics

A recursive (unidirectional) path analysis was conducted to test the relationships posited

in the model presented in Figure 1-1 using the statistical software package LISREL 8.0.

Observed scores for the latent variables were used in the analysis, and error variances were fixed

using coefficient a for each measure. Error variances for the perceived self-efficacy variables

were permitted to correlate, as were the error variances for the perceived self-control variables.

Preliminary analyses were conducted to examine to examine the relationships among the

variables. The correlation matrix, as well as means and standard deviations for each of the

variables used in the path analysis are presented in Table 3-1.

As hypothesized, the total number of students' role models were significantly positively

correlated with the number of balanced possible selves (r = .39, p < .05). In addition, all of the

perceived self-efficacy measures were significantly positively correlated. Also as hypothesized,

the perceived self-efficacy measures were significantly positively correlated to the respective

self-control measures. Specifically, Perceived Self-Efficacy for Delay of Gratification was

significantly positively correlated with Perceived Delay of Gratification (r = .38, p < .05),

Perceived Self-Efficacy for Planful Thinking was significantly positively correlated with

Perceived Planful Thinking (r = .37, p < .05), and Perceived Self-Efficacy for Emotion

Regulation was significantly positively correlated with Perceived Emotion Regulation (r = .42, p

< .05). However, each of the perceived self-efficacy measures was also significantly positively

correlated with other perceived self-control skills, although to lesser extents. Perceived Self-

Efficacy for Delay of Gratification was significantly positively correlated with Perceived Planful

Thinking (r = .18, p < .05) and Perceived Emotion Regulation (r = .16, p < .05); Perceived Self-









selves. This study will provide information elucidating the connection between total number of

role models and the number of balanced possible selves.

Theory and research also suggest that individuals' social relationships influence their

perceptions of self-efficacy and self-control abilities. Social cognitive theory emphasizes the

roles of observation and modeling in human learning and development, and Bandura (1997)

suggested that most human behavior is learned observationally through modeling the behaviors,

attitudes, and emotional reactions of others. In this study, I extended this line of research by

examining the relationships between total number of role models, the number of balanced

possible selves, perceived self-efficacy for self-control skills, and perceived self-control skills

(including delay of gratification, emotion regulation, and planful thinking). This research may

elucidate the mechanisms through which role models ultimately influence students' self-control

and academic achievement, which may provide insight into ways that role models can help

students develop academic and social competence. Specifically, I hypothesized that the number

of balanced possible selves and perceptions of self-efficacy would mediate the relationships

between individuals' role models and their perceptions of self-control and GPA.

Finally, some theorists have already suggested that possible selves serve as the

underpinning of perceptions of self-efficacy, self-control, and competence (Markus & Ruvolo,

1989; Ruvolo & Markus, 1992), but none have directly examined these relationships. I extended

this line of research by explicitly examining the relationships among college students' possible

selves, perceived self-efficacy, perceived self-control, and GPA.

In sum, the proposed model integrates critical concepts from social cognitive theory,

motivation theory, self-concept theory, and achievement theory. Results from this study offer

insights into the relationships among role models, possible selves, perceived self-efficacy for









Future research should more closely examine the characteristics of the social relationships that

are most likely to have positive effects.

In conclusion, this study provided some support for the path model relating role models,

possible selves, perceived self-efficacy, self-control, and GPA. Researchers may be able to use

this model as a starting point or reference for their own studies, making modifications as

necessary. Future studies should test the model using measures that are academically-focused, as

it is likely that the predictive power of these variables would be improved if they are context

specific.









In the spaces below, please list and describe your current negative role models (people you do
not want to be like). If you need more room, use the back of this paper. You do not have to
provide names, but please consider the following questions:

What do you not admire about this individual?
What undesirable qualities does he or she possess?
What is his or her relation to you?
In what ways does this person motivate you?


My current negative role models
What I am doing to avoid becoming like this
(People who I do not want to be like and the person:
undesirable qualities they possess):
(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)









LIST OF FIGURES


Figure page

1-1 Theoretical model of the relationships among role models, possible selves, perceived
self-efficacy, perceived self-control, and GPA ............................................................. 27









majority of social cognitive theorists agree that individuals are most likely to model the

behaviors and qualities of individuals with whom they identify, which depends on the degree to

which they perceive the models to be similar to themselves, and the degree of emotional

attachment that is felt toward the models (Thomas, 1990; Woodward, 1982).

Theorists have consistently maintained that identification with role models is critical to

professional, academic, and emotional development (Bandura, 1977; Erikson, 1985). Bandura

suggested that role models serve both informational and motivational functions. Similarly,

Lockwood, Sadler, Fyman, and Tuck (2004) suggested that individuals may use both positive

and negative role models simultaneously as a means of effectively channeling their motivation.

Thus, individuals observe their role models for information regarding how to act and the

consequences of such actions and then use this information as a basis for their decisions about

how to act in the future. Individuals are motivated to model behaviors and qualities to the extent

that they lead to desirable consequences, and individuals are motivated to avoid behaviors and

qualities to the extent that they lead to undesirable consequences. Researchers have repeatedly

demonstrated that role models are sources of motivation and inspiration (Lockwood & Kunda,

1999; Lockwood, Jordan, & Kunda, 2002; Lockwood et al., 2004). To date, most of the research

on role models focuses on positive role models (also referred to as mentors and exemplars); but it

seems more likely that the combination of positive and negative role models has a more powerful

relationship to self-control than positive role models alone. In this study I explored the

relationship between the total number of positive and negative role models and self-control by

examining several potential mediators that might account for the relationship, specifically

possible selves, and perceived self-efficacy for self-control. The rationales for these linkages are

explained in the following sections.









better monitor their progress toward positive or negative outcomes. Oyserman, Bybee, and Terry

(2006) later found that youth with balanced academic possible selves spent more time engaging

in self-control behaviors related to academic achievement (i.e., spent more time doing

homework, were less behaviorally disruptive, and more behaviorally engaged in classroom

activities).

Of special relevant to this study, Cross and Markus (1994) proposed that possible selves

may link effective steps and strategies for solving problems with beliefs about one's ability and

competence in the domain. Similarly, Ruvolo and Markus (1992) suggested that the

"underpinnings of a sense of efficacy, control, and competence are specific, self-relevant

thoughts and feelings, particularly images and conceptions of the self in the future, desired

states" (p. 97). Thus, they proposed that possible selves provide the foundation for perceived

self-efficacy, control, and competence. Bandura (1997) defined perceived self-efficacy as

"beliefs in one's capabilities to organize and execute the courses of action required to produce

given attainments" (p. 3), and studies have revealed that perceived self-efficacy has a positive

impact on individuals' confidence, motivation, perseverance, and success (Bandura, 1997;

Schunk, 1984). As Schunk (2003) observed, when compared with their less efficacious

counterparts, "those who feel efficacious for learning or performing a task participate more

readily, work harder, persist longer when they encounter difficulties, and achieve at a higher

level" (p. 161). Thus, the extent to which individuals believe that they are capable of regulating

their behavioral responses may predict their ability to do so. Bandura further suggested that most

human behavior is learned observationally through modeling the behaviors, attitudes, and

emotional reactions of others and that perceived self-efficacy could be increased by vicariously

experiencing the successes of role models. Hence, the extent to which parents, peers, and other









CHAPTER 2
METHOD

Participants

The sample consisted of 163 college students (124 women, 39 men). Age ranged from

18 to 26 with a mean of 19.89 and a standard deviation of 1.66. The ethnic backgrounds

represented were as follows: 65% White, 13% Black, 12% Hispanic, 5% Asian, and 5% other.

The majority of students were sophomores (37%), followed by juniors (24%), seniors (22%), and

freshman (17%). Participants were recruited from two sources: (a) the Educational Psychology

subject pool (specifically, students in three educational psychology classes, EDF 3110 Human

Growth and Development, EDF 3210 Educational Psychology, or EDF 3135 The Adolescent),

and (b) other undergraduate courses offered by the Educational Psychology Department at the

University of Florida. For students recruited from the Educational Psychology subject pool,

participation in this study fulfilled a research requirement for the course (students who elected

not to participate were given an alternate assignment by their instructor to fulfill the research

requirement). Participants who were recruited from other educational psychology courses

received either extra credit (not to exceed 1% of their grade) or class participation credit,

depending on the preference of the course instructors.

Measures

Role Models

An open-ended question was created to collect information regarding students' role

models (see Appendix A). The measure consists of a brief definition of role models (including

definitions of positive versus negative role models), and students are asked to list and describe in

detail their positive and negative role models. The total number of role models generated by each

participant was tallied and used in the analyses.









B POSSIBE SELVES QUESTIONNAIRE ........................................ ......................... 47

C PERCEIVED SELF-EFFICACY MEASURE ............................................ ............... 49

D DEMOGRAPHIC INFORMATION .......................................................... ............... 50

R E F E R E N C E S .......................................................................... 5 1

B IO G R A PH IC A L SK E T C H .............................................................................. .....................57














































6









Rosenbaum, M. (1983). Learned resourcefulness as a behavioral repertoire for the self-regulation
of internal events: Issues and speculations. In M. Rosenbaum, C. Franks, & Y. Jaffe
(Eds.), Perspectives on behavior therapy in the eighties (pp. 54-73). New York:
Springer.

Rosenbaum, M., & Ben-Ari, K. (1985). Learned helplessness and learned resourcefulness:
Effects of noncontingent success and failure on individuals differing in self-control skills.
Journal of Personality and Social Psychology, 48,198-215.

Rosenbaum, M., & Jaffe, Y. (1983). Learned helplessness: The role of individual differences in
learned resourcefulness. British Journal of Social Psychology, 22, 215-225.

Rosenbaum, M., & Palmon, N. (1984). Helplessness and resourcefulness in coping with
epilepsy. Journal of Consulting and Clinical Psychology, 52, 244-253.

Rosenbaum, M., & Rolnick, A. (1983). Self-control behaviors and coping with seasickness.
Cognitive Therapy and Research, 9, 79-90.

Rubin, K. H., Coplan, R. J., Fox, N. A., & Calkins, S. D. (1995). Emotionality, emotion
regulation, and preschoolers' social adaptation. Development andPsychopathology, 7, 49-
62.

Ruvolo, A. P., & Markus, H. R. (1992). Possible selves and performance: The power of self-
relevant imagery. Social Cognition, 10, 95-124.

Schunk, D. H. (1984). Self-efficacy perspective on achievement behavior. Educational
Psychologist, 19, 48-58.

Schunk, D. H. (2003). Self-efficacy for reading and writing: Influence of modeling, goal setting,
and self-evaluation. Reading and Writing Quarterly, 19, 159-172

Shoda, Y., Mischel, W., & Peake, P. K. (1990). Predicting adolescent cognitive and self-
Regulatory competencies from preschool delay of gratification: Identifying diagnostic
conditions. Developmental Psychology, 26(6), 978-986.

Simons, A. D., Lustman, P. J., Wetzel, R. D., & Murphy, G. E. (1985). Predicting response to
cognitive therapy of depression: The role of learned resourcefulness. Cognitive Therapy
and Research, 9, 79-89.

Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts good
adjustment, less pathology, better grades, and interpersonal success. Journal of
Personality, 72, 271-324.

Thomas, R. M. (1990) Social learning theory. In R. M. Thomas (Ed.), Encyclopedia of
human development and education: Theory, research, and studies (pp. 75-78). New
York: Pergamon Press.









difficulty" (p. 962). In addition, Markus and Nurius have proposed that possible selves are

important motivators because they provide specific, self-relevant goals to work toward or to

avoid, thereby energizing and organizing individuals' behaviors. Further, Cross and Markus

(1991) noted that possible selves help individuals make more direct connections between their

goals and their strategies for attaining them by allowing individuals to simulate their futures,

which enables them to organize and integrate information and strategies relevant to their goals

and to judge the extent to which they are approaching (or avoiding) desired (or undesired)

outcomes.

Markus and Nurius (1986) also proposed that possible selves serve as standards for

comparison and evaluation of the current self. That is, individuals can monitor the status and

development of their current self by envisioning their desired and undesired future selves.

Markus and Nurius further suggested that positive possible selves may be encouraging because

they foster hope and optimism, whereas negative possible selves may be discouraging because

"their associated affect and expectations may stifle attempts to change or develop" (p. 963).

However, negative possible selves may also be motivating to the extent that they highlight the

strategies necessary to avoid undesired outcomes. Research suggests that individuals who have

balanced possible selves appear to have more motivation and control over their behavior than

individuals without such balance. In one study, public school youth had significantly more

balanced possible selves than delinquent youth (Oyserman & Markus, 1990), and balance in

possible selves has been found to have a positive relationship to school persistence (Oyserman &

Markus, 1990; Oyserman, Gant, & Ager, 1995). On the basis of these results, Oyserman and

Markus suggested that individuals with more balance among their possible selves have more

motivational resources because they can envision a greater array of potential outcomes and can









REFERENCES


ACT. (2007a). National collegiate retention and persistence to degree rates. Retrieved December
7, 2007, from http://www.act.org/path/policy/pdf/retain_2007.pdf

ACT. (2007b). 2007 retention/completion summary tables. Retrieved December 7, 2007, from
http://www.act.org/path/policy/pdf/retain_trends.pdf

Akgun, S. (2004). The effects of situation and learned resourcefulness on coping responses.
Social Behavior and Personality, 32, 441-448.

Akgun, S., & Ciarrochi, J. (2003). Learned resourcefulness moderates the relationship between
academic stress and academic performance. Journal of Educational Psychology, 23, 287-
294.

Anderman, E. M., Anderman, L. H., & Griesinger, T. (1999). The relation of present and
possible selves during early adolescence to grade point average and achievement goals.
Elementary School Journal, 100, 3-17.

Anthis, K. S., Dunkel, C. S., & Anderson, B. (2003). Gender and identity status differences in
late adolescents' possible selves. Journal ofAdolescence, 27, 147-152.

Bahniuk, M., Dobos, J., & Hill, S. (1990). The impact of mentoring, collegial support, and
information adequacy on career success: A replication. Journal of Social Behavior and
Personality, 5, 431-451.

Bandura, A. (1977). Social learning theory. New York: General Learning Press.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
Englewood Cliffs: NJ: Prentice-Hall.

Bandura, A. (1991). Social cognitive theory of self-regulation. Organizational Behavior and
Human Decision Processes, 50, 248-287.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.

Bryant, A. L., & Zimmerman, M. A. (2003). Role models and psychosocial outcomes among
African American adolescents. Journal ofAdolescent Research, 18, 36-67.

Calkins, S. D. (1994). Origins and outcomes of individual differences in emotion regulation.
Monographs of the Society for Research in Child Development, 59, 53-72.

Carey, M. P., Carey, K. B., Carnrike, C. L., & Meisler, A. W. (1990). Learned resourcefulness,
drinking, and smoking in young adults. Journal ofPsychology: Interdisciplinary and
Applied, 124, 391-395.









significantly predicted SAT scores years later. More recent studies found that individuals with

high self-control had significantly better grades than individuals with low self-control (Feldman,

Martinez-Pons, & Shaham, 1995; Tangney et al., 2004) and that self-control significantly

predicted grade point average among college students (Wolfe & Johnson, 1995). In this study, I

further examined the relationships among three perceived self-control skills (delay of

gratification, emotion regulation, and planful thinking) and college students' GPA. Furthermore,

I sought to determine whether total number of role models, balanced possible selves, perceived

self-efficacy, and perceived self-control predicted GPA.

Measurement of Self-Control

The most promising of the few measures of self-control available is Rosenbaum's

(1980a) Self-Control Schedule, which was designed to measure learned resourcefulness.

Rosenbaum (1985) defined learned resourcefulness as "an acquired repertoire of cognitive-

behavioral skills by which a person self-regulates internal responses (such as emotions, pain, and

cognitions) that would otherwise interfere with the smooth execution of a target behavior" (p.

200). The Self-Control Schedule is a self-report measure that assesses individuals' general

repertoire of self-control behaviors and their tendencies to use these behaviors when faced with

everyday problems (Akgun & Ciarrochi, 2003). As conceived by Rosenbaum, learned

resourcefulness is multi-faceted and incorporates four aspects: (a) the use of cognitions and self-

instructions to cope with emotional and physiological responses, (b) the application of problem-

solving strategies, (c) the ability to delay immediate gratification, and (d) a general belief in one's

ability to self-regulate internal events (i.e., perceived self-efficacy) (Rosenbaum, 1983).

In sum, the research using Rosenbaum's Self-Control Schedule suggests that individuals

with high resourcefulness believe in their ability to deal with aversive stimuli, use more

beneficial coping and problem solving strategies, and are better able to minimize the negative









and "If I had the pills with me, I would take a tranquilizer whenever I felt tense and nervous")

were replaced with two new items written by the researchers ("I have a hard time waiting for

something that I really want," and "If I have a choice between a smaller reward now or a bigger

reward later, I would choose the smaller reward so that I could have it now"). These two items

were added to increase the number of items referring to delay of gratification, because only a few

items seemed to reflect this construct. The analysis of 176 undergraduate students' responses to

the items on the Self-Control Schedule yielded seven self-control factors: Emotion Regulation,

Perceived Self-Efficacy, Ability to Control Physiological Responses, Planful Thinking, Problem-

Solving Ability, Ability to Control Unwanted Thoughts, and Delay of Gratification. Three of

these factors were used to measure self-control skills in this study because they were well-

represented by the items: Emotion Regulation (a = .72), Planful Thinking (a = .62), and Delay of

Gratification (a = .59).

The Proposed Model and Purpose of the Study

In summary, theoretical accounts and the research literature provide support for the

following predictions represented in the conceptual model in Figure 1-1: (a) the total number of

role models will have a direct relationship to the number of balanced possible selves, (b) the

number of balanced possible selves will have a direct relationship to the perceived self-efficacy

variables (Perceived Self-Efficacy for Delay of Gratification, Perceived Self-Efficacy for

Emotion Regulation, and Perceived Self-Efficacy for Planful Thinking), (b) the perceived self-

efficacy variables will have direct relationships with the corresponding perceived self-control

variables (Perceived Delay of Gratification, Perceived Emotion Regulation, and Perceived

Planful Thinking), (c) the perceived self-control variables will have direct relationships to GPA.

It is also hypothesized that the total number of role models will predict GPA through the

mediating variables of balanced possible selves, perceived self-efficacy, and perceived self-









self-control is perceived self-efficacy, which plays a fundamental role because of its influence on

thought, affect, motivation, and action (Bandura, 1991, 1997). Recent research suggests that self-

control predicts many important outcomes including achievement, adjustment, substance abuse,

emotional stability, and quality of interpersonal relationships (Eisenberg & Fabes, 1992; Shoda,

Mischel, & Peake, 1990; Tangney, Baumeister, & Boone, 2004; Wills & Stoolmiller, 2002). In

light of the importance that self-control plays in psychological functioning, research into the

factors that contribute to the development of self-control and the processes that explain its

widespread influence is needed. Few studies of these issues have been conducted. The purpose of

this study is to examine factors that may predict self-control and grade point average (GPA) in

college students. Specifically, this study will examine the extent to which students' role models

and possible selves affect their perceptions of self-efficacy for self-control, reported self-control

behaviors, and GPA.

Self-Control

Rosenbaum (1980a) offered a promising conception of self-control in his development of

the Self-Control Schedule, a measure that assesses individuals' repertoire of self-control

behaviors and tendencies to use these behaviors when faced with problems (Akgun & Ciarrochi,

2003). As conceived by Rosenbaum, self-control consists of four components: (a) using

cognitions and self-instructions to cope with emotional and physiological distress, (b) applying

problem-solving strategies, (c) the ability to delay immediate gratification, and (d) confidence

(i.e., perceived efficacy) in one's ability to self-regulate thoughts and feelings (Rosenbaum,

1983). The purpose of the present study is to identify factors that predict self-control in college

students.

Research with the Self-Control Schedule has shown consistently that individuals scoring

high on Rosenbaum's (1980a) measure engage in healthier coping tendencies, have higher levels









Role Models and Possible Selves

Lockwood and colleagues (Lockwood, 2002; Lockwood et al., 2004; Lockwood &

Kunda, 1997, 1999) have suggested that role models are motivating to the extent that they point

to plausible positive and negative possible selves and the strategies for achieving or avoiding

them. Possible selves are defined as cognitive self-representations concerning what one expects

to become, what one hopes to become, and what one fears becoming (Markus & Nurius, 1986).

Positive possible selves include those self-representations and goals that one hopes to realize (for

example, one might desire to be wealthy or to be known as a person of integrity) and negative

possible selves refer to those self-representations that one hopes to avoid or fears becoming (for

example, one might fear becoming a college dropout or being known as a liar). Positive possible

selves may be classified as promotion goals because they reflect the personal goals that students

hope to achieve, whereas negative possible selves may be classified as prevention goals because

they reflect outcomes that students hope to avoid. Thus, a mixture of positive and negative role

models may provide the basis for the development of positive and negative possible selves and

the motivation for achieving (or avoiding) them.

Theorists have suggested that possible selves are shaped by social, cultural, and

environmental factors and can be facilitated or hindered according to the quality of these factors.

Although an individual is free to imagine a vast array and unlimited number of possible selves,

the actual collection of possible selves that an individual imagines for himself or herself is

"derived from the categories made salient by the individual's particular sociocultural and

historical context and from models, images, and symbols provided by the media and by the

individual's immediate social experiences" (Markus & Nurius, 1986, p. 954). For example,

children who grow up in abusive households may envision themselves in abusive relationships

later in life; they may have difficulty forming positive possible selves concerning their









class or outside of class depending on the instructor's preference. Volunteers were asked to sign

letters of informed consent prior to completing the questionnaire. Participants completed the

measures in the same order. The entire procedure took approximately 30-45 minutes. The final

sample consisted of 163 participants, after 22 questionnaires were eliminated because of

incomplete data.









APPENDIX D
DEMOGRAPHIC INFORMATION

Please answer the following questions by marking the appropriate number on your Scantron
answer sheets.

1. Gender: If female, mark 0; if male, mark the number 1.

2. Your age: For example, if you are 21, mark the number 2 on item #38 on the Scantron sheet
and mark the number 1 on item #39.

3. Class: 1 Freshman 2 Sophomore 3 Junior 4 Senior
5 Other (please describe)


4. Ethnicity: White = 0, Black = 1, Hispanic = 2, Asian = 3, Other = 4

5. GPA: Please estimate your GPA to two digits. For example, if your GPA is 3.5, mark 3 on
line 42 and 5 on line 43.









of perceived self-efficacy and performance, are more likely to persist after being confronted with

repeated failures, demonstrate greater ability to tolerate pain and seasickness and to cope with

seizures, and exhibit fewer symptoms of stress and depression than individuals low in self-

control (Rosenbaum, 1980a and b; Rosenbaum & Ben-Ari, 1985; Rosenbaum & Jaffe, 1983;

Rosenbaum & Palmon, 1984; Rosenbaum & Rolnick, 1983). In sum, the findings of Rosenbaum

and his colleagues have suggested that individuals high in self-control are better able to minimize

the negative effects of situational stressors on their performance and overall psychological

functioning compared to individuals low in self-control.

Other researchers have found similarly positive results with Rosenbaum's (1983)

measure of self-control. Ginter, West, and Zarski (1989) investigated the relationship between

self-control and coping strategies and found that those high in self-control used more beneficial,

problem-focused coping strategies and reported significantly fewer symptoms of stress than

individuals low in self-control. Later research supported these findings, as Akgun (2004)

reported that persons high in self-control used more positive reappraisal, were more likely to

seek social support, and less likely to use escape-avoidance coping strategies. Akgun also found

that those with high self-control had higher levels of perceived self-efficacy regarding their

abilities to effectively cope with stress. Other investigations into the link between self-control,

stress, and academic performance have suggested that self-control moderates the effect of

academic stress on academic performance. As Rosenbaum and Jaffe (1983) noted, considerable

evidence shows that individuals high in self-control are better able to tolerate and cope with

uncontrollable aversive stimulation.

In addition, several researchers examined the relationship between self-control and

depression. Rosenbaum and Palmon (1984) reported that patients high in self-control were









Table 3-1. Intercorrelations, means, and standard deviations of variables in path analysis (N= 163)
Variable 1 2 3 4 5 6 7 8 9
1. Total role models --
2. Balanced possible selves .39* --
3. Self-Efficacy for Delay of Gratification .00 -.04 --
4. Self-Efficacy for Emotion Regulation -.01 -.01 .41 --
5. Self-Efficacy for Planful Thinking .06 .11 .58* .36* --
6. Perceived Delay of Gratification .00 .09 .38* .22* .22* --
7. Perceived Emotion Regulation .12 .07 .18* .42* .17* .05
8. Perceived Planful Thinking .08 .15 .16* .07 .37* .24* .26* --
9. GPA .11 .01 .09 .13 .11 .20* .08 .19*

M 4.91 1.13 10.54 10.88 8.21 8.38 13.38 9.94 3.32
SD 3.12 1.15 3.91 3.38 2.69 3.82 4.93 4.09 0.42
*p<.05.









In addition to expectations and expected goals, we all have images or pictures of what we don't
want to be like, what we don't want to do, or want to avoid being. First, think a minute about
ways you would not like to be in the future-i-ingg you are concerned about or want to avoid
being like.
Write those concerns or selves to-be-avoided in the spaces below.
In the space next to each concern or to-be-avoided self, mark X in the NO column if you
are not currently working on avoiding that concern or to-be-avoided self and mark X in
the YES column if you are currently doing something so that this will not happen in the
future.
For each concern or to-be-avoided self that you marked YES, use the space at the end of
each line to write what you are doing this year to reduce the chances that this will
describe you in the future. Use the first row for the first concern, the second row for the
second concern and so on.

Am I doing If yes,
In the future, I want to avoid something to What I am doing now to avoid being that
avoid this way in the future
NO YES
1.

2.

3.

4.

5.

6.

7.

8.

9.

10.









Gruber, V. A., & Wildman, B. G. (1987). The impact of dysmenorrhea on daily activities.
Behavior Research and Therapy, 25(2), 123-128.

Hart, D., Atkins, R., & Ford, D. (1998). Urban America as a context for the development of
moral identity in adolescence. Journal of Social Issues, 54, 513-530.

Ibarra, H. (1999). Provisional selves: Experimenting with image and identity in professional
adaptation. Administrative Science Quarterly, 44, 764-791.

Kao, G. (2000). Group images and possible selves among adolescents: Linking stereotypes to
expectations by race and ethnicity. Sociological Forum, 15, 407-430.

Katz, R. C., & Singh, N. (1986). A comparison of current smokers and self-cured quitters on
Rosenbaum's Self-Control Schedule. Addictive Behaviors, 11, 63-65.

Kemper, T. (1968). Reference groups, socialization, and achievement. American Sociological
Review, 33, 31-45.

Kennett, D., Morris, E., & Bangs, A. (2006). Learned resourcefulness and smoking. cessation
revisited. Patient Education and Counseling, 60, 206-211.

Knox, M., Funk, J., Elliott, R., & Bush, E. G. (1998). Adolescents' possible selves and their
relationship to global self-esteem. Sex Roles, 39, 61-80.

Knox, M., Funk, J., Elliott, R., & Bush, E. G. (2000). Gender differences in adolescents' possible
selves. Youth & Society, 31, 287-309.

Kulik, C. T., & Ambrose, M. L. (1992). Personal and situational determinants of referent choice.
Academy ofManagement Review, 17, 212-237.

Leonardi, A., Syngollitou, E., & Kiosseoglou, G. (1998). Academic achievement, motivation,
and future selves. Journal ofAdolescence, 21, 219-222.

Lewinsohn, P. M., & Alexander, C. (1990). Learned resourcefulness and depression. In: M.
Rosenbaum (Ed.), Learned resourcefulness: On coping skills, self-control, and adaptive
behavior (pp. 202-217). New York: Springer.

Lockwood, P. (2002). Could it happen to you? Predicting the impact of downward comparisons
on the self. Journal ofPersonality and Social Psychology, 82, 343-358.

Lockwood, P., Jordan, C., & Kunda, Z. (2002). Motivation by positive or negative role models:
Regulatory focus determines who will best inspire us. Journal ofPersonality and Social
Psychology, 83, 854-864.

Lockwood, P., & Kunda, Z. (1997). Superstars and me: Predicting the impact of role models
on the self. Journal ofPersonality and Social Psychology, 73, 91-103.









may be able to identify students who are more likely to struggle academically or to drop out of

college.





























To my grandparents









The others to which we are continually looking to for information and comparing

ourselves are often referred to as role models. Unfortunately, little consensus exists on a

definition of role models, and the term role model is often used synonymously with terms like

mentor and exemplar. Hence, as Gibson (2004) suggested, "the construct of role models remains

a popularly used but vaguely defined notion" (p. 135). For the purposes of clarification, I have

adopted Gibson's definition of role model as "a cognitive construction based on the attributes of

people in social roles an individual perceives to be similar to him or herself to some extent" and

desires to either increase perceived similarity by emulating those attributes or to decrease

perceived similarity by avoiding those attributes (p. 136). Further, Gibson (2003) defined the

process of role modeling as "a cognitive process in which individuals actively observe, adapt,

and reject attributes of multiple role models" (p. 593). These definitions imply that role models

may be positive or negative. Positive role models are individuals who have achieved success,

who are considered competent in a relevant domain, who exhibit prestige and power, and who

inspire others to emulate certain qualities. Negative role models are individuals who have

experienced some kind of failure or misfortune, who possess undesirable qualities, and who

motivate others to avoid similar adversity (Lockwood, 2002).

An individual's repertoire of role models may range from close relatives and friends, to

coworkers, superstars, historical figures, and even fictional characters (Gibson, 2004; Ibarra,

1999; Kemper, 1968; Lockwood & Kunda, 1997). Situational factors (such as proximity) and

person factors (such as gender and age) affect the number and types of role models available to

an individual (Kulik & Ambrose, 1992). Traditionally, it has been suggested that individuals

emulate models who exhibit four primary characteristics: competence, gender appropriateness,

prestige and power, and individual relevance (Bandura, 1977; Thomas, 1990). However, the









TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ................................................................. ........... ............. .....

L IST O F T A B L E S .............................................................................................

LIST OF FIGURES .................................. .. ..... ..... ................. .8

A B S T R A C T ......... ....................... .................. .......................... ................ .. 9

CHAPTER

1 INTRODUCTION ............... .............................. ........................ .... 11

State ent of the Problem .................. ................................... .... .. ...............11
S elf-C o n tro l ....................................................................................12
R ole M odels and Self-Control ............................................................ ............... 15
R ole M odels and Possible Selves ........................................................ .. .............. 18
Balanced Possible Selves and Perceived Self-Efficacy for Self-Control........................19
Perceived Self-Efficacy for Self-Control and Academic Achievement..........................22
M easurem ent of Self-C control .................................................... ................................24
The Proposed M odel and Purpose of the Study....................................................................26
Theoretical Significance ...................................................... ...... ....... 27
Practical Significance .................................. .. ... .... ...... .. ............29

2 M E T H O D .............. .... ............................................................... 3 1

P artic ip an ts ......................................................................... 3 1
M e a su re s ................... ................... ...................1..........
R o le M o d els ............................................................................... 3 1
P o ssib le S elv es ................................................................3 2
P perceived Self-C control ..............................................................32
Perceived Self-Efficacy ................................... ......... .................. 33
GPA ............ ..................... ............... ...............33
Procedures ...................... ................................. .........33

3 R E SU L T S .............. ... ................................................................ 35

Preliminary Analyses and Descriptive Statistics .......................................35
Analysis of the Proposed Model ........................ .................36

4 DISCU SSION ........... .... ..... ... ........... ....................... ........ 38

APPENDIX

A ROLE MODELS QUESTIONNAIRE .................................................45









Lockwood, P., & Kunda, Z. (1999). Salient best selves can undermine inspiration by
outstanding role models. Journal ofPersonality and Social Psychology, 76, 214-228.

Lockwood, P., Sadler, P., Fyman, K., & Tuck, S. (2004). To do or not to do? Using positive and
negative role models to harness motivation, Social Cognition, 22, 422-450.

Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41, 954-969.

Markus, H., & Ruvolo, A. (1989). Possible selves: Personalized representations of goals. In L. A.
Pervin (Ed.), Goal concepts in personality and social psychology (pp. 211-241).
Hillsdale: Erlbaum.

Marshik, T. (2007). Exploratory factor analysis of the Self-Control Schedule (SCS). Unpublished
manuscript.

Mischel, W., Shoda, Y., & Peake, P. K. (1988). The nature of adolescent competencies predicted
by preschool delay of gratification. Journal ofPersonality and Social Psychology, 54,
687-696.

National Center for Education Statistics. (2005). Student effort and educational progress:
Postsecondary persistence and progress. Retrieved October 23, 2007, from
http://nces.ed.gov/programs/coe/2004/section3/indicatorl9.asp

Oyserman, D., Bybee, D., & Terry K. (2006). Possible selves and academic outcomes: How and
when possible selves impel action. Journal ofPersonality and Social Psychology, 91,
188-204.

Oyserman, D., Gant, L., & Ager, J. (1995). A socially contextualized model of African
American identity: Possible selves and school persistence. Journal ofPersonality and
Social Psychology, 69, 1216-1232

Oyserman, D., & Markus, H. R. (1990). Possible selves and delinquency. Journal ofPersonality
and Social Psychology, 59, 112-125.

Oyserman, D., Terry, K., & Bybee, D. (2002). A possible selves intervention to enhance school
involvement. Journal ofAdolescence, 24, 313-326.

Redden, E. M., Tucker, R. K., & Young, L. (1983). Psychometric properties of the Rosenbaum
schedule for assessing self-control. Psychological Record, 33, 77-86.

Rosenbaum, M. (1980a). A schedule for assessing self-control behaviors: Preliminary findings.
Behavior Therapy, 11, 109-121.

Rosenbaum, M. (1980b). Individual differences in self-control behaviors and tolerance of painful
stimulation. Journal ofAbnormal Psychology, 89, 581-590.









students with less balance. However, the number of balanced possible selves did not significantly

predict perceived self-efficacy for self-control or GPA. In other words, students with more

balance in their possible selves did not feel more efficacious for engaging in self-control, nor did

they have a higher GPA. These findings do not lend support to the claims that possible selves

provide the foundation for perceptions of self-efficacy (Cross & Markus, 1994; Ruvolo &

Markus, 1992), and these results seem to contrast with reports that possible selves are

significantly related to improved GPA (Anderman et al., 1999) and achievement test scores

(Leonardi et al., 1998). There are several possible interpretations of these findings (or the lack

thereof). First, the number of balanced possible selves may predict GPA through other

mechanisms, such as motivation (i.e., the extent to which individuals' possible selves motivate

them to engage in self-regulatory behaviors may predict academic achievement), future time

perspective (i.e., the extent to which individuals see the contingency between their current

actions and their future goals may predict academic achievement), and plausibility (i.e., the

extent to which possible selves are likely to be achieved or avoided given the strategies that are

being used to attain them may predict academic achievement). Second, it is possible that the

balance measure used in this study was too general. In this study, balance was measured by

tallying the number of positive possible selves that had matching negative possible selves in the

same domain, but the final balance score was a sum of balance across all domains. Hence, the

balance measure was not specific to the domains of self-control or achievement (more

specifically, academic achievement). It is also possible that the self-efficacy and self-control

measures were too general. Although balance was indirectly related to delay of gratification, it

may have been more prudent to use specific measures of academic perceived self-efficacy and

academic self-control. Oyserman et al. (2006) used measures of academic possible selves and









Efficacy for Planful Thinking was significantly positively correlated with Perceived Delay of

Gratification (r = .22 p < .05) and Perceived Emotion Regulation (r = .17, p < .05), and

Perceived Self-Efficacy for Emotion Regulation was significantly positively correlated with

Perceived Delay of Gratification (r = .22, p < .05). Finally, both Perceived Delay of Gratification

and Perceived Planful Thinking were significantly positively related to GPA (r = .20, p < .05 and

r = .19, p < .05, respectively).

Analysis of the Proposed Model

According to the path analysis of the proposed model, the goodness of fit test indicates

that 2 (20) = 28.17, p = .11. Thus, the chi-square statistic is not significant, indicating that the

model fits the data. The goodness of fit indices are consistent with this conclusion: the

comparative fit index (CFI) = .96, the root mean square error of approximation (RMSEA) = .05,

the non-normed fit index (NNFI) = .92, and the root mean squared residual (RMR) = .06. Table

3-2 presents the total, direct, and indirect effects specified in the model. All effects were

expected to be positive, and directional hypothesis tests were conducted. Significance was

determined using a .05 Type I error rate.

The total number of role models significantly predicted the number of balanced possible

selves (y = .39, p < .05), and this effect was entirely direct. There were no significant direct or

indirect effects of total role models or the number of balanced possible selves on the perceived

self-efficacy variables. In predicting the perceived self-control factors (Delay of Gratification,

Emotion Regulation, and Planful Thinking), the direct effects of the corresponding perceived

self-efficacy variables were significant. Specifically, Perceived Self-Efficacy for Delay of

Gratification predicted Perceived Delay of Gratification (f = .55, p < .05), Perceived Self-

Efficacy for Emotion Regulation predicted Perceived of Emotion Regulation (l = .60, p < .05),

and Perceived Self-Efficacy for Planful Thinking predicted Perceived Planful Thinking (f = .35,









Cassidy, J. (1994). Emotion regulation: Influences on attachment relationships. Monographs of
the Society for Research in Child Development, 59, 228-249.

Cross, S., & Markus, H. (1991). Possible selves across the lifespan. Human Development, 34,
230-255.

Cross, S., & Markus, H. R. (1994). Self-schemas, possible selves, and competent performance.
Journal ofEducational Psychology, 86, 423-438.

Darwin, C. (1981). The descent of man, and selection in relation to sex. Princeton, NJ:
Princeton University Press. (Original work published 1871)

Day, J. D., Borkowski, J. G., Punzo, D., & Howsepian, B. (1994). Enhancing possible selves in
Mexican American students. Motivation and Emotion, 18, 79-103.

Dunkel, C. S., & Anthis, K. S. (2001). The role of possible selves in identity formation: A
short-term longitudinal study. Journal ofAdolescence, 24, 765-776.

Edwards, D., & Riordan, S. (1994). Learned resourcefulness in Black and White South African
university students. Journal of Social Psychology, 134(5), 665-675.

Eisenberg, N., & Fabes, R. A. (1992). Emotion regulation and the development of social
competence. In M. S. Clark (Ed.), Emotion and social behavior: Vol. 14. Review of
personality and socialpsychology (pp. 119-150). Newbury Park, CA: Sage.

Eisenberg, N., Fabes, R. A., Shepard, S. A., Murphy, B. C., Guthrie, I. K., Jones, S., Friedman,
J., Poulin, R., & Maszk, P. (1997). Contemporaneous and longitudinal prediction of
children's social functioning from regulation and emotionality. ChildDevelopment, 68,
642-664.

Erikson, E. H. (1985). Childhood and society (35th ed.). New York: Norton.

Feldman, S. C., Martinez-Pons, M., & Shaham, D. (1995). The relationship of self-efficacy, self-
regulation, and collaborative verbal behavior with grades: Preliminary findings.
Psychological Reports, 77, 971-978.

Freud, S. (1989). The ego and the id (J. Riviere, Trans.). NewYork: Norton. (Original work
published 1923)

Gibson, D. E. (2004). Role models in career development: New directions for theory and
research. Journal of Vocational Behavior, 65, 134-156.

Ginter, G. G., West, J. D., Zarski, J. J. (1989). Learned resourcefulness and situation-specific
coping with stress. Journal ofPsychology, 123, 295-304.









BIOGRAPHICAL SKETCH

Tesia Marshik majored in psychology and philosophy as an undergraduate. She graduated

cum laude and received a B.S. degree from John Carroll University in Ohio in May of 2005. She

began her graduate career at the University of Florida in August of 2005. She plans to continue

on to earn a Ph.D. in a combined program of developmental and educational psychology. She

currently teaches courses in the Educational Psychology Department at UF. Upon graduation she

plans to pursue a career as a professor at a liberal arts university, where she will conduct

research, teach, and advise undergraduate students.









strongly associated with depression (serving as a risk factor, possible trigger, and concomitant of

depressive symptomology).

In sum, it appears that individuals with high self-control believe in their ability to deal

with aversive stimuli, use more beneficial coping and problem solving strategies, and are better

able to minimize the negative effects of aversive stimuli compared to individuals with low self-

control. From the studies reviewed, it is clear that self-control is related to a variety of important

self-regulatory processes. Despite the important outcomes associated with this construct, the

factors that contribute to the development of self-control are still unclear. Social cognitive theory

(Bandura, 1997) suggests several important factors that may contribute to the development of

self-control. One of the most important may be the relationships individuals develop with other

people, particularly when the people in these relationships become models of behavior for the

individuals. The potential impact of these models on self-control is a major focus of this study.

Role Models and Self-Control

Some researchers have reported that individuals' social relationships influence their

ability to control their impulses and behaviors (Calkins, 1994; Cassidy, 1994). This notion is

consistent with social cognitive theory (Bandura, 1986), which emphasizes the roles of

observation and modeling in human learning and development. Many theorists have proposed

that the processes by which individuals develop self-understanding are inherently social because

individuals continually compare themselves with others to identify their unique characteristics

and to evaluate their own abilities. According to Bandura (1986, 1991), most human behavior is

learned observationally through modeling the behaviors, attitudes, and emotional reactions of

others. In other words, humans have the capacity to learn from observing others' successes and

failures.









APPENDIX B
POSSIBLE SELVES QUESTIONNAIRE

What will you be like in the future? Probably everyone thinks about the future, and when doing
so, we usually think about the kinds of experiences that are in store for us and the kinds of people
we might possibly become. Each of us has some image or picture of what we will be like and
what we want to avoid being like in the future. Think about your future-imagine what you'll be
like, and what you'll be doing.
In the spaces below, write what you expect you will be like and what you expect to be
doing in the future.
After each expected goal, mark X in the NO column if you are not currently working on
that goal or doing something about that expectation and mark X in the YES column if
you are currently doing something to get to that expectation or goal.
For each expected goal that you marked YES, use the space to the right to write what you
are doing to attain that goal. Use the first row for the first expected goal, the second row
for the second expected goal and so on.

Am I am doing If yes,
In the future, I expect to be something to be What I am doing now to be that way in the
that way future
NO YES
1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

































2008 Tesia T. Marshik









self-control, perceptions of self-control, and GPA that may be useful in the development of

further research and theory on these important social, emotional, and cognitive concepts.

Practical Significance

This study elucidates important relationships, which has several practical implications for

future experimental research to determine approaches for increasing college students' ability to

engage in self-control and to improve their academic performance. First, it will provide

information about the possible predictors of self-control. The construct of self-control has

received increasing attention over the years due to its associations with psychopathology. The

concept of self-control is especially popular among researchers and clinicians seeking to

elucidate individual differences in the development of psychopathology. Several studies provide

evidence that poor self-control skills are related to lower levels of social competence (Eisenberg

& Fabes, 1992; Eisenberg et al., 1997) and to internalizing and externalizing disorders (Hart,

Atkins, & Ford, 1998; Rubin, Coplan, Fox, & Calkins, 1995). In addition, research suggests that

self-control predicts many important outcomes including achievement, adjustment, substance

abuse, emotional stability, and quality of interpersonal relationships (Eisenberg & Fabes, 1992;

Shoda et al., 1990; Tangney et al., 2004; Wills & Stoolmiller, 2002). Social-cognitive theory

offers a number of potentially important constructs that may contribute to academic achievement

including role models, possible selves, perceived efficacy for self control, and self-control

behaviors. If researchers can identify mechanisms that underlie perceived self-efficacy and self-

control, they will be better able to help individuals develop appropriate and beneficial self-

control skills, which will increase the likelihood of their success. In addition, this study provides

information about factors that predict academic achievement (as measured by GPA) in college

students. If these factors are shown to increase academic achievement in experimental studies,

important new interventions can be developed to improve academic achievement and educators









ROLE MODELS, POSSIBLE SELVES, PERCEIVED SELF-EFFICACY, AND PERCEIVED
SELF-CONTROL AS PREDICTORS OF GPA IN COLLEGE STUDENTS



















By

TESIA T. MARSHIK


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS IN EDUCATION

UNIVERSITY OF FLORIDA


2008









significant social influences model appropriate self-control skills may affect the development of

individual individuals' possible selves that in turn affect their perceived self-efficacy for self-

control, which in turn affects their ability to regulate their behavioral responses. Some evidence

supports this notion, as researchers have linked positive role models to increased perceived self-

efficacy, psychological well-being, career success, and overall positive self-concepts (Bahniuk,

Dobos, & Hill, 1990; Bryant & Zimmerman, 2003; Turner, 1996; Weinberg, Grove, & Jackson,

1992). In addition, indirect evidence supports the relationship between possible selves and

perceived self-efficacy for self-control. Some researchers have found that positive possible

selves were significantly related to improved grade point average (Anderman, Anderman, &

Griesinger, 1999) and Leonardi, Syngollitou, and Kiosseoglou (1998) reported that the overall

quality of students' possible selves was related to school achievement and task persistence. As

perceptions of self-efficacy are typically found to predict achievement, it is plausible to

extrapolate from these findings that possible selves may predict perceived self-efficacy for self-

control. Consequently, in this study, I extended this line of research by examining whether the

relationship between the number of role models and self-control was mediated by number of

balanced possible selves and perceived self-efficacy for self-control.

Perceived Self-Efficacy for Self-Control and Academic Achievement

One context in which perceived self-efficacy and self-control skills are especially

important is in colleges and universities, and this study will provide information about factors

that might predict academic achievement (as measured by GPA) in college students. The

identification of significant predictors of GPA is important because universities across the nation

are struggling to increase retention rates. Although recent data on dropout rates in colleges are

not available, several studies have revealed that between one third and one half of students who

enter college do not complete their programs. ACT (2007a and b) administered surveys showing









CHAPTER 1
INTRODUCTION

I count him braver who overcomes his desires than him who conquers his enemies, for
the hardest victory is the victory over self.
-Aristotle

He who conquers himself is the mightiest warrior.
-Confucius

He who reigns within himself, and rules passions, desires, and fears, is more than a king.
-John Milton

Statement of the Problem

The ability to control one's passions, thoughts, and behaviors is a fundamental element of

human agency. Self-control has long been heralded by poets, philosophers, and religious leaders.

The failure of individuals to control themselves, to act without reason, or to act contrary to that

which rationality prescribes was considered a major problem even during ancient times when

great thinkers such as Plato, Socrates, and Aristotle took it upon themselves to study akrasia, or

weakness of the will. Freud (1923/1989) later suggested that the human psyche is in a state of

perpetual conflict as the ego struggles to regulate the unrelenting impulses of the id while

balancing the demanding pressures of the superego. Darwin (1871/1981) promoted the role of

self-control when he suggested that "the highest possible stage in moral culture is when we

recognize that we ought to control our thoughts" (p. 123). Recently, the study of self-control in

psychology has garnered increasing attention due to its conjectured role in both internalizing and

externalizing disorders. Self-control refers to those processes (both conscious and nonconscious)

involving the organization and mobilization of resources that enable individuals to guide their

goal-directed behaviors over time and across multiple, dynamic contexts. Aspects of self-control

include the monitoring and regulation of one's thoughts, emotions, and impulses through the use

of self-instructions, self-motivation, self-evaluation and delay of gratification. Another aspect of









CHAPTER 4
DISCUSSION

The results of this study indicate that the total number of role models significantly

predicts balanced possible selves. This finding lends support to theorists' claim that possible

selves are shaped and influenced by social factors. Specifically, this result suggests that

individuals with more role models, including both positive role models (people they admire and

try to emulate) and negative role models (people they try to avoid being like), are better able to

envision the positive and negative possibilities associated with their future goals. Researchers

should further investigate this relationship and should examine whether positive and negative

role models are differentially related to possible selves. For example, it is unclear whether both

positive and negative role models are necessary to predict balanced possible selves. Positive role

models may contribute to the formation of positive possible selves, whereas negative role models

may contribute to the formation of negative possible selves. Alternatively, positive role models

may also contribute to the formation of negative possible selves and negative role models may

also contribute to the formation of positive possible selves. Researchers need to examine whether

having more positive (or negative role) models affects balanced possible selves. Finally,

researchers should also examine the qualities of the relationships between students and their role

models to determine what aspects contribute to the production of possible selves.

Oyserman and Markus (1990) suggested that individuals who have balanced possible

selves may have more motivation and control over their behavior than individuals without such

balance. This study lends partial support to this claim, as balance significantly predicted

students' perceptions of their ability to delay of gratification (through an indirect effect mediated

by perceived self-efficacy for delay of gratification). Hence, students who reported more

balanced possible selves reported more self-control (in terms of delaying their gratification) than




Full Text

PAGE 1

ROLE MODELS, POSSIBLE SELVES, PERCEI VED SELF-EFFICACY, AND PERCEIVED SELF-CONTROL AS PREDICTORS OF GPA IN COLLEGE STUDENTS By TESIA T. MARSHIK A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN EDUCATION UNIVERSITY OF FLORIDA 2008

PAGE 2

2008 Tesia T. Marshik

PAGE 3

To my grandparents

PAGE 4

4 ACKNOWLEDGMENTS It is a pleasure to thank the m any people who helped make this thesis possible. The support, encouragement, and wisdom of the professors in the Educational Psychology and Psychology Departments have inspired me to do my best. It is difficult to overstate my gratitude to my advisor and chair, Dr. Patricia Asht on, for her instruction, guidance, patience, and encouragement throughout this process. I would ha ve been lost without her. I am also greatly indebted to Dr. James Algina for his excellent instruction and advice regarding statistical procedures and interpretation, and for spending ho urs responding to my qu estions and reviewing my computer programs. I would also like to th ank my professor and friend, Dr. John Bengston, for his stimulating conversation, challenging questions, ready ear, a nd incisive mind. The aforementioned individuals have contributed most to my development as a student, and as a person. They practice what they teach and are exem plars for their students. Furthermore, I would like to thank my student colleagues for their stim ulating contributions to class discussions and for their support and encouragement. I am also grateful to Elaine Green and Linda Parsons for helping the department run smoothly, for always ta king care of things, an d for assisting me in many different ways, especially with reminders about deadlines and requirements. Lastly, and most importantly, I acknowledge with gratitude the enduring support of my family. My husband, Jesse, has listened with patience to the ramblings of my ideas a nd endured countless nights of me working frantically on the computer. He has always encouraged me to pursue my academic career to its fullest potential and he continually lif ts my spirit. I also wish to thank my siblings, especially my sister, for being ex cellent role models and teachers throughout my life and for their perpetual support and encouragement. Finally, I am grateful to my parents and grandparents for raising me, supporting me, teaching me, loving me, and encouraging me to strive towards excellence.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................8 ABSTRACT.....................................................................................................................................9 CHAP TER 1 INTRODUCTION..................................................................................................................11 Statement of the Problem....................................................................................................... .11 Self-Control.....................................................................................................................12 Role Models and Self-Control......................................................................................... 15 Role Models and Possible Selves....................................................................................18 Balanced Possible Selves and Perceived Self-Efficacy for Self-Control........................ 19 Perceived Self-Efficacy for Self-Control and Academic Achievement.......................... 22 Measurement of Self-Control.......................................................................................... 24 The Proposed Model and Purpose of the Study...................................................................... 26 Theoretical Significance....................................................................................................... ..27 Practical Significance......................................................................................................... ....29 2 METHOD......................................................................................................................... ......31 Participants.............................................................................................................................31 Measures.................................................................................................................................31 Role Models.....................................................................................................................31 Possible Selves................................................................................................................32 Perceived Self-Control.................................................................................................... 32 Perceived Self-Efficacy................................................................................................... 33 GPA.................................................................................................................................33 Procedures..................................................................................................................... ..........33 3 RESULTS...............................................................................................................................35 Preliminary Analyses and Descriptive Statistics .................................................................... 35 Analysis of the Proposed Model.............................................................................................36 4 DISCUSSION.........................................................................................................................38 APPENDIX A ROLE MODELS QUESTIONNAIRE................................................................................... 45

PAGE 6

6 B POSSIBE SELVES QUESTIONNAIRE............................................................................... 47 C PERCEIVED SELF-EFFICACY MEASURE....................................................................... 49 D DEMOGRAPHIC INFORMATION...................................................................................... 50 REFERENCES..............................................................................................................................51 BIOGRAPHICAL SKETCH.........................................................................................................57

PAGE 7

7 LIST OF TABLES Table page 3-1 Intercorrelations, means, and standard de viations of variables in path analysis ( N = 163) ....................................................................................................................................43 3-2 Total, direct, and indirect effects in the proposed model ( N = 163) ..................................44

PAGE 8

8 LIST OF FIGURES Figure page 1-1 Theoretical model of the relationships am ong role models, possible selves, perceived self-efficacy, perceived self-control, and GPA.................................................................. 27

PAGE 9

9 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Ma ster of Arts in Education ROLE MODELS, POSSIBLE SELVES, PERCEI VED SELF-EFFICACY, AND PERCEIVED SELF-CONTROL AS PREDICTORS OF GPA IN COLLEGE STUDENTS By Tesia T. Marshik May 2008 Chair: Patricia Ashton Major: Educational Psychology Self-control predicts many important deve lopmental outcomes including achievement, emotional stability, and successful social relationships. The potenti al for self-control to predict academic achievement in college students is particularly important because few good predictors of academic achievement in college students exist and college retention and graduation rates are decreasing. The purpose of the present study was to identify factors that pr edict self-control and GPA in college students. Participants included 163 undergraduate students at the University of Florida who completed take-home self-report questionnaires. A recurs ive path analysis was used to test a model relating students role models, possible selves, perceive d self-efficacy, perceived self-control skills, and GPA. Partial support for the model wa s found, as total role models predicted the number of balan ced possible selves; the number of balanced possible selves predicted Delay of Gratification; perceived self-efficacy predicted perceived self-control; and Planful Thinking predicted GPA. These results are consistent with social cognitive theory, as role models influenced students possible selves, which in turn influenced students self-control skills and academic achievement. However, the lack of relationships among some of the variables indicates that more re search is needed to examine th e relationships among role models, possible selves, and self-control. In particular, researchers should use domain-specific measures

PAGE 10

10 of academic possible selves and academic se lf-control to predict students academic achievement.

PAGE 11

11 CHAPTER 1 INTRODUCTION I count him braver who overcomes his desire s than him who conquers his enemies, for the hardest victory is the victory over self. Aristotle He who conquers himself is the mightiest warrior. Confucius He who reigns within himself, and rules passio ns, desires, and fears, is more than a king. John Milton Statement of the Problem The ability to control ones passions, thoughts, and behaviors is a f undamental element of human agency. Self-control has long been heralded by poets, philosophers, and religious leaders. The failure of individuals to cont rol themselves, to act without reas on, or to act contrary to that which rationality prescribes was considered a major problem even during ancient times when great thinkers such as Plat o, Socrates, and Aristotle took it upon themselves to study akrasia, or weakness of the will. Freud (1923/1989) later sugges ted that the human psyche is in a state of perpetual conflict as the ego struggles to regu late the unrelenting impulses of the id while balancing the demanding pressures of the supe rego. Darwin (1871/1981) promoted the role of self-control when he suggested that the highest possible stage in moral culture is when we recognize that we ought to control our thoughts (p. 123). Recently, the st udy of self-control in psychology has garnered increasing a ttention due to its conjectured role in both internalizing and externalizing disorders. Self-control refers to those processes (both conscious and nonconscious) involving the organization and mobi lization of resources that enab le individuals to guide their goal-directed behaviors over time and across mu ltiple, dynamic contexts. Aspects of self-control include the monitoring and regul ation of ones thoughts, emoti ons, and impulses through the use of self-instructions, self-motivati on, self-evaluation and delay of gratification. Another aspect of

PAGE 12

12 self-control is perceived self-effi cacy, which plays a fundamental role because of its influence on thought, affect, motivation, and action (Bandura, 1991, 1997). Recent research suggests that selfcontrol predicts many important outcomes includ ing achievement, adjustment, substance abuse, emotional stability, and quality of interpersonal relationships (E isenberg & Fabes, 1992; Shoda, Mischel, & Peake, 1990; Tangney, Baumeister, & Boone, 2004; Wills & Stoolmiller, 2002). In light of the importance that se lf-control plays in psychological functioning, research into the factors that contribute to the development of self-control and the processes that explain its widespread influence is needed. Few studies of these issues have been conducted. The purpose of this study is to examine factors that may predic t self-control and grade point average (GPA) in college students. Specifically, th is study will examine the extent to which students role models and possible selves affect their perceptions of self-efficacy for self-control, reported self-control behaviors, and GPA. Self-Control Rosenbaum (1980a) offered a promising concepti on of self-control in his development of the Self-Control Schedule, a measure that asse sses individuals repertoire of self-control behaviors and tendencies to use these behavior s when faced with problems (Akgun & Ciarrochi, 2003). As conceived by Rosenbaum, self-control consists of four components: (a) using cognitions and self-instructions to cope with emotional and phys iological distress, (b) applying problem-solving strategies, (c) th e ability to delay immediate gratification, and (d) confidence (i.e., perceived efficacy) in one's ability to se lf-regulate thoughts and feelings (Rosenbaum, 1983). The purpose of the present study is to identif y factors that predict self-control in college students. Research with the Self-Control Schedule has shown consistently th at individuals scoring high on Rosenbaums (1980a) measure engage in h ealthier coping tendencies, have higher levels

PAGE 13

13 of perceived self-efficacy and performance, are more likely to persist after being confronted with repeated failures, demonstrate greater ability to tolerate pain and seasickness and to cope with seizures, and exhibit fewer symptoms of stress and depression than individuals low in selfcontrol (Rosenbaum, 1980a and b; Rosenbaum & Ben-Ari, 1985; Rosenbaum & Jaffe, 1983; Rosenbaum & Palmon, 1984; Rosenbaum & Rolnic k, 1983). In sum, the findings of Rosenbaum and his colleagues have suggested that individuals high in self-cont rol are better able to minimize the negative effects of situa tional stressors on th eir performance and overall psychological functioning compared to indivi duals low in self-control. Other researchers have found similarly pos itive results with Rosenbaums (1983) measure of self-control. Ginter, West, and Zars ki (1989) investigated the relationship between self-control and coping strategies and found that t hose high in self-control used more beneficial, problem-focused coping strategies and reported significantly fewer symptoms of stress than individuals low in self-contro l. Later research supported these findings, as Akgun (2004) reported that persons high in self-control used more positive reappraisal, were more likely to seek social support, and less lik ely to use escape-avoidance co ping strategies. Akgun also found that those with high self-contro l had higher levels of perceive d self-efficacy regarding their abilities to effectively cope with stress. Other investigations into the link between self-control, stress, and academic performance have suggested that self-control moderates the effect of academic stress on academic performance. As Ro senbaum and Jaffe (1983) noted, considerable evidence shows that individuals hi gh in self-control are better able to tolerate and cope with uncontrollable aversive stimulation. In addition, several researchers examined the relationship between self-control and depression. Rosenbaum and Palmon (1984) reported that patients high in self-control were

PAGE 14

14 significantly less depressed, coped better with their disability (epilepsy), and maintained a stronger belief in their control over their health and thei r seizures compared to individuals low in self-control. Simons, Lustman, Wetzel, and Murp hy (1985) examined self-control and depressed patients response to cognitive behavioral ther apy and reported that pa tients entering cognitive therapy with relatively high Self -Control Schedule scores responded more favorably to cognitive therapy than patients with low scores. In cont rast, Lewinsohn and Alexan der (1990) reported that adolescents low in self-control exhibited an increased probability of becoming depressed. Finally, some researchers have reported relationships between self -control (as measured by the Self-Control Schedule) and self-reported pa tterns of alcohol consumption and tobacco use in young adults (Carey, Carey, Ca rnrike, & Meisler, 1990; Katz & Singh, 1986). Carey et al. compared undergraduates self-re ported patterns of alcohol consumption and found that the heaviest drinkers had the lowest scores and that individuals who drank infr equently or not at all had the highest scores on the Self-Control Schedule. Several studies also suggest that smokers have significantly lower scores on the Self-Control Schedule th an non-smokers (Carey et al., 1990; Kennett, Morris, & Bangs, 2006). Similarl y, Katz and Singh (1986 ) reported that exsmokers scored significantly higher on the Self-Control Schedule than did smokers who had attempted to quit but failed. On the basis of their results, the authors sugge sted that ability of smokers to quit smoking may have been attributab le to the better coping skills associated with high self-control. Kennett et al. replicated these results and re ported that individuals who quit smoking exhibited just as much self-control as individuals who ne ver smoked, even after controlling for age differences. Taken together, th ese studies suggest that self-control skills may protect against substance abuse. Furthermore, these results s upport the notion th at self-control skills may serve as a protective factor against depr ession, given that substance use (and abuse) is

PAGE 15

15 strongly associated with depressi on (serving as a risk factor, possible trigger, and concomitant of depressive symptomology). In sum, it appears that indivi duals with high self-control beli eve in their ability to deal with aversive stimuli, use more beneficial copi ng and problem solving strategies, and are better able to minimize the negative effects of aversive stimuli compared to individuals with low selfcontrol. From the studies reviewed, it is clear that self-control is related to a variety of important self-regulatory processes. Despite the important outcomes associated with this construct, the factors that contribute to the development of self-control are s till unclear. Social cognitive theory (Bandura, 1997) suggests several important factor s that may contribute to the development of self-control. One of the most important may be the relationships individuals develop with other people, particularly when the people in these re lationships become models of behavior for the individuals. The potential impact of these models on self-control is a major focus of this study. Role Models and Self-Control Som e researchers have reporte d that individuals social relationships influence their ability to control thei r impulses and behaviors (Calkins, 1994; Cassidy, 1994). This notion is consistent with social cognitive theory (B andura, 1986), which emphasizes the roles of observation and modeling in human learning and development. Many theorists have proposed that the processes by which indi viduals develop self-understanding are inherently social because individuals continually compare themselves with others to identify their unique characteristics and to evaluate their own abil ities. According to Bandura (1986, 1991), most human behavior is learned observationally through mode ling the behaviors, attitudes, and emotional reactions of others. In other words, humans have the capacity to learn from observing others successes and failures.

PAGE 16

16 The others to which we are continually looking to for information and comparing ourselves are often referred to as role models Unfortunately, little consensus exists on a definition of role models, and the term role model is often used synonymously with terms like mentor and exemplar. Hence, as Gibson (2004) sugge sted, the construct of role models remains a popularly used but vaguely defi ned notion (p. 135). For the purpos es of clarification, I have adopted Gibsons definition of role model as a cognitive construc tion based on the attributes of people in social roles an individu al perceives to be similar to hi m or herself to some extent and desires to either increase perceived similarity by emulating those attributes or to decrease perceived similarity by avoiding those attribut es (p. 136). Further, Gibson (2003) defined the process of role modeling as a cognitive process in which individuals actively observe, adapt, and reject attributes of multiple role models ( p. 593). These definitions imply that role models may be positive or negative. Positive role models are individuals who have achieved success, who are considered competent in a relevant domain, who exhibit pres tige and power, and who inspire others to emulate certain qualities. Negative role models are individuals who have experienced some kind of failure or misfortune, who possess undesira ble qualities, and who motivate others to avoid sim ilar adversity (Lockwood, 2002). An individuals repertoire of role models may range from close relatives and friends, to coworkers, superstars, historical figures, and even fictional characters (Gibson, 2004; Ibarra, 1999; Kemper, 1968; Lockwood & Kunda, 1997). Situ ational factors (such as proximity) and person factors (such as gender and age) affect the number and types of role models available to an individual (Kulik & Ambrose, 1992). Traditi onally, it has been suggest ed that individuals emulate models who exhibit four primary characteristics: competence, gender appropriateness, prestige and power, and individual relevan ce (Bandura, 1977; Thomas, 1990). However, the

PAGE 17

17 majority of social cognitive theorists agree th at individuals are most likely to model the behaviors and qualities of individua ls with whom they identify, which depends on the degree to which they perceive the models to be simila r to themselves, and the degree of emotional attachment that is felt toward the models (Thomas, 1990; Woodward, 1982). Theorists have consistently maintained that identification with role models is critical to professional, academic, and emotional deve lopment (Bandura, 1977; Erikson, 1985). Bandura suggested that role models serve both inform ational and motivational functions. Similarly, Lockwood, Sadler, Fyman, and Tuck (2004) sugg ested that individuals may use both positive and negative role models simultaneously as a means of effectively ch anneling their motivation. Thus, individuals observe their role models for information regarding how to act and the consequences of such actions and then use this information as a basis for their decisions about how to act in the future. Indivi duals are motivated to model beha viors and qualities to the extent that they lead to desirable consequences, and individuals are motivated to avoid behaviors and qualities to the extent that they lead to undesi rable consequences. Resear chers have repeatedly demonstrated that role models are sources of motivation and inspiration (Lockwood & Kunda, 1999; Lockwood, Jordan, & Kunda, 2002; Lockwood et al., 2004). To date, most of the research on role models focuses on positive role models (also referred to as mentors and exemplars ); but it seems more likely that the combination of positive and negative role models has a more powerful relationship to self-con trol than positive role models al one. In this study I explored the relationship between the total number of positive and negative role models and self-control by examining several potential medi ators that might account for th e relationship, specifically possible selves, and perceived self-efficacy for se lf-control. The rationales for these linkages are explained in the following sections.

PAGE 18

18 Role Models and Possible Selves Lockwood and colleagues (Lockwood, 2002; Lockwood et al., 2004; Lockwood & Kunda, 1997, 1999) have suggested that role m odels are motivating to the extent that they point to plausible positive and negative possible selves and the strategies for achieving or avoiding them. Possible selves are defined as cognitive self-representations concerning what one expects to become, what one hopes to become, and wh at one fears becoming (Markus & Nurius, 1986). Positive possible selves include those self-represen tations and goals that one hopes to realize (for example, one might desire to be wealthy or to be known as a person of integrity) and negative possible selves refer to those self -representations that one hopes to avoid or fears becoming (for example, one might fear becoming a college dropout or being known as a liar). Positive possible selves may be classified as promotion goals becau se they reflect the personal goals that students hope to achieve, whereas negative possible selves may be classifi ed as prevention goals because they reflect outcomes that students hope to avoid. Thus, a mixture of positive and negative role models may provide the basis for the developm ent of positive and negative possible selves and the motivation for achieving (or avoiding) them. Theorists have suggested that po ssible selves are shaped by social, cultural, and environmental factors and can be facilitated or hind ered according to the quality of these factors. Although an individual is free to imagine a vast array and unlimited number of possible selves, the actual collection of possible selves that an individual imagin es for himself or herself is derived from the categories made salient by the individuals particul ar sociocultural and historical context and from models, images and symbols provided by the media and by the individuals immediate soci al experiences (Markus & Nu rius, 1986, p. 954). For example, children who grow up in abusive households may envision themselves in abusive relationships later in life; they may have difficulty fo rming positive possible selves concerning their

PAGE 19

19 relationships with others. Se veral researchers (Anthis, Dunkel, & Anderson, 2003; Day, Borkowski, Punzo, & Howsepian, 1994; Kao, 2000; Knox, Funk, Elliott, & Bush, 2000) have demonstrated the influence of sociocultural fact ors on childrens possibl e selves, reporting that prevalent gender, racial, and ethni c stereotypes were reflected in the types of possible selves children and adolescents envision ed. Thus, given that possible se lves are greatly influenced by social and cultural factors, it is likely that individuals positive and negative role models influence the types of possible se lves they construct. However, no studies have specifically examined the relationship between individuals role models and their possible selves. In this study, I examined the relationship between the total number of students role models (including positive and negative role models) and their balanced possible selves. Balance means having both a positive and negative aspect of a fu ture goal, or having a hoped-for self and a corresponding feared self in the same dom ain (Oyserman & Markus, 1990). The balance measure of possible selves incorporates both positive and negative possible selves and should therefore capture the effects of both positive and ne gative role models. On the basis of previous studies, I hypothesized that indi viduals with more role models have more balanced possible selves because they are able to envision the pos itive and negative possibilities associated with their future goals. The next link in the proposed model is from balan ced possible selves to perceived self-efficacy for self-control. Balanced Possible Selves and Perceive d Self-E fficacy for Self-Control Research into the likely roles of possible selv es suggests that (similar to role models) they function as incentives for future behavior, and they provide an evalua tive and interpretive context for the current self-con cept. Markus and Nurius (1986) suggested, The efficient performance of almost any task, whether rela tively mundane, or more complex, requires the construction of the possible self that carries out the action, comp letes the task, or masters the

PAGE 20

20 difficulty (p. 962). In addition, Markus and Nuri us have proposed that possible selves are important motivators because they provide specific self-relevant goals to work toward or to avoid, thereby energizing and or ganizing individuals behaviors. Further, Cross and Markus (1991) noted that possible selves help individual s make more direct connections between their goals and their strategies for attaining them by allowing individuals to simulate their futures, which enables them to organize and integrate info rmation and strategies relevant to their goals and to judge the extent to which they are ap proaching (or avoiding) desired (or undesired) outcomes. Markus and Nurius (1986) also proposed that possible selves serve as standards for comparison and evaluation of the current self. Th at is, individuals can monitor the status and development of their current self by envision ing their desired and undesired future selves. Markus and Nurius further suggested that posit ive possible selves may be encouraging because they foster hope and optimism, whereas negativ e possible selves may be discouraging because their associated affect and expectations may s tifle attempts to change or develop (p. 963). However, negative possible selves may also be mo tivating to the extent that they highlight the strategies necessary to avoid undesired outcomes. Research sugge sts that individuals who have balanced possible selves appear to have more motivation and cont rol over their behavior than individuals without such bala nce. In one study, public school youth had significantly more balanced possible selves than delinquent youth (Oyserman & Markus, 1990), and balance in possible selves has been found to have a positiv e relationship to school persistence (Oyserman & Markus, 1990; Oyserman, Gant, & Ager, 1995). On the basis of these results, Oyserman and Markus suggested that individuals with more balance among their possi ble selves have more motivational resources because they can envision a greater array of potential outcomes and can

PAGE 21

21 better monitor their progress toward positive or negative outcomes. Oyserman, Bybee, and Terry (2006) later found that y outh with balanced academic possibl e selves spent more time engaging in self-control behaviors re lated to academic achievement (i.e., spent more time doing homework, were less behaviorally disruptive, and more behaviorally engaged in classroom activities). Of special relevant to this study, Cross and Markus (1994) proposed that possible selves may link effective steps a nd strategies for solving problems with beliefs about ones ability and competence in the domain. Similarly, Ruvolo and Markus (1992) suggested that the underpinnings of a sense of efficacy, control, and competence are specific, self-relevant thoughts and feelings, particularly images and conceptions of the self in the future, desired states (p. 97). Thus, they proposed that possi ble selves provide the foundation for perceived self-efficacy, control, and competence. Bandura (1997) defined perceived self-efficacy as beliefs in ones capabilities to organize and execute the courses of action required to produce given attainments (p. 3), and studies have re vealed that perceived self-efficacy has a positive impact on individuals confidence, motivati on, perseverance, and success (Bandura, 1997; Schunk, 1984). As Schunk (2003) observed, when compared with their less efficacious counterparts, those who feel efficacious for l earning or performing a task participate more readily, work harder, persist longer when they encounter difficulties, and achieve at a higher level (p. 161). Thus, the extent to which indivi duals believe that they are capable of regulating their behavioral responses may predict their abil ity to do so. Bandura further suggested that most human behavior is learned obs ervationally through modeling th e behaviors, attitudes, and emotional reactions of others and that perceive d self-efficacy could be increased by vicariously experiencing the successes of role models. Hence, the extent to which parents, peers, and other

PAGE 22

22 significant social influences model appropriate self-control skills may affect the development of individual individuals possible selves that in tu rn affect their perceived self-efficacy for selfcontrol, which in turn affects their ability to regulate their behavioral responses. Some evidence supports this notion, as researchers have linked po sitive role models to in creased perceived selfefficacy, psychological well-being, career success, and overall pos itive self-concepts (Bahniuk, Dobos, & Hill, 1990; Bryant & Zimmerman, 2003; Turner, 1996; Weinberg, Grove, & Jackson, 1992). In addition, indirect evidence supports the relationship between possible selves and perceived self-efficacy for self-control. Some researchers have found that positive possible selves were significantly related to improve d grade point average (Anderman, Anderman, & Griesinger, 1999) and Leonardi, Syngollitou, and Kiosseoglou (1998) repo rted that the overall quality of students possible selves was related to school achievement and task persistence. As perceptions of self-efficacy are typically found to predict achievement, it is plausible to extrapolate from these findings th at possible selves may predict pe rceived self-efficacy for selfcontrol. Consequently, in this study, I extended this line of research by examining whether the relationship between the number of role mode ls and self-control was mediated by number of balanced possible selves and percei ved self-efficacy for self-control. Perceived Self-Efficacy for Self-Control and Academic Achievement One context in which perceived s elf-efficacy and self-control skills are especially important is in colleges and universities, and this study will provide information about factors that might predict academic achievement (as measured by GPA) in college students. The identification of significant pred ictors of GPA is important because universities across the nation are struggling to increase retention rates. Alt hough recent data on dropout rates in colleges are not available, several studies have revealed that between one third and on e half of students who enter college do not complete their programs. ACT (2007a and b) administered surveys showing

PAGE 23

23 that, from 1983 to 2007, approximately one third of the students who attended college dropped out before their second year. In addition, the report reve aled that non-return rates were increasing and that little has changed in the last two decades regarding 5-year graduation rates, still hovering around 50%. The National Center for E ducation Statistics (2005) reported similar results for the years between 1989 and 1995 and not ed that 5-year graduation rates have not changed despite increased access to colleges. In a more recent study, the U.S. Census Bureau (2000) reported that one in three Americans drops out of college, and that this number appears to be steadily increasing. Clearly, th e numbers of students who drop out of college demonstrate the need to identify factors that contribute to this problem. Perceived self-efficacy has been shown to have a strong link to academic achievement (Bandura, 1997), hence, it is important for researchers to identify potential predictors of perceived self-efficacy and academic achievement at the college level. It is likely that individuals w ho have confidence in their abilit y to engage in self-control have higher levels of self-control than individu als with lower perceptions of self-efficacy for self-control. Furthermore, it is likely that indivi duals with higher self-control achieve at higher rates because they are better able to organize their resources, plan ahead, engage in more effective problem solving strategi es, and control factors that pote ntially interfere with successful performance (such as unwanted thoughts and em otional distractions). These skills may be particularly important and useful in academic settings, where students are regularly required to complete assignments and tasks within time cons traints, and some evidence suggests self-control is a significant predictor of academic achieve ment and school performance. For example, Mischel and colleagues (Mischel, Shoda, & Peake, 1988; Shoda, Mische l, & Peake, 1990) conducted a series of studies a nd reported that childrens dela y of gratification at age 4

PAGE 24

24 significantly predicted SAT scores years later. More recent studi es found that individuals with high self-control had significantly better grades than individuals with lo w self-control (Feldman, Martinez-Pons, & Shaham, 1995; Tangney et al., 2004) and that self-control significantly predicted grade point average among college stud ents (Wolfe & Johnson, 1995). In this study, I further examined the relationships among thr ee perceived self-control skills (delay of gratification, emotion regulation, and planful thi nking) and college student s GPA. Furthermore, I sought to determine whether total number of ro le models, balanced possible selves, perceived self-efficacy, and perceived self-control predicted GPA. Measurement of Self-Control The most promising of the few measures of self-control available is Rosenbaums (1980a) Self-Control Schedule, which was design ed to measure learned resourcefulness. Rosenbaum (1985) defined learned resourcefulness as an acquired repertoire of cognitivebehavioral skills by which a person self-regulates internal responses (such as emotions, pain, and cognitions) that would otherwise interfere with the smooth execution of a target behavior (p. 200). The Self-Control Schedule is a self-repor t measure that assesse s individuals general repertoire of self-control behaviors and their tendencies to use th ese behaviors when faced with everyday problems (Akgun & Ciarrochi, 2003) As conceived by Rosenbaum, learned resourcefulness is multi-faceted and incorporates f our aspects: (a) the use of cognitions and selfinstructions to cope with emo tional and physiological responses, (b) the application of problemsolving strategies, (c) the ability to delay immediate gratification, a nd (d) a general belief in one's ability to self-regulate internal events (i.e ., perceived self-efficacy) (Rosenbaum, 1983). In sum, the research using Rosenbaums Self -Control Schedule suggests that individuals with high resourcefulness believ e in their ability to deal with aversive stimuli, use more beneficial coping and problem solving strategies and are better able to minimize the negative

PAGE 25

25 effects of aversive stimuli compared to individua ls with low resourcefulness. From these studies, it is clear that learned resourcefulness involves a variety of important self-regulatory processes. Yet despite the promising results associated with this construct, research provides inconsistent conclusions regarding the factor ial structure of the Self-Contro l Schedule. For example, Gruber and Wildman (1987) conducted an exploratory fact or analysis and subseq uently reported that only three significant factors emerged: problem -focused coping, mood and pain control, and externality (which the authors concluded is the reverse of self-efficacy). Other researchers conducted factor analyses across groups. Edwa rds and Riordan (1994) performed separate varimax rotations for Black and White students a nd reported 14 and 12 factor s, respectively, that were difficult to interpret. Redden, Tucker, a nd Young (1983) also used varimax rotation and obtained six factors. The author s, however, cautioned that inte rpretation of the factors should proceed tentatively because of a lack of a clear, strong factor structure (pp. 84-85). Thus, the factor structure of the Self-Cont rol Schedule is still unclear, though research suggests that the items on the Self-Control Schedu le load on more than the four factors that Rosenbaum proposed as composing the construct of learned resourcefulness (albeit the factors that emerged in previous studies do reflect similar themes). From the information available, it appears that all previously published factor an alyses were conducted using varimax rotation. Constraining the factors of the Self-Control Schedule to an orthogonal solution when theory and the nature of the items suggest they should be co rrelated likely led to erroneous conclusions. To address these concerns, I conducted an exploratory factor analys is of the Self-Control Schedule using promax rotation, which allowed the factors to correlate (Marshik, 2007). Prior to data collection, two items were eliminated due to thei r sensitive content. Th ese items ( If I would smoke two packages of cigarettes a day, I proba bly would need outside help to stop smoking,

PAGE 26

26 and If I had the pills with me, I would take a tranquilizer whenever I felt tense and nervous) were replaced with two new items written by th e researchers (I have a hard time waiting for something that I really want, and If I have a choice between a smaller reward now or a bigger reward later I would choose the smaller re ward so that I could have it now). These two items were added to increase the number of items refe rring to delay of gratification, because only a few items seemed to reflect this construct. The an alysis of 176 undergraduate students responses to the items on the Self-Control Schedule yielded seven self-control fact ors: Emotion Regulation, Perceived Self-Efficacy, Ability to Control Physio logical Responses, Planful Thinking, ProblemSolving Ability, Ability to Control Unwanted Thoughts, and Delay of Gr atification. Three of these factors were used to measure self-contro l skills in this study because they were wellrepresented by the items: Emotion Regulation ( = .72), Planful Thinking ( = .62), and Delay of Gratification ( = .59). The Proposed Model and Purpose of the Study In summary, theoretical acc ounts and the research literat ure provide support for the following predictions represented in the conceptual model in Figure 1-1: (a) the total number of role models will have a direct relationship to the number of balanced possible selves, (b) the number of balanced possible selves will have a di rect relationship to the perceived self-efficacy variables (Perceived Self-Effi cacy for Delay of Gratificatio n, Perceived Self-Efficacy for Emotion Regulation, and Perceived Self-Efficacy fo r Planful Thinking), (b) the perceived selfefficacy variables will have direct relationshi ps with the corresponding perceived self-control variables (Perceived Delay of Gratification, Perceived Emo tion Regulation, and Perceived Planful Thinking), (c) the perceive d self-control variable s will have direct relationships to GPA. It is also hypothesized that the total number of role mode ls will predict GPA through the mediating variables of balanced possible selv es, perceived self-efficac y, and perceived self-

PAGE 27

27 control. Specifically, the model depicts the follo wing predictions regarding mediation effects: (a) the number of balanced possible selves will medi ate the relationship between the total number of role models and the perceived self-efficacy variables, (b) the self-efficacy variables will mediate the relationship between the number of balanced possible selves and the perceived self-control variables, and (c) the perceived self-control variables will medi ate the relationship between the perceived self-efficacy variables and GPA. In the conceptual model depicted in Figure 1-1, ellipses indicate latent variables and r ectangles indicate obs erved variables. Figure 1-1. Theoretical model of the relationships among role models, possible selves, perceived self-efficacy, perceived self-control, and GPA. Theoretical Significance The results of this study will serve to in tegrate several research areas and will help psychologists better understand the links between individuals role models, possible selves, selfcontrol skills, and academic achievement. Mar kus and Nurius (1986) have suggested that possible selves are shaped by and greatly infl uenced by social, cultural, and environmental factors and can thus be facilitated or hindered acco rding to the quality of these factors. However, few studies have examined the connections between individuals social relationships and their possible selves, and none have examined the as sociations between role models and possible Role Models (total number) Balanced Possible Selves GPA Perceived Delay of Gratification Perceived Emotion Regulation Perceived Planful Thinking Perceived SelfEfficacy for Emotion Regulation Perceived Self-Efficacy for Planful Thinking Perceived SelfEfficacy for Delay of Gratification

PAGE 28

28 selves. This study will provide information eluc idating the connection be tween total number of role models and the number of balanced possible selves. Theory and research also suggest that indi viduals social relationships influence their perceptions of self-efficacy and self-control abil ities. Social cognitive theory emphasizes the roles of observation and mode ling in human learning and de velopment, and Bandura (1997) suggested that most human beha vior is learned observationall y through modeling the behaviors, attitudes, and emotional reactions of others. In this study, I exte nded this line of research by examining the relationships between total numbe r of role models, the number of balanced possible selves, perceived self-effi cacy for self-control skills, and perceived self-c ontrol skills (including delay of gratificati on, emotion regulation, and planful thinking). This research may elucidate the mechanisms through which role mode ls ultimately influence students self-control and academic achievement, which may provide insi ght into ways that role models can help students develop academic and social competen ce. Specifically, I hypothesized that the number of balanced possible selves and perceptions of self-efficacy would mediate the relationships between individuals role models and thei r perceptions of self-control and GPA. Finally, some theorists have already sugge sted that possible selves serve as the underpinning of perceptions of self-efficacy, self-control, and competence (Markus & Ruvolo, 1989; Ruvolo & Markus, 1992), but none have direc tly examined these relationships. I extended this line of research by explicitly examining the relationships among college students possible selves, perceived self-efficacy, pe rceived self-control, and GPA. In sum, the proposed model integrates criti cal concepts from social cognitive theory, motivation theory, self-concept th eory, and achievement theory. Results from this study offer insights into the relationships among role mode ls, possible selves, perceived self-efficacy for

PAGE 29

29 self-control, perceptions of self -control, and GPA that may be useful in the development of further research and theory on these important social, emotional, and cognitive concepts. Practical Significance This study elucidates important relationships, which has severa l practical im plications for future experimental research to determine appro aches for increasing college students ability to engage in self-control and to improve their academic performance. First, it will provide information about the possible predictors of se lf-control. The construc t of self-control has received increasing attention ove r the years due to its associat ions with psychopathology. The concept of self-control is es pecially popular among researcher s and clinicians seeking to elucidate individual differences in the development of psychopathology. Several studies provide evidence that poor self-cont rol skills are related to lower leve ls of social competence (Eisenberg & Fabes, 1992; Eisenberg et al., 1997) and to internalizing and external izing disorders (Hart, Atkins, & Ford, 1998; Rubin, Coplan, Fox, & Calkin s, 1995). In addition, re search suggests that self-control predicts many important outcomes including achievement, adjustment, substance abuse, emotional stability, and quality of interp ersonal relationships (E isenberg & Fabes, 1992; Shoda et al., 1990; Tangney et al., 2004; Wills & Stoolmiller, 2002). Social-cognitive theory offers a number of potentially important construc ts that may contribute to academic achievement including role models, possible selves, perceive d efficacy for self control, and self-control behaviors. If researchers can identify mechanisms that underlie perceived self-efficacy and selfcontrol, they will be better able to help individuals develop appropriate and beneficial selfcontrol skills, which will increas e the likelihood of thei r success. In addition, this study provides information about factors that predict academic achievement (as measured by GPA) in college students. If these factors are s hown to increase academic achieve ment in experimental studies, important new interventions can be developed to improve academic achievement and educators

PAGE 30

30 may be able to identify students who are more likely to struggle academically or to drop out of college.

PAGE 31

31 CHAPTER 2 METHOD Participants The sam ple consisted of 163 college stude nts (124 women, 39 men). Age ranged from 18 to 26 with a mean of 19.89 and a standard deviation of 1.66. The ethnic backgrounds represented were as follows: 65% White, 13% Black, 12% Hispan ic, 5% Asian, and 5% other. The majority of students were sophomores (37%), followed by juniors (24%), seniors (22%), and freshman (17%). Participants were recruited from two sources: (a) the Educational Psychology subject pool (specifically, stude nts in three educational psyc hology classes, EDF 3110 Human Growth and Development, EDF 3210 Educationa l Psychology, or EDF 3135 The Adolescent), and (b) other undergraduate courses offered by the Educational Psychology Department at the University of Florida. For students recruite d from the Educational Psychology subject pool, participation in this study fulfilled a research requirement for the course (students who elected not to participate were given an alternate assignment by their instructor to fulfill the research requirement). Participants who were recru ited from other educational psychology courses received either extra credit (not to exceed 1% of their grade) or class participation credit, depending on the preference of the course instructors. Measures Role Models An open-ended question was created to coll ect infor mation regard ing students role models (see Appendix A). The measure consists of a brief definition of role models (including definitions of positive versus negative role models), and students are asked to list and describe in detail their positive and negative role models. The total number of role models generated by each participant was tallied and used in the analyses.

PAGE 32

32 Possible Selves An open-ended questionnaire m odeled after Oyserman and Markus (1990) was used to obtain information about students possible selves (see Appendix B). On the measure, individuals are asked to genera te their hoped-for and feared possible selves and list their strategies for obtaining or avoidi ng them The measure cons ists of a descripti on of possible selves and instructs students to think about their own possible selv es and then to list all of their negative and positive possible selves. Cross and Markus (1991) found that college students responses to open-ended measures of possible selves ranged fr om simple one-word desc riptions to elaborate and vivid descriptions of both hoped-for and feared possible selves. The results of additional studies have shown that open-ende d measures of possible selves el icit unique and diverse sets of individual responses (Knox, Funk, Elliott, & Bu sh, 1998; Knox et al., 2000; Leonardi et al., 1998; Oyserman & Markus, 1990). Following the pro cedures used by Oyserman and Markus (1990), Oyserman, Terry, and Bybee (2002), and Dunkel and Anthis (2 001), participants positive and negative possible selves were coded into six categories ( achievement interpersonal relationships personality traits material and lifestyles physical and health-related and negative) and balance was assessed by tallying the number of connections between students positive and negative possible selves in the same domain. Possible selves were double coded and interrater agreement was 94% (all disagreements were discussed to agreement). Perceived Self-Control The Self-Control Schedule (Rosenbaum 1980a) was used to assess participants selfcontrol. The Self-Control Schedule is composed of 36 items, and each item is scored on a 6-point Likert-type scale, with re sponses that range from very uncharacteristic of me to very characteristic of me The Self-Control Schedule score is the sum of the individual items after reversing the scores of some ite ms. A high score indicates a high le vel of self-control. Originally,

PAGE 33

33 the score for each item could range from -3 to + 3, with no neutral response at 0, and total scores ranging from -108 to +108. However, the scaling method was modified in this study so that items were scored from 0 to 5, and the to tal score potentially ranged from 0 to 180. A preliminary exploratory factor analysis of 176 undergraduate stude nts responses to the items on the Self-Control Schedule yielded seven factors, three of which were examined in the present study because they had the highest reliability es timates and were well represented by the items: Emotion Regulation, Planful Thinking, and Delay of Gratification. In the sample obtained for this study, coefficient alpha for these va riables were .72, .62, and .59, respectively. Perceived Self-Efficacy A nine-item measure of perc eived self-efficacy created for this study was used to assess participants confidence in their ability to control their thinking, emotions, and behavior (see Appendix C). These questions have Likert -type response options ranging from 0 ( cannot do ) to 6 ( certain can do ). A high score indicates high perceived self-efficacy for self-control. An exploratory factor analysis of 176 undergraduate students responses to the items on the measure yielded three factors: Perceive d Self-Efficacy for Delay of Grat ification, Perceived Self-Efficacy for Emotion Regulation, and Perceived Self-Efficacy for Planful Thinking. In this study, internal consistency estimates using coefficien t alpha were .77, .68, and .79, respectively GPA As part of a dem ographic questionnaire, student s were asked to report their current overall GPA (see Appendix D). Students GPAs ranged from 2.3 to 4.0, with a mean of 3.32 and a standard deviation of .42. Procedures With the permission of course instructors, participants were recruited from courses in educational psychology. Volunteers were asked to complete the questionnaires either during

PAGE 34

34 class or outside of class depending on the instructors preference. Volunteers were asked to sign letters of informed consent prior to completi ng the questionnaire. Participants completed the measures in the same order. The entire pro cedure took approximately 30-45 minutes. The final sample consisted of 163 participants, after 22 questionnaires were eliminated because of incomplete data.

PAGE 35

35 CHAPTER 3 RESULTS Preliminary Analyses and Descriptive Statistics A recursive (unidirectio nal) path analysis was conducted to test the relationships posited in the m odel presented in Figure 1-1 using the statistical software package LISREL 8.0. Observed scores for the latent variables were used in the analysis, and error variances were fixed using coefficient for each measure. Error variances fo r the perceived self-efficacy variables were permitted to correlate, as were the error variances for the perceived self-control variables. Preliminary analyses were conducted to exam ine to examine the relationships among the variables. The correlation matrix, as well as means and standard deviations for each of the variables used in the path analysis are presented in Table 3-1. As hypothesized, the total number of students role models were si gnificantly positively correlated with the number of balanced possible selves ( r = .39, p < .05). In addition, all of the perceived self-efficacy measures were significantly positively correlated. Also as hypothesized, the perceived self-efficacy measures were significantly positively correlated to the respective self-control measures. Specifically, Perceived Self-Efficacy for Delay of Gratification was significantly positively correlated with Perceived Delay of Gratification ( r = .38, p < .05), Perceived Self-Efficacy for Planful Thinking wa s significantly positively correlated with Perceived Planful Thinking ( r = .37, p < .05), and Perceived Self-Efficacy for Emotion Regulation was significantly pos itively correlated with Perceived Emotion Regulation ( r = .42, p < .05). However, each of the perceived self-effi cacy measures was also significantly positively correlated with other perceived self-control skills, although to le sser extents. Perceived SelfEfficacy for Delay of Gratification was significan tly positively correlated with Perceived Planful Thinking ( r = .18, p < .05) and Perceived Emotion Regulation (r = .16, p < .05); Perceived Self-

PAGE 36

36 Efficacy for Planful Thinking was significantly po sitively correlated with Perceived Delay of Gratification ( r = .22 p < .05) and Perceived Emotion Regulation (r = .17, p < .05), and Perceived Self-Efficacy for Emotion Regulation wa s significantly positively correlated with Perceived Delay of Gratification ( r = .22, p < .05). Finally, both Percei ved Delay of Gratification and Perceived Planful Thinking were signi ficantly positively related to GPA ( r = .20, p < .05 and r = .19, p < .05, respectively). Analysis of the Proposed Model According to the path analysis of the propos ed model, the goodness of fit test indicates that 2 (20) = 28.17, p = .11. Thus, the chi-square statistic is not significant, i ndicating that the model fits the data. The goodness of fit indices are consistent with this co nclusion: the comparative fit index (CFI) = .96, the root mean square error of a pproximation (RMSEA) = .05, the non-normed fit index (NNFI) = .92, and the root mean squared residual (RMR) = .06. Table 3-2 presents the total, direct, and indirect effects specified in the model. All effects were expected to be positive, and directional hypothesis tests were c onducted. Significance was determined using a .05 Type I error rate. The total number of role models significan tly predicted the number of balanced possible selves ( = .39, p < .05), and this effect was entirely direct There were no significant direct or indirect effects of total role models or the num ber of balanced possible selves on the perceived self-efficacy variables. In predicting the perceive d self-control factors (D elay of Gratification, Emotion Regulation, and Planful Thinking), the direct effects of the corresponding perceived self-efficacy variables were significant. Speci fically, Perceived Self -Efficacy for Delay of Gratification predicted Perceived Delay of Gratification ( = .55, p < .05), Perceived SelfEfficacy for Emotion Regulation predicte d Perceived of Emotion Regulation ( = .60, p < .05), and Perceived Self-Efficacy for Planful Thinki ng predicted Perceived Planful Thinking ( = .35,

PAGE 37

37 p < .05). There were no significant indirect e ffects on Perceived Emotion Regulation or Perceived Delay of Gratification. However, there was a significant indirect effect of the number of balanced possible selves on Pe rceived Delay of Gratification ( = .08 p < .05), which was mediated by Perceived Self-Efficacy for Delay of Gratification. Finally, the only significant direct effect on GPA was Pe rceived Planful Thinking ( = .22, p < .05). Support was also found for a significant indirect e ffect of Perceived Self-Efficacy for Planful Thinking on GPA ( = .10, p < .05 ) which was mediated by Per ceived Planful Thinking.

PAGE 38

38 CHAPTER 4 DISCUSSION The results of this study i ndicate that the total number of role models significantly predicts balanced possible selves. This finding lends support to theorist s claim that possible selves are shaped and influenced by social fa ctors. Specifically, this result suggests that individuals with more role mode ls, including both positive role models (people they admire and try to emulate) and negative role models (people they try to avoid being lik e), are better able to envision the positive and negative possibilities asso ciated with their future goals. Researchers should further investigate this relationship and should examine whether positive and negative role models are differentially related to possible selves. For example, it is unclear whether both positive and negative role models are necessary to predict balanced possible selves. Positive role models may contribute to the formation of positive possible selves, whereas negative role models may contribute to the formation of negative possible selves. Alternatively, positive role models may also contribute to the formation of negativ e possible selves and negative role models may also contribute to the formation of positive possible selves. Researchers need to examine whether having more positive (or negative role) models affects balanced possible selves. Finally, researchers should also examine the qualities of th e relationships between students and their role models to determine what aspects contribu te to the production of possible selves. Oyserman and Markus (1990) suggested that individuals who have balanced possible selves may have more motivation and control over their behavior than in dividuals without such balance. This study lends partia l support to this claim, as balance signific antly predicted students perceptions of their abi lity to delay of gratification (thr ough an indirect effect mediated by perceived self-efficacy for delay of gratif ication). Hence, stude nts who reported more balanced possible selves reported more self-control (in terms of de laying their gratification) than

PAGE 39

39 students with less balance. However, the number of balanced possible selv es did not significantly predict perceived self-efficacy fo r self-control or GPA. In othe r words, students with more balance in their possible selves did not feel more efficacious for engaging in self-control, nor did they have a higher GPA. These findings do not le nd support to the claims that possible selves provide the foundation for perceptions of se lf-efficacy (Cross & Markus, 1994; Ruvolo & Markus, 1992), and these results seem to cont rast with reports that possible selves are significantly related to improve d GPA (Anderman et al., 1999) and achievement test scores (Leonardi et al., 1998). There are several possible interp retations of these fi ndings (or the lack thereof). First, the number of balanced possible selves may predict GPA through other mechanisms, such as motivation (i.e., the extent to which individuals possible selves motivate them to engage in self-regulatory behaviors may predict academic achievement), future time perspective (i.e., the extent to which individu als see the contingency between their current actions and their future goals may predict acad emic achievement), and plausibility (i.e., the extent to which possible selves are likely to be achieved or avoided given the strategies that are being used to attain them may predict academic achievement). Second, it is possible that the balance measure used in this study was too ge neral. In this study, balance was measured by tallying the number of positive possible selves th at had matching negative possible selves in the same domain, but the final balance score was a sum of balance across all domains. Hence, the balance measure was not specific to the domai ns of self-control or achievement (more specifically, academic achievement). It is also possible that the self-e fficacy and self-control measures were too general. Although balance was indirectly related to de lay of gratification, it may have been more prudent to use specific m easures of academic perceived self-efficacy and academic self-control. Oyserman et al. (2006) us ed measures of academic possible selves and

PAGE 40

40 academic self-control and found a significant re lationship between them (such that higher balance predicted higher levels of perceived self -control). Furthermore, Oyserman et al. reported that balance in academic possible selves signifi cantly predicted GPA in high school students. Future studies should use measures of these va riables that refer specifically to academic achievement. As hypothesized, the perceived self-efficacy measures predicted their corresponding perceived self-control factors. Specifically, students perceive d self-efficacy for delay of gratification predicted their perceptions of their ab ility to delay gratifica tion; students perceived self-efficacy for emotion regulation predicted their perceptions of their ability to regulate their emotions, and students perceived self-efficacy for planful thinking predicted their perceptions of their ability to engage in planfu l thinking. Thus, the extent to wh ich students believed that they were capable of regulating their behavioral responses predicted th eir self-reports of their ability to do so. These results are consistent with soci al cognitive theory (Bandura, 1997), which posits that human behavior is learned observationally through modeling the behaviors, attitudes, and emotional reactions of others, and that perc eived self-efficacy can be affected by social influences. These results also support Banduras conception of perceived self-efficacy as being domain-specific, although all of the self-efficacy factors were significantly, positively correlated (presumably because they all dea lt with self-control in general). Furthermore, perceived selfefficacy for planful thinking significantly predic ted GPA (through an indir ect effect mediated by planful thinking). This result is c onsistent with an extensive amount of literature indicating that perceived self-efficacy has a positive impact on individuals success and achievement (Bandura, 1997; Schunk, 1984, 2003).

PAGE 41

41 Finally, the only perceived self -control factor that predic ted GPA was planful thinking. This finding suggests that student s who report that they are better able to plan ahead and envision how they will approach and solve problems achieve at higher levels than individuals who report that they lack this self-control skill. It is not clear why the other factors (delay of gratification and emotion regulation) were not significantly related to GPA, but upon closer inspection of the perceived self-efficacy items, the items for planfu l thinking all deal with problem-solving ability (a skill that is especially rele vant to the academic setting). The items for emotion regulation and delay of gratification are more general and context-free. Future studies should utilize a more specific, multidimensional measure of academic se lf-control in order to better predict GPA. In sum, this study identified some predictors of self-control skills and academic achievement. These findings are relevant to rese archers attempting to identify college students who are likely to succeed and those who are likel y to struggle academicall y. In particular, this study identified potential targets of intervention for individuals who are struggling academically or who lack certain self-control sk ills. The results of this study s uggest that an intervention aimed at improving the balance of students possible se lves may improve some self-control skills (i.e., planful thinking), which may improve GPA. Oyse rman et al. (2006) conducted an intervention aimed at improving the quality of students possible selves. The intervention was successful in that students academic initiative, grades, and st andardized test scores improved, while absences and school misconduct declined. These results suggest that parent s, teachers, and other role models may influence students academic achievement. Specificall y, parents, teachers, and other significant social influences may influence studen ts possible selves, which in turn may affect their ability to regulate their behaviors and resp onses, which may ultimately affect achievement.

PAGE 42

42 Future research should more closely examine the ch aracteristics of the soci al relationships that are most likely to have positive effects. In conclusion, this study provided some suppor t for the path model relating role models, possible selves, perceived self-effi cacy, self-control, and GPA. Re searchers may be able to use this model as a starting point or reference for their own studies, ma king modifications as necessary. Future studies should test the model us ing measures that are academically-focused, as it is likely that the predictive power of these va riables would be improved if they are context specific.

PAGE 43

43Table 3-1. Intercorrelations, means, and standard deviations of variables in path analysis (N = 163) Variable 1 2 3 4 5 6 7 8 9 1. Total role models -2. Balanced possible selves .39* -3. Self-Efficacy for Delay of Gratification .00 -.04 -4. Self-Efficacy for Emotion Regulation -.01 -.01 .41* -5. Self-Efficacy for Planful Thinking .06 .11 .58* .36* -6. Perceived Delay of Gratification .00 .09 .38* .22* .22* -7. Perceived Emotion Regulation .12 .07 .18* .42* .17* .05 -8. Perceived Planful Thinking .08 .15 .16* .07 .37* .24* .26* -9. GPA .11 .01 .09 .13 .11 .20* .08 .19* -M 4.91 1.13 10.54 10.88 8.21 8.38 13.38 9.94 3.32 SD 3.12 1.15 3.91 3.38 2.69 3.82 4.93 4.09 0.42 p < .05.

PAGE 44

44Table 3-2. Total, direct, and indir ect effects in the proposed model ( N = 163) Variable Effect 1 2 3 4 5 6 7 8 9 1. Total role models Total ---------Direct ---------Indirect ---------2. Balanced possible selves Total .39* --------Direct .39* --------Indirect ---------3. Self-Efficacy for Delay of Gratification Total .06 -.16 -------Direct --.16 -------Indirect -.06 --------4. Self-Efficacy for Emotion Regulation Total -.02 -.06 -------Direct ---------Indirect -.02 --------5. Self-Efficacy for Planful Thinking Total .05 .14 -------Direct ---------Indirect .05 --------6. Perceived Delay of Gratification Total .03 .08* .55* ------Direct .03 -.55* ------Indirect -.08* -------7. Perceived Emotion Regulation Total .02 .04 -.60* -----Direct ---.60* -----Indirect .02 .04 -------8. Perceived Planful Thinking Total -.02 -.06 --.35* ----Direct ----.35* ----Indirect -.02 -.06 -------9. GPA Total .00 .00 .09 -.03 .10* .16 .05 .22* -Direct -----.16 .05 .22* -Indirect .00 .00 .09 -.03 .10* ----Note -means the effect is not in the model. p < .05.

PAGE 45

45 APPENDIX A ROLE MODELS QUESTIONNAIRE Who influences the way you act and the type of person you try to be? We all probably observe other people who inspire us to behave in certai n ways. These people whom we desire to be like, or to avoid being like, can be thought of as our role models. Role models may range from close relatives and friends, to coworkers, superstars, hi storical figures, and even fictional characters. Positive role models are individuals who possess desirable qualities and who inspire others to emulate these qualities. Negative role models are individuals who possess undesirable qualities and who motivate others to avoid these qualities. Think about your current role models. Who do you want to be like? Who do you want to avoid being like? In the spaces below, please list and describe your current positive role models (people you desire to be like). If you need more room, use the back of this paper. You do not have to provide names, but please consider th e following questions: What do you admire about this individual? What positive qualities doe s he or she possess? What is his or her relation to you? In what ways does this person motivate you? My current positive role models (People who I want to be like and the desirable qualities they possess): What I am doing to become like this person: (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

PAGE 46

46 In the spaces below, please list and describe your current negative role models (people you do not want to be like). If you need more room, us e the back of this paper. You do not have to provide names, but please cons ider the following questions: What do you not admire about this individual? What undesirable qualities does he or she possess? What is his or her relation to you? In what ways does this person motivate you? My current negative role models (People who I do not want to be like and the undesirable qualities they possess): What I am doing to avoid becoming like this person: (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

PAGE 47

47 APPENDIX B POSSIBE SELVES QUESTIONNAIRE What will y ou be like in the future? Probably ev eryone thinks about the future, and when doing so, we usually think about the kinds of experiences that are in stor e for us and the kinds of people we might possibly become. Each of us has some image or picture of what we will be like and what we want to avoid being li ke in the future. Think about your futureimagine what youll be like, and what youll be doing. In the spaces below, write what you expect you will be like and what you expect to be doing in the future. After each expected goal, mark X in the NO column if you are not currently working on that goal or doing something about that expe ctation and mark X in the YES column if you are currently doing something to ge t to that expectation or goal. For each expected goal that you marked YES, use the space to the right to write what you are doing to attain that goal. Use the first ro w for the first expected goal, the second row for the second expected goal and so on. In the future, I expect to be Am I am doing something to be that way If yes, What I am doing now to be that way in the future NO YES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

PAGE 48

48 In addition to expectations and e xpected goals, we all have images or pictures of what we dont want to be like, what we dont want to do, or wa nt to avoid being. First, think a minute about ways you would not like to be in the future things you are concerned about or want to avoid being like. Write those concerns or selves to -be-avoided in the spaces below. In the space next to each c oncern or to-be-avoided self, mark X in the NO column if you are not currently working on avoiding that conc ern or to-be-avoided se lf and mark X in the YES column if you are currently doing someth ing so that this will not happen in the future. For each concern or to-be-avoided self that you marked YES, use the space at the end of each line to write what you are doing this year to reduce the chances that this will describe you in the future. Use the first row for the first concern, the second row for the second concern and so on. In the future, I want to avoid Am I doing something to avoid this If yes, What I am doing now to avoid being that way in the future NO YES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

PAGE 49

49 APPENDIX C PERCEIVED SELF-EFFICACY MEASURE For each of the following items, indicate how certain you are that you could perform the following tasks by choosing the appropriate number (1 7) on the following scale and marking it on the Scantron sheet. ANSWER SCALE: 0-6 0 1 2 3 4 5 6 Cannot do Moderately certain can do Certain can do 1. When I am feeling down, I make myself feel better by thinking positive thoughts. 2. When I feel pain, I keep myself from thinking about it by thinking of other things. 3. When I fail, I stop worrying about it by thinki ng of how I can be suc cessful in the future. 4. When I am faced with a difficult problem, I solve it by taking a step-by-step approach. 5. When I have a lot of work to do, I create a plan to complete it effectively. 6. When I have a bad habit, I ov ercome it by first identif ying everything that supports the habit. 7. When I have to complete an unpleasant task I do it right away. 8. When I can choose a small reward immediately or a larger reward later, I choose to wait for the larger reward. 9. When I have a difficult job to do, I do it ri ght away even though I would rather be doing something else.

PAGE 50

50 APPENDIX D DEMOGRAPHIC INFORMATION Please answ er the following questions by marking the appropriate number on your Scantron answer sheets. 1. Gender: If female, mark 0; if male, mark the number 1. 2. Your age: For example, if you are 21, mark the number 2 on item #38 on the Scantron sheet and mark the number 1 on item #39. 3. Class: 1 Freshman 2 Sophomo re 3 Junior 4 Senior 5 Other (please describe) ____________________________ 4. Ethnicity: White = 0, Black = 1, Hi spanic = 2, Asian = 3, Other = 4 5. GPA: Please estimate your GPA to two digi ts. For example, if your GPA is 3.5, mark 3 on line 42 and 5 on line 43.

PAGE 51

51 REFERENCES ACT. (2007a). Nation al collegiate retention and persistence to de gree rates. Retrieved December 7, 2007, from http://www.act.org/path/policy/pdf/retain_2007.pdf ACT. (2007b). 2007 retention/completion summary tables. Retrieved December 7, 2007, from http://www.act.org/path/po licy/pdf/retain_trends.pdf Akgun, S. (2004). The effects of situation and learned resourcefulness on coping responses. Social Behavior and Personality, 32, 441-448. Akgun, S., & Ciarrochi, J. (2003). Learned reso urcefulness moderates the relationship between academic stress and academic performance. Journal of Educational Psychology, 23, 287294. Anderman, E. M., Anderman, L. H., & Griesi nger, T. (1999). The re lation of present and possible selves during early adolescence to grade point average and achievement goals. Elementary School Journal, 100, 3-17. Anthis, K. S., Dunkel, C. S., & Anderson, B. (2003). Gender and identity status differences in late adolescents possible selves. Journal of Adolescence, 27, 147-152. Bahniuk, M., Dobos, J., & Hill, S. (1990). The impact of mentoring, collegial support, and information adequacy on car eer success: A replication. Journal of Social Behavior and Personality, 5, 431-451. Bandura, A. (1977). Social learning theory New York: General Learning Press. Bandura, A. (1986). Social foundations of thought and acti on: A social cognitive theory Englewood Cliffs: NJ: Prentice-Hall. Bandura, A. (1991). Social cognitiv e theory of self-regulation. Organizational Behavior and Human Decision Processes, 50, 248-287. Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman. Bryant, A. L., & Zimmerman, M. A. (2003). Role models and psychosocial outcomes among African American adolescents. Journal of Adolescent Research, 18, 36-67. Calkins, S. D. (1994). Origins and outcomes of individual differences in emotion regulation. Monographs of the Society for Re search in Child Development, 59, 53. Carey, M. P., Carey, K. B., Carnrike, C. L., & Meisler, A. W. (1990). Learned resourcefulness, drinking, and smoking in young adults. Journal of Psychology: Interdisciplinary and Applied 124 391-395.

PAGE 52

52 Cassidy, J. (1994). Emotion regulation: In fluences on attachment relationships. Monographs of the Society for Research in Child Development 59 228. Cross, S., & Markus, H. (1991). Po ssible selves ac ross the lifespan. Human Development, 34 230-255. Cross, S., & Markus, H. R. (1994). Self-schemas possible selves, and competent performance. Journal of Educational Psychology, 86, 423-438. Darwin, C. (1981). The descent of man, and selection in relation to sex Princeton, NJ: Princeton University Press. (Original work published 1871) Day, J. D., Borkowski, J. G., Punzo, D., & Hows epian, B. (1994). Enhancing possible selves in Mexican American students. Motivation and Emotion, 18, 79-103. Dunkel, C. S., & Anthis, K. S. (2001). The role of possible selves in identity formation: A short-term longitudinal study. Journal of Adolescence, 24, 765-776. Edwards, D., & Riordan, S. (1994). Learned re sourcefulness in Black a nd White South African university students. Journal of Social Psychology 134(5), 665-675. Eisenberg, N., & Fabes, R. A. (1992). Emotion regulation and the de velopment of social competence. In M. S. Clark (Ed.), Emotion and social behavior: Vol. 14. Review of personality and social psychology (pp. 119-150). Newbury Park, CA: Sage. Eisenberg, N., Fabes, R. A., Shepard, S. A., Murphy, B. C., Guthrie, I. K., Jones, S., Friedman, J., Poulin, R., & Maszk, P. (1997). Contem poraneous and longitudinal prediction of childrens social functioning from regulation and emotionality. Child Development 68 642-664. Erikson, E. H. (1985). Childhood and society (35th ed.). New York: Norton. Feldman, S. C., Martinez-Pons, M., & Shaham, D. (1995). The relationship of self-efficacy, selfregulation, and collaborative verbal behavior with grades: Preliminary findings. Psychological Reports 77, 971. Freud, S. (1989). The ego and the id (J. Riviere, Trans.). NewYo rk: Norton. (Original work published 1923) Gibson, D. E. (2004). Role models in career development: New directions for theory and research. Journal of Vocational Behavior 65, 134-156. Ginter, G. G., West, J. D., Zarski, J. J. (1989) Learned resourcefulness and situation-specific coping with stress. Journal of Psychology, 123, 295-304.

PAGE 53

53 Gruber, V. A., & Wildman, B. G. (1987). Th e impact of dysmenorrhea on daily activities. Behavior Research and Therapy, 25(2), 123-128. Hart, D., Atkins, R., & Ford, D. (1998). Urban America as a context for the development of moral identity in adolescence. Journal of Social Issues 54, 513-530. Ibarra, H. (1999). Provisional selves: Experimen ting with image and iden tity in professional adaptation. Administrative Science Quarterly 44 764. Kao, G. (2000). Group images and possible selves among adolescents: Linking stereotypes to expectations by race and ethnicity. Sociological Forum, 15, 407. Katz, R. C., & Singh, N. (1986). A comparison of current smokers and se lf-cured quitters on Rosenbaum's Self-Control Schedule. Addictive Behaviors, 11, 63-65. Kemper, T. (1968). Reference groups, socialization, and achievement. American Sociological Review, 33 31-45. Kennett, D., Morris, E., & Bangs, A. (2006). Learned resourcefulness and smoking. cessation revisited. Patient Education and Counseling 60, 206-211. Knox, M., Funk, J., Elliott, R., & Bush, E. G. ( 1998). Adolescents possible selves and their relationship to global self-esteem. Sex Roles, 39, 61-80. Knox, M., Funk, J., Elliott, R., & Bush, E. G. (2000) Gender differences in adolescents possible selves. Youth & Society, 31, 287-309. Kulik, C. T., & Ambrose, M. L. (1992). Personal a nd situational determinants of referent choice. Academy of Management Review, 17 212. Leonardi, A., Syngollitou, E., & Kiosseoglou, G. (1998). Academic achievement, motivation, and future selves. Journal of Adolescence, 21, 219-222. Lewinsohn, P. M., & Alexander, C. (1990). Learned resourcefulness and depression. In: M. Rosenbaum (Ed.), Learned resourcefulness: On copi ng skills, self-control, and adaptive behavior (pp. 202-217). New York: Springer. Lockwood, P. (2002). Could it happen to you? Pred icting the impact of downward comparisons on the self. Journal of Personality and Social Psychology, 82, 343-358. Lockwood, P., Jordan, C., & Kunda, Z. (2002). Mo tivation by positive or negative role models: Regulatory focus determines who will best inspire us. Journal of Personality and Social Psychology, 83, 854. Lockwood, P., & Kunda, Z. (1997). Superstars and me: Predicting the impact of role models on the self. Journal of Personality and Social Psychology, 73, 91.

PAGE 54

54 Lockwood, P., & Kunda, Z. (1999). Salient best selves can undermine inspiration by outstanding role models. Journal of Personality and Social Psychology, 76, 214-228. Lockwood, P., Sadler, P., Fyman, K., & Tuck, S. (2004). To do or not to do? Using positive and negative role models to harness motivation, Social Cognition 22, 422-450. Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41 954-969. Markus, H., & Ruvolo, A. (1989). Possible selves: Personalized representa tions of goals. In L. A. Pervin (Ed.), Goal concepts in personality and social psychology (pp. 211-241). Hillsdale: Erlbaum. Marshik, T. (2007). Exploratory factor analysis of the Self-Control Schedule (SCS). Unpublished manuscript. Mischel, W., Shoda, Y., & Peake, P. K. (1988). Th e nature of adolescent competencies predicted by preschool delay of gratification Journal of Personality and Social Psychology 54, 687. National Center for Education Statistics. (2005). Student effo rt and educational progress: Postsecondary persistence and progre ss. Retrieved October 23, 2007, from http://nces.ed.gov/programs/co e/2004/section3/indicator19.asp Oyserm an, D., Bybee, D., & Terry K. (2006). Po ssible selves and academic outcomes: How and when possible selves impel action. Journal of Personality and Social Psychology 91, 188-204. Oyserman, D., Gant, L., & Ager, J. (1995). A socially contextualized model of African American identity: Possible selves and school persistence. Journal of Personality and Social Psychology, 69 1216-1232 Oyserman, D., & Markus, H. R. (1990). Possible selves and delinquency. Journal of Personality and Social Psychology, 59 112. Oyserman, D., Terry, K., & Bybee, D. (2002). A po ssible selves intervention to enhance school involvement. Journal of Adolescence 24, 313-326. Redden, E. M., Tucker, R. K., & Young, L. (1983) Psychometric properties of the Rosenbaum schedule for assessing self-control. Psychological Record 33, 77-86. Rosenbaum, M. (1980a). A schedule for assessing self-control behaviors: Preliminary findings. Behavior Therapy, 11, 109-121. Rosenbaum, M. (1980b). Individual differences in self-control behaviors an d tolerance of painful stimulation. Journal of Abnormal Psychology, 89, 581-590.

PAGE 55

55 Rosenbaum, M. (1983). Learned resourcefulness as a behavioral repertoire for the self-regulation of internal events: Issues and speculations In M. Rosenbaum, C. Franks, & Y. Jaffe (Eds.), Perspectives on behavior therapy in the eighties (pp. 54-73). New York: Springer. Rosenbaum, M., & Ben-Ari, K. (1985). Learned helplessness and learned resourcefulness: Effects of noncontingent success and failure on individuals differing in self-control skills. Journal of Personality and Social Psychology, 48, 198-215. Rosenbaum, M., & Jaffe, Y. (1983). Learned helplessn ess: The role of individual differences in learned resourcefulness. British Journal of Social Psychology, 22, 215-225. Rosenbaum, M., & Palmon, N. (1984). Helpless ness and resourcefulness in coping with epilepsy. Journal of Consulting and Clinical Psychology 52, 244-253. Rosenbaum, M., & Rolnick, A. (1983). Self-contro l behaviors and coping with seasickness. Cognitive Therapy and Research, 9, 79-90. Rubin, K. H., Coplan, R. J., Fox, N. A., & Ca lkins, S. D. (1995). Emotionality, emotion regulation, and preschoolers' social adaptation. Development and Psychopathology, 7, 4962. Ruvolo, A. P., & Markus, H. R. (1992). Possible selves and performance: The power of selfrelevant imagery. Social Cognition 10 95-124. Schunk, D. H. (1984). Self-efficacy pers pective on achievement behavior. Educational Psychologist, 19, 48-58. Schunk, D. H. (2003). Self-efficacy for reading a nd writing: Influence of modeling, goal setting, and self-evaluation. Reading and Writing Quarterly, 19, 159-172 Shoda, Y., Mischel, W., & Peake, P. K. ( 1990). Predicting adolescent cognitive and selfRegulatory competencies from preschool dela y of gratification: Identifying diagnostic conditions. Developmental Psychology 26(6), 978-986. Simons, A. D., Lustman, P. J., Wetzel, R. D ., & Murphy, G. E. (1985). Predicting response to cognitive therapy of depression: Th e role of learned resourcefulness. Cognitive Therapy and Research 9, 79-89. Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts good adjustment, less pathology, better grades, and interpersonal success. Journal of Personality, 72, 271-324. Thomas, R. M. (1990) Social learning theory. In R. M. Thomas (Ed.), Encyclopedia of human development and education: Theory, research, and studies (pp. 75-78). New York: Pergamon Press.

PAGE 56

56 Turner, S. (1996). Big brothers: Impact on litt le brothers self-con cepts and behaviors. Adolescence, 31, 875-882. U.S. Census Bureau. (2000). Educational at tainment 2000. Retrieved October 10, 2007, from http://www.census.gov/pr od/2003pubs/c2kbr-24.pdf Weinberg, R., Grove, R., & Jackson, A. (1992). Strategies for building self-efficacy in tennis players: A comparative analysis of Australian and American coaches. Sport Psychologist, 6, 3-13. Wills, T. A., & Stoolmiller, M. (2002). The role of self-control in early escalation of substance use: A time-varying analysis. Journal of Consulting and Clinical Psychology 70, 986997. Wolfe, R. N., & Johnson, S. D. (1995). Personal ity as a predictor of college performance. Educational & Psychological Measurement 55, 177. Woodward, W. R. (1982). The "dis covery" of social behavioris m and social learning theory, 1870-1980. American Psychologist, 37, 396-410.

PAGE 57

57 BIOGRAPHICAL SKETCH Tesia Marshik m ajored in psychology and philo sophy as an undergraduate. She graduated cum laude and received a B.S. degree from John Carroll University in Ohio in May of 2005. She began her graduate career at the University of Florida in August of 2005. She plans to continue on to earn a Ph.D. in a combined program of developmental and educational psychology. She currently teaches courses in the Educational Psychology Department at UF. Upon graduation she plans to pursue a career as a professor at a liberal arts university, where she will conduct research, teach, and advise undergraduate students.