‘—

"

 

    
     
   

-u-I‘-
a

WWW «4|:ch L rm

‘u—m.‘

'iL‘P «I‘l'éf‘a'

,

1.72;?»

u-iiliwywmun NIT 1-3.1; AM

 

 

’(0 1r. F'r’ilr’g KM'TJJIZ‘“ >4

2. ..

 

 

Ar 11'3““ A

2“. .

J-

r:
1

'F.
\—

f" {1551; «7.1-1.1

-, 2.. .
‘2”:

3-22.32.

u.‘
..

- —..

. .u-{igq

ﬁrm-up

1' 37V?"
. xi.

i
:1

a
.u.‘
--»...__V

n...‘
—-.-.
~.

0
(.H- .
~ .. .1..- .

1 r

Thesis
"i

(A

 

6:003
5Y917é70
LIBRARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

PERFECTIONISM, GENDER, AND STRESS AS PREDICTORSOF
ALCOHOL CONSUMPTION, ALCOHOL RELATED PROBLEMS AND
DRINKING TO COPE AMONG COLLEGE STUDENTS: A TEST

OF THE DIATHESIS-STRESS MODEL OF PERFECTIONISM

presented by

Benjamin Neal Cohen

has been accepted towards fulﬁllment
of the requirements for

Ph.D. degree in Counseling Psychology

 

 

Major professor

Date August 13, 2003

 

MSUi: an Afﬁrmative Action/Equal Opportunity Institution 0-12771

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE
'1‘?!" ‘* " ’ Ii

Ace. ‘ 1‘51
(e;

3‘3
CD
2:;
0

 

 

VU

a tram) ffr m:

 

 

 

0 A
0409 12

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDuopGS-p. 15

PERFECTIONISM, GENDER, AND STRESS AS PREDICTORS OF ALCOHOL
CONSUIVIPTION, ALCOHOL RELATED PROBLEMS, AND DRINKING TO COPE
AMONG COLLEGE STUDENTS: A TEST OF THE DIATHESIS-STRESS MODEL

OF PERFECTIONISM

By

Benjamin Neal Cohen

A DISSERTATION
Submitted to

Michigan State University
in partial fulﬁllment of the requirement
for the degree of

DOCTOR OF PHILOSOPHY

Department of Counseling, Education Psychology,
and Special Education

2003

ABSTRACT
Perfectionism, Gender, and Stress as Predictors of Alcohol Consumption, Alcohol

Related Problems, and Drinking to Cope Among College Students: A Test of the
Diathesis-Stress Model of Perfectionism

By

Benjamin Neal Cohen
The purpose of this research was to investigate the utility of the Diathesis—Stress Model of
Perfectionism to predict alcohol consumption, alcohol related problems, and drinking
alcohol to cope among college students. It was hypothesized that gender and
perfectionism (i.e., self-oriented, self-discrepant, socially-prescribed) would moderate the
relationship between perceived stress and alcohol involvement. Results of logistic
regression analyses (n = 198) found that self-oriented perfectionism, a two-way
interaction (self-oriented perfection x perceived stress), and a three-way interaction (self-
oriented perfectionism x perceived stress x gender) were associated with a change in the
likelihood of consuming alcohol. Interpretation of the two-way interaction suggested that
low self-oriented perfectionists who experienced low perceived stress were more likely to
consume alcohol than low self-oriented perfectionist who experienced high or moderate
perceived stress. Interpretation of the three-way interaction suggested a similar
relationship among female college students. However, among male college students, low
self-oriented perfectionists who experienced low perceived stress were less likely to
consume alcohol than high self—oriented perfectionists who experienced low perceived
stress. Results of two logistic regression models regarding alcohol related problems and

drinking alcohol as a means to c0pe were not signiﬁcant. Results suggest a possible

Stress-Buﬂer Model of Perfectionism among male college students. Clinical implications

and limitation of the research are discussed.

© Copyright by
BENJAMIN NEAL COHEN
2003

Dedicated to Gabrielle and Eli.

May all your dreams come true.

ACKNOWLEDGEMENTS

I would like to acknowledge all individuals who have assisted in this project.
First and foremost, I would like to thank Dr. Kenneth Rice for his dedication, support,
responsiveness, patience, and encouragement during this project, as well as throughout
my graduate studies. He has been an incredible exemplar of the psychologist as
“scientist-practitioner.” He is a wonderful mentor as a clinician, researcher, educator,
advisor, statistician, supervisor, clinical director, and administrator. His “adaptive”
perfectionism has been an essential asset to my professional and personal development. It
has been a great pleasure to work with him.

I would also like the thank the additional members of my dissertation committee
which include: Drs. Dave Novicki, Mark Reckase, and Gloria Smith. Their feedback,
comments, and suggestions were invaluable in the development, implementation, and
reporting of this project.

I would like to express my gratitude to my parents, Robert and Carol Cohen, as
well as my parents-in-law, Anthony and Angela Pilgrim, for their very generous and
continuous support throughout this process. A special thanks to Joy, Lon, and Heather
Newby. Their generosity serves as a model for all. I also need to thank my daughter,
Gabrielle, and my son, Eli, for the helpful distraction they have provided in recent years.
I would also like to acknowledge my wife, Allison, who accommodated to my absent
presence during my years in graduate school. This dissertation could not have been

accomplished without her love and support.

vi

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... viii
LIST OF FIGURES ............................................................................................................ ix
CHAPTER I: INTRODUCTION ....................................................................................... 1
CHAPTER II: LITERATURE REVIEW ........................................................................... 8
Perfectionism and Negative Psychological Outcomes ............................................ 8
Perfectionism and Alcohol Involvement ................................................................. 9
Alcohol Abuse Among College Students .............................................................. 13
Diathesis-Stress Model of Perfectionism .............................................................. 15
Stress-Response Dampening ................................................................................. 27
Stress-Vulnerability Model (SVM) of Alcohol Consumption .............................. 33
General Summary .................................................................................................. 37
Hypotheses ............................................................................................................ 38
Clinical Implications ............................................................................................. 41
CHAPTER 1]]: METHOD ............................................................................................... 44
Procedure ............................................................................................................... 44
Participants ............................................................................................................ 46
Instruments ............................................................................................................ 52
CHAPTER IV: RESULTS ............................................................................................... 63
Data Cleaning ........................................................................................................ 63
Descriptive Statistics and Preliminary Analyses ................................................... 63
Analyses ................................................................................................................ 68
CHAPTER V: DISCUSSION .......................................................................................... 84
Limitations ............................................................................................................ 93
Clinical Implications ............................................................................................. 95
Conclusion and Recommendations for Future Studies ......................................... 96
REFERENCES ................................................................................................................ 100
APPENDD( I (Informed Consent) .................................................................................. 128
APPENDIX 11 (Recruiting Script) .................................................................................. 131
APPENDIX III (Demographic Information) ................................................................... 133
APPENDIX IV (Questionnaires) .................................................................................... 138

vii

LIST OF TABLES

TABLE 1

Demographic Characteristics ................................................................................ 49
TABLE 2

Demographic Characteristics of Parents ............................................................... 51
TABLE 3

Descriptive Statistics ............................................................................................. 64
TABLE 4

Correlation Matrix of Variables ............................................................................ 65
TABLE 5

Hierarchical Logistic Regression to Predict Alcohol Consumption Group .......... 70
TABLE 6

Hierarchical Logistic Regression to Predict Alcohol Related Problems Group 79

TABLE 7
Hierarchical Logistic Regression to Predict Drinking Alcohol to Cope Group.... 81

viii

LIST OF FIGURES

FIGURE 1
Interaction of Self-Oriented Perfectionism and Perceived Stress
in the Predicted Odds of Consuming Alcohol ....................................................... 72

FIGURE 2
Interaction of Self-Oriented Perfectionism and Perceived Stress
in the Predicted Odds of Alcohol Related Problems ............................................. 74

FIGURE 3
Interaction of Self-Oriented Perfectionism and Perceived Stress
in the Predicted Odds of Drinking Alcohol to Cope ............................................. 76

CHAPTER I: INTRODUCTION

Over the past two decades, perfectionism has become an important individual
differences variable that has been linked to a multitude of negative psychological
outcomes (Blatt, 1995; Hamachek, 1978; Hollender, 1978; Pacht, 1984). Researchers
have found that perfectionism is related to anxiety (Alden, Bieling, & Wallace, 1994),
chronic pain (Hewitt, Flett, & Mikail, 1995; Liebman, 1978; Van Houdenhove, 1986),
depression (Hewitt & Dyck, 1986; Hewitt & Flett, 1990), eating disorders (Axtell &
Newlon, 1993; Garner, Olmstead, & Polivy, 1983), obsessive-compulsive disorders
(Frost, Steketee, Cohn, & Gn‘ess, 1994), personality disorders (Hewitt, Flett, & Tumbull,
1992), psychosomatic disorders (Fonnan, Tosi, & Rudy, 1987) and suicide (Adkins &
Parker, 1996). Additional studies have also linked perfectionism to alcohol consumption
in adults (Hewitt & Flett, 1991; Nerviano & Gross, 1983). For example, Hewitt and
Flett (1991) have found that certain types of perfectionism are signiﬁcantly related to
alcohol consumption and have suggested that alcohol consumption may stem from
having to cope with a perceived failure to achieve perfection due to high self-standards
and self-critical reactions. They have also suggested that alcohol consumption may stem
from being unable to meet perceived social pressures to be ﬂawless that are imposed on
them by signiﬁcant others and by society.

Studies of perfectionism using college student samples have paralleled the
ﬁndings of studies using adults. Studies of perfectionism among college students have
also linked perfectionism to a variety of similar negative psychological outcomes,

including academic adjustment problems (Rice & Mirzadeh, 2000), anxiety

(Deffenbacher, Zwemer, Whisman, Hill, & Sloan, 1986; Dunkley, Blankstein, Halsall,
Williams, & Winkworth, 2000; Flett, Hewitt, & Dyck, 1989; Goldfried & Sobocinski,
1975; Johnson & Slaney, 1996), depression (Rice, Ashby, & Slaney, 1998), daily hassles
(Dunkley, Blankstein, Halsall, Williams, & Winkworth, 2000), low self-esteem (Rice et
al., 1998), suicide (Chang, 1998), and alcohol abuse (Cohen & Rice, 2003). The results
of the only study to examine the relationship between perfectionism and alcohol abuse
among college students found that perfectionism, when combined with low anxiety, was
predictive of alcohol abuse (Cohen & Rice, 2003). Interestingly, perfectionism by itself
was not predictive of alcohol abuse.

A theoretical link between perfectionism and alcohol involvement has been
described in the literature, however few studies to date have empirically investigated this
relationship. The few studies that have investigated this relationship empirically have
found evidence of a theoretical link in adult samples, however, these studies used
different instruments to assess perfectionism and thus operationalized perfectionism
differently. No study to date has simultaneously examined multiple measures of
perfectionism and their possible link to perfectionism. In addition, only one study has
investigated this link using a college sample. This is surprising, given the current
problematic alcohol consumption and alcohol related problems among college students.

Research suggests that late adolescence and early adulthood is associated with the
highest prevalence of alcohol consumption (Grant et al., 1994; Johnston, O’Malley, &
Bachman, 1996; Kessler, et al., 1997). In particular, studies indicate that alcohol
consumption and abuse are widespread on college campuses, ﬁnding that over 85% of

college students drink and up to 50% of college students report that they have engaged in

binge drinking with the intention of getting drunk (Wechsler, Davenport, Dowdall,
Moeykens, & Castillo, 1994; Wechsler, Dowdall, Mainner, Gledhill-Hoyt, & Lee, 1998).
In addition, psychosocial problems associated with heavy drinking, such as rapes,
unplanned sexual intercourse, and violent crimes on campus have also increased in recent
years (Schudkit, Klein, Twitchell, & Springer, 1994; Wechsler et al., 1994). Thus, given
the high prevalence of alcohol consumption and alcohol related problems among college
students as well as the empirical link between perfectionism and alcohol consumption in
adult samples, an examination of the relationship between alcohol consumption and
perfectionism among college students seems warranted.

Most recently, research on perfectionism has begun to examine the Diathesis-
Stress Model of Perfectionism to predict negative psychological outcomes (Hewitt &
Flett, 2002). This model may be useful to predict alcohol involvement among college
students. This model proposes that perfectionists are more vulnerable to negative
psychological outcomes because perfectionism exacerbates the influence of stress on
negative psychological outcomes (Hewitt & Flett, 1991; Hewitt, Flett, & Ediger, 1996).
Consistent with this model, there has been some evidence to suggest that perfectionism
does indeed intensify the aversive effect of stress on negative psychological outcomes,
such as anxiety (Chang & Rand, 2000; Joiner & Schmidt, 1995), depression (Flett,
Hewitt, Blankstein, & Mosher, 1995; Hewitt & Dyck, 1986; Hewitt & Flett, 1993;
Hewitt, Flett, & Ediger, 1996; Joiner & Schmidt, 1995), and suicide (Hewitt, Flett, &
Weber, 1994). However, no study to date has investigated the utility of this model to
predict alcohol involvement among college students. Given the prevalence of both

perfectionism (Chang & Rand, 2000; Dunkley, Blankstein, Halsall, Williams, &

Winkworth, 2000; Rice, Ashby, Slaney, 1998) and stress (D'Zurilla & Sheedy, 1991;
Gadzella, 1994; Hirsch & Ellis, 1996; Ross, Neibling, & Heckert, 1999; Towbes &
Cohen, 1996) among college students, this may be a useful model to predict alcohol
involvement among college students.

An additional argument for the use of the DSM to predict alcohol consumption
among college students derives from studies that have found that stress is associated with
alcohol consumption. Speciﬁcally, the Stress-Response Dampening (SRD) model (Sher
& Levinson, 1982) posits that individuals will increase their use of alcohol during times
of stress.

To date, the empirical evidence for the SRD model of alcohol consumption is
mixed. Despite an intuitive appeal, recent reviews of empirical research of the SRD
indicate that the relationship between stress and alcohol consumption is equivocal in
college students. Perkins (1999) found that stress reduction was a motivating factor to
drink among college students and recent college graduates. However, other studies have
found no relationship between the number of stressful life events and increased alcohol
consumption (Corcoran & Parker, 1991). Additional studies have found that college
students' alcohol consumption decreases during times of stress, such as before exams
(Noel & Cohen, 1998).

One reason for the discrepant ﬁndings may be due to the variety of instruments
used to measure stress. According to some researchers (Chang & Rand, 2000; Cohen,
Kamarck, & Mermelstein, 1983; Cohen, 1986; Hammen, 1992; Hewitt, Flett, & Mosher,
1992) instruments that assess stress using an endorsement of speciﬁc daily hassles or life

events may be insensitive to chronic stress, stress occurring in the lives of close friends

and family, stress from expectations about the future, and stress from events not listed on
the measure. Therefore, the mixed results may be due to inconsistent or inaccurate
operationalization of stress.

Another reason for the inconsistent findings between alcohol consumption and
stress may be due to important moderating variables that signiﬁcantly inﬂuence the
relationship between stress and alcohol involvement. To account for these inconsistent
ﬁndings, researchers have attempted to identify theoretically signiﬁcant individual-
difference variables that moderate the relationship between stress and alcohol
involvement (Cooper, Russell, Skinner, Frone, & Mudar, 1992; Greeley & Oei, 1999).
Gender is one individual—difference variable that has been found to moderate the
relationship between stress and alcohol consumption, alcohol related problems, and the
motivation to drink alcohol to cope with negative emotions. However, to date no study
has examined if personalin characteristics, such as perfectionism, may also be important
individual difference variables that may moderate the relationship between stress and
alcohol related outcomes. This is surprising given research ﬁndings that indicate both
perfectionism and gender moderate the relationship between stress and other negative
psychological outcomes (Chang & Rand, 2000; Flett, Hewitt, Blankstein, & Mosher,
1995; Hewitt, Flett, & Ediger, 1996; Joiner & Schmidt, 1995).

To summarize, theory and research have linked perfectionism to alcohol
involvement (Cohen & Rice, 2003; Hewitt & Flett, 1991). Theory and research have also
linked stress to alcohol involvement (i.e., SRD), however this relationship is inconsistent
and may be moderated by signiﬁcant individual-difference variables, such as gender or

personality characteristics, that make individuals more vulnerable to the effects of stress

(SVM). Interestingly, theory and research have also recently suggested that
perfectionism may be an important individual-difference variable that may make
individuals more vulnerable to the effects of stress and more likely to suffer from an
exacerbation of negative psychological outcomes (DSM). It is notable, however, that no
research to date has investigated the DSM to predict alcohol involvement. In particular,
no study has investigated if perfectionism interacts with stress and gender to predict
alcohol consumption, alcohol related problems, and the motivation to drink alcohol to
cope. In addition, no study has examined these variables as they relate to college student
alcohol involvement. Thus, the DSM may be a useful model to predict alcohol
consumption, alcohol related problems, and the motivation to drink alcohol to cope
among college students.

The purpose of this study is to investigate the relationship between perfectionism,
stress, gender, and their interactions in the prediction of alcohol consumption, alcohol
related problems, and the motivation to drink alcohol to cope. Based upon theoretical
models as well as previous research, I hypothesize that perfectionism, stress, and gender
will predict alcohol consumption, alcohol related problems, and the motivation to drink
alcohol to cope. I also hypothesize that perfectionism and gender will moderate the
relationship between: (a) stress and alcohol consumption, (b) stress and alcohol related
problems, and (c) stress and the motivation to drink alcohol to cope. More speciﬁcally, I
hypothesize that male gender will predict alcohol consumption, alcohol related problems,
and the motivation to drink to cope with negative emotions, particularly when men report
high levels of perfectionism (i.e., socially prescribed perfectionism, self-oriented

perfectionism, and self-discrepant perfectionism) and report high levels of perceived

stress. In contrast, I hypothesize that female gender will not be predictive of alcohol
consumption, alcohol related problems, and the motivation to drink alcohol to cope,
despite reports of high perfectionism and perceived stress. The chapter that follows will
critically review contemporary literature regarding perfectionism and negative
psychological outcomes, alcohol consumption and alcohol related problems among
college students, the Diathesis-Stress Model of Perfectionism (DSM), the Stress-
Response Dampening (SRD) model, and the Stress-Vulnerability Model (SVM). I will
then make an argument for utilizing the DSM as an integrative model to predict college
student alcohol consumption, alcohol related problems, and motivation to drink to cope.
The review will conclude with a discussion of speciﬁc research hypotheses of this study.
These hypotheses will address the prediction of alcohol consumption, alcohol related

problems, and motivation to drink alcohol.

CHAPTER II: LITERATURE REVIEW

In this chapter I present an organized review of the literature relevant to the
current study. First, I critically review contemporary literature regarding perfectionism
and negative psychological outcomes, including alcohol consumption among adults.
Next, I examine the current problematic abuse of alcohol consumption among college
students and the possible link between perfectionism and college alcohol drinking. Then,
I examine the empirical evidence for the Diathesis-Stress Model of Perfectionism (DSM)
and suggest its potential application toward a better understanding of alcohol
involvement (i.e., alcohol consumption, alcohol related problems, and the motivation to
drink to cope among college students). Next, I present the equivocal empirical results of
the Stress—Response Dampening (SRD) model among college students and potential
reasons for those equivocal results. The Stress- Vulnerability Model (SVM) will then be
presented and examined. I then make an argument for the use of the DSM as an
integration of the SRD and the SVM models to gain a better understanding and prediction
of college student alcohol involvement. I conclude the chapter with a discussion of the
speciﬁc research hypotheses based upon an integration of these theoretical models.
These hypotheses will address the prediction of alcohol consumption, alcohol related

problems, and drinking alcohol to cope among college students.

Perfectionism and Negative Psychological Outcomes
Perfectionism has historically been of interest to psychologists (Adler, 1956;
Freud, 1926/1959; Hamachek, 1978; Horney, 1950). In fact, historians of psychology

believe that the ﬁrst psychological book ever written was entitled, Psychologia hoc est,

de hominus perfectione, or Psychology: About the perfectibiliry of man [sic], and was
published in 1590 (Lapointe, 1970).

Psychologists have become increasingly interested in perfectionism in the last two
decades and have come to regard perfectionism as an important individual difference that
is theoretically linked to a multitude of negative psychological outcomes (Blatt, 1995;
Hamachek, 1978; Hollender, 1978; Pacht, 1984). Researchers have also found empirical
evidence that links perfectionism to negative psychological outcomes, including anxiety
(Alden, Bieling, & Wallace, 1994), chronic pain (Hewitt, Flett, & Mikail, 1995; Liebman,
1978; Van Houdenhove, 1986), depression (Hewitt & Dyck, 1986; Hewitt & Flett, 1990),
eating disorders (Axtell & Newlon, 1993; Garner, Olmstead, & Polivy, 1983), obsessive-
compulsive disorders (Frost, Steketee, Cohn, & Griess, 1994), personality disorders
(Hewitt, Flett, & Tumbull, 1992), psychosomatic disorders (Forman, Tosi, & Rudy,
1987) and suicide (Adkins & Parker, 1996; Blatt, 1995). Additional literature has also
theoretically linked perfectionism to alcohol involvement. For example, Lombardi,
Florentino, and Lombardi (1998) have theorized that alcohol involvement is one of the
many negative psychological outcomes of perfectionism. Literature has also found an
empirical link between perfectionism and alcohol consumption (Hewitt & Flett, 1991;

Nerviano & Gross, 1983).

Perfectionism and Alcohol Involvement
Nerviano and Gross (1983) conducted a thorough review of the empirical
literature regarding the relationship between personality and alcohol consumption. They

concluded that, for some individuals, "destructive drinking seems to occur in the context

of obsessive perfectionism. These inhibited individuals are atypical of the alcoholic
stereotype, but can be easily detected with intensive clinical effort and lack of prejudice"
(p. 847). Thus, they argue that perfectionists may not ﬁt the typical clinical proﬁle of an
alcoholic, however, their personality pattern may put them at greater risk for problematic
drinking. Furthermore, these authors seem to suggest that perfectionists who do have
destructive drinking patterns may not be diagnosed because of clinicians' bias against
probing for drinking problems.

Hewitt and Flett (1991) found empirical evidence for a link between
perfectionism and alcohol involvement. Based upon the self-discrepancy theory of
alcohol abuse (Hull, 1981; Hull, 1987), Hewitt and Flett hypothesized that "excessive
drinking is an attempt to alleviate the negative affect associated with discrepancies
between the actual and ideal self" (p. 467). In order to test this theory, Hewitt and Flett
(1991) operationalized three aspects of perfectionism. "Self-oriented perfectionism" is a
personal need for the self to be ﬂawless. "Socially prescribed perfectionism" is the
individual's perception that others expect him or her to be ﬂawless. "Other-oriented
perfectionism" is the individual's expectation that others should be ﬂawless. Hewitt and
Flett (1991) found that alcohol consumption was signiﬁcantly related to two aspects of
perfectionism in a sample of 77 adult psychiatric patients. Alcohol consumption was
signiﬁcantly related to self-oriented perfectionism (r = .22, p < .05) and socially
prescribed perfectionism (r = .27, p < .05), but was not signiﬁcantly related to other-
oriented perfectionism (r = .20, p >.05). However, it is worth noting that the effect size
associated with other-oriented perfectionism is comparable to the effect size associated

with self-oriented perfectionism and socially prescribed perfectionism.

lO

Interestingly, this study also found that when men and women were analyzed
separately, self-oriented perfectionism was signiﬁcantly related to alcohol consumption
for men, and not for women, whereas socially prescribed perfectionism was signiﬁcantly
related to alcohol consumption for women, and not for men. This led Hewitt and Flett
(1991) to speculate that alcohol consumption in men stems from a discrepancy between
their desired achievement and their actual achievement, and thus involves self-standards
and self-critical reactions. In contrast, Hewitt and Flett (1991) speculated that alcohol
involvement in women stems from a discrepancy between their desired interpersonal
competence of their actual interpersonal competence, and thus involves needing to meet
unrealistic social pressures and standards believed to be imposed by others. Thus, these
researchers concluded that gender may be an important variable in understanding the
relationship between alcohol involvement and perfectionism. However, one limitation of
this study was that that the instrument used to operationalize perfectionism was not
theoretically consistent with the self-discrepant aspect of perfectionism that they
proposed is linked to alcohol abuse. In addition, the outcome measure assessed alcohol
abuse. No outcome measure was used to assess perfectionists' underlying motivation for
using alcohol. That is, Hewitt and Flett (1991) failed to directly assess if perfectionists
were indeed motivated to drink alcohol in order to cope with negative emotions, as their
theory suggests.

Cohen and Rice (2003) conducted the only study that has investigated the
association between alcohol involvement and perfectionism among college students (n =
135). Based upon the Tension-Reduction Hypothesis (Conger, 1956), they investigated if

anxiety, perfectionism, and the interaction of these variables were predictive of alcohol

11

abusers and non-alcohol abusers. In this study, perfectionism was assessed using the
Almost Perfect Scale-Revised (APS-R; Slaney, Rice, Mobley, Trippi, & Ashby, 2001). In
particular, the Discrepancy subscale of the APS-R was used, which assesses the
difference between a person's ideal standards and a person's perceived performance. The
rationale for the use of this instrument, instead of the Multidimensional Perfectionism
Scale used in the Hewitt and Flett (1991) study, was that it operationalized perfectionism
in a manner that was more consistent with the self-discrepancy theory of alcohol
involvement described by Hull (1981) and Hewitt and Flett (1991). Results of a logistic
regression analysis indicated that the neither perfectionism nor anxiety were predictive of
drinking classiﬁcation. However, a signiﬁcant interaction effect was found, indicating
that alcoholic classiﬁcation was more likely under conditions of low anxiety and high
perfectionism. The results of this study were unexpected, given the theoretical and
empirical link between perfectionism and alcohol involvement. However, these results
suggest that perfectionism, especially discrepancies between ideal standards and actual
performance, may play a critical role moderating alcohol involvement in college students.
Unfortunately, these researchers did not examine the possible moderating effects of
gender, as suggested by Hewitt and Flett (1991). This would have provided a more
complete examination of the gender speciﬁc self-discrepancy theory of alcohol
involvement (Hewitt & Flett, 1991). In addition, this study was limited because it did not
ascertain if self-discrepant perfectionism was associated with alcohol related problems or
drinking to cope with negative emotions.

Empirical research on the relationship between perfectionism and alcohol

involvement is limited to two studies, one of which found evidence for a direct link

12

between some aspects of perfectionism and alcohol abuse in a sample of psychiatric
patients (Hewitt & Flett, 1991). This study also suggested the possible moderating
effects of gender. Another study by Cohen and Rice (2003) found evidence for
perfectionism as moderating alcohol involvement among a sample of college students.
Although these studies utilized different types of samples and different measures of
perfectionism, they seem to suggest that perfectionism may be linked to alcohol
involvement, especially in the context of other moderating variables. However, the
Hewitt and Flett (1991) model that associates perfectionism with alcohol involvement has
not been fully examined among college students. This is surprising, given the pervasive

and problematic use of alcohol among college students.

Alcohol Involvement Among College Students

Recently, a task force of the National Advisory Council on Alcohol Abuse and
Alcoholism reviewed the current literature regarding college campus alcohol abuse
(Malloy & Goldman, 2002). The ﬁndings from this review are alarming. Approximately
85% of college students drink alcohol (W echsler, Davenport, Dowdall, Moeykens, &
Castillo, 1994) and 31% meet diagnostic criteria for alcohol abuse (Knight et al., 2002).

Research indicates that college students tend to consume drink more alcohol and
experience more negative consequences of alcohol during their first two years of college
than during their last two years of school, suggesting that college students may “mature
out” of negative drinking behavior (Klein, 1994; Wechsler et al., 1998). Also notable is
the fact that, unlike the general adult population, female college students consume

approximately an equal amount of alcohol when compared to their male counterparts yet

have been ignored and underrepresented in the literature (Crandall, 1995; Martin &
Hofffman, 1993; O’Hare, 1998; Perkins, 1992; Wechsler, Molnar, Davenport, & Baer,
1999).

This is a significant problem because college women who abuse alcohol are at a
much greater risk of being the victim of alcohol related problems including rape, assault,
violent crime, unprotected sexual intercourse, and unplanned sexual intercourse (Cooper,
2002; Hingson et al., 2002; Lo, 1996; O’Hare, 1998; Perkins et al., 2002; Wechsler et al.,
1994). In particular, a national survey conducted by Wechsler et al. (1994) found that
alcohol is a contributing factor in 90% of student rapes, 41% of unplanned sexual
encounters, and 22% of unprotected sexual encounters.

It is also noteworthy that signiﬁcant psychosocial problems associated with heavy
drinking on campuses have increased in recent years (Schudkit, Klein, Twitchell, &
Springer, 1994). An estimated 500,000 college students between the ages of 18-24 are
unintentionally injured each year under the inﬂuence of alcohol and an additional 1400
students die from such injuries (Hingson et al., 2002), making alcohol related accidents
and injuries the leading cause of death in this age group (National Institute of Alcohol
Abuse and Alcoholism, 1984). In addition, 1.5% of the college student population per
year attempts suicide due to drinking (Presley et al., 1998).

Academic performance in college is also signiﬁcantly hindered by alcohol abuse.
About 25% of college students report negative academic consequences of their drinking
including missed classes, failure to complete assignments, performing poorly on exams,
receiving lower grades and failing college courses (Berkowitz, & Perkins, 1986; Engs et

al., 1996, Engs & Hanson, 1985; Presley et al., 1996; Wechsler, 2002).

14

In conclusion, the task force recommended that brief motivational or skill-based
treatment interventions are most effective to reduce and prevent alcohol abuse, if such
treatments are aimed at high—risk students. However, relatively few risk factors have
been empirically identiﬁed to date (e. g., freshmen, members of fratemities/sororities,
athletes), which has led researchers to recommend that future studies identify additional
risk factors for college student alcohol abuse (Larimer & Cronce, 2002; Presley,

Weilman, & Leichleiter, 2002).

Diathesis-Stress Model of Perfectionism

Most recently, research on perfectionism has begun to examine the Diathesis-
Stress Model of Perfectionism (DSM) to predict negative psychological outcomes (Hewitt
& Flett, 2002). This model may be useful to predict alcohol involvement among college
students. Hewitt and Flett (2002) described the DSM as follows: "Perfectionism can play
a moderating role in producing psychopathological states by enhancing or exacerbating
the aversiveness of experienced stressors or failures” (p. 258). In particular, the
Diathesis-Stress Model of Perfectionism proposes that perfectionism magniﬁes the
effects of stress because perfectionism is associated with self-defeating styles of
cognitive appraisal (i.e., overgeneralization, personalization, etc.), maladaptive coping,
and poor problem-solving skills (Hewitt & Dyck, 1986; Hewitt & Flett, 1993; Hewitt &

Flett, 2002).

15

Consistent with this model, there has been some evidence to suggest that

perfectionism does indeed intensify the aversive effect of stress1 on negative
psychological outcomes, such as anxiety (Chang & Rand, 2000; Joiner & Schmidt, 1995),
depression (Flett et al., 1995; Hewitt & Dyck, 1986; Hewitt & Flett, 1993; Hewitt et al.,
1996; Joiner & Schmidt, 1995), hopelessness (Chang & Rand, 2000), and suicide (Hewitt

etal., 1994).

Diathesis-Stress Model of Perfectionism and Negative Psychological Outcome
Hewitt and Dyck (1986) conducted what appears to be the earliest study of the

DSM. A sample of 105 undergraduates completed measures of depression, stressful life
events, and perfectionism on two occasions that were two months apart. Individuals were
categorized into groups of either perfectionists or non-perfectionists based upon a median
split of their scores on the lO-item Perfectionism Scale (Burns, 1980). The results
indicated that for perfectionists, stress and depression were positively related on both
occasions. For non-perfectionists, stress and depression were not signiﬁcantly related on
either occasion. Using multiple regression analyses, results indicated that the most
powerful predictor of depression was a history of previous depression, which accounted
for 31% of the variance. On the second occasion, perfectionism scores signiﬁcantly
predicted depression, after controlling for depression scores on the ﬁrst occasion, which

accounted for an additional 7% of the variance. Contrary to the DSM, however, the

 

' Stress is deﬁned as any circumstances that threatens or perceives to threaten an individual's well-being,
thereby taxing the individual's coping ability (Whitehead, 1994).

16

interaction of stress and perfectionism was not predictive of depression. These
researchers conjectured that the failure of the interaction of stress and perfectionism to
predict depression was related to a poor measurement of stress. The authors reported the
test-retest correlation of .63 across a two-month span and an internal consistency
coefﬁcient alpha of .70. I would further suggest that the failure to ﬁnd evidence for the
interaction of stress and perfectionism to predict depression was related to a poor
measure of perfectionism with questionable reliability and validity. These authors state
that the Perfectionism Scale “carries some reliability and validity as a measure of
perfectionistic thinking” (p. 139) but report only a modest correlation between the
measure and endorsed perfectionistic adjectives (r = .23, p < .001). Fortunately, more
contemporary research on the DSM has beneﬁted from the development of
psychometrically valid measurements of perfectionism and stress. Hewitt and Flett
(1993) examined a more circumscribed test of the DSM, which they termed the Speciﬁc
Vulnerability Hypothesis. They hypothesized that self-oriented

perfectionism, which is concerned with personal achievement, would interact with
achievement related stressors (e. g., work load) to predict depression. They also
hypothesized that socially prescribed perfectionism, which is concerned with meeting the
expectations of others, would interact with interpersonal stressors (e.g., relationship
problems) to predict depression. In a sample of depressed (n = 51) and non-

depressed (n = 94) clients, they tested whether dimensions of perfectionism (i.e., self-
oriented, other-oriented, and socially prescribed perfectionism), stress (i.e., achievement
and interpersonal hassles), and their interaction were predictive of depression. Consistent

with the Specific Vulnerability Hypothesis, a hierarchical regression analysis indicated

17

that self-oriented perfectionism interacted with achievement hassles to predict depression
in both samples. That is, as achievement hassles increased, individuals with high self-
oriented perfectionism tended to be more depressed. In contrast, as achievement hassles
increased for individuals with low self-oriented perfectionism, there was no
corresponding change in depression. In the depressed sample, the interaction of socially
prescribed perfectionism and interpersonal hassles was also predictive of depression. As
interpersonal hassles increased, individuals with high socially prescribed perfectionism
tended to be more depressed. In contrast, as interpersonal hassles increased for
individuals with low socially prescribed perfectionism, there was no corresponding
change in depression. Additional results found that in the non-depressed sample, the
interaction of socially prescribed perfectionism and achievement hassles was predictive
of depression. As achievement hassles increased, individuals with high socially
prescribed perfectionism tended to be more depressed, however, as achievement hassles
increased for individuals with low socially prescribed perfectionism, there was no
corresponding change in depression. The researchers concluded that this study yields
partial- though not full- support the Speciﬁc Vulnerability Hypothesis of perfectionism.
The most consistent ﬁnding was that self-oriented perfectionism interacted with
achievement hassles to predict depression. However, most importantly, this study
demonstrates evidence for the more general DSM, which suggests that perfectionism
interacts with stress to yield more negative psychological outcomes, such as depression.

It is worth noting that one of the limitations of this study was that the instrument
used to measure stress (Hassles Scale; Delongis, Folkman, & Lazarus, 1988) has

questionable psychometric properties. For example, some researchers have claimed that

18

the instrument is confounded by psychiatric symptoms (Dohrenwend & Shrout, 1985),
negative affect (Watson, 1990), and neuroticism (Costa & McCrae, 1990). Other
researchers have criticized the measure for failing to assess the degree of perceived stress
that is experienced, relying too much upon speciﬁc stressful events (Chang & Rand,
2000; Cohen et al., 1983; Cohen, 1986; Hammen, 1992; Hewitt, Flett, & Mosher, 1992).
These researchers believe this instrument is not a reliable assessment of chronic stress,

interpersonal stress, stress related to the future, and stress related to events that are not

 

included on the instrument. Therefore, one reason for the mixed results may be due to
inconsistent or inaccurate measurement of stress.

Flett et a1. (1995) also tested the DSM to predict depression in two samples of
college students. Consistent with this theory, they hypothesized that self-oriented
perfectionism and socially prescribed perfectionism would interact with stress to produce
higher levels of depression. Sample 1 consisted of 374 undergraduate students who
completed measures of perfectionism, stress, and depression on one occasion. Sample 2
consisted of 173 undergraduate students who completed the same measures, on two
occasions that were three months apart. Results of hierarchical regression analyses found
that self-oriented perfectionism, life stress, and the interaction of these two variables were
predictive of depression at Time 1 in both samples. Speciﬁcally, they found that for both
samples, high levels of stress and high self-oriented perfectionism were associated with
higher levels of depression among the undergraduates. In contrast, high levels of stress
as well as low self-oriented perfectionism were not associated with depression. Another
hierarchical regression analysis from the second sample found that increases in

depression from Time 1 to Time 2 were associated with greater self-oriented

perfectionism. Results of other possible interactions between perfectionism and stress

were not reported. No significant effects were found for socially prescribed perfectionism

in the prediction of depression. The signiﬁcant interactions between self-oriented
perfectionism by life stress in the prediction of depression suggested the moderating
effects of self-oriented perfectionism. In fact, the authors of this study concluded that
self-oriented perfectionism may be relatively adaptive in situations of low stress but may
be relatively maladaptive when faced with high levels of stress, resulting in a greater
vulnerability to negative psychological outcomes, such as depression. This would be
consistent with the DSM. However, the authors also stated that there is less evidence for
the inclusion of socially prescribed perfectionism in the DSM.

Hewitt et a]. (1996) conducted a second study that investigated the Speciﬁc
Vulnerability Hypothesis, which is the more circumscribed theory of the DSM. In this
model, self-oriented perfectionism was hypothesized to interact with achievement-
oriented stress to exacerbate depression whereas socially prescribed perfectionism is
thought to interact with interpersonal stress to exacerbate depression. These authors
conducted a longitudinal study using a sample of undergraduates (n = 103), in which
depression, stress type (i.e., achievement and interpersonal stress), and perfectionism
were measured on two occasions, four months apart. Results of hierarchical regression
analysis indicated that depression at Time 2 was predicted by achievement stress and the
interaction of achievement stress by self-oriented perfectionism. An analysis of this
interaction indicated that as achievement stress increased, only high self-oriented
perfectionists experienced an increase in depression. Results of a second hierarchical

regression analysis indicated that depression at Time 2 was predicted by socially

2O

 

.V-Tl" In-
t

prescribed perfectionism and interpersonal stress, but not by the interaction of these
terms. Thus, the results of this study were similar to the results of the Flett et al. (1995)
study. Both studies found signiﬁcant interactions for self-oriented perfectionism by
achievement stress in the prediction of depression, but neither study found signiﬁcant
interactions for socially prescribed perfectionism by interpersonal stress, thus only
obtaining partial support for the Specific Vulnerability Hypothesis. However, in my
opinion, both studies support the more general DSM, which proposes that perfectionism
(i.e., self-oriented and socially prescribed perfectionism) interacts with non-speciﬁc stress
to cause negative psychological consequences, such as depression. However, the failure
to ﬁnd a signiﬁcant interaction between socially prescribed perfectionism and
interpersonal stress could also be attributed to the poor quality of the measurement of
stress, as described earlier. To add to that critique, neither study used the measurement of
stress (Life Events Questionnaire; Cochrane & Robertson, 1973) as it was originally
designed. Instead, they had three graduate students categorize a list of 55 stressful events
as related to either achievement or interpersonal stress and both studies report a
surprisingly high 100% interrater agreement. This failure to ﬁnd a signiﬁcant interaction
for socially prescribed perfectionism might also be attributed to the small sample size and
subsequent lack of statistical power to adequately test this hypothesis.

Joiner and Schmidt (1995) conducted a similar study that investigated the DSM
and the more reﬁned Speciﬁc Vulnerability Hypothesis. Using a sample of
undergraduates (n = 174), these researchers measured perfectionism (i.e., self-oriented
and socially prescribed perfectionism), stress type (i.e., achievement stress and

interpersonal stress), depression, and anxiety on two occasions, three weeks apart.

2]

Similar to previous studies, this study used multiple regression analyses to investigate if
an interaction between perfectionist type (i.e., socially prescribed and self-oriented) and
speciﬁc stress (i.e., interpersonal and achievement) would predict depression (at Time 2).
Based upon gender socialization theory, they believed that socially prescribed
perfectionism and self-oriented perfectionism are more problematic for men, and used
this theory as a justiﬁcation for exploring possible moderating effects of gender. They
conducted a second hierarchical regression analysis to predict anxiety and used the
identical independent variables as they had used in the prediction of depression.
Interestingly, these authors found a different result from previous studies (Hewitt & Flett,
1993; Hewitt et al., 1996). For men, they found that socially prescribed perfectionism
interacted with both interpersonal stress and achievement stress in the prediction of
depression. For example, men in the sample with high interpersonal stress and high
socially prescribed perfectionism tended to have the highest level of depression, whereas
men in the sample with low interpersonal stress and high socially prescribed
perfectionism had the lowest level of depression. In contrast, women with high
interpersonal stress and low socially prescribed perfectionism had the highest level of
depression whereas women with low interpersonal stress and low socially prescribed
perfectionism had the lowest level of depression. Contrary to Hewitt and Flett (1993) and
Hewitt et a1. (1996), they also found that the interaction of self-oriented perfectionism by
achievement stress was not predictive of depression. In general, these results provide
evidence for the DSM, but are contrary to the Speciﬁc Vulnerability Hypothesis.

In this study, they also investigated if the Specific Vulnerability Hypothesis could

predict anxiety scores. Using hierarchical multiple regression, these researchers found

22

that among men only, the interaction of socially prescribed perfectionism by
interpersonal stress was signiﬁcantly predictive of scores on anxiety. (The nature of this
interaction was not provided by the authors). This would be consistent with the Stress-
Vulnerability Hypothesis. However, the interaction of self-oriented perfectionism by
either type of stress was not predictive of anxiety scores for either gender.

As noted earlier, the results of this study contrasted the findings of other studies
(Hewitt & Flett, 1993; Hewitt etal., 1996). One possible reason for this could be that
different measures of perfectionism and stress were used, which might account for some
of the discrepant ﬁndings between the studies. For example, the constructs of self-
oriented perfectionism and socially prescribed perfection were measured by selecting six-
items from the Eating Disorders Inventory (Garner, Olmstead, & Polivy, 1983) that were
judged to measure the respective constructs. A clear rationale for the use of this
instrument was not provided. Despite this limitation, the authors of this study concluded
that the DSM received support among males but a more speciﬁc Stress-Vulnerability
Hypothesis was not supported. Ultimately, the authors stated that both self-oriented and
socially prescribed perfectionism may exacerbate both achievement stress and
interpersonal stress, however, they suggested that gender role identity may be an
important moderating variable in understanding this complex interaction.

Most recently, Chang and Rand (2000) have found support for the DSM in a
sample of 215 undergraduates. Consistent with this model and similar to previous
studies, perfectionism (i.e., self-oriented perfectionism, other oriented perfectionism,
socially prescribed perfectionism), perceived stress, and the interaction of these variables

were hypothesized to predict negative psychological outcome. However, instead of

23

predicting depression as other studies have done, Chang and Rand (2000) used this model
to predict general psychological symptoms (Symptoms Check List-90-R; Derogatis,
1983) and hopelessness (Beck Hopelessness Scale; Beck, Weissman, Lester, & Texler,
1974). The procedure required that participants complete these measures on two
occasions, 4 to 5 weeks apart. In addition, this study differed from previous studies
because it measured perceived stress, as opposed to daily hassles or stressful life events.
The rationale for this was to overcome the conceptual limitations associated with these
instruments. As mentioned previously, several researchers have found that the
endorsement of stressful events does not adequately capture the extent to which those
events were perceived as being stressful, and such lists would most certainly fail to
include all possible stressful life events that can be experienced (Chang & Rand, 2000; S.
Cohen; 1986; Hammen, 1992; Hewitt, Flett, & Mosher, 1992).

Results of a series of hierarchical multiple regressions found that perceived stress
was a signiﬁcant predictor of psychological symptoms at Time 2, accounting for 21% of
the variance. Socially prescribed perfectionism was a signiﬁcant predictor of
psychological symptoms at Time 2, accounting for 2% of the variance. Surprisingly,
neither self-oriented perfectionism nor other-oriented perfectionism was predictive of
psychological symptoms. There was also a signiﬁcant interaction for socially prescribed
perfectionism by perceived stress in the prediction of psychological symptoms at Time 2,
accounting for 2% of the variance. An analysis of this interaction found that under high
levels of perceived stress, high socially prescribed perfectionism was associated with
more psychological symptoms than low socially prescribed perfectionism. When under

low levels of perceived stress, low socially prescribed perfectionism did not signiﬁcantly

24

 

differ from high socially prescribed perfectionism in measures of psychological
symptoms.

A second hierarchical multiple regression analysis was conducted to predict
hopelessness at Time 2. Results of this analysis found that perceived stress, self-oriented
perfectionism, and socially prescribed perfectionism were predictive of hopelessness,
accounting for 32%, 2.5%, and 2.5% of the variance, respectively. Other oriented
perfectionism was not predictive of h0pelessness. Similar to the previous analysis, a
signiﬁcant interaction effect was found for social prescribed perfectionism by perceived
stress in the prediction of hopelessness, accounting for 3% of the variance. An analysis of
this interaction revealed that at high levels of stress, high socially prescribed
perfectionism was associated with higher levels of hopelessness than low socially
prescribed perfectionism.

Chang and Rand (2000) concluded that this study provides incremental validity
for the DSM. That is, socially prescribed perfectionism moderated the relationship
between perceived stress and negative psychological outcomes beyond what would be
predicted by perceived stress and perfectionism. However one notable limitation of this
study is that these authors did not measure perceived stress at Time 2 when distress was
measured. Nevertheless, these results are similar to another study that has found that only
socially prescribed perfectionism was an important moderating variable in the DSM
(Hewitt & Flett, 1993). Surprisingly, Chang and Rand (2000) failed to ﬁnd support for a
DSM that involved self-oriented perfectionism. Thus, these ﬁndings are contrary to the
results of other studies that have found evidence for the moderating inﬂuence of both

self-oriented perfectionism and socially prescribed perfectionism in the DSM (Hewitt &

25

 

Flett, 1993; Hewitt et al., 1996). The results of Chang and Rand (2000) are also in
opposition to Flett et a1. (1995) who found that only self-oriented perfectionism, and not
socially prescribed perfectionism, was an important moderating variable in the DSM.

One possible reason for the inconsistent findings among these studies concerning
the DSM may be related to the failure to account for the moderating effects of gender.
Only two studies to date have found evidence for both self-oriented perfectionism and
socially prescribed perfectionism in a DSM (Hewitt & Flett, 1993; Joiner & Schmidt,
1995). Notably, Joiner and Schmidt (1995) were the only researchers who considered the
moderating effects of gender and found evidence for the moderating effects of both self-
oriented perfectionism and socially prescribed perfectionism only among men, and not
among women. As a result, these researchers have extended the DSM by suggesting that
"perfectionistic men may be more vulnerable to psychological distress than
perfectionistic women because departures from perfectionism may be less compatible
with stereotypical masculine gender role than with stereotypical feminine gender roles"
(Joiner & Schmidt, 1995, p. 180). Hence, they argued that future studies on this topic
should include gender in their models, in order to further examine its possible moderating
effects.

Although the results of studies related to the DSM are inconsistent in terms of the
moderating effects of socially prescribed perfectionism and self-oriented perfectionism,
the general trend of this research strongly suggests that when under high stress, high
perfectionism is associated with an exacerbation of negative psychological symptoms.
This may be particularly true among men. However, one of the limitations of research

related to the DSM is that this model has only been used to predict depression, and to a

26

 

lesser extent, anxiety, hopelessness, and general negative psychological symptoms. No
study to date has investigated the utility of the DSM to predict other negative
psychological outcomes, such as alcohol involvement or alcohol related behaviors. This
is surprising, given both the theoretical and empirical link between perfectionism and

alcohol involvement (Hewitt & Flett, 1991; Nerviano & Gross, 1983).

Stress-Response Dampening

 

An additional argument for the use of the DSM to predict alcohol involvement L
and alcohol related behaviors derives from studies that have found that stress can lead to
both alcohol involvement and alcohol related behaviors. Speciﬁcally, the Stress-Response
Dampening (SRD) model proposed by Sher and Levinson (1982) posits that stress leads
individuals to drink more alcohol as a means to cope with the negative effects of stress.

Perkins (1999) has argued that the SRD model is an especially useful model in
understanding college drinking. He believes that the college environment includes
stressors (e. g., demanding academic environment, transition toward adult autonomy) as
well as a perceived social norm that encourages drinking. He believes that these factors
lead many college students to use drinking as a means to cope with stress.

Several studies among college students have found evidence in support of the
SRD model. That is, they have found evidence that college students consume more
alcohol when they are under stress. In a recent survey (n = 163), 36% of college students
reported that drinking to alleviate stress was an acceptable use of alcohol (McCormack,
1996). Surprisingly, this percentage increased from 23% in a similar survey conducted

four years previously at the same university. It is also noteworthy that 41% of the men

27

and 26% of the women believed that drinking to alleviate stress was an acceptable use of
alcohol. This suggests that men and women may differ in their motivation to use alcohol
to COpe with stress. Although data were obtained from only one university, the authors
suggested that using alcohol to alleviate stress is becoming increasingly popular on
college campuses.

Perkins (1999) conducted a survey of motivations for drinking among college
students. Based upon a sample of undergraduates in 1991 (n = 926), Perkins (1999) found
that 57% of men and 54% of women reported that stress was a signiﬁcant motivation to
drink alcohol. Interestingly, Perkins also found that this percentage of stress-motivated
drinking seems to decrease after graduation from college. He found that twelve to thirteen
years after graduation, about 39% of men and 40% of women reported that stress was a
signiﬁcant motivation to drink alcohol.

Noel and Lisman (1980) conducted an early study that investigated the
relationship between alcohol involvement and stress among college students. These
researchers used a series of experiments in which female participants were randomly
assigned to two conditions. In the ﬁrst condition, participants were presented with
anagram problems that were able to be solved. In the second condition, participants were
presented with anagram problems that were not able to be solved. After attempting to
solve the anagram puzzles, participants were then asked to "taste" an alcoholic drink.
Results of this study indicate that participants who were presented unsolvable anagrams
consumed signiﬁcantly more alcohol than participants who were presented solvable
anagrams. The researchers concluded that the reason those participants drank more

alcohol was because the unsolvable problems provoked more stress, which in turn lead to

28

 

more alcohol involvement. However, a major limitation of this research was a failure of
the researchers to assess the level of stress the participants were experiencing, thus
leaving this conclusion to be speculative.

Rabow and Neuman (1984) also obtained evidence for the SRD model among 76
undergraduates. These participants were asked to record their alcohol consumption and
stress level in a journal for one month. Using a qualitative analysis, these authors found
that binge drinking was signiﬁcantly associated with stressful life events. In particular,
they found that number of alcoholic beverages consumed, as well as the frequency of
consumption, increased signiﬁcantly following academically stressful times, such as mid-
term and ﬁnal exams. Thus, the results of this study would seem to be consistent with the
SRD model.

Orcutt and Harvey (1991) found modest support for the SRD model among 328
undergraduate students. These researchers investigated college-drinking behavior in
four-hour blocks during the week and weekend. Results of their study indicated support
for the SRD during the weekday but not during the weekend. That is, they found that
individuals during the week who reported more stress during the day were found to drink
more alcohol later that evening. However, the relationship between stress and subsequent
alcohol consumption was not found during the weekend. The researchers had difﬁculty
explaining this inconsistency and concluded that drinking on weekdays may be deemed a
culturally appropriate "time out" from stress (e.g., "happy hour"). Presumably, drinking
during the weekend to alleviate stress is not deemed as culturally acceptable.

Another study conducted by Kidorf and Lang (1999) used an experimental design

to investigate the relationship between induced stress and alcohol consumption. Eighty-

29

four undergraduates participated in a two-part experiment. In the first part of the
experiment, participants consumed alcohol for thirty minutes. This level of alcohol
consumption served as an indication of alcohol under low-stress condition. The
participants then returned two to ﬁve days later. At this time, participants were told that
they were to consume alcohol for thirty minutes and then they would be required to give
a speech concerning "their most undesirable characteristic." Participants were told that
this speech would be videotaped and evaluated by faculty and students. This condition
was regarded as a high-stress condition. Although the participants were never actually
required to give such a speech, their degree of stress and alcohol consumption was
measured in anticipation of such a speech. Results indicated that the undergraduates'
consumption of alcohol increased in the stressful condition when compared to the low-
stress condition. The researchers concluded that this was evidence for the SRD model.
In contrast, additional studies have found that college students' alcohol
consumption seems to decrease during times of stress. This would be contrary to the SRD
model of alcohol involvement. For example, Noel and Cohen (1998) had a sample of 73
undergraduates record their level of alcohol consumption "during a typical week" and
"during an exam week." These researchers hypothesized that alcohol consumption would
increase during exam week, when students were presumed to be under more stress.
Contrary to their hypotheses, the undergraduate students actually consumed signiﬁcantly
less alcohol per day during exam week when compared to the daily consumption of
alcohol during a typical week. This is despite the ﬁnding that the undergraduates
experienced more reported stress during exam week than during a typical week. This is

the opposite of what was expected based upon the SRD model. Hence, the researchers

30

 

concluded that the decreased drinking during stressful times occurred because the
undergraduates used "effective alternatives to drinking" in order to cope with their stress.

Other studies have found no relationship between the number of stressful life
events and increased alcohol involvement. For example, Corcoran and Parker (1991)
conducted an experiment with 69 undergraduates in which one group was exposed to a
low-stress condition (i.e., brief essay on a favorite leisure activity) and another group was
exposed to a high-stress condition (i.e., a speech on their most embarrassing body part).
They then measured the amount of alcohol that the undergraduates consumed. Contrary
to the SRD model, these researchers failed to ﬁnd any signiﬁcant difference in the
amount of alcohol consumed between the two levels of stress.

McCreary and Sadava (1998) also investigated the SRD model in a sample of 288
undergraduates. These researchers used three different measurements of stress, which
assessed perceived stress, daily hassles, and negative life events. Outcome measurements
included alcohol consumption and adverse consequences of alcohol involvement (i.e.,
alcohol related problems). These researchers used a hierarchical multiple regression
analysis to predict alcohol consumption. Results of this analysis found that none of the
measures of stress were signiﬁcant predictors of alcohol consumption. The failure to ﬁnd
that any of the three measures of stress were predictive of alcohol consumption was
surprising and contrary to the SRD model and the previous cited research. A second
hierarchical multiple regression analysis found that daily hassles were a signiﬁcant
predictor of alcohol related problems. Daily hassles accounted for approximately 9% of
the variance, after controlling for the effects of alcohol consumption and gender. This

may suggest that the SRD model may be most useful in predicting alcohol related

31

 

problems among college students, rather than alcohol consumption per se. Results of this
study also found only a moderate relationship between alcohol consumption and alcohol
related problems (r = .54), suggesting that alcohol consumption and alcohol related
problems may be related, but distinct constructs, each worthy of investigation .

Based upon these ﬁnding, McCreary and Sadava (1998) speculated that their
results did not support the SRD model was that "the stress-alcohol association may be
salient in only certain contexts" (p. 254). Unfortunately, these authors did not adequately
clarify the "contexts" in which they believed this association would be most salient. It
might be that signiﬁcant moderating variables may signiﬁcantly inﬂuence the
relationship between stress and alcohol involvement. This speculation has been supported
by a thorough review of the research related to the SRD model (Greeley & Oei, 1999).
After considering the mixed results of many studies, Greeley and Oei (1999) concluded
that, "These studies demonstrate that there are many factors that act as moderators. . .of
alcohol as a SRD agent. A simple SRD model of alcohol involvement is inadequate to
account for most problem-drinking behavior. Interactive models that incorporate the
effects of individual differences on stress. . .have supplanted the simpler, single-factor

explanatory models such as ...SRD" (p. 38).

Stress-Vulnerability Model (SVM) of Alcohol Involvement
Stress-Vulnerability Model (SVM) is an example of an interactive model that
incorporates how individual differences interact with stress to predict alcohol
consumption and alcohol related problems. Among the moderating variables that this

model explores is gender (Cooper, Frone, Russell, & Peirce, 1997; Cooper, Russell,

32

Skinner, Frone, & Mudar, 1992). To date, research strongly suggests that men consume
more alcohol and experience more alcohol related problems when compared to women
(Lemle & Mishkind, 1989; McCreary, Newcomb, & Sadava, 1999; Perkins, 1999),
however the SVM extends this line of research by theorizing that gender moderates the
relationship between stress and alcohol involvement. In particular, this model proposes
that men are more prone than women to extemalize their stress by using alcohol and
exhibiting alcohol related problems when exposed to high stress. This model also
theorizes that men are more likely to be motivated to drink alcohol to cope with negative
emotions.

One of the most interesting aspects of SVM is that it is not only used to predict
alcohol consumption and alcohol related problems, but it is also used to predict the
underlying motivation to drink alcohol. In contrast to the predominant view of
researchers, Cooper and her colleagues believe that "drinking behavior is not a unitary
phenomenon but instead represents multiple psychologically distinct behaviors deﬁned
by the different underlying functions they serve" (Cooper, Frone, Russell, & Mudar,
1995, p. 990). Speciﬁcally, Cooper and her colleagues have described four fundamentally
different motivations to drink alcohol. These motives include coping with negative
emotions, improving socialization, enhancing mood, and conforming to peer pressure.
Individuals who drink to cope with negative emotions or enhance positive emotions, have
presumably learned to do so because they lack more adaptive strategies to self-regulate
emotions. In contrast, individuals who are motivated to drink to socialize or conform to
peer pressure are engaging in normative behavior and may not present the same risk for

chronic alcohol dependency. The work of Cooper and her colleagues have made an

33

important contribution to the literature because it is an interactive model that not only
predicts alcohol consumption and alcohol related problems, but also predicts the
underlying motivations to drink alcohol.

In one of the first studies to investigate the Stress-Vulnerability Model of alcohol
involvement, Cooper et al. (1992) sampled 1316 adult drinkers. They hypothesized that
alcohol consumption, alcohol related problems, and the motive to drink to cope with
negative emotions would be predicted by stress, gender and the interaction of stress and
gender. In particular, they hypothesized that men who were high in stress would be most
likely to consume alcohol, have alcohol related problems, and acknowledge the motive to
drink to cope. Results of hierarchical multiple regressions found that gender, stress, and
the interaction of stress and gender were indeed signiﬁcantly predictive of alcohol
consumption and alcohol related problems. An analysis of the interactions found that
men under high stress were most likely to consume alcohol and experience alcohol
related problems. In contrast, differences in the level of stress were not predictive of
alcohol consumption or alcohol related problems for women. With regard to drinking to
cope, results indicated that male gender and high stress were signiﬁcant predictors,
however, no signiﬁcant interaction of these terms was found. The authors concluded that
the results of this study found partial support for SVM. Based upon these results, Cooper
et al. (1992) concluded that the SRD model of alcohol involvement is "overly broad and
that individual characteristics must be considered in order to account for stress-related
effects on alcohol use" (p. 148). That is, the results of this study suggest that some
individual characteristics, such as gender, may indeed make some individuals more

vulnerable to using alcohol when they are stressed.

34

Another study by Laurent, Catanzaro, and Callan (1995) examined the SVM
using an adolescent sample of high school students ( n 2184; average age = 17 years
old). Based upon the SVM described by Cooper et al. (1992), they hypothesized that
alcohol consumption, alcohol related problems, and the motive of drinking to cope with
negative emotions would be predicted by stress, gender, and the interaction of stress and
gender. They predicted that men under stress would be particularly vulnerable to alcohol
consumption, alcohol related problems, and drinking alcohol to cope. Interestingly, the
results of this study differed from the ﬁndings of Cooper et al. (1992). Hierarchical
multiple regressions found that gender was not predictive of alcohol consumption,
alcohol related problems, or drinking alcohol to cope. Stress was predictive of drinking
to cope and alcohol related problems but was not predictive of alcohol consumption.
However, the interaction of stress by gender was predictive of alcohol consumption,
accounting for 5% of the variance. An analysis of this interaction suggested that under
high stress, adolescent girls reported more alcohol consumption than adolescent boys.
Surprisingly, the interaction of stress by gender was not predictive of either drinking to
cope or alcohol related problems. Similar to Cooper et a1. (1992), these researchers
concluded that gender is one individual difference variable that moderates the
relationship between stress and alcohol consumption, thus providing evidence for the
SVM of alcohol consumption. However, these researchers found it difﬁcult to interpret
the unexpected ﬁnding that under high stress, it is adolescent girls-not adolescent boys-
who consume more alcohol. This is contrary to what would be expected based upon the
SVM (Cooper, Russell, & Frone, 1990; Cooper et al., 1992; Cooper et al., 1997). These

results are provocative because they suggest the nature of the moderating effects of

35

gender on alcohol consumption may by inﬂuenced by age. The researchers did conclude
that "in the future, it may be useful to examine additional variables, such as drinking
motives beyond drinking to cope" (p. 650) and recommended that future research on the
SVM use the Drinking Motives Questionnaire (Cooper, 1994) to predict additional
motives for alcohol involvement.

In summary, C00per and colleagues (Cooper, Russell, & Frone, 1990; Cooper et
al., 1992; Cooper et al., 1997) have investigated an interactive model in which speciﬁc
individual differences make individuals more vulnerable to abusing alcohol when under
stress. Cooper and colleagues have found evidence for a Stress-Vulnerability Model
(SVM) in which adult men who experience high stress are more likely to consume
alcohol and experience more alcohol related problems. However, this model is limited
because these results may not be consistent across age groups (e.g., Laurent et al., 1995).
In addition, the SVM has yet to consider additional individual differences, such as
personality, that may make individuals more vulnerable to alcohol consumption when
under stress. In particular, this model has yet to consider how perfectionism may
moderate this relationship, despite empirical evidence that directly links self-oriented
perfectionism and socially prescribed perfectionism to alcohol involvement (Hewitt &
Flett, 1991). In addition, the DSM indicates that self-oriented and socially prescribed
perfectionism may interact with gender to make individuals more vulnerable to stress,
and thus be associated with other negative psychological outcomes such as depression,
anxiety hopelessness, and negative psychological symptoms (Chang & Rand, 2000; Flett
et al., 1995; Hewitt et al., 1996; Hewitt & Flett, 1993; Joiner & Schmidt, 1995). Thus, it

seems plausible that perfectionism may be an additional individual difference variable

36

that may moderate the relationship between stress and alcohol consumption and alcohol

related problems.

General Summary

Despite theoretical and intuitive appeals to link high stress with alcohol
involvement (i.e., SRD), empirical research has failed to find a consistent link between
these variables. The equivocal results might be due in part to the differences in the
operationalization of stress, however, these results might also be due important
moderating variables that inﬂuence the effect stress has in the prediction of alcohol
involvement. In fact, researchers who have investigated the Stress- Vulnerability Model of
alcohol involvement have identiﬁed gender as a signiﬁcant individual difference variable
that moderates the relationship between stress and alcohol involvement. A general trend
of the research is that adult men, more so than adult women, tend to report alcohol
consumption, alcohol related problems, and the motivation to drink alcohol to cope with
negative emotions when under stress (Cooper et al., 1992; Greeley & Oei, 1999).
However, research also suggests that adolescent girls, more so than adolescent boys, tend
to consume more alcohol when under stress. However, to date no study has examined if
personality characteristics may also be an important individual difference variables that
moderate the relationship between stress and these alcohol related outcomes. This is
surprising, given research ﬁndings that indicate both perfectionism (Hewitt & Flett,
1991) and gender (Lemele & Mishkind, 1989; McCreary, Newcomb, & Sadava, 1999)
predict alcohol involvement. Furthermore, according to the DSM, perfectionism has also

been found to be an individual difference variable that makes individuals more vulnerable

37

to a multitude of negative psychological outcomes such as depression (Flett, Hewitt,
Blankstein, & Mosher, 1995; Hewitt, Flett, & Ediger, 1996; Joiner & Schmidt, 1995),
anxiety (Joiner & Schmidt, 1995), hopelessness (Chang & Rand, 2000), and general
negative psychological symptoms (Chang & Rand, 2000).

To date, no study has investigated if perfectionism interacts with stress and
gender to predict alcohol consumption, alcohol related problems, and motivation to drink
alcohol to cope. In addition, no study has examined these variables as they relate to
college student alcohol involvement. Thus, an integration of these models may yield a
useful model in predicting alcohol consumption, alcohol related problems, and the
motivation to drink alcohol to cope.

Therefore, the purpose of this study is to investigate if perfectionism interacts
with stress to predict alcohol consumption, alcohol related problems, and the motivation
to drink alcohol to cope. Because gender has been found to be a signiﬁcant moderating
variable in both the DSM and the Stress-Vulnerability Model (SVM) of alcohol

involvement, gender as a possible second moderating variable will also be explored.

Hypotheses
Based upon theoretical models and previous research, the following hypotheses
were derived for this study:
1. Among college students, male gender will increase the likelihood of

consuming alcohol, having alcohol related problems, and using alcohol to

cope.

38

2. After statistically controlling for gender, perfectionism will increase the
likelihood of consuming alcohol, having alcohol related problems, and using
alcohol to cope among college students. More speciﬁcally, socially
prescribed perfectionism, self-oriented perfectionism, and self-discrepant
perfectionism will be associated with an increased likelihood of consuming
alcohol, having alcohol related problems, and using alcohol to cope.

3. After statistically controlling for gender and perfectionism, perceived stress
will be associated with an increased likelihood of consuming alcohol, having
alcohol related problems, and using alcohol to cope.

4. Perfectionism will moderate the relationship between stress and alcohol
involvement among college students. That is, socially prescribed
perfectionism, self-oriented perfectionism, and self-discrepant perfectionism
will interact with perceived stress to increase the likelihood of consuming
alcohol, having alcohol related problems, and using alcohol to cope. More
speciﬁcally, self-reports of high perfectionism (i.e., socially prescribed
perfectionism, self-oriented perfectionism, and self-discrepant perfectionism)
and high perceived stress will be associated with an increased likelihood of
consuming alcohol, having alcohol related problems, and using alcohol to
cope. In contrast, self-reports of low perfectionism and high perceived stress
will not differ from self-reports of low perfectionism and low perceived stress
in terms of the increased likelihood of consuming alcohol, having alcohol

related problems, and using alcohol to cope.

39

5.

Gender will interact with perceived stress to increase the likelihood of
consuming alcohol, having alcohol related problems, and using alcohol to
cope. More speciﬁcally, men with high-perceived stress will be most likely to
consume alcohol, have alcohol related problems, and drink alcohol as a
motive to cope. In contrast, women with hi gh-perceived stress will not differ
from women with low perceived stress in terms of the likelihood of
consuming alcohol, having alcohol related problems, and drinking alcohol as a
motive to cope.

Both gender and perfectionism will moderate the relationship between stress
and alcohol involvement among college students. That is, male gender and
perfectionism (i.e., socially prescribed perfectionism, self-oriented
perfectionism, and self-discrepant perfectionism) will interact with perceived
stress to increase the likelihood of consuming alcohol, having alcohol related
problems, and using alcohol to cope. More speciﬁcally, I hypothesize that
men with high perfectionism and high-perceived stress would be the most
likely to consume alcohol, have alcohol related problems, and use alcohol to
cope. In contrast, among women who are high in perfectionism, perceived
stress level would not be associated with consuming alcohol, having alcohol

related problems, and using alcohol to cope.

4O

Clinical Implications

This study has significant implications for the treatment of alcohol consumption
among college students. Currently, one of the predominant approaches to treating
alcohol involvement is to train clients to better manage their stress through healthy
behaviors such as cognitive restructuring, coping skills, and problem solving skills
(Marlatt, 1996; Marlatt & Gordon, 1985). In fact, Marlatt and colleagues have developed
a brief cognitive-behavioral harm-reduction program for college students called the Brief
Alcohol Screening and Intervention for College Students (BASIC; Dimeff, Baer,
Kivlahan, & Marlatt, 1999). This program includes harm-reduction principles (e. g.,
model of addiction, nutrition/exercise information), motivational enhancement, and
cognitive-behavioral skills training (e.g., relaxation training, setting drinking limits,
assertiveness training, relapse-prevention strategies). This treatment was recently shown
to be more effective for college students than other treatment approaches, such as a
psychoeducational intervention and an assessment-only control group (Murphy et al.,
2001). Other clinicians have recommended that alcohol involvement can be decreased
with additional approaches to stress reduction, such as autogenic training, biofeedback,
and meditation (Yost & Mines, 1985). However, research related to the effectiveness of
these stress-reducing interventions has yielded equivocal results, suggesting possible
moderating variables (Baer, Marlatt, Kivlahan, Fromme, Larimer, & Williams, 1992;
Hawkins, Catalano & Wells, 1986; Hitchcox, 1979; Kivlahan, Marlatt, Fromme, Coppel,
& Williams, 1990; Krummel, 1977; Myers & Brown, 1990; Richter, Brown, & Mott,

1991).

41

If perfectionism is found to predict increased alcohol involvement, it would
suggest that long-term therapy focused on changing underlying personality may be
appropriate. For example, Blatt and Zuroff (2002) summarize their empirical research on
the treatment of perfectionism and conclude that long-term, intensive,
psychodynamically-oriented psychotherapy may be necessary in the treatment of
perfectionism because it involves changing underlying personality structure.

Furthermore, if perfectionism is found to be a variable that exacerbates the effects
of stress on alcohol involvement, clinicians may want to use treatment strategies that
address issues of perfectionism in the context of stress. For example, existing
intervention programs that address issues of both perfectionism and stress in the
transition to college (e.g., Moore & Barrow, 1986) may be adapted to prevent alcohol
misuse and alcohol related problems. These treatment approaches suggest that reducing
alcohol involvement for perfectionists may not be straightforward and may involve long-
term therapy to address longstanding characterological traits.

It is worth noting that literature indicates that perfectionism may potentially
interfere with the psychological treatment of clinical alcohol abuse. Literature provided
by Alcoholics Anonymous (i.e., the Twelve-Step treatment for alcohol abuse and
dependency) suggests that perfectionism impedes several stages of the alcohol recovery
process. Most notably, Alcoholic Anonymous proponents such as Brian (1985), argue
that perfectionists will have difﬁculty with issues of humility (“accepting limitations as
imperfect beings”) and surrender (“admit we are powerless to control addictions”).

This study will hopefully help clinicians gain a better understand of how the

various aspects of perfectionism may moderate perceived stress, and thereby

42

differentially inﬂuence alcohol involvement, alcohol related problems and motives for
using alcohol. A better understanding of how this complex relationship may differ

between men and women would also be of interest to clinicians.

43

CHAPTER III: METHOD

This chapter outlines the methodology utilized in the current study. First, it
presents a description of the procedures utilized to recruit participants, obtain informed
consent, and collect data. The chapter then provides details of the demographic
characteristics of the sample that was obtained. Then the chapter describes the
demographic questionnaire and the six self-reported measures that were used to assess the
variables of interest. Psychometric information regarding the reliability and validity of
the instruments is provided.

Procedures

This study was conducted at a large, public university in the southeast that has a
national reputation for signiﬁcant alcohol involvement among its students. For example,
this university was ranked as the ﬁfth best “party school” in the recent survey of US
colleges and universities conducted by the Princeton Review (Franek, 2002). This
ranking was based upon a 70-item questionnaire that was randomly distributed to
100,158 students at 345 colleges. The criteria for a “party school” emphasized alcohol
use and other indicators of alcohol involvement on campus. Based on this reputation, it
is reasonable to suspect that students may be attracted to this university, perhaps in part,
because of their interest in drinking.

To order to obtain a diverse sample of undergraduates that are representative of
the population of college students, prospective participants were selected from a random
list of 1000 undergraduate college students generated by the Academic Affairs Ofﬁce of
the university. Based upon demographic characteristics provided on the list, participants

were selected for inclusion in the study if they met all of the following criteria: (1) were

enrolled full-time (12 credits or more), (2) were freshmen or sophomores, (3) were 18 or
19 years old, and (4) lived in a residence hall (i.e., dormitory room). The rationale for
limiting the sample to traditionally-aged freshmen and sophomores was based on
research that indicates that the ﬁrst two years of college among traditionally-aged
colleges students is associated with the most alcohol consumption and the highest
frequency of alcohol related problems (Klein, 1994; Wechsler et al., 1994).

A total of 391 individuals on the random list of 1000 met the speciﬁed criteria.
These individuals were contacted via phone during weekday evenings. A total of ﬁve
attempted phone calls were made before a potential participant was dropped. Potential
recruits were asked if they would like to participate in a study with the purpose of
“learning more about the characteristics of college students that contribute to college
drinking habits." The recruitment script, informed consent, and research questionnaire
packet were approved by the research institutional review boards at the University of
Florida (#2002-702) and at Michigan State University (#02-228). A total of 200
participants (men = 106, women = 94) were offered (and subsequently paid) $10 for their
participation in the study. The researcher arranged for participants to complete research
instruments in a medium-sized classroom at a time that did not conﬂict with scheduled
courses. Participants completed their questionnaire packets during the months of October
and November as well as the ﬁrst week of December, 2002. These dates were selected in
order to avoid measuring drinking behavior associated with the adjustment and transition
to the start and ﬁnish of the academic semester.

After obtaining informed consent from the participants and after providing the

participants with a copy of the informed consent, participants were asked to complete a

45

packet of questionnaires that contained the instruments described below. Three versions
of the packet of questionnaires were generated, in which the order of the questionnaires
was varied. (A total of 66 participants completed Version 1, 77 completed Version 2, and
65 completed Version 3). In addition, participants were asked to complete a questionnaire
regarding their background and demographic characteristics as well as some background
and demographic characteristics of their parents (e. g., marital status, education,
employment type). Socioeconomic status was determined based upon parent education
and occupational prestige (Stevens & Hoisington, 1987). To ensure anonymity, the
questionnaire packets did not request any identifying information (e.g., name, social
security number, student number, etc.). All questionnaire packets were coded so that the
information obtained from the participants was not be able to be matched with the person

who had completed it.

Participants

Table 1 and Table 2 present the demographic characteristics of the sample of 198
participants. By design, the age of participants in the sample ranged from 18 to 19 years
(M = 18.38, SD = .49). About 52% (n = 104) of the sample was male and about 48% (n =
94) of the sample was female. Approximately 58% of the sample was White/European
American, 15% was Black/African American, 14% was Hispanic/Latino, 6% was
Asian/Asian-American, 2% was Multiracial, 1.5% was Native-American, and 0.5% was
Paciﬁc Islander. Demographic characteristics of ﬁrst and second year college students at
the University of Florida, who were enrolled as full-time students in the Fall of 2002,

were calculated based upon data provided by the university (http://www.ir.uﬂ.edu/

46

factbook/ipedsenr.pdf). Demographic characteristics of the population indicated that
approximately 70% of the population was White/European American, 10% was
Black/African American, 12% was Hispanic/Latino, 7% was Asian/Asian-
American/Paciﬁc Islander, 0.5% was Native-American, and 1% was of “unknown
race/ethnicity.” In addition, 44% of the population were men and 56% were women. In
general, demographic characteristics of the sample approximated the demographic
characteristics of the population of interest, although demographic characteristics of the
obtained sample seemed to include more racial/ethnic minorities and men. These
differences between the demographic characteristic of the obtained sample and the
population may be accounted for by the selection criteria of the study, which included
only 18 or 19 years-old students who lived in residential halls.

College majors were varied but the most frequent reported majors were Business
Administration (21%), Engineering (19%), Journalism and Communications (12%),
Liberal Arts and Sciences (8%), and Natural Resources and Environment (7%).
Approximately 14% were undecided about their major. On average, students had
completed one semester of college. Average reported GPA was 3.41 (SD = .52), though
GPA was only available for 55% of participants who had completed at least one semester
of college. Most participants’ parents were married and living together (68%) although
17% reported that their parents had divorced and remarried and about 7% indicated that
their parents divorced and remained unmarried. As can be seen in Table 2, participants’
mothers and fathers were well educated, responding that 57% of mothers and 66% of
fathers had a college education or higher. Approximately 66% of mothers and 83% of

fathers worked full-time. Parent occupational prestige score is an index of

47

socioeconomic status. Higher prestige scores reﬂect higher socioeconomic status. In this
study the mean prestige score for participants’ fathers was 51.72. This degree of prestige
would be consistent with occupations such as social workers, funeral directors, and
electricians. The mean prestige score for participants’ mothers was 43.70. This degree of
prestige would be consistent with occupations such as nursing assistants, bank tellers, and

insurance agents.

48

Table 1.
Demographic Characteristics

@de Point Average
Mean
SD
Range
Number of Completed Semesters
Mean
SD
Range
Gender 1%)
Men
Women

Age 1%)
18 years—old

19 years-old

Race 1%)
Asian/Asian American

Black/African-American

Hispanic, Latino, Mexican-American
Paciﬁc Islander

Native American/American Indian
White, European-American
Multicultural/Multiracial

Other

49

3.41
0.52
1.0 - 4.0

1.04
1.23
0—4

52.5
47.5

62.1
37.9

6.1
15.2
13.6

0.5

1.5
58.1

2.5

2.5

College Maior (%)

Agriculture & Life Sciences 3.5
Liberal Arts & Sciences 8.1
Business Administration 21.2
Journalism and Communications 11.6
Education 3
Engineering 19.2
Health and Human Performance 6.6
Health Professions 0.5
Design, Construction, & Planning 2
Natural Resources and Environment 7.1
Pharmacy 0.5
Nursing 2.5
Undecided (No preference) 14.1

Parents' Marital Status 1%)

Married and living together 67.5
Separated 3.6
Divorced and neither parent remarried 7.1
Divorced and one or both parent remarried 16.8
Widowed, or one parent deceased 2.5
Single-parent (never been married) 2.5

50

Table 2.
Demographic Characteristics of Parents

Education of Parent 1%)
Less than high school

High school degree/GED

Post high school/Technical school
Associate's degree/some college

College Degree

Some graduate or post-bachelor's training
Graduate or post-bachelor's training

Current Employment Status of Parent (%)
Works full-time
Works part-time
Does not work outside of the home
Laid off/unemployed
Disabled
Retired
Other

Prestige Score of Parent's OccupaLtion
Mean
SD

Range

Mother

5.6
10.2
8.1
19.3
32.5
5.6
18.8

65.5
12.2
12.2
2.5

4.6

43.70
15.25
16.08-81.09

51

Father

4.6
8.6
7.1
13.7
26.9
5.1
34.0

82.6
0.5
3.1
6.2
6.7
0.5
0.5

51.72
15.30
17.5-81.09

Instruments

Alcohol Consumption. The Quantity-Frequency (QF) instrument was used to
assess alcohol consumption in this study. The most recent version of this instrument was
used in the Harvard School of Public Health College Alcohol Study (Wechsler et al.,
1994; Wecshler, 1998; Wecshler & Kuo, 2000). This study was a national survey of
college alcohol consumption on 140 colleges campuses and included 17, 592 college
students. The instrument standardizes the criteria for a "drink" (i.e., 12-oz can of beer, 4-
oz glass of wine, 12-oz can/bottle of wine cooler, or 1.25—oz shot of liquor). The QF
instrument is composed of two items. The ﬁrst item assesses how many occasions the
respondent drank in the last 30 days. The six possible responses include a range: (1) O,
(2) 1-2, (3) 3-5, (4) 6-9, (5) 10-19, (6) 20-39, and (7) 40 or more. For the purpose of
analysis, the midpoint of each of the response categories is used to deﬁne how often the
students drank in the last 30 days. For example, the midpoint for response (2) would be
1.5 and the midpoint for response (3) would be 4, etc. In order to reduce skewness
caused by extreme scores, “40 or more” drinks is deﬁned as 40 drinks. The second item
on the QF assesses the volume of alcohol consumed: “In the past 30 days, on those
occasions when you drank alcohol, how many drinks did you usually have?” The
response choices include 0,1, 2, 3, 4, 5, 6, 7, 8, 9+. Again, in order to reduce skewness
caused by extreme scores, “9+” drinks is deﬁned as 9 drinks. The total number of drinks
consumed during the 30 days is calculated by multiplying the number of occasions of
drinking during the 30 days by the number of drinks per occasion during those 30 days.

Thus, the total number of drinks in the past 30 days could range from 0 to 360.

52

 

 

This quantity x frequency method of alcohol consumption was first introduced by
Jessor, Graves, Hanson, and Jessor (1968) and has been widely used in the assessment of
adult alcohol consumption (Cooper, Russell, Skinner, Frone, & Mudar, 1992; Wilsnack,
Klassen, & Wilsnack, 1984) as well as college alcohol consumption (Greenﬁeld, 1986;
McCreary & Sadava, 1998; Wechsler et al., 1994; Wechsler, 1998; Wecshler & Kuo,
2000).

In a review of the literature, Sobell, Sobell, Klajner, Pavan, and Basian (1986)
concluded that measures of self-reported alcohol consumption are both reliable and valid.
In particular, the QF instrument demonstrates adequate psychometric properties. With
regard to reliability, Wood et. a1 (2001) obtained an or coefﬁcient of .89 using this
instrument by combining the two-item questionnaire with similar questions that assess
the quantity and frequency of alcohol use in the past year as well as quantity and
frequency of heavy alcohol use in the past week. In the current study, only the quantity x
frequency product term was used. Cronbach’s coefficient or is based on a sum of items
whereas the QF is based on a product of two items. Therefore Cronbach’s coefﬁcient or is
not an appropriate index of internal consistency for this measure. With regard to validity,
Stacy, Widman, Hays, and DiMatteo (1985) found adequate evidence for construct
validity using a mulitrait-multimethod matrix (Campbell & Fiske, 1959). Convergent
validity was demonstrated by a strong positive relationship (r = .85) between self-ratings
of alcohol involvement (i.e., degree to which an individual rates self as alcohol user) and
self-reports of alcohol involvement (i.e., amount of alcohol an individual reports to
consume). These researchers compared self-reports of marijuana and cigarette

consumption to self-reports of alcohol consumption, and found only a modest

53

 

relationship between self—reported use of these substances. Thus, these researchers
argued that this demonstrated adequate evidence for discriminant validity of the QF
instrument. Concurrent validity is reported by Room (1991) who found a positive
association between the responses on the QF instrument and long-term health effects of
alcohol (e. g., cirrhosis of the liver) and the sale of alcohol. Concurrent validity is also
supported in a study by Williams, Aitken, and Malin (1985) who found that the QF
instrument was signiﬁcantly associated with alcohol use in daily diary entries (r = .74). In
addition, the QF instrument was signiﬁcantly predictive of subsequent self-reported
alcohol involvement in daily diary entries (r = .73), thus indicating the predictive validity
of the instrument. More recently, Nystrom, Perasalo, and Salaspuro (1993) found
evidence for concurrent validity. These researchers found a signiﬁcant relationship
between the QF instrument and alcohol intoxication frequency for both men (r = .73) and
women (r = .82). N ystrom, Perasalo, and Salaspuro (1993) also found a signiﬁcant
relationship between the QF instrument and the degree of alcohol related problems for
both men (r = .51) and women (r = .62).

Alcohol Related Problems. The Young Adult Alcohol Problems Screening Test
(YAAPST) is a 27-item questionnaire that assesses the negative consequences of alcohol
involvement among college students (Hurlbut & Sher, 1992). Participants respond to the
items based upon a 9-point scale that indicates the frequency of occurrence of the
problems (0: No, never; 9 = 40 or more times per year). The YAAPST is speciﬁcally
designed for use with college students and includes negative consequences that are
typically associated with alcohol involvement among the general population (e. g.,

headaches, blackouts, intoxicated driving), as well as negative consequences that are

54

typically associated with alcohol involvement among college students (e.g., missing
classes, low grades on exams, regretful sexual situations). This instrument allows for
various scoring options in order to assess lifetime alcohol problems, recent alcohol
problems (in the last year), or past-year severity of alcohol problems. For the past-year
severity of alcohol problems, items are weighted according to the severity of the problem.
For this study, the past-year severity of alcohol problems scoring criteria will be used.

The psychometric properties of the instrument were based upon a sample of 482
undergraduate students that included both abusive and non-abusive drinkers (Hurlbut &
Sher, 1992). A factor analysis of the 27-items yielded a one-factor solution, suggesting
the YAAPST is a one-dimensional instrument. Internal consistency has also been
demonstrated in this study, yielding a coefﬁcient a of .84 for past-year severity scores.
Test-retest reliability has also been demonstrated. The intraclass correlation coefﬁcients
during a nine-month period was .78 for the past-year severity scores.

Concurrent validity for the YAAPST has also been demonstrated (Hurlbut &
Sher, 1992). The scores on the YAAPST were compared with other measures of alcohol
involvement and alcohol related problems, including the Short Form of the Michigan
Alcohol Screening Test, the Quantity-Frequency measure of alcohol consumption in the
past year, the Quantity—Frequency measure of alcohol consumption in the past month, and
the heavy drinking composite. Correlations with these instruments were .43, .46, .57., and
.65, respectively. Construct validity was assessed by comparing the YAAPST to variables
that are theorized to be related to alcohol problems. The YAAPST was found to be
signiﬁcantly correlated with alcohol expectancies, including tension reduction (r = .47),

social lubrication (r = .50), enhanced activities (r = .54), and enhanced performance (r =

55

.37). The YAAPST was also found to be signiﬁcantly correlated with motives for
drinking, including affect regulation (r = .57) and social drinking (r = .39).

The YAAPST was also compared to diagnoses of alcohol abuse obtained from a
semi-structured clinical interviews. Using these diagnoses as the criterion variable, a cut
off score of 35 on the past-year severity index yields a sensitivity of approximately 90%
and a speciﬁcity of approximately 60%. The authors of the instrument conclude that this
instrument has demonstrated greater reliability and validity than other measures of
alcohol related problems among college students and have recommended the instrument
be used as both a research and clinical instrument (Hurlbut & Sher, 1992).

Motives for Drinking. The motivation to drink alcohol was measured by the
Drinking Motives Questionnaire (DMQ; Cooper, 1994). This instrument is a 20-item
instrument designed to measure four different motivations to drink alcohol. The items are
rated on a 6-point Likert-Type scale that ranges from 1 (never) to 6 (almost always). The
four motivational subscales include social motives ("How often do you drink because it
helps you enjoy a party?"), coping motives ("How often do you drink because it helps
you when you feel depressed or nervous?"), enhancement motives ("How often do you
drink because you like the feeling?"), and conformity motives ("How often would you
say you drink to ﬁt in with a group you like?").

The DMQ was studied in an initial sample of 1243 older adolescents, whose
average age was 17.3 years and who had all reported that they had consumed alcohol. In
fact, 95% reported that they drank alcohol in the past 6 months. A factor analysis of the
20-items yielded a four-factor solution, suggesting the DMQ is a multidimensional

instrument with four relatively independent subscales. This four-factor solution did not

56

vary across gender, race, or age. Internal consistency has also been demonstrated in this
study, yielding a coefﬁcient on of .85, .84, .88, and .85, for social motives, coping
motives, enhancing motives, and conformity motives, respectively (Cooper, 1994).
Correlations among these subscales ranged from r = .16 (Enhancement and Conformity)
to r = .68 (Enhancement and Social). Cross-time stability was been reported by Copper
(1994). They compared data from their late adolescent sample to data obtained from an
adult sample (Cooper etal., 1992) and reported "remarkably similar patterns" in the
motives to drink alcohol, suggesting that motives to drink are relatively stable across
time. However, these measures were not completed by the same individuals across this
time span and to date, no test-retest reliability has been reported in the literature.

Concurrent validity for the DMQ has also been demonstrated by comparing it
with other with other measures of alcohol involvement and alcohol related problems,
including a quantity of alcohol consumed in the past six months, a frequency of alcohol
consumed in the six months, a heavy drinking composite index, and a drinking problems
index. Hierarchical multiple regression analyses indicated that all four subscales were
predictive of the quantity of alcohol consumed in the past six months and the frequency
of alcohol consumed in the six month. Coping motives, enhancement motives, and
conformity motives were also predictive of heavy drinking and drinking problems,
however, social motives were not.

Concurrent validity for the DMQ is also established by ﬁndings that suggest that
each motive to drink alcohol was associated with a distinct drinking context. That is
social motives were negatively associated with drinking at home, at a friends house, and

with family and was positively associated with drinking at parties. Coping motives was

57

 

negatively associated with drinking at parties and with family and was positively
associated with drinking at home. Enhancement motives were negatively associated with
drinking at parties and positively associated with drinking at a friend's house, drinking at
bars, and drinking with same-sex friends. Conformity motives were negatively
associated with drinking at home and at bars, and positively associated with drinking at
parties (Cooper, 1994).

Cooper and colleagues assert that the DMQ is a valid and reliable instrument that
yields four empirically based distinct motivations to drink alcohol. These four patterns
represent psychologically distinct behaviors deﬁned by the different underlying functions
that they serve (Cooper, 1994; C00per et al., 1995). Cooper and colleagues add that the
instrument can be utilized in research and clinical assessment.

M. Stress was measured using the Perceived Stress Scale (PSS; Cohen,
Kamarck, & Mermelstein, 1983). This instrument is a l4-item measure of stress, as
perceived by the respondent (e. g., "In the last month, how often have you felt nervous
and 'stressed'?"). The items are rated on a 5-point Likert-Type scale from 0 (never) to 4
(very often). Higher scores indicate higher perceived stress in the last month. Unlike
other measures of stress, this measure is not dependent upon endorsement of speciﬁc
stressful daily events but rather purports to measure the overall perceived degree of stress
(Cohen et al., 1983).

The instrument has demonstrated good reliability. Internal consistency estimates
of the instrument using two independent college samples were or = .84 and or =.85 (Cohen
et al., 1983). The test-retest correlation coefficient for two days was r = .85 whereas the

test-retest correlation coefﬁcient for six weeks was .55 (Cohen et al., 1983). This is

58

theoretically consistent with the construct of stress, because stress is anticipated to stay
relatively stable for short periods of time, but is not necessarily expected to remain stable
for long periods of time because it is inﬂuenced by changes in daily hassles, major life
events, and ability to cope with stress. Construct validity has also been demonstrated
using this instrument. The PSS was found to be positively and signiﬁcantly related to
instruments that assess the number of life events and the perceived impact of these life
events (Cohen, 1986). The PSS has been found to be signiﬁcantly positively related to
depression, utilization of health services, and physical complaints. In addition, the PSS
has been found to independently predict social anxiety and smoking-reduction
maintenance (Cohen et al., 1983). Hewitt, Flett, and Mosher (1992) have also found this
instrument to be signiﬁcantly related to depression [r (94) = .57, p < .001] and have
found that scores on the PSS were higher for clinical samples than for non-clinical
samples (e.g., college students), thus supporting the construct validity of the instrument.
Perfectionism. Two instruments were used to measure perfectionism. The ﬁrst
instrument is the Multidimensional Perfectionism Scale (MPS; Hewitt & Flett, 1991),
which is a 45-item instrument in which respondents are asked to rate their agreement to
statements based on a Likert-type scale. Typically, the instrument ranges from 1
(Strongly Agree) to 7 (Strongly Disagree), with a lower numerical values indicate greater
agreement with the item (i.e., more perfectionistic traits). However, for the purposes of
this study, the Likert-type scale was reversed so that higher numerical values indicate
greater agreement with the item (i.e., 1: Strongly Disagree and 7 = Strongly Agree).
This scale was modiﬁed in order to be consistent with the scaling of other instruments

and to aid in statistical interpretation.

59

The MPS assesses three theoretically distinct dimensions of perfectionism. The
Self-Oriented Perfectionism (MPS-Self) scale assesses high personal expectations that
individuals hold for themselves (e. g., "It makes me uneasy to see an error in my work").
The Socially Prescribed Perfectionism (MPS-Social) scale assesses the concerns about
meeting the perfectionistic standards of other people (e. g., "My family expects me to be
perfect"). The Other-Oriented Perfectionism (MPS-Other) scales assesses the degree to
which individuals have perfectionistic standards for others (e.g., "I have high
expectations for the people who are important to me").

A principal-component factor analysis of data obtained from a college sample has
indicated a three-factor solution, comprising the MPS-Self scale (15 items), MPS-Social
scale (15 items), and the MPS-Other scale (15 items). Hewitt and Flett (1991) used a
college sample and also found high internal consistency for the three scales of the MPS
(MPS-Self or: .89; MPS-Social or: .86; MPS-Other or: .79). Research on the MPS has
also demonstrated good test-retest reliability over a three-month period (MPS-Self r =
.88; MPS-Social r = .75; MPS-Other r = .85).

The MPS has also demonstrated good concurrent validity, indicated by the scales
correlating with other measures in the expected manner. Speciﬁcally, Hewitt, Flett,
Tumbill, and Mikail (1991) found that the three scales of the MPS were signiﬁcantly
positively correlated with other measures of perfectionism, such as Burns' (1983)
Perfectionism Scale and Frost, Marten, Lahart; and Rosenblate's (1990) Multidimensional
Perfectionism Scale. Hewitt, Flett, Tumbill, and Mikail (1991) also found that the MPS
was not signiﬁcantly related to measures of social desirability (Marlowe-Crowne Social

Desirability Scale; Crowne & Marlowe, 1960).

60

The second instrument used to measure perfectionism was the Almost Perfect
Scale-Revised (APS-R; Slaney, Rice, & Ashby, 2002; Slaney, Rice, Mobley, Trippi, &
Ashby, 2001). This instrument consists of 23 items that are responded to based upon a
Likert-type scale that ranges from 1: Strongly Disagree to 7: Strongly Agree. A
conﬁrmatory factor analysis by Slaney et al. (2001) has yielded a three-factor solution for
the items on the APS-R. This three-factor solution was later cross-validated on an
independent sample. The three scales that comprise the APS-R are Standards (APS-R
Standards; 7 items that measure personal standards; "I have high expectations for
myself"), Discrepancy (APS-R Discrepancy; 12 items that measure the discrepancy
between performance and standards; "My performance rarely measures up to my
standards"), and Order (APS- R Order; 4 items that measure organization and need for
order; "I am an orderly person"). Factor analyses by Slaney et al. (2001) have given
further evidence for the convergent and discriminant validity of these scales. They found
that the correlations between the scales of the APS-R were modest or insigniﬁcant
(Standards and Order, r = .42; Standards and Discrepancy, r = -.12, Order and
Discrepancy, r = -.03). In addition, the APS-R was correlated with the three scales of the
MPS. Results indicated a modest to non-signiﬁcant correlation between the scales. The
highest correlations were between Discrepancy and Socially Prescribed Perfectionism (r
= .50) and between Standards and Self-Oriented Perfectionism (r = .55). In addition, the
scales of the APS-R and the MPS were found to relate in different ways to depression,
self-esteem, and Grade Point Average. This indicates that the APS-R and the MPS have

scales that are tapping unique dimensions of perfectionism. Cronbach's coefﬁcient alphas

61

for the subscales have been or: .85 (High Standards), or: .92 (Discrepancies), or: .86

(Order).

62

CHAPTER IV : RESULTS
Data Cleaning

Analyses for this study were conducted using SPSS 10.1. Prior to scoring and
analysis, the data were screened for accuracy of data entry, missing values and ﬁt
between their distribution and the underlying statistical assumptions of multiple
regression (Fox, 1991; Tabachnik & Fidel], 1996). First, data from 200 participants were
examined for data entry errors and missing values by visual inspection and by comparing
the actual range of data with the possible values. One case had missing values for the
Drinking Alcohol as a Motive to Cope and one case had missing values for the
Multidimensional Perfectionism Scale. These two cases were deleted in subsequent
analyses, yielding a total of 198 participants. The calculation of the scores of the various
scales were computed using an SPSS syntax program, according to the scoring criteria
described in the literature. The scoring of the instruments’ scales were checked for

accuracy by comparing them to two cases calculated by hand.

Descriptive Statistics and Preliminary Analyses
Scale means, standard deviations, ranges, and reliability estimates of all measures
are presented in Table 3. The scale means, standard deviations, ranges, and reliability
estimates of the measures are generally consistent with the those obtained in previous
studies. The mean scores and standard deviations of the perfectionism measures in this
study are comparable to others studies involving college students. For example, Suddath
and Slaney (2001) report Discrepancy mean scores and standard deviations of 23.20 (SD

= 14.04). They also report Self-Oriented Perfectionism mean scores and standard

63

deviations of 68.70 (SD = 14.55) as well as Socially Prescribed Perfectionism mean
scores and standard deviations of 50.10 (SD = 12.61). With regard to Perceived Stress
Scale, Cohen et al. (1983) reported a mean score of 23.18 and a standard deviation of
7.31 among a college student sample. This is also comparable to the results of this study.
With regard to the alcohol related instruments, Cooper et al. (1994) reports a sample
mean of 5.60 and a standard deviation of 1.75 among a college student sample, which is
somewhat lower than the means and standard deviations obtained in this sample. As far
as alcohol consumption is concerned, Wechsler et al. (1999) report a mean score of 21.21
drinks consumed per month. They also report the average mean score of “binge drinkers”
which was 77.57 drinks consumed per month. The results of this study found a higher
mean score of alcohol consumption (36.36) among this sample, although this mean was
not as high as the mean score of alcohol consumption for “binge drinkers” in the

Wechsler et al. (1999) study.

Table 3.
Descriptive Statistics

 

 

 

 

APS-R Discrepancy 12-84 12-79 39.49 14.66 .91
MPS- Self Oriented Perfectionism 15-105 31-105 71.92 15.54 .89
MPS- Socially Prescribed Perfectionism 15-105 24-91 54.66 13.19 .83
P88 0-56 5-49 24.18 8.77 .86
QF 0360 02655 36.36 58.23 N/A
YAAPST 0-134 0-77 12.44 16.34 .90
DMQ- Coping 5-30 5-30 8.9 5.1 .88

 

No_te_: APS-R = Almost Perfect Scale-Revised;
DMQ-Coping = Drinking Motives Questionnaire-Coping;
MPS = Multidimensional Perfectionism Scale; PSS = Perceived Stress Scale;
QF = Quantity-Frequency;YAAPST = Young Adult Alcohol Problems Screening Test

Pearson product—moment correlations among the variables are presented in Table
4. Based upon this correlational matrix, there is no indication of multicollinearity because

no correlation was greater than .80 (Tabachnik & Fidel], 1996).

Table 4.

Correlation Matrix of Variables

 

Measure 1 2 3 4 5 6
1. QF -
2. YAAPST .45“ -
3. DMQ-Coping .55M -.52** -
4. P88 —.l8* .06 .04 -
5. APS-R Discrepant -.08 —.00 .00 .61“ -
6. MPS- Socially Prescribed -.09 .02 -.02 .50“ .46** -
7. MPS- Self-Oriented -.20** -.ll -.14* .25" .30“ .59**

 

Note: N=198 *p<.05. **p<.Ol.

APS-R = Almost Perfect Scale-Revised

DMQ-Coping = Drinking Motives Questionnaire-Coping
MPS = Multidimensional Perfectionism Scale

PSS = Perceived Stress Scale

QF = Quantity-Frequency
YAAPST = Young Adult Alcohol Problems Screening Test

To check for normal distribution of the variables, frequency distributions of the
independent and dependent variables, as well as the distribution of the residual error of
the dependent variables (i.e., the difference between the predicted and obtained DV
values) derived from multiple regression analyses, were conducted. Results of the
frequency distribution of the independent variables (MPS-Social, MPS-Self, APSR-
Discrepancy, PSS) indicated that the distributions were approximately normal. However,

the frequency distribution and a scatterplot of the three dependent variables (QF,

65

YAAPST, DMQ-Coping) were significantly skewed in a positive direction, resulting in
31%, 28%, and 40% of participants in the sample yielding the lowest possible total score
on the QF, the YAAPST, and the DMQ-Coping, respectively. The dependent variables
were transformed using various techniques (e.g., square root, logarithm, inverse
transformations) yet this did not result in a normal distribution of the residual errors.
Therefore, a categorization was created such that individuals who had the lowest possible
score on the QF (QF = 0) would compose the group of non-alcohol consumers whereas
QF scores higher than 0 would compose a group of alcohol consumers. In a similar
manner, students with the lowest possible score on the YAAPST (YAAPST = 0) would
compose the group of students with no alcohol related problems whereas YAAPST
scores greater than 0 would compose the group of students with alcohol related problems.
In addition, students with the lowest possible score on the DMQ-Coping (DMQ-Coping =
5) would compose the group of students who did not drink alcohol to cope whereas
students with DMQ-Coping scores greater than 5 would compose the group of students
who did drink alcohol to cope.

Chi-square analyses were conducted to see if there were signiﬁcant differences in
the frequencies of the dependent variables (alcohol consumption, alcohol related
problems, and drinking alcohol to cope) between the two age groups (18 and 19 year
olds) in the sample. Results of a chi-square analysis of age and alcohol consumption
indicated no statistically signiﬁcant association, x2 (1, 198) = .361, p = .548. Similarly,
there was no statistically signiﬁcant association between age and alcohol related

problems, x2 (1, 198) = .074, p = .785. In addition, results of a chi-square analysis

66

between age and drinking alcohol to cope indicated no statistically significant
association, x2 (l, 198) = 3.30, p = .069.

Chi-square analyses were also conducted to test the association between the
frequencies of the dependent variables (alcohol consumption, alcohol related problems,
and drinking alcohol to cope) and the number of semesters in college. Results of a chi-
square analysis between semester in college and alcohol consumption indicated no
statistically signiﬁcant association, x2 (4, 198) = 4.35, p = .36. Similarly, results revealed
no statistically signiﬁcant association between semester in college and alcohol related

problems, x2 (4, 198) = 2.40, p = .66]. Results also indicated no signiﬁcant association

between age and drinking alcohol to cope, xz (4, 198) = .64, p = .960.

Following recommendations from J accard (2001), predictor variables were
“centered” or transformed into deviation score form by subtracting the respective group
mean score from individual scores, yielding a mean of approximately 0. These predictor
variables include the perceived stress scale from the PSS, one subscale of the APS-R
(APS-R Discrepancy), and the two subscales of the MPS (MPS-Self and MPS-Socially
Prescribed). This transformation yields an intercept that is more interpretable than
would be the case for non-centered variables. In addition, this approach maximizes
variability, reduces multicollinearity, and increases the likelihood of detecting signiﬁcant
interaction effects. However, centering does not change the value of the regression

coefﬁcients (J accard, 2001).

67

Analyses
Hierarchical Logistic Regression: Prediction of Alcohol Consumption Group
Hierarchical logistic regression analysis was used to predict alcohol consumption
group membership (QF score greater than 0). The variable entered ﬁrst in the model was
gender (1 = male; 0 = female) to test for possible differences between men and women
(Hypothesis 1). Step 1 was not statistically signiﬁcant [x2 (1,N=l98) = 2.42, p = .12]. In
the next block, the three centered perfectionism subscales (i.e., self-oriented
perfectionism, socially prescribed perfectionism, and self-discrepant perfectionism) were
entered. In addition, centered perceived stress from the PSS scale was also entered into
this block (Hypothesis 2). Step 2 was statistically signiﬁcant [x2 (4, N=198,= 8.59, p = .04].
The following block consisted of the multiplicative term of centered perceived stress by
each of the three centered subscales of perfectionism (Hypothesis 3). Step 3 was
statistically signiﬁcant [)6 (3.11:193): 4.02, p = .045]. The fourth block was composed of
the multiplicative term of centered perceived stress by gender (Hypothesis 4). Step 4 was
not statistically signiﬁcant [x2 (1, N=193) = .17, p = .68]. The ﬁnal step consisted of a three-
way interaction of perceived stress by the three scales of perfectionism by gender
(Hypothesis 5). Step 5 was statistically signiﬁcant [762 (3. N=l98) = 10.61, p = .01]. The
results of the overall logistic regression model which included all variables (GM) was
statistically signiﬁcant [712 (12, ":19,” = 26.46, p = .01]. In addition, the overall, the model
correctly classiﬁed 72.2% of the cases.
The Hosmer and Lemeshow Test (HLT) is an indication of the unexplained
variance in the model, wherein statistical signiﬁcance and large values are indicative of a

poorly ﬁtting model. In this model, the HLT was not signiﬁcant. This result indicated that

68

the overall model, which included all variables, was a reasonable fit for the data, x2 (12,
M193): 5.86, p = .66.

Hosmer and Lemeshow (1989) have proposed a measure of the pseudo-variance
explained by logistic regression models. Speciﬁcally, RLZ is the proportional reduction in
)6 when the predictor variables are included in the model compared to the model when
these variables are not included. This is derived by dividing the chi-square for the model
by the initial Log-Likelihood measure (26.46 -:- 228.71 = .116). Thus, this overall model
accounted for about a 12% reduction in explained variance when compared to a model
with no predictor variables (Menard, 1995). Nagelkerke (1991) has proposed a
comparable adjusted estimate of pseudo-variance. In this model, the Nagelkerke R2 was
.176. Therefore, this model explained about 17.6 % of the variance.

Table 5 shows the logistic regression coefﬁcients ([3), standard errors (SE), Wald
statistics (W 2), and odd ratios (OR) for each variable in the model. The Wald statistics for
centered MPS-Self in Step 2 was statistically signiﬁcant, x2 (i, N=193) = 10.58, p = .001.
The OR for the MPS-Self variable represents the multiplying factor by which the odds of
drinking are predicted to change, given a one-unit increase in self-oriented perfectionism,
when the other variables in the model equal 0 (e.g., when perceived stress is average and
gender is female). The OR for MPS-Self was .950. Subtracting one from this term and
multiplying by 100 indicates the percentage change in the odds of being classiﬁed in the
group that consumes alcohol, as this term changes by one unit. Thus, there is a reduction
in the odds of being classiﬁed in the group that consumes alcohol by 5%. Thus, self-
oriented perfectionists were less likely to be classified in the group who consumed

alcohol, after controlling for the other variables in the model (Hypothesis 2).

69

Table 5.

Hierarchical Logistic Regression to Predict Alcohol Consumption Group

 

Variable B S.E. Wald p Exp (B) 95% CI Exp (B)
Lower Upper
Step 1 Gender .410 .396 1.07 .30 1.51 .693 3.278
Step 2 MPS-Self -.051 .016 10.58 .001** 0.95 .921 0.980
MPS-Social .024 .019 1.61 .204 1.03 .987 1.06
APS-R Discrepancy .016 .016 1.03 .309 1.02 .985 1.05
PSS -.056 .037 2.34 .13 0.95 .880 1.016
Step 3 MPS-Self x PSS .006 .003 5.14 .023* 1.01 1.00 1.012
MPS-Social x PSS -.001 .003 .276 .599 0.99 .993 1.004

APS-R Discrepancy x PSS -.001 .002 .283 .595 0.99 .995 1.003
Step 4 Gender x PSS .004 .045 .006 .936 1.00 .919 1.095
Step 5 Gender x PSS x MPS-Self -.009 .004 5.90 .015* .991 .984 0.998
Gender x PSS x MPS Social .001 .004 .021 .844 .999 .992 1.007

GenderxPSSxDiscrepancy. .004 .003 1.78 .183 1.004 .998 1.009

Intercept .687 .268 6.55 .010* 1.988

 

Note: * = p < .05

**=p<.001

The Wald statistic for the interaction of PSS x MPS-Self in Step 3 (Hypothesis 3)
was statistically signiﬁcant [x2 (1, Nags, = 5.14, p = .023]. The OR for PSS x MPS-Self
reﬂects the two-way interaction contrast between self— oriented perfectionism and

perceived stress when other variables in the model equal 0 (i.e., gender is female).

70

Jaccard (2001) and Cohen, Cohen, West, and Aiken (2003) suggest a method of plotting
the predicted odds of interactions in logistic regression. In order to facilitate the
interpretation of interactions, these authors provide the following formula: logit (it) = or +
[31X + [322 + [33 XZ, where in this case logit (it) = the log odds of being categorized as an
alcohol consumer, or = constant, B] = coefﬁcient associated with variable X (i.e.,
Perceived Stress), B; = coefﬁcient associated with variable Z (i.e., Self-Oriented
Perfectionism), and B3 = coefﬁcient associated with the interaction of X2. This equation
can be re-written as: logit (it) = [[31 + B32] X + [1322 + a] or with regard to the speciﬁc
test variables, logit (n) = [[31 4» B3 (Self-Oriented Perfectionism)] [Perceived Stress] + [132
(Self-Oriented Perfectionism) + constant]. Therefore, the equation for the interaction of
Self-Oriented Perfectionism by Perceived Stress (Hypothesis 3) was: logit (7!) = [.687 +
.0006 Z] X = [0.051 Z + 0.687]. Values equal to the centered mean (Average), one
standard deviation above the centered mean (High), and one standard deviation below the
centered mean (Low) of Perceived Stress and Self-Oriented Perfectionism were
substituted into the equation above. The resulting log odds of being categorized as an
alcohol consumer was then transformed into predicted odds of being categorized as
consuming alcohol by using the exponent function. Figure 1 graphically represents the

nature of this interaction.

71

Figure 1

Figure Caption. Interaction of Self-Oriented Perfectionism and Perceived Stress in the

Predicted Odds of Consuming Alcohol.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Self-Oriented Perfectionism x
Perceived Stress
g: 18
'g 16 3
= \
g 14
O 12 \
0
'5 '5 \ +Low Stress
.: 10
3 8 8 \ —A— Avg. Stress
3 < 6 \ - I- 'ngh Stress
E 4 A. \ \
._ \
0- 0 l .
Low Average High
Self-Oriented Perfectionism

 

 

 

A graphical representation of the two-way interaction of MPS-Self and PSS
(Figure 1) suggests that among college students with low MPS-Self scores, low scores on
the PSS are associated with a 16.8 fold increase in the odds of alcohol consumption.

Among college students with low MPS-Self scores, average and high scores on the PSS

72

are associated with a 4.7 and 1.3 fold increase in the odds of alcohol consumption.
Among college students with high MPS-Self scores, average and low PSS scores are
associated with a slight decrease in the odds of alcohol consumption (.83 and .57,
respectively). This suggests that college students who are low in self-oriented
perfectionism are more likely to consume alcohol when they experience low and
moderate degrees of perceived stress compared to when they experience a high degree of
perceived stress. In contrast, college students with high MPS-Self scores are less likely
to consume alcohol than their counterparts, regardless of scores on the PSS.

As reported previously, the three-way interaction of centered PSS, MPS-Self, and
gender (Hypothesis 5) was statistically signiﬁcant in Step 5 [7c2 (1,N=198)= 5.90, p = .015].
In order to facilitate the interpretation of this three-way interaction, the data ﬁle was
“split” and the logistic regression models to predict alcohol consumption were analyzed
separately for each gender. The resulting coefﬁcients associated with PSS, MPS-Self, and
the interaction between PSS and MPS-Self were derived and were used to generate
equations for each gender. As described previously, values equal to the centered mean
(Average), one standard deviation above the centered mean (High), and one standard
deviation below the centered mean (Low) of Perceived Stress, and Self-Oriented
Perfectionism were substituted into the equation. The resulting log odds of being
categorized as consuming alcohol was then transformed into predicted odds of being
categorized as consuming alcohol by using the exponent function.

The equation derived for the interaction of Self-Oriented Perfectionism x
Perceived Stress among women was: logit (n) = [-.039 + (.005) Z] X = [(-.071) Z +

.771]. Figure 2 graphically represents the nature of this interaction.

73

Figure 2

Figure Caption. Self-Oriented Perfectionism and Perceived Stress Among Women in the

Predicted Odds of Consuming Alcohol.

 

Predicted Odds of

 

Self-Oriented Perfectionism x
Perceived Stress (Women)

 

 

 

 

 

 

 

+ Low Stress

 

—A— Avg. Stress

 

 

- I- 'High Stress

 

 

 

 

 

 

20 ‘
-18 —.
gm \\ ,-_
214 ———-——
<12 — \ +~~—«
m \

E 8
a 6 k \
5 4 \\
O 2 .-.-‘>:\“I
-i~.
0 . T

 

 

 

Low Average

High

Self-Oriented Perfectionism

 

A graphical representation of the interaction of self-oriented perfectionism and

perceived stress (Figure 2) is virtually identical to the relationships described above.

Among college students with low MPS-Self scores, low scores on the PSS are associated

with a 18.8 fold increase in the odds of alcohol consumption. Among college students

with low MPS-Self scores, average and high scores on the PSS are associated with a 7.1

and 2.6 fold increase in the odds of consuming alcohol. Among college students with

74

 

high MPS-Self scores, high, average and low PSS scores are associated with a slight
decrease in the odds of alcohol consumption (.46, .66, and .94, respectively). This
suggests that college students who are low in self-oriented perfectionism are more likely
to consume alcohol when they experience low and moderate degrees of perceived stress
compared to when they experience a high degree of perceived stress. In contrast, college
students with high MPS-Self scores are less likely to consume alcohol than their
counterparts, regardless of scores on the PSS.

The equation derived for the interaction of Self-Oriented Perfectionism x
Perceived Stress among men was: logit (n) = [-.066 + (-.003) Z] X + [(-.014) Z + 1.25].
Figure 3 graphically represents the nature of this interaction. Values equal to the
centered mean (Average), one standard deviation above the centered mean (High), and
one standard deviation below the centered mean (Low) of Perceived Stress and Self-
Oriented Perfectionism were substituted into this equation. The resulting log odds of
being categorized as consuming alcohol was then transformed into predicted odds of

being categorized as consuming alcohol by using the exponent function.

75

Figure 3.
Figure Caption. Interaction of Self-Oriented Perfectionism and Perceived Stress Among

Men in the Predicted Odds of Consuming Alcohol.

 

Self-Oriented Perfectionism x
Perceived Stress (Men)

 

i
l
i

 

 

 

/ —0— Low Stress

—A—Avg. Stress
L - I- High Stress

 

 

 

 

 

 

 

Alcohol

O-‘NOD-bO'IOJVmCD

 

 

Predicted Odds of Consuming

 

 

 

Low Average High
Self-Oriented Perfectionism

 

 

 

Among male college students with high scores on the MPS-Self and were
associated with signiﬁcant increases in the likelihood of consuming alcohol. Speciﬁcally,
among male college students with high scores on the MPS-Self, low scores on the PSS

have a 7.6 fold increase in likelihood alcohol consumption. This is compared with male

76

college students with high scores on the MPS-Self and high scores on the PSS, who were
only 1.1 times more likely to consume alcohol. In contrast, among male college students
with low scores on the MPS-Self, students with low PSS scores have a 5.2 fold increase
in the likelihood of alcohol consumption whereas students with low scores on the MPS-
Self and high scores on the PSS were 3.5 times more likely to consume alcohol. This
suggests that high self-oriented perfectionism may buffer the effects of stress on alcohol

consumed among male college students.

Hierarchical Logistic Regression: Prediction of Alcohol Related Problems Group
In order to predict group membership of college students with alcohol related

problems (YAAPST score greater than 0), the same step sequence of variables was used
as in the previous model involving alcohol consumption. The variable entered ﬁrst in the
model was gender (1 = male; 0 = female) to test for possible differences between genders
(Hypothesis 1). This step was not statistically signiﬁcant [x2 (1, N=193) = 2.41, p = .12].
In the next block, the three centered perfectionism subscales (i.e., self—oriented
perfectionism, socially prescribed perfectionism, and self-discrepant perfectionism) were
entered. In addition, centered perceived stress from the PSS scale was entered into the
same block (Hypothesis 2). Step 2 was not statistically signiﬁcant [x2 (4, N=193) = .08, p =
.784]. The following block consisted of the multiplicative term of centered perceived
stress by each of the three centered subscales of perfectionism (Hypothesis 3). Step 3
was not statistically signiﬁcant [x2 (3, N=193) = .98, p = .81]. The fourth block was
composed of the multiplicative term of centered perceived stress by gender (Hypothesis

4). Step 4 was not statistically signiﬁcant [)12 (l, N=|98) = .00, p = .99]. The ﬁnal step

77

consisted of a three-way interaction of perceived stress by the three scales of

perfectionism by gender (Hypothesis 5). Step 5 was not statistically signiﬁcant [x2 (3,
N=193) = 5.11, p = .164]. The results of the logistic regression of the full model (GM),
which included all variables, was not statistically signiﬁcant, x2 (12, N=|93) = 16.90, p =

.153. Table 6 indicates the results associated with each variable in the model.

78

Table 6.

Hierarchical Logistic Regression to Predict Alcohol Related Problems Group

 

 

95% CI.
Variables B S.E. Wald p Exp (B) Lower Upper
Step 1 Gender 0.624 0.40 2.426 0.119 1.866 0.851 4.089
Step 2 Self-Perfectionism -0.028 0.015 3.523 0.061 0.972 0.944 1.001
Social Perfectionism -0.016 0.019 0.694 0.405 0.984 0.947 1.022
Discrepant Perfectionism 0.01 0.016 0.389 0.533 0.984 0.979 1.042
Perceived Stress -0.009 0.034 0.063 0.803 0.991 0.927 1.061
Step 3 Self-Perfectionism x
Perceived Stress 0.004 0.003 1.734 0.188 1.004 0.998 1.009
Social Perfectionism x
Perceived Stress -0.002 0.003 0.333 0.564 0.998 0.993 1.004
Discrepant Perfectionism x
Perceived Stress 0.00 0.002 0.03 0.863 1.00 0.997 1.004
Step 4 Gender x
Perceived Stress 0.013 0.044 2.6535 0.105 0.994 0.93 1.104
Step 5 Gender x
Perceived Stress x
Self-Perfectionism -0.006 0.004 2.65 0.105 0.994 0.93 1.104
Gender x
Perceived Stress x
Social Perfectionism -0.001 0.004 0.05 0.824 0.999 0.992 1.006
Gender x
Perceived Stress x
Discrepant Perfectionism 0.003 0.003 1.048 0.306 1.003 0.997 1.008
Intercept 0.686 0.262 6.861 0.009 1.985
Note: * = p < .05
** = p < .01

79

Logistic Regression: Prediction of Drinking Alcohol to Cope Group

This logistic regression analysis model predicted group membership of college
students who drink alcohol to cope (DMQ score greater than 5). The order of entry of the
variables remained the same as in the previous model. The variable entered ﬁrst in the
model was gender (1 = male; 0 = female) to test for possible differences between genders
(Hypothesis 1). Step 1 was not statistically signiﬁcant [x2 (1, N=193) = 1.03, p = .31]. In the
next block, the three centered perfectionism subscales (i.e., self-oriented perfectionism,
socially prescribed perfectionism, and self-discrepant perfectionism) were entered. In
addition, centered perceived stress from the PSS scale was entered into this block
(Hypothesis 2). Step 2 was not statistically signiﬁcant [x2 (4, (0:193) = 1.10, p = .30]. The
following block consisted of the multiplicative term of centered perceived stress by each
of the three centered subscales of perfectionism (Hypothesis 3). Step 3 was not
statistically signiﬁcant [x2 (3, N=198) = 4.28, p = .23]. The fourth block was composed of
the multiplicative term of centered perceived stress by gender (Hypothesis 4). Step 4 was
not statistically signiﬁcant [7(2 (1_N=193)= 1.44, p = .23]. The ﬁnal step consisted of a
three-way interaction of perceived stress by the three scales of perfectionism by gender
(Hypothesis 5). Step 5 was not statistically signiﬁcant [x2 (1,139,): 4.46, p = .22]. The
results of the overall logistic regression model (GM), which included all variables, was
not statistically signiﬁcant [x2 (.2 , N=193) = 17.87, p = .12]. Table 7 shows the signiﬁcance

level of each variables.

80

Table 7.

Hierarchical Logistic Regression to Predict Drinking Alcohol to Cope Group

 

95% CI.
Variables E SE. Wald p Exp (B) Lower Upper
Step 1 Gender 0.498 0.372 1.787 0.181 1.645 0.793 3.415
Step 2 Self-Perfectionism -0.029 0.014 4434 0.035 0.972 0.946 .998
Social Perfectionism 0.001 0.018 0.002 0.966 1.001 0.966 1.036
Discrepant Perfectionism -0.002 0.015 0.013 0.909 0.998 0.969 1.028
Perceived Stress 0.007 0.032 0.049 0.824 1.007 0.947 1.071
Step 3 Self-Perfectionism x
Perceived Stress .000 0.002 0.026 0.873 1.00 0.966 1.005
Social Perfectionism x
Perceived Stress .000 0.003 0.001 0.981 1.00 0.995 1.005
Discrepant Perfectionism x
Perceived Stress 0.001 0.002 0.362 0.547 1.001 0.998 1.004
Step 4 Gender x
Perceived Stress 0.056 0.04 1.933 0.164 1.057 0.977 1.144
Step 5 Gender x
Perceived Stress x
Self-Perfectionism -0.003 0.003 0.985 0.321 0.997 0.99 1.003
Gender x
Perceived Stress x
Social Perfectionism -0.003 0.004 0.778 0.378 0.997 0.99 1.004
Gender x
Perceived Stress x
Discrepant Perfectionism 0.004 0.003 1.81 0.178 1.004 0.998 1.01
Intercept 0.153 0.247 0.385 0.535 1.166

 

Note: * = p < .05

**=p<.01

81

Exploratory Analyses

Several predicted hypotheses were not supported in this study. Therefore, in order to
ascertain the reasons why these hypotheses were not supported, a number of exploratory
analyses were conducted. In particular, it is possible that students pursuing different
majors could experience different degrees of stress, have different tendencies toward
perfectionism, and have different academic environments where drinking alcohol is more
or less normative. To that end, an examination of the association between college major
and the variables in this study was conducted. College students in this sample reported a
total of 12 different majors. However, ﬁve categories of majors contained fewer than 10
students, and were therefore not included in these analyses. Only the seven categories of
majors that included at least 10 or more students were examined. A series of three chi—
square analyses were conducted to determine the association between drinking
involvement across the seven majors. The association between college major and alcohol
consumption group (QF) was not statistically signiﬁcant, x2 (6, 198) = 1.89, p = .93. The
association between college major and alcohol related problems group (YAAPST) was
not statistically signiﬁcant, if (6, 198) = 5.03, p = .54. Similarly, the association
between college major and drinking alcohol to cope group (DMQ-Coping) was not
statistically signiﬁcant, x2 (6, 198) = 5.03, p = .54.

In addition, a series of univariate analyses of variance (ANOVA) were conducted to
determine if the study variables were signiﬁcantly different across college majors.
Results of the univariate ANOVAs did not reveal signiﬁcant mean differences between
the study variables. Results of an ANOVA did not reveal signiﬁcant differences in the

mean scores of perceived stress (PSS) among college students, F (6,198) = 1.79, p = .11.

82

Results of an ANOVA also did not reveal significant differences in the mean scores of
Discrepancy, F (6,198) = .85, p = .534. In addition, there were no significant differences
between college majors on self-oriented perfectionism, F (6,198) = .64, p = .70.
Similarly, no Signiﬁcant differences were found on socially-prescribed perfectionism, F
(6,198) = .66, p = .67.

As a result of these exploratory analyses, the substantive ﬁndings, or lack thereof,
were not likely attributable to systematic differences between students pursuing different
college majors. Said differently, in this sample, there do not appear to be culture-speciﬁc

norms for drinking, perfectionism, or stress as reﬂected by the students’ major course of

study.

83

CHAPTER V: DISCUSSION

Based upon previous research conducted on college students (Lemle & Mishkind,
1989; McCreary, Newcomb, & Sadava, 1999; Perkins, 1999), I hypothesized that male
gender would signiﬁcantly increase the likelihood of being categorized as being involved
with alcohol. In contrast to Hypothesis 1, male gender was not found to be associated
with a greater probability of being categorized as consuming alcohol, having alcohol
related problems, or drinking alcohol to cope among this college sample. Failing to ﬁnd
a signiﬁcant gender difference in alcohol involvement is consistent with recent research
conducted at a Midwestern and Southern university conducted by Lo (1996). Perkins
(1999), however, makes a valid point when he reminds readers that failing to ﬁnd a
signiﬁcant gender difference in the likelihood of consuming alcohol does not suggest that
the degree of intoxication is equivalent between the genders. Although the mean
consumption of alcohol among men and women may not signiﬁcantly differ, women who
consume an equal amount of alcohol as men may achieve a much higher degree of blood
alcohol concentration because of biological differences in body weight, fat-to-water
ratios, and metabolic processing. Thus, although male gender was not predictive of
alcohol involvement in this study, this does not suggest that gender would not be
predictive of intoxication levels.

Consistent with Hypothesis 2, the results of this study found that self-oriented
perfectionism was associated with a slight decrease in the likelihood of being categorized
as consuming alcohol (after statistically controlling for the effects of gender). However,
self-oriented perfectionism was not associated with a greater or lesser likelihood of being

categorized as having alcohol related problems or using alcohol to cope. In addition,

84

other measures of perfectionism, such as self-discrepant perfectionism and socially-
prescribed perfectionism, were not associated with any of the three measures of alcohol
involvement. Furthermore, the ﬁnding that self-oriented perfectionism was associated
with a 5% decrease in the likelihood of being categorized as consuming alcohol was not
in the expected direction. Based upon previous literature among adult samples (i.e.,
Hewitt & Flett, 1991; Nerviano & Gross, 1983), it was hypothesized that self-oriented
perfectionism would be associated with an increased likelihood of being categorized as
consuming alcohol. The ﬁnding that self-oriented perfectionism was not associated with
such an increase among a college sample suggests that the relationship between self-
oriented perfectionism and alcohol consumption among the college population may differ
from adult clinical samples.

The reason why self-oriented perfectionism is associated with a decrease in
alcohol consumption among college student but is associated with an increase in alcohol
consumption among adult clinical samples is not obvious. Self-oriented perfectionism is
conceptualized as having high personal standards and a strong motivation to be ﬂawless.
A recent review of the literature regarding the MPS-Self has suggested that it may
measure a relatively adaptive trait, associated with self-related personality traits rather
than the negative psychological symptoms or characteristics associated with other
measures of perfectionism (Enn & Cox, 2002). The notion that self-oriented
perfectionism is a relatively adaptive trait would be consistent with the ﬁnding that self-
oriented perfectionism was associated with a Slight decrease in the likelihood of being
categorized in the group of alcohol consumers. It is conceivable that self-oriented

perfectionists are less likely to consume alcohol because they believe that doing so would

85

not be consistent their high personal standards or may impede with their strong
motivation to perform ﬂawlessly on tasks.

The results of this study failed to ﬁnd evidence that perceived stress increases the
likelihood of being categorized as consuming alcohol, having alcohol related problems,
or using alcohol to cope, which was also predicted (Hypothesis 2). Thus, this study did
not ﬁnd evidence for the Stress-Response Dampening model proposed by Sher and
Levinson (1982). This is consistent with previous literature that did not ﬁnd an
association between stress and alcohol involvement (Corcoran & Parker, 1991; McCreary
& Sadava, 1998). However, this study did ﬁnd a statistically signiﬁcant (inverse)
correlation between perceived stress and being categorized as a consumer of alcohol (r =
-.l8). This ﬁnding is inconsistent with the Stress-Response Dampening model, but would
be consistent with an alternative to the SRD model of alcohol consumption proposed by
Krause (1991). He theorizes that stress can lead to a decrease in alcohol consumption
rather than an increase. He asserts that individuals may choose to decrease their
consumption of alcohol depending upon the nature and source of the Stress. For
example, he believes that health-related stressors may cause people to reduce their
drinking because of a fear that it might exacerbate health problems. Similarly, Krause
believes that ﬁnancial stressors may cause people to reduce their drinking because they
cannot afford to purchase alcohol. Therefore, it may be that the particular type of stress
experienced by the college students in this sample (e.g., transition to college, college
coursework, peer relationships, etc.) may be associated with a decrease in alcohol

consumption.

86

Hypothesis 3 stated that perfectionism would interact with perceived stress to
increase the likelihood of being categorized as consuming alcohol, having alcohol related
problems, and using alcohol to cope. In particular, high perfectionism and high perceived
stress was hypothesized to be associated with an increased likelihood of being
categorized as consuming alcohol, having alcohol related problems, and using alcohol to
cope. The results of this study did ﬁnd a significant interaction between perceived stress
and self-oriented perfectionism. Although contrary to Hypothesis 3, low self-oriented
perfectionists with low perceived stress had the highest predicted odds of alcohol
consumption when compared with low self-oriented perfectionists with average and high
perceived stress. In addition, college students with low self-oriented perfectionism and
low perceived stress also had a higher predicted odds of consuming alcohol when
compared to high self-oriented perfectionists with low, average, or high perceived stress.
This ﬁnding is consistent with a previous study that found that college students with low
self-discrepant perfectionism and low anxiety were most predictive of alcohol
consumption (Cohen & Rice, 2003). However, results from this study are the opposite of
predictions based upon the Diathesis-Stress Model of Perfectionism (DSM), which
proposes that perfectionism magniﬁes the effects of stress and becomes relatively
maladaptive because it is associated with self-defeating styles of cognitive appraisal,
maladaptive coping and poor problem-solving skills (Hewitt & Dyck, 1986; Hewitt &
Flett, 1993; Hewitt & Flett, 2002). In addition, the results of this study are not consistent
with several previous research studies that linked the DSM to depression and
psychological distress (Chang & Rand, 2000; Hewitt & Dyck, 1986; Hewitt & Flett,

1993; Flett, Hewitt, Blankstein, & Mosher, 1995; Joiner & Schmidt, 1995). It is

87

interesting to note that among college students with low self-oriented perfectionists, the
predicted odds of consuming alcohol decreases only as perceived stress increases. In
contrast, among high self-oriented perfectionists, the predicted odds of consuming
alcohol does not increase, regardless of the perceived stress level. This might indicate
that high self-oriented perfectionists are less likely to consume alcohol, regardless of
perceived stress level, because they believe that any degree of alcohol consumption may
interfere with their ability to be perfect or perform ﬂawlessly at tasks (i.e., tests,
classroom assignments, studying). However, the same may not be true of college
students with low self-oriented perfectionism. Such individuals may only become
concerned about task performance (or perhaps task completion) when they perceive a
high degree of stress. It is at this time that they choose to decrease their alcohol
consumption.

It was hypothesized that male gender would interact with perceived stress to
increase the likelihood of being categorized as consuming alcohol, having alcohol related
problems, or using alcohol to cope (Hypothesis 4). According to the Stress- Vulnerability
Model (SVM) proposed by Cooper et al. (1992), male college students were expected to
be more prone than female college students to extemalize their stress by consuming
alcohol, having alcohol related problems, and drinking alcohol to cope when exposed to
high stress. Unlike previous studies, this study failed to ﬁnd evidence for the moderating
effects of gender on stress in the prediction of alcohol involvement (Cooper et al., 1992;
Laurent, Catanzaro, & Callan, 1995). Cooper et al. (1992) found that among adult, male
gender moderated the relationship between stress and alcohol involvement. However,

Laurent et al. (1995) found that among adolescents, female gender moderated the

88

relationship between stress and alcohol involvement. The lack of a signiﬁcant interaction
between gender and alcohol involvement in the current study may be due to the age of the
participants, which was slightly older than the mean age in the adolescent sample but
younger than the mean age in the adult sample. Perhaps during this transitional period
between adolescence and adulthood, the relative impact of gender on stress may not be
signiﬁcantly different. Alternatively, as described below, the impact of a third moderating
variable (i.e., self-oriented perfectionism) may signiﬁcantly moderate the relationship
between gender and stress in the likelihood of alcohol involvement.

Hypothesis 5 proposed that among men with high perfectionism, high-perceived
stress and would be most likely to consume alcohol, have alcohol related problems, and
use alcohol to cope. In contrast, among women who are high in perfectionism, perceived
stress level would not be predictive of alcohol consumption, having alcohol related
problems, and using alcohol to cope. Results of this study are partially consistent with
Hypothesis 5, in that among women with high self—oriented perfectionism, the degree of
perceived stress was not predictive of alcohol consumption. Similar to the two-way
interaction described previously, among women with low self-oriented perfectionism,
high and average perceived stress was associated with the highest predicted odds of
alcohol consumption. In contrast, among high self-oriented perfectionists, the predicted
odds of consuming alcohol increased, regardless of the perceived stress level. Again, this
may indicate that women with high self-oriented perfectionism are less likely to consume
alcohol, regardless of perceived stress level, because they believe that any degree of
alcohol consumption may interfere with their ability to be perfect or perform ﬂawlessly at

tasks. Female college students with low self-oriented perfectionism may only decrease

89

their alcohol consumption after they become concerned that it may hamper their ability to
performance a task under intense pressure or stress.

The relationship does hold true of male college students and the results for men
were not consistent with Hypothesis 5. Results of the two-way interaction between self-
oriented perfectionism and perceived stress among male college students indicated that
high self-oriented perfectionists with low perceived stress was most predictive of alcohol
consumption when compared to high self-oriented perfectionists with average to high
perceived stress. The ﬁnding that among male college students, high self-oriented
perfectionists with high perceived stress consume less alcohol than high self-oriented
perfectionists with low perceived stress suggests that college males with high self-
oriented perfectionism may monitor their use of alcohol consumption and may reduce
their alcohol involvement when they experience stress, electing to use adaptive coping.
Indeed, there is evidence in the literature to suggest that self—oriented perfectionism is
associated with adaptively coping with stress, such as positive problem-solving
orientations and learned resourcefulness (Flett, Hewitt, Blankstein, & O’Brien, 1991;
Flett, Hewitt, Blankstein, Solnik, & Van Brunschot, 1996). This may suggest that among
male college students, a new theory of perfection may be in order that proposes that
relatively adaptive aspects of perfectionism (i.e., self-oriented perfectionism) may buffer
the effects of stress on alcohol consumption. That is, as stress increases, male self-
oriented perfectionists are able to draw upon their adaptive coping skills when confronted
with stress and are less prone to using maladaptive coping skills, such as consuming
alcohol. I propose that this theory could be referred to as the Stress-Buffer Model of

Perfectionism (SBM). This indicates that future studies could examine if self-oriented

9O

perfectionism among male college students interacts with coping responses and stress
level to predict alcohol involvement.

In general, these results suggest that among college students, gender seems to
moderate the relationship between perfectionism (i.e., self-oriented perfectionism) and
perceived stress with regard to alcohol consumption. This suggests that the Stress-Buffer
Model of Perfectionism related to alcohol consumption may be a useful model to
understand alcohol consumption among male college students (see Hypothesis 3), but not
as useful in understanding alcohol consumption among female college students.
Therefore, models should take into account the effects of gender, which may
differentially affect the interaction between self-oriented perfectionism and perceived
stress regarding alcohol consumption.

One surprising ﬁnding of the study was that only self-oriented perfectionism was
related to alcohol involvement whereas other aspects of perfectionism (i.e., self-
discrepant and socially-prescribed perfectionism) were not signiﬁcantly related to alcohol
involvement. The reason for this is not clear. As discussed previously, self-oriented
perfectionism involves high personal standards and is regarded by some researchers as a
relatively adaptive trait (Enn & Cox, 2002). In contrast, self-discrepant perfectionism is
conceptualized as the person’s perception of the extent to which perfectionistic standards
have been achieved. Slaney et al. (2001) argue that this perceived discrepancy is
maladaptive and is related to signiﬁcant psychological distress. Socially-prescribed
perfectionism involves the perception that unrealistically high standards are being
imposed by others. Likewise, it has been regarded as a relatively maladaptive trait (Enn

& Cox, 2002) and has been found to be associated with alcohol consumption among

91

adults (Hewitt & Flett, 1991). Therefore, a rationale as to why self-discrepant
perfectionism and socially-prescribed perfectionism were not associated with an increase
in alcohol involvement among colleges students may be because college students, unlike
their adults counterpart, do not tend to use alcohol to cope with the psychological distress
associated with their perfectionism. They may utilize other means to cope with their
distress.

It was also surprising that of the models proposed, the only dependent variable
that was predicted by the independent variables in the study was alcohol consumption.
Although self-oriented perfectionism was signiﬁcantly (negatively) associated with
drinking alcohol to cope (r = -. l4), gender, perfectionism, perceived stress, and their
respective interactions were not predictive of either alcohol related problems or drinking
alcohol to cope. I believe that it is most likely that the young age of the participants
inﬂuenced lack of association between alcohol related problems as well as the motivation
to drink alcohol to cope with negative emotions. It is worth reminding readers that the
college students in the study were intentionally sampled because they were young (18
and 19 year-olds) and had experienced only a few semesters at the university. Therefore,
none of the adolescents in the sample were of legal drinking age, which might make it
difficult to detect important associations between the independent and dependent
variables. These indicators of alcohol involvement may not emerge until later in the
students’ academic career or until the later stages of adulthood. For example, college
students during this developmental stage were perhaps able to curtail their alcohol
consumption before it led to signiﬁcant alcohol related problems. This argument is

consistent with the results of a recent study by Turrisi, Padilla, and Wierma (2000), in

92

which drinking related problems, as assessment by the YAAPST, were signiﬁcantly
fewer among traditionally-aged college freshmen and sophomores when compared to
nontraditionally-aged college freshmen and sophomores as well as upperclassmen (i.e.,
juniors and seniors).

In addition, the motivation to drink alcohol to cope with negative emotions may
be more of a characteristic of later stages of development when it is associated with
clinical alcohol dependence. Perhaps other motives to drink, such as social motives,
enhancement motives, and conformity motives described by Cooper (1994), may be more
predictive of college student drinking. These and other possibilities could be examined in

future studies.

Limitations

This study had several notable limitations. First, the use of a self-reported stress
questionnaire has been criticized by some researchers. Critics have argued that such
measures can be biased and inaccurate because of participants' faulty memories of events
and their subsequent misperception of those events (McQuaid et al., 1992; Raphael,
Cloitre, & Dohrenwend, 1991). Some researchers have suggested that interview based
measures of stress can provide more accurate measures of stressful events when
compared to self-report measures (e.g., Hewitt et al., 1996; Hewitt & Flett, 2002).
Therefore, future studies may beneﬁt from assessing stress via clinical interviews.

In addition, the results obtained from college students at this university may not
be consistent with college students at other universities. It is quite possible that a self-

selection process occurred in which the college students at the university where the data

93

were gathered had selected the university because their perception of alcohol
involvement on campus. Speciﬁcally, the university from which the data were obtained
has a reputation for being a “party school,” which would tend to attract college students
who were interested in drinking. Therefore, the extent of alcohol consumption, alcohol
related problems, and motivation to cope with negative emotions via alcohol
consumption may not generalize to other universities.

Another limitation of this study was that the methodology only addressed the
relationships among average levels of the variables of interest. Kenny, Bolger, and
Kashy (1999) and Bryk and Raudenbush (1992) remind researchers that within-person
associations can differ in both magnitude and direction from between-person
associations. Thus, future research would beneﬁt from a methodology design, such as a
daily process design, that assesses the pertinent variables on a daily basis. This would
allow for a multilevel hierarchical linear regression model that could account for both
between-person and within-person variance, thus yielding a better understanding of the
complex relationship between stress, perfectionism, and alcohol involvement. In
particular, Dowdall and Wechsler (2002) suggest that multilevel hierarchical linear
regression model may be particularly useful when analyzing drinking related data from
college students at several universities because such an analysis could statistically control
for characteristics shared by students attending the same university.

This study is also limited because of the reliance on retrospective self-reports of
alcohol consumption, stress, and perfectionism. This may be a methodological barrier
because it fails to capture the rapid and dynamic relationship between these variables.

Such an approach might also be hindered by the participants' tendency to not

94

acknowledge the extent of their alcohol consumption. One method of overcoming the
retrospective self-report bias would be to utilize an ecological momentary assessment
(Stone & Shiffman, 1994) in which participants record their responses to questionnaires
on a daily basis. In particular, this methodological approach is utilized more frequently in
studies that investigate alcohol consumption (Collins, Morsheimer, Shiffman, Paty, Gyns,

& Papandonatos, 1998; Swendson, Tennen, Carney, Afﬂeck, Willard & Hromi, 2000).

Clinical Implications

This study has several implications for the treatment of alcohol involvement
among college students. Currently, one of the predominant approaches to treating
alcohol consumption is to train clients to better manage their stress through healthy
behaviors (Marlatt, 1996; Marlatt & Gordon, 1985) and stress reduction (Yost & Mines,
1985). Research indicates that these approaches do indeed decrease the level of alcohol
consumption and can prevent relapse (Hawkins, Catalano & Wells, 1986; Murphy et al.,
2001; Myers & Brown, 1990; Richter, Brown, & Mott, 1991). However, the ﬁnding that
perceived stress by itself was not predictive of alcohol involvement among this sample of
college students suggest that additional interventions strategies may be warranted.

The ﬁnding that self-oriented perfectionism, regarded by some researchers as an
adaptive characteristic, was associated with a decreased likelihood of being categorized
as a consumer of alcohol, suggests that these are approaches that clinicians may want to
utilize strategies to enhance some adaptive aspects of perfectionism, such as high
personal standards and a motivation to be ﬂawless, as a means to decrease, or perhaps

prevent, alcohol consumption among college students. This approach may serve to

95

deve10p additional resilient factors in the prevention of alcohol involvement. Rather than
being a negative and intransi gent personality trait that needs to be modiﬁed by long-term,
intensive, psychodynamically-oriented psychotherapy (e. g., B1att& Zuroff, 2002), some
perfectionistic traits should be enhanced in psychotherapy. This may be especially
helpful for male college students, for whom the adaptive trait of perfectionism seems to
buffer the effects of stress on alcohol consumption. In addition, among female college
students, a high degree of (self-oriented) perfectionism was associated with a signiﬁcant
decrease in the likelihood of consuming alcohol at all levels of stress, suggesting that
enhancing the adaptive characteristics of perfectionism may be a particularly useful
therapeutic intervention for female college students. These clinical implications are
speculative at this point and would necessitate an empirical research to assess the validity

of such treatment approaches.

Conclusions and Future Directions for Research

In summary, this research investigated the utility of the Diathesis-Stress Model of
Perfectionism to predict alcohol consumption, alcohol related problems, and drinking
alcohol to cope. The study hypothesized that both gender and perfectionism (i.e., self-
oriented, self-discrepant, socially-prescribed) would moderate the relationship between
perceived stress and alcohol involvement among college students. Although gender did
not moderate the relationship between perceived stress and alcohol involvement, the
results of a logistic regression analysis (n = 198) found that self-oriented perfectionism, a
two-way interaction (self-oriented perfection x perceived stress), and a three-way

interaction (self-oriented perfectionism x perceived stress x gender) were associated with

96

a change in the likelihood of consuming alcohol. Interpretation of the two-way
interaction suggested that low self-oriented perfectionists who experienced low perceived
stress were about 16 times more likely to consume alcohol than low self-oriented
perfectionists who experienced high perceived Stress. Interpretation of the three-way
interaction suggested a similar relationship among female college students. However,
among male college students, low self-oriented perfectionists who experienced low
perceived stress were less likely to consume alcohol than high self-oriented perfectionists
who experienced low perceived stress. Results of two logistic regression models
regarding alcohol related problems and drinking alcohol as a means to cope were not
signiﬁcant, perhaps due to the age of the participants and differing motives to drinking
alcohol. Results of this study suggest a possible Stress-Buffer Model of Perfectionism
among male college students.

Because the association between perceived stress, self-oriented perfectionism, and
alcohol consumption seems to differ between men and women, future studies may beneﬁt
from further investigation of academic adjustment variables as well. These may include
additional measures of academic performance (i.e., GPA), social adjustment, and
psychological well-being. For example, the low level of stress that appeared to be
associated with an increased likelihood of being classiﬁed into a drinking or drinking-
related problem category under certain conditions may not be indicative of the absence of
other college-relevant adjustment difﬁculties. It may be that those students are unlikely to
persist in their studies and more likely to experience academic difficulties and

prematurely withdraw from the university.

97

It is evident from these results that future research would also benefit from a
further exploration of the Stress-Buffer Hypothesis of Peg’ectionism, particularly among
male college students. The nature of the seemingly beneﬁcial effects of self-oriented
perfectionism among college men could be examined. Theories, such as gender
socialization, which account for differences in the buffering effects of stress among
perfectionistic men could be further explored.

The results of this study also suggest that future research may want to include a
wider range of age of the participants. Turrisi et al. (2000) have found that upperclassmen
(i.e., juniors and seniors) tend to consume more alcohol than both traditionally-aged and
nontraditionally-aged freshmen and sophomores. In addition, this study found that
traditionally-aged freshmen and sophomores tend to have fewer alcohol related problems
than nontraditionally-aged freshmen and sophomores as well as upperclassmen (i.e.,
juniors and seniors). However, this ﬁnding is not consistent across all studies (i.e., Klein,
1994; Wechsler et al., 1998). This suggests that future studies would beneﬁt from a
longitudinal study that compares drinking behaviors across several different age groups
(e.g., traditionally-age compared with non-traditionally-aged; upper classmen compared
with lower classmen).

Furthermore, future studies should measure differences in college student’s
susceptibility to alcohol intoxication in a laboratory setting, in addition to measuring
alcohol consumption, alcohol related problems, and motives to drink alcohol. This would
allow researchers to assess how the various combinations of perfectionism, stress, and
gender relate to possible differences in alcohol intoxication. As Perkins (1999) reminds

readers, women are more prone to becoming intoxicated after they consume an equal

98

quantity of alcohol than their male counterparts. Are various combinations of these
factors associated with greater levels of intoxication? Future studies could also expand
upon this research by including additional measures to assess additional aspects of
perfectionism (e.g., other-oriented perfectionism, high personal standards, high need for
order) as well as additional motives to drink alcohol (social motives, enhancement
motives, conformity motives, etc.). In summary, this study has prompted enough

potential avenues for research to last a career.

99

REFERENCES

Adkins, K.K. & Parker, W. (1996). Perfectionism and suicidal preoccupation. Journal of

Personality, 64, 529-543.

Adler, A. (1956). The neurotic disposition. In H.L Ansbacher & R.R. Ansbacher (Eds),

The individual psychology of Alfred Alder (pp. 239-262). New York: Harper.

Aiken, L.S. & West, S.G. (1991). Multiple regression: Testing and interpreting

interactions. Newbury Park, CA: Sage.

Alden, L.E.. Bieling, P.J., & Wallace, ST. (1994). Perfectionism in an interpersonal
context: A self-regulation analysis of dysphoria and social anxiety. Cognitive

Therapy and Research, 18, 297-316.

Axtell, A. & Newlon, BJ. (1993). An analysis of Adlerian life themes of bulimic women.
Individual Psychology: Journal of Adlerian Theory, Research, & Practice, 49,

58-67.

Baer, J.S., Marlatt, G.A., Kivlahan, D., Fromme, K., Larimer, M., & Williams, E. (1992).

An experimental test of three methods of alcohol risk reduction with young

adults. Journal of Consulting and Clinical Psychology, 60, 974-979.

100

Beck, A.T., Weissman, A., Lester, D., & Trexler, L. (1974). The measurement of
pessimism: The Hopelessness Scale. Journal of Consulting and Clinical

Psychology, 42, 861-865.

Berkowitz, A.D., & Perkins, H.W. (1986). Problem drinking among college students: A

review of recent research. Journal of American College Health, 35, 1-28.

Blatt, SJ. (1995). The destructiveness of perfectionism: Implications for the treatment of

depression. American Psychologist, 50, 1003-1020.

Blatt, S.J. & Zuroff, DC. (2002). Perfectionism in the therapeutic process. In G.L. Flett
& P.L. Hewitt (Eds) (pp. 393-406). In Perfectionism: Theory, Research, and

Treatment. Washington, DC: American Psychological Association.

Brian, L. (1985). Perfectionism. Center City, MN: Hazelden.

Bryk, A.S. & Raudenbush, SW. (1992). Hierarchical linear models: Applications and

data analysis methods. Newbury Park: Sage.

Burns, DD. (1980). The perfectionist's script for self-defeat. Psychology Today, 34-44.

Burns, DD. (1983). The spouse who is a perfectionist. Medical Aspects of Human

Sexuality, 17, 219-230.

101

Campbell, D.T., & Fiske, D.W. (1959). Convergent and discriminant validation by the

multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.

Carnegie Foundation for the Advancement of Teaching (1990). Campus Life: In Search

of Community. Princeton, NJ: Princeton University Press.

Champoux, J., & Peters, W. (1987). Form, effects size and power in moderated

regression analysis. Journal of Occupational Psychology, 60, 243-255.

Chang, EC. (1998). Cultural differences, perfectionism, and suicide risk: Does social

problem solving still matter?. Cognitive Therapy and Research, 74, 1109-1120.

Chang, E.C., & Rand, KL. (2000). Perfectionism as a predictor of subsequent
adjustment: Evidence for a speciﬁc diathesis-stress mechanism among college

students. Journal of Counseling Psychology, 47, 129-137.

Cochrane, R., & Robertson, A. (1973). The Life Events Inventory: A measure of the

relative severity of psycho-social stressors. Journal of Psychosomatic Research,

17, 135-139.

102

Cohen, B.N., & Rice, KG. (2003). Perfectionism and anxiety as predictors of alcohol
abuse in college students: Test of an integrative moderator model. Manuscript

submitted for publication.

Cohen, J. (1983). Applied multiple regression/correlation analysis for the behavioral

sciences (Rev. ed.). Hillsdale, NJ: Lawrence Erlbaum.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed).

Hillsdale, NJ: Lawrence Erlbaum Associates.

Cohen, J ., Cohen, R, West, S.G., & Aiken, LS. (2003). Applied multiple
regression/correlation analysis for the behavioral sciences (3rd ed.). Hillsdale,

NJ: Lawrence Erlbaum Associates.

Cohen, S. (1986). Contrasting the Hassles Scale and the Perceived Stress Scale: Who is

really measuring appraised stress?. American Psychologist, 41, 716-718.

Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress.

Journal of Health and Social Behavior, 24, 386-396.

Colings, R., Morsheimer, E., Shiffman, S., Pary, J ., Gyns, M., & Papandonatos, G.
(1998). Ecological momentary assessment in a behavioral drinking moderation

training program. Experimental and Clinical Psychopharmacology, I I , 306-315.

103

 

Conger, J. J. (1956). Alcoholism: Theory, problem and challenge: H. Reinforcement
theory and the dynamics of alcoholism. Quarterly Jounzal of Studies on Alcohol,

13, 296-305.

Cooper, ML. (1994). Motivations for alcohol use among adolescents: Development and

validation of a four-factor model. Psychological Assessment, 6, 117-128.

Cooper, ML. (2002). Alcohol use and risky sexual behavior among college students and

youth: Evaluating the evidence. Journal of Studies on Alcohol, 14, 101-117.

Cooper, M.L., Frone, M.R., Russell, M., & Mudar, P. (1995). Drinking to regulate
positive and negative emotions: A motivational model of alcohol use. Journal of

Personality and Social Psychology, 69, 990-1005.

Cooper, M.L., Frone, M.R., Russell, M., & Peirce, RS. (1997). Gender, stress, coping,
and alcohol use. In R.W. Wilsnack and SC. Wilsnack (Eds), Gender and
alcohol: Individual and social perspectives (pp. 199-224). New Brunswikck, NJ:

Rutgers Center for Alcohol Studies.

Cooper, M.L., Russell, M., & Frone, MR. (1990). Work stress and alcohol effects: A

test of stress-induced drinking. Journal of Health and Social Behavior, 31, 260-

276.

104

 

Cooper, M.L., Russell, M., Skinner, J .B., Frone, M.R., & Mudar, P. (1992). Stress and
alcohol use: Moderating effects of gender, coping, and alcohol expectancies.

Journal of Abnormal Psychology, 101 , 139-152.

Corcoran, K.J. & Parker, PS. (1991). Alcohol expectancy questionnaire tension
reduction scale as a predictor of alcohol consumption in a stressful situation.

Addictive Behaviors, I 6, 129-137.

Costa, P.T., Jr. & McCrae, RR. (1990). Personality: Another" hidden factor" in stress

research. Psychological Inquiry, I, 22-24.

Crowne, D., & Marlow, D. (1960). A new scale of social desirability independent of

psychopathology. Journal of Consulting Psychology, 24, 349-354.

Deffenbacher, J .L., Zwemer, W.A., Whisman, M.A., Hill, R.A. & Sloan, RD. (1986).

Irrational beliefs and anxiety. Cognitive Therapy and Research, 10, 281-292.

Delongis, A., Folkman, S., Lazarus, RS. (1988). The impact of daily stress on health and

mood: Psychological and social resources as mediators. Journal of Personality

and Social Psychololgy, 54, 486-495.

105

Derotatis, LR. (1983). The SCL-90-R: Administration, scoring, and procedures manual

II. Baltimore: Clinical Psychometric Research.

Dimeff, L.A., Baer, J .S., Kivlahan, D.R., & Marlatt, G.A. (1999). Brief alcohol
screening and intervention for college students: A harm reduction approach.

New York: Guilford Press.

Dohrenwend, B.P., & Shrout, PE. (1985). "Hassles" in the conceptualization and

measurement of life stress variables. American Psychologist, 40, 780-785.

Dowdall, G.W., & Wechselr, H. (2002). Studying college alcohol use: Widening the

lens, sharpening the focus. Journal of Studies on Alcohol, 14, 14-22.

Dunkley, D.M., Blankstein, K.R., Halsall, J ., Williams, M., & Winkworth, G. (2000).
The relation between perfectionism and distress: Hassles, coping, and Perceived
social support as mediators and moderators. Journal of Counseling Psychology,

4 7, 437-453.
D'Zurilla, T.J., & Sheedy, CF. (1991). Relation between social problem-solving ability

and subsequent level of psychological stress in college students. Journal of

Personality and Social Psychology, 61 , 841-846.

106

 

Eng, R.C., Diebold, B.A., & Hansen, DJ. (1996). The drinking patterns and problems of
a national sample of college students, 1994. Journal of Alcohol and Drug

Education, 41, 13-33.

Engs, R.C. & Hanson, DJ. (1985). The drinking-pattems and problems of college

students: 1983. Journal of Alcohol and Drug Education, 31, 65-82.

Enns, M.W. & Cox, BJ. (2002). The nature and assessment of perfectionism: A critical
analysis. In G.L. Flett & P.L. Hewitt (Eds), Perfectionism: Theory, Research,
and Treatment (pp. 33-62). Washington, DC: American Psychological

Association.

Flett, G.L., Hewitt, P.L., Blankstein, K. & O’Brien, S. (1991). Perfectionism and learned
resourcefulness in depression and self-esteem. Personality and Individual

Differences, 12, 61-68.

Flett, G.L., Hewitt, P.L., Blankstein, K.R., Solnik, M., & Van Brunschot, M. (1996).

Perfectionism, social problem-solving ability, and psychological distress. Journal

of Rational-Emotive and Cognitive-Behavior Therapy, 14, 245-275.

Flett, G.L., Hewitt, P.L., & Dyck, DC. (1989). Self-oriented perfectionism, neuroticism,

and anxiety. Personality and Individual Differences, 10, 731-735.

107

Flett, G.L., Hewitt, P.L., Blankstein, K.R., & Mosher, SW. (1995). Perfectionism, life

events, and depressive symptoms: A test of the diathesis—stress model. Current

Psychology, 14, 112-137.

Forrnan, M.A., Tosi, D.J., & Rudy, DR. (1987). Common irrational beliefs associated
with the psychophysiological conditions of low back pain, peptic ulcer and
migraine headache: A multivariate study. Journal of Rational-Emotive Therapy,

5, 255-265.

Franck, R. (2002). The best 345 colleges: 2003 edition. New York: Random House.

Freud, S. (1959). Inhibitions, symptom and anxiety. In J. Strachey (Ed), The standard
edition of the complete psychological works of Sigmund Freud (Vol. 20, pp. 77-

175). London: Hogarth Press. (Original work published 1926).

Frost, R.O., Marten, P.A., Lahart, C., & Rosenblate, R. (1990). The dimensions of

perfectionism. Cognitive Therapy and Research, 14, 449-468.

Frost, R.O., Steketee, G., Cohn, L., & Griess, K. (1994). Personality traits in subclinical

and non-obsessive-compulsive volunteers and their parents. Behavior Research

and Therapy, 32, 47-56.

108

 

Gadzella, BM. (1994). Student-life stress inventory: Identiﬁcation of and reactions to

stressors. Psychological Reports, 74, 395-402.

Garner, D.M., Olmstead, M.P., & Polivy, J. (1983). Development and validation of a
multidimensional eating disorder inventory for anorexia nervosa and bulimia.

International Journal of Eating Disorders, 2, 15-34.

Glass, T.A. Prigerson, H., Karsl, S.V., & Mendes de Leon, CF. (1995). The effects of
negative life events on alcohol consumption among older men and women.

Journal of Gerontology, SOB, 8205-8216.

Goldfried, M.R. & Sobocinski, D. (1975). Effects of irrational beliefs on emotional

arousal. Journal of Consulting and Clinical Psychology, 43, 504-510.

Grant, B.F., Harford, T.C., Dawson, D.A., Chou, P., Dufor, M., & Pickering, R. (1994).
Prevalence of DSM-IV alcohol and dependence: United States, 1992. Special

focus: Women and alcohol. Alcohol Health and Research World, 18, 243-248.

Greenﬁeld, T. (1986). Quantity per occasion and consequences of drinking: A

reconsideration and recommendation. International Journal of the Addictions, 21,

1059- 1079.

109

 

Greenley, J ., & Oei, T. (1999). Alcohol and Tension Reduction. In K.E. Leonard & H.T.
Blane (Eds) (pp. 14-53). Psychological Theories of Drinking and Alcoholism

( 2nd edition). New York: Guilford Press.

Hamachek, DE. (1978). Psychodynamics of normal and neurotic perfectionism.

Psychology, 15, 27-33.

Hammen, C. (1992). Life events and depression: The plot thickens. American Journal of

Community Psychology, 20, 179-193.

Hawkins, D.J., Catalano, R.F. & Wells, EA. (1986). Measuring effects of a skill training
intervention for drug abusers. Journal of Consulting and Clinical Psychology, 54,

661-664.

Hewitt, P.L. & Flett, G.L. (1991). Perfectionism in the self and social context:
Conceptualization, assessment, and association with psychotherapy. Journal of

Personality and Social Psychology, 60, 456-470.

Hewitt, P.L., & Dyck, DC. (1986). Perfectionism, stress, and vulnerability to depression.

Cognitive Therapy and Research, 10, 137-142.

Hewitt, P.L., & Flett, G.L. (1990). Perfectionism and depression: A multidimensional

analysis. Journal of Social Behavior and Personality, 5, 423-438.

110

Hewitt, P.L., & Flett, G.L. (1993). Dimensions of Perfectionism, daily stress, and
depression: A test of the speciﬁc vulnerability hypothesis. Journal of Abnormal

Psychology, 102, 58-65.

Hewitt, P.L., & Flett, G.L. (2002). Perfectionism and Stress Processes in
Psychopathology. In G.L. Flett & P.L. Hewitt (Eds) (pp. 255-284).
Perfectionism: Theory, Research, and Treatment. Washington, DC: American

Psychological Association.

Hewitt, P.L., Flett, G.L., & Edi ger, E. (1996). Perfectionism and depression:
Longitudinal assessment of a speciﬁc vulnerability hypothesis. Journal of

Abnormal Psychology, 105, 276-280.

Hewitt, P.L., Flett, G.L., & Mikail, SF. (1995). Perfectionism and relationship
adjustment in pain patients and their spouses. Journal of Family Psychology, 9,

335-347.

Hewitt, P.L., Flett, G.L., & Mosher, SW. (1992). The perceived stress scale: Factor

structure and relation to depression symptoms in a psychiatric sample. Journal of

Psychopathology and Behavioral Assessment, 14, 247-257.

111

Hewitt, P.L., Flett, G.L., & Tumbull, B. (1992). Perfectionism and mulitphasic
personality inventory (MMPI) indices of personality disorder. Joumal of

Psychopathology and Behavioral Assessment, 14, 323-335.

Hewitt, P.L., Flett, G.L., & Tumbull-Donovan, W., & Mikail, SF. (1991). A
multidimensional perfectionism scale: Reliability, Validity, and Psychometric
properties in psychiatric samples. Psychological Assessment: A Journal of

Consulting and Clinical Psychology, 3, 464-468.

Hewitt, P.L., Flett, G.L., & Weber, C. (1994). Dimensions of perfectionism and suicide

ideation. Cognitive Therapy and Research, 18, 439-460.

Hingson, R.W., Heeren, T., Zakocs, R.C., Kopstein, A. (2002). Magnitude of alcohol-
related mortality and morbidity among US. college students ages 18-24. Journal

of Studies on Alcohol, 14, 136-144.

Hirsch, J.K., & Ellis, J .B. (1996). Differences in life stress and reasons for living among

college suicide ideators and non-ideators. College Student Journal, 30, 377-384.
Hitchcox, CF. (1979). EMG feedback training in women alcoholics: An alternative

coping response to stress and helplessness. Dissertation Abstracts International,

39, 3516-3517.

112

Hollender, M.H. (1978). Perfectionism: A neglected personality trait. Journal of Clinical

Psychiatry, 39, 384.

Homey, K. (1950). Neurosis and human growth. New York: Norton.

Hull, J .G. (1981). A self-awareness model of the causes and effects of alcohol

consumption. Psychology Journal of Abnormal, 90, 586-600.

Hull, J .G. (1987). Self-awareness model. In H.T. Blane & K.E. Leonard (Eds).
Psychological theories of drinking and alcohol consumption (pp. 272-304). New

York: Guilford Press.

Hurlbut, S.C. & Sher, KJ. (1992). Assessing alcohol problems in college students.

College Health, 41, 49-58.

J accard, J. (2001). Interaction Effects in Logistic Regression. Sage University Paper
Series on Quantitative Applications in the Social Sciences, 07-135. Thousand

Oaks, CA: Sage.

Jessor, R., Graves, T.D., Hanson, R.C., & Jessor, S.L. (1968). Society, personality, and

deviant behavior: A study of a tri-ethnic community. New York: Holt, Rinehart,

& Winston.

113

Johnson, D.P., & Slaney, RB. (1996). Perfectionism: Scale development and a Study of
perfectionistic clients in counseling. Journal of College Student Development, 37,

29-41.

Johnston, L.D., O’Malley, P.M. & Bachman, J .G. (1996). National survey results on drug
use form the Monitoring the Future Studies, 1975-1994. Volume H. College

students and young adults. Rockville, MD: National Institute on Drug Abuse.

Joiner, T.E., & Schmidt, NE. (1995). Dimensions of perfectionism, life stress, and
depressed and anxious symptoms: Prospective support for Diathesis-Stress but
not Speciﬁc Vulnerability among male undergraduates. Journal of Social and

Clinical Psychology, 14, 165-183.

Kasl, S.V., Ostfeld, A.M., Berkman, L.F., & Jacobs, SC. (1987). Stress and alcohol
consumption: The role of selected social and environmental factors. In E.
Gottheil, K.A. Druley, S. Pashko, & S.P. Weinstein (Eds), Stress and Addiction

(pp. 40-60). New York: Brunner/Mazel.

Kenny, D.A., Bolger, N., & Kashy, DA. (1999). Traditional methods for estimating
multilevel models. In D.S. Moskowitz & Hershberger (Eds), Modeling
intraindividual variability with repeated measures data: Method and

applications. Englewood Cliffs, NJ: Erlbaum.

114

Kessler, R.C., Crum, R.M., Warren, L.A., Nelson, C.B., Schulenber, J. & Anthony, J .C.
(1997). Lifetime co-occurrence of DSM-III-R alcohol abuse and dependence
with other psychiatric disorders in the national comorbidity survey. Archives of

General Psychiatry, 54, 313-321.

Kidorf, M. & Lang, AR. (1999). Effects of social anxiety and alcohol expectancies on

stress-induced drinking. Psychology of Addictive Behaviors, 13, 134-142.

Klein, H. (1994). Changes in college students’ use and abuse of alcohol, and in their
attitudes toward drinking over the course of their college years. Journal of Youth

and Adolescences, 23, 251-269.

Knight, J .R., Weschsler, H., Kuo, M., Seibring, M., Weitzman, E.R., & Schuckit, M.
(2002). Alcohol abuse and dependence among US. college students. Journal of

Studies on Alcohol, 63, 263-270.

Krause, N. (1991). Stress, religiosity and abstinence from alcohol. Psychology of Aging,

6, 134-144.

Krummel, RP. (1977). An investigation into selected effects of adding relaxation and

assertion training to an existing treatment program for alcoholics. Dissertation

Abstracts International, 38, 4990-4991.

115

Lapointe, EH. (1970). Origin and evolution of the term "psychology". American

Psychologist, 25, 640-646.

Larimer, M.E. & Cronce, J .M. (2002). Identification, prevention, and treatment: A
review of individual-focused strategies to reduce problematic alcohol
consumption by college student. Journal of Studies on Alcohol, Supplement No.

14, 148-163.

Laurent, J ., Catanzaro, S.J., & Callan, MK. (1997). Stress, alcohol-related expectancies
and coping preferences: A replication with adolescents of the Cooper et al.

(1992) Model. Journal of Studies on Alcohol, 58, 644-651.

Lemle, R. & Mishkind, ME. (1989). Alcohol and masculinity. Journal of Substance

Abuse Treatment, 6, 213-222.

Liebman, W.M. . (1978). Recurrent abdominal pain in children: A retrospective study of

119 patients. Clinical Pediatrics, 1 7, 149-153.

Lo, CC. (1996). Are women heavier drinkers than we thought they were? Journal of

Studies on Alcohol, 57, 531-535.

Lombardi, D.N., Florentino, M.C., & Lombardi, A]. (1998). Perfectionism and abnormal

behavior. Journal of Individual Psychology, 54, 61-71.

116

Malloy, E.A. & Goldman, M. (2002). A call to action: Changing the culture of drinking
at US. colleges. (NIH Publication No. 02-5010). Washington, DC: US.

Government Printing Ofﬁce.

Marlatt, G.A. (1996). Reducing college student binge drinking: A harm-reduction
approach. In R]. Resnick and RH. Rozensky (Eds), Health Psychology Through
the Life Span: Practice and Research Opportunities (pp. 377-393). Washington,

DC: American Psychological Association.

Marlatt, G.A., & Brown, SA. (1985). Relapse Prevention: A Self-Control Strategy for

the Maintenance of Behavior Change. New York: Guilford Press.

McClelland, G., & Judd, C. (1993). Statistical difﬁculties of detecting interactions and

moderator effects. Psychological Bulletin, 114, 376-390.

McCormack, AS. (1996). Drinking in stressful situations: College men under pressure

(1990 an d 1994). College Student Journal, 30, 65-68.
McCreary, D.R. & Sadava, SW. (1998). Stress, drinking, and the adverse consequences

of drinking in two samples of young adults. Psychology of Addictive Behaviors,

12, 247-261.

117

McCreary, D.R., Newcomb, M.D., & Sadava, SW. (1999). The male role, alcohol use,
and alcohol problems: A structural modeling examination in adult women and

men. Journal of Counseling Psychology, 46, 109-124.

McQuaid, J ., Monroe, S., Roberts, J ., Johnson, S., Garamoni, G., Kuppfer, D., & Frank,
E. (1992). Toward the standardization of life stress assessment: Deﬁnition

discrepancies and inconsistencies in methods. Stress Medicine, 8, 47-56.

Moore, C.A., & Barrow, LC. (1986). Perfectionistic thinking in university students:
Implications for individual treatment. In J .E. Talley & W.J.K. Rockwell (Eds),
Counseling and psychotherapy with college students (pp. 100-112). New York:

Praeger.

Murphy, J .G., Duchnick, J .J ., Vuchinich, R.E., Davison, J .W., Karg, R.S., Olson, A.M.,
Smith, AR, & Coffey, T.T. (2001). Relative efﬁcacy of a brief motivational
intervention for college student drinkers. Psychology of Addictive Behaviors, 15,

373-379.

Myers, M.G., & Brown, SA. (1990). Coping responses and relapse among adolescent

substance abusers. Journal of Substance Abuse, 2, 177- 189.

National Institute on Alcohol Abuse and Alcoholism. (1984). Report of the 1983

Prevention Planning Panel. Rockville, MD: Author.

118

Nerviano, V.J., & Gross, H.W. (1983). Personality types of alcoholics on objective

inventories: A review. Journal of Studies on Alcohol, 44, 837-851.

Noel, N.E. & Cohen, DJ. (1998). Changes in substance use during times of stress:
College students the week before exams. Journal of Drug Education, 27, 363-

372.

Noel, N E & Lisman, SA. (1980). Alcohol consumption by college women following
exposure to unsolvable problems: Learned helplessness or stress induced

drinking?. Behavior Research and Therapy, 18, 429-440.

Nystrom, M., Perasalo, J ., & Salaspuro, M. (1993). Screening for heavy drinking and
alcohol-related problems in young university students: The CAGE, the Mm-
MAST, and the Trauma Score Questionnaires. Journal of Studies on Alcohol, 54,

528-533.

Orcutt, J .D. & Harvey, L.K. (1991). The temporal patterning of tension reduction: Stress

and alcohol use on weekdays and weekends. Journal of Studies on Alcohol, 52,

415-424.

Pacht, AR. (1984). Reﬂections on perfectionism. American Psychologist, 39, 386-390.

119

Perkins, H.W. (1999). Stress-motivated drinking in collegiate and postcollegiate young

adulthood: Life course and gender patterns. Journal of Studies on Alcohol, 60,

219-227.

Perkins, H.W. (2002). Surveying the damage: A review of research on consequences of
alcohol misuse in college p0pulations. Journal of Studies on Alcohol, Supplement

No.14, 91-100.

Presley, C.A. Leichliter, M.A., & Meilman, PW. (1998). Alcohol and drugs on
American college campuses: A report to college presidents: Third in a series,

1995, I996, I997. Carbondale, IL: Core Institute, Southern Illinois University.

Presley, C.A., Meilman, P.W., & Cashin, JR. (1996). Alcohol and drugs on American
college campuses: Use, consequences, and perceptions of the campus

environment, Volume IV: 1992-1994. Carbondale, IL: Core Institute, Southern

Illinois University.

Presley, C.A., Weilman, P.W., & Leichleiter, J .S. (2002). College factors that inﬂuence

drinking. Journal of Studies on Alcohol, Supplement No. 14, 82-90.

Quitkin, F.M., Rifkin, A., Kaplan, J ., & Klein, DE (1972). Phobic anxiety syndrome
complicated by drug dependence and addiction. Archives of General Psychiatry,

27, 159-162.

120

Rabow, J. & Neuman, CA. (1984). Saturday night live: Chronicity of alcohol
consumption among college students. Substance and Alcohol Actions/Misuse, 5,

1-7.

Raphael, K.G., Cloitre, M., & Dohrenwend, BR (1991). Problems of recall and
misclassiﬁcation with checklist methods of measuring stressful life events. Health

Psychology, 10, 62-74.

Rice, K.G. & Dellwo, J .P. (2001). Within-semester stability and adjustment correlates of
the Multidimensional Perfectionism Scale. Measurement and Evaluation in

Counseling and Development, 34, 146-156.

Rice, K. G., & Dellwo, J. P. (2002). Perfectionism and self-development: Implications for

college adjustment. Journal of Counseling and Development, 80, 188-196.

Rice, K.G., & Lapsley, D.K. (2001). Perfectionism, coping, and emotional adjustment.

Journal of College Student Development, 42, 157-168.

Rice, K.G., & Mirzadeh, SA. (2000). Perfectionism, attachment, and Adjustment.

Journal of Counseling Psychology, 47, 238-250.

121

Rice, K.G., Ashby, J. S., & Slaney, R. B. (1998). Self-esteem as a mediator between
perfectionism and depression: A structural equations analysis. Journal of

Counseling Psychology, 45. 304-314.

Richter, S.S., Brown, S.A., & Mott, MA. (1991). The impact of social support and self-
esteem on adolescent substance abuse treatment outcome. Journal of Substance

Abuse, 3, 371—385.

Room, R. (1991). Measuring alcohol consumption in the U.S.: Methods and rationales.
In W.B. Clark & M.E. Hilton (Eds), Alcohol in America: Drinking practices and

problems (pp. 26-50). Albany: State university of New York Press.

Ross, S.E., Neibling, B.C., & Heckert, TM. (1999). Sources of stress among college

students. College Student Journal, 33, 312-317.

Schudkit, M.A., Klein, J .L., Twitchell, G.R., & Springer, L.M. (1994). Increases in
alcohol-related problems for men on a college campus between 1980 and 1992.

Journal of Studies on Alcohol , 55, 739-742.

Sher, K.J. & Levinson, R.W. (1982). Risk for alcoholism and individual differences in

the stress-response dampening effect of alcohol. Journal of Abnormal

Psychology, 91, 350-367.

122

Slaney, R.B., Rice, K.G., & Ashby, J .S. (2002) A programmatic approach to measuring
perfectionism: The Almost Perfect Scale. In G.L. Flett & P.L. Hewitt (Eds) (pp.
63-88). In Perfectionism: Theory, Research, and Treatment. Washington, DC:

American Psychological Association.

Slaney, R.B., Rice, K.G., Mobley, M., Trippi, J ., & Ashby, J .S. (2001). The revised
Almost Perfect Scale. Measurement and Evaluation in Counseling and

Development, 34, 130-145.

Sobell, M.B., Sobell, L.C., Klajner, F., Pavan, D., & Basian, E. (1986). The reliability of
a timeline method for assessing normal drinker college students' recent drinking

history: Utility for alcohol research. Addictive Behaviors, I I, 149-161.
Stacy, A.W., Widman, K.F., Hays, R., & DiMatteo, MR. (1985). Validity of self-reports
of alcohol and other drug use: A multitrait-multimethod assessment. Journal of

Personality and Social Psychology, 49, 219-232.

Stevens, G., & Hoisington, E. (1987). Occupational prestige and the 1980 US. labor

force. Social Science Research, I 6, 74-105.

Stone, A.A., & Shiffman, S. (1994). Ecological momentary assessment (EMA) in

behavioral medicine. Annals of Behavioral Medicine, 16, 199-202.

123

Suddath, B. H. & Slaney, RB. (2001). An investigation of the dimensions of
perfectionism in college students. Measurement and Evaluation in Counseling

and Development, 34, 157-165.

Swendson, J., Tennen, H., Carney, M.A., Afﬂeck, G., Willard, A. & Hromi, A. (2000).
Mood and alcohol consumption: An experience sampling test of the self-

medication hypothesis. Joumal of Abnormal Psychology, 109, 198-204.

Tabachnick, B.G., & Fidel], LS. (1996). Using Multivariate Statistics ( 3rd ed. ). New

York: HarperCollins College Publishers.

Towbes, L.C., & Cohen, LR (1996). Chronic stress in the lives of college students:
Scale development and prospective prediction of distress. Journal of Youth and

Adolescence, 25, 199-217.

Turrisi, R., Padilla, K.K., & Wiersman, K.A. (2000). College student drinking: An
examination of theoretical models of drinking tendencies in freshmen and

upperclassmen. Journal of Studies on Alcohol, 61, 598-602.

Van Houdenhove, B. (1986). Prevalence and psychodynamic interpretation of premorbid

hyperactivity in patients with chronic pain. Psychotherapy and Psychosomatics,

45, 195-200.

124

Watson, D. (1990). On the dispositional nature of stress measures: Stable and nonspecific

inﬂuences on self-reported hassles. Psychological Inquiry, I, 34-36.

Wechsler, H., & Kuo, M. (2000). College students define binge drinking and estimate its
prevalence: Results of a national survey. Journal of American College Health,

49, 57-64.

Wechsler, H., Dowdall, G.W., Maenner, G., Glendhill-Hoyt, J ., and Lee, H. (2000).
College binge drinking in the 19905: A continuing problem Journal of the

American College of Health, 48, 199-210.

Wechsler, H., Dowdall, G.W., Maenner, G., Gledhill-Hoyt, J. & Lee, H. (1998). Changes
in binge drinking and related problems among American college students between
1993 and 1997: Results of the Harvard School of Public Health college of alcohol

study. Journal of American College Health, 47, 57-68.

Wechsler, H., Davenport, A., Dowdall, G.W., Moeykens, B. & Castillo, S. (1994).
Health and behavioral consequences of binge drinking in college: A national
survey of students at 140 campuses. Journal of the American Medical

Association, 272, 1672-1677.

Wechsler, H., Lee, J.B., Kuo, M., Seibring, M., Nelson, T.F. & Lee, HP. (2002). Trends

in college binge drinking during a period of increased prevention efforts:

125

Findings from four Harvard School of Public Health study surveys, 1993-2001.

Journal of American College of Health, 50, 203-217.

Wechsler, H., Molnar, B.E., Dvaenport, A.E., & Baer, J .S. (1999). College alcohol use:

A full or empty glass? Journal of American College Health, 4 7, 247-252.

Whitehead, W.E. (1994). Assessing the effects of stress on physical symptoms. Health

Psychology, 13, 99-102.

Williams, G.D., Aitken, S.S., & Malin, H. (1985). Reliability of self-reported alcohol
consumption in a general population survey. Journal of Studies on Alcohol, 46,

223-227.

Wilsnack, S.C., Klassen, A.D., & Wilsnack, R.W. (1984). Drinking and reproductive
dysfunction among women in a 1981 national survey. Alcoholism: Clinical and

Experimental Research, 8, 451-458.

Wood, M.D., Read, J .P., Palfai, T.P., & Stevenson, RF. (2001). Social inﬂuence

processes and college student drinking: The mediational role of alcohol outcome

expectancies. Journal of Studies on Alcohol, 62, 32-43.

126

Yost, J .K. & Mines, RA. (1985). Stress and Alcoholism. In T.E. Bratter & G.G. Forrest
(Eds), Alcoholism and substance abuse: Strategies for clinical interventions (pp.

74-103). New York: Free Press.

127

Appendix I
Informed Consent
SUBJECT RESEARCH INFORMATION AND _§ONSENT FORM

You are being asked to participate in a research study. This consent form provides
you with information about the study. The Principal Investigators (people in charge of
this research) and their representatives will describe this study to you and answer all of
your questions. Read the information below and ask questions about anything you do not
understand before deciding whether or not to take part.

Your participation is entirely voluntary and youican refuse to participate without
penalty or loss of beneﬁts to which you are otherwise entitled. You may take home an
unsigned copy of this consent form to think about or discus with family or friends before
making your decision. Note, you must be 18 years old or older to participate in this
research study.

What is this study? The purpose of this study is to learn more about the
characteristics of college students that contribute to college drinking habits.

What will be done if you take part in this study? If you choose to participate in
this study, you will complete paper-and-pencil questionnaires that contain questions
about drinking habits, personality, and stress. It will take about 30 minutes to complete
the questionnaires. There are no right or wrong responses to the items on the measures.

Where will participants go to complete the research questionnaires? Participants
will complete the research questionnaires in a medium-sized classroom (about 45 seats)

in the psychology building.

128

What are the possible discomforts and risks? There are no known risks involved
in completing the questionnaires and many students ﬁnd that they learn something about
themselves from answering the items. Nonetheless, if answering the questions makes
you feel uncomfortable, you may consider speaking to a counselor who may be able to
help you with your reactions. (Contact information for campus counseling services
appear at the end of this form).

What are the possible benefits to you and to others? You may beneﬁt by
participating in this study through increased awareness and self-understanding. You will
also be contributing to knowledge regarding researchers ability to understand college
student drinking habits.

Will you receive payment for your participation in this study? You will be paid
$10.00 for your participation in this research.

How will your privacy and confidentiality of your research records be
protected? Your privacy will be protected to the maximum extent permissible by law.
The following precautions will be taken to assure conﬁdentiality. Information that you
provide will not be associated with your name or any other identifying information. Your
name will not appear on any of the questionnaires and your name will not be included in
any written report. Responses will only be associated with a participant identiﬁcation
number, which is based upon the order in which the instrument was administered.
Knowledge of your participation in this study is limited to the principal investigators,
however, we will only be aware of your participation in the study and we will not know
which questionnaires you completed. Questionnaire responses and demographic

information will be stored on the Principal Investigators' personal computers and will be

129

kept in password-protected files. The computers will be kept in locked offices of the
Principal Investigators.

Who would you call if you have any questions? In the future, you may have
questions about your study participation. If you have any questions, you may contact the
Principal Investigators, Dr. Ken Rice or Ben Cohen, at 392-0601, ext. 246. Should you
have any questions or concerns about your rights as a research participant, you may
contact the ofﬁce of the University of Florida Institutional Review Board (Dr. C. Michael
Levy, UFIRB Ofﬁce, PO. Box 112250, Gainesville, FL 32611-2250, 98A Psychology
Building, 392-0433).

If you have read the information above and consider yourself to be informed about
this study's purpose, procedures, possible beneﬁts and risks, please sign your name below
indicating your agreement to participate in this research on a purely voluntary basis.
Keep the other copy of the consent form for your records.

I voluntarily agree to participate in the procedure and I have received a copy of

this description.

 

(Sign your name here) (Date)

Campus Counseling Resource: University of Florida Counseling Center, P301

Peabody Hall, 392-0433 Sudent Mental Health Services, 245 Inﬁrmary Building, 392-

1171

130

Appendix 11
Recruiting Script (Phone)

Hello, my name is Ben Cohen and I am conducting research at the University of
Florida. The purpose of my research is learn more about the characteristics of college
students that contribute to college drinking habits. I am calling to see if you may be
interested in participating in this research. Are you 18 years old or older? {If response is
"no", say "Ok. I am sorry but you can not participate in this research unless you are 18
years old or older. Thank you. Have a nice day." Do not continue with recruitment. If
response is "yes", continue...}. Would you like me to tell you more about what will be
done if you participate in the study? {If response is "no", say "Ok. Thank you.
Have a nice day." Do not continue with recruitment. If response is "yes", continue... }.
If you choose to participate in this study, you will complete paper-and-pencil
questionnaires that contain questions about drinking habits, personality, and stress. It
will take about 30 minutes to complete the questionnaires. There are no right or wrong
responses to the items on the questionnaires. You will be paid $10.00 for your
participation in this research.

There are no known risks involved in completing the questionnaires and many
students ﬁnd that they learn something about themselves from answering the items. You
may beneﬁt by participating in this study through increased awareness and self-
understanding. You will also be contributing to knowledge regarding researchers ability
to understand college student drinking habits.

Your privacy will be protected to the maximum extent permissible by law. The

following precautions will be taken to assure conﬁdentiality. Information that you

131

provide will not be associated with your name or any other identifying information. Your
name will not appear on any of the questionnaires and your name will not be included in
any written report. Responses will only be associated with a participant identiﬁcation
number, which is based upon the order in which the questionnaire was administered.
Knowledge of your participation in this study is limited to the principal investigators,
however, we will only be aware of your participation in the study and we will not know
which measures you completed. Questionnaire responses and demographic information
will be stored on the Principal Investigators' personal computers and will be kept in
password-protected ﬁles. The computers will be kept in locked ofﬁces of the Principal
Investigators.

Do you think that you may want to participate in this research? {If response is "no,"
say "Ok. Thank you. Have a nice day." Do not continue with recruitment. If response
is "yes," continue... }. OK. Here are the days and times that are available. {Give at least
four choices of days/times during which instrument will be administered}. Are you able
to participate during one of those times? {If "no" say, "OK, we will contact you in about
two weeks when we have some different times available. Would that be ok?" If "yes,"
say...) "OK. That will be in room __ of the Psychology Building."

If you have any questions, you can contact the Principal Investigators, Dr. Ken Rice
or Ben Cohen, at 392-3684. Should you have any questions or concerns about your
rights as a research participant, you may contact the ofﬁce of the University of Florida

Institutional Review Board at 392-0433.

132

 

Appendix 1]]
Demographic Information
1. Please circle the number next to your gender:
(1) MALE
(2) FEMALE
2. How old are you? _ years old

3. Please circle the number next to your Race/Ethnicity or please describe the specific

 

group that you identify with the most in the blank next to your ethnicity
(for example, Chinese American, German, Navajo, Alaskan Aleut)

(1) Asian or Asian-American

 

(2) Black, African-American

(3) Hispanic, Latino, Mexican-American

 

(4) Pacific Islander

(5) Native American or American Indian

 

(6) White, European American

(7) Multicultural/Multiracial

 

(8) Other, please specify

 

133

4. Please circle the number next to your college:

(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)

Agriculture & Life Sciences
Liberal Arts & Sciences

Business Administration
Journalism and Communications
Education

Engineering

Health and Human Performance
Health Professions

Design, Construction, & Planning

(10) Natural Resources and Environment

(1 1) Pharmacy

(12) Social Sciences

(13) Veterinary Medicine

(14) Nursing

(15) Undecided (No preference)

5. How many semesters have you completed of college?

Semesters

(do not count the current semester, put 0 if you are a ﬁrst semester student)

6. Please indicate your undergraduate Grade Point Average:

(skip this if you do not have a GPA yet)

134

GPA (4.0 scale)

 

7. Please circle the number next to your parents' marital status:
(1) Married and living together
(2) Separated
(3) Divorced and neither parent remarried
(4) Divorced and one or both parent remarried
(5) Widowed, or one parent deceased

(6) Single-parent (never been married)

 

*****The following Questions ask about your MOTHER or the person who is or was

the primapy female caregiver in your family.

8. How much education has your mother (or primary female caregiver in the family)
completed? (please circle)
(1) less than high school
(2) high school degree (or GED)
(3) post high school (e.g., trade, technical, secretarial)
(4) some college (e.g., one year, associate's degree)
(5) completed college (e.g., bachelor's degree)
(6) some graduate or post-bachelor's training

(7) completed graduate or post-bachelor's training

135

 

9. What is her current occupation? (Please name job or describe what she does or did,

even you she is laid off, disabled, or retired).

 

10. What is her current employment status? (please circle)
(1) she works full-time
(2) she works part-time (less than 30 hours per week)
(3) she does not work outside of the home because she is employed full-time
in home-making
(4) she does not work because she is laid off or unemployed
(5) she does not work because she is disabled
(6) she does not work because she is retired

(7) other (describe)

 

136

 

*****The following guestiﬂrs ask about meATHER 01' the p__erson who is

or was the primary male caregiver in your family.

 

11. How much education has your father (or primary male caregiver in the family)
completed? (please circle)
(1) less than high school
(2) high school degree (or GED)
(3) post high school (e.g., trade, technical, secretarial)
(4) some college (e.g., one year, associate's degree)
(5) completed college (e.g., bachelor's degree)
(6) some graduate or post-bachelor‘s training
(7) completed graduate or post-bachelor's training
12. What is his current occupation? (Please name job or describe what he does or did,

even you he is laid off, disabled, or retired).

 

13. What is his current employment status? (please circle)
(1) he works full-time
(2) he works part-time (less than 30 hours per week)
(3) he does not work outside of the home because he is employed full-time in
home-making
(4) he does not work because he is laid off or unemployed he does not work
because he is disabled

(5) he does not work because he is retired other (describe)

137

 

Appendix IV
9.13
For the next two questions, 9% is deﬁned as:
o 12 ounce (360 mL) can/ bottle of beer
0 5 ounce (120 mL) glass of wine
0 12 ounce (360 mL) can/ bottle of wine cooler
o 1.25 ounce (37mL) shot of liquor (straight or mixed)

Please answer the next two guestions by circling the letter next to your resmnse:
1). On how many occasions did you drink alcohol in the past 30 days?
a). 0 occasions
b). 1-2 occasions
c). 3-5 occasions
(1). 6-9 occasions
e). 10-19 occasions
f). 20-39 occasions

g). 40 or more occasions

138

 

2). In the past 30 days, on those occasions when you drank alcohol, how many
drinks did you usually have?

a). Not applicable

b). l

c). 2

d). 3

f). 5
8). 6

h). 7

j). 9+

139

 

YAAPST
Below is a list of problems related to alcohol use. Please indicate how frequently you
experienced the particular problem in the last year by recording the corresponding
number in the space next to the item. It is important to be as honest as possible.
For Questions # 1 to #8:
0 = No, never
1 = Yes, but not in the past year
2 = One time in the past year
3 = Two times in the past year
4 = Three times in the past year
5 = Four to six times in the past year
6 = Seven to eleven times in the past year
7 = Twelve to twenty times in the past year
8 = Twenty-one to thirty-nine times in the past year
9 = Forty or more times in the past year
_1. Have you driven a car when you knew you had too much to drink to drive safely?
__2. Have you ever had a headache (hangover) the morning after you had been
drinking?
3. Have you ever felt very sick to your stomach or thrown up after drinking?
4. Have you showed up late for work or school because of drinking, a hangover, or
an illness caused by drinking?
5. Have you not gone to work or missed classes at school because of drinking, a

hangover, or an illness caused by drinking?

140

 

6. Have you gotten into physical fights when drinking?
7. Have you ever gotten trouble at work or school because of drinking?
8. Have you ever been ﬁred from a job or suspended or expelled from school

because of your drinking?

For Questions #9 to #20:

0 = No, never

1 = Yes, but not in the past year
2 = One time in the past year

3 = Two times in the past year

4 = Three or more times in the past year

_9. Have you damaged property, set off a false alarm, or other things like that after
you had been drinking?
___10. Has your boyfriend/girlfriend (spouse), parent(s), or other near relative ever
complained to you about your drinking?
__11. Has your drinking ever created problems between you and your
boyfriend/girlfriend (or spouse) or another near relative?
12. Have you ever lost friends (including boyfriends or girlfriends) because of your
drinking?
13. Have you ever neglected your obligations, your family, your work, or school
work for two or more days in a row because of your drinking?

14. Has drinking ever gotten you into sexual situations which you later regretted?

141

 

15.

l6.

l7.

l8.

19.

20.

Have you ever received a lower grade on an exam or paper than you should
have because of your drinking?

Have you ever been arrested for drunken driving, driving while intoxicated, or
driving under the inﬂuence of alcohol?

Have you ever been arrested, even for a few hours, because of other drunken
behaviors?

Have you awakened the morning after a good bit of drinking and found that you
could not remember a part of the evening before?

Have you ever had "the shakes" after stopping or cutting down on drinking (for
example, your hands shake so that your coffee cup rattles in the saucer or you
have trouble lighting a cigarette)?

Have you ever felt like you needed a drink just after you'd gotten up (that is,

before breakfast)?

142

 

For Questions #21 to #27:

0 = No, never

1 = Yes, but not in the past year
2 = Yes, in the past year

21.

22.
23.
24.
25.

26.

27.

 

Have you ever found you needed larger amounts of alcohol to feel any effect, or
that you could no longer get high or drunk on the amount that used to get you I 7
high or drunk?

Have you ever felt that you needed alcohol or were dependent on alcohol?

Have you ever felt guilty about your drinking?

Has a doctor ever told you that your drinking was harming your health?

Have you ever gone to anyone for help to control your drinking?

Have you ever attended a meeting of Alcoholics Anonymous because of your

concern about your drinking?

Have you ever sought professional help for your drinking (for example, spoken

to a physician, psychologist, psychiatrist, alcoholism counselor, clergyman

about your drinking)?

143

DMQ
Below is a list of reasons people give for drinking alcohol. There are no right or wrong
answers to these questions. We just want to know about the reasons why you usually

drink when you do.

Thinking now of all the times you drink, and using the following scale record your

answer in the space next to the question.

 

1 = Never '
2 = Almost never

3 = Some of the time

4 = About half of the time

5 = Most of the time

6 = Almost always

1. How often do you drink because it helps you enjoy a party?

2. How often do you drink because you like the feeling?

3. How often would you say you drink to be sociable?

4. How often do you drink to forget your worries?

5. How often do you drink because it make social gathering more fun?

6. How often do you drink because it helps you when you feel depressed or
nervous?

7. How often do you drink to cheer up when you're in a bad mood?

144

16.

17.

18.

19.

20.

How often do you drink because your friends pressure you to drink?
How often do you drink because it's fun?

How often do you drink because it's exciting?

. How often do you drink it improves parties and celebrations?

How often do you drink so that others won't kid you about not drinking?

. How often do you get drunk to get high?

How often do you drink to celebrate a special occasion with friends?
How often do you drink because you feel more self-conﬁdent or sure of
yourself?

How often do you drink so you won't feel left out?

How often do you drink to forget about your problems?

How often do you drink because it gives you a pleasant feeling?

How often would you say you drink to ﬁt in with a group you like?

How often do you drink to be liked?

145

 

‘
ll

PSS
The questions in this scale ask you how you about your feelings and thoughts
during the last month. In each case, you will be asked to indicate how often you felt or
thought a certain way. Although some of the questions are similar, there are differences
between them and you should treat each one as a separate question. The best approach is
to answer each question fairly quickly. This is, don't try to count up the number of times

you felt a particular way, but rather indicate the alternative that seems like a reasonable

 

estimate. r _.

For each of the questions, choose from the following alternatives:

0: Never
=almost never

2=sometimes

3=fairly often

4=very often

1. In the last month, how often have you been upset because of something
that happened unexpectedly?

2. In the last month, how often have you felt that you were unable to
control the important things in you life?

3. In the last month, how often have you felt nervous and "stressed"?

146

__6.

_7.

8.

In the last month, how often have you dealt successfully with irritating
life hassles?

In the last month, how often have you felt that you were effectively
COping with important changes that were occurring in your life? p.
In the last month, how often have you conﬁdent in your ability to handle

your personal problems?

 

In the last month, how often have felt that things were going your way? F1

In the last month, how often have you found that you could not cope

with all the things that you had to do?

9.

10.

11.

l2.

l3.

14.

In the last month, how often have you been able to control irritations in
your life?

In the last month, how often have you felt that you were on top of
things?

In the last month, how often have you been angered because of things
that happened that were outside of your control?

In the last month, how often have you found yourself thinking about
things that you have to accomplish?

In the last month, how often have you been able to control the way

you spend you time?

In the last month, how often have you felt difficulties were piling up so

high that you could not overcome them?

147

MPS

Listed below are a number of statements concerning personal characteristics and traits.

Read each item and decide whether you agree or disagree and to what extent. In the

space next to the statement, mark "1" if you strongly disagree, mark "7" if you strongly

agree. If you feel somewhere in between, mark any one of the numbers between 1 and 7;

if you feel neutral or undecided, mark the midpoint, "4."

 

Strongly Disagree Slightly Neutral Slightly Agree Strongly
Disagree Disagree Agree Agree
1 2 3 4 5 7

 

 

 

 

 

 

 

 

 

 

_ 1. When I am working on something, Icannot relax until it is perfect.

2. I am not likely to criticize someone for giving up too easily.

3. It is not important that the people I am close to are successful.

4. I seldom criticize my friends for accepting second best.

5. I ﬁnd it difﬁcult to meet others' expectations of me.

6. One of my goals is to be perfect in everything I do.

7. Everything that others do must be of top-notch quality.

8. I never aim for perfection in my work.

_9. Those around me readily accept that I can make mistakes too.

__10. It doesn't matter when someone close to me does not do their absolute
best.

1 1. The better I do, the better I am expected to do.

12. I seldom feel the need to be perfect.

148

13.

14.

15.

l6.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

Anything I do that is less than excellent will be seen as poor work by
those around me.

I strive to be as perfect as I can be.

It is important that I am perfect in everything I attempt.

I have very high expectations for the people who are important to me.
I strive to be the best at everything I do.

The people around me expect me to succeed at everything I do.

I do not have very high standards for those around me.

I demand nothing less than perfection of myself.

Others will like me even if I don't excel at everything.

I can't be bothered with people who won't strive to better themselves.
It make me uneasy to see an error in my work.

I do not expect a lot from my friends.

Success means that I must work even harder to please others.

IfI ask someone to do something, I expect it to be done ﬂawlessly.
Icannot stand to see people close to me make mistakes.

I am perfectionistic in setting my goals.

The people who matter to me should never let me down.

Others think I am okay, even when I do not succeed.

I feel that people are too demanding of me.

I must work to my full potential at all time.

Although they may not show it, other people get very upset with me

when I slip up.

149

34. I do not have to be the best at whatever I am doing.

35. My family expects me to be perfect.

36. I do not have very high goals for myself.

37. My parents rarely expected me to excel in all aspects of my life.

38. I respect people who are average.

39. People expect nothing less than perfection from me.

40. I set very high standards for myself.

41. People expect more from me than I am capable of giving.

42. I must always be successful at school or work.

43. It does not matter to me when a close friend does not try their hardest.

44. People around me think I am still competent even if I make a mistake.

45. I seldom expect others to excel at whatever they do.

150

ALS-B
The following items are designed to measure certain attitudes people have toward
themselves, their performance, and toward others. It is important that your answers be
true and accurate for you. In the space next to the statement, please enter a number from
"1" (strongly disagree) to "7" (strongly agree) to describe your degree of agreement with

each item.

 

 

Strongly Disagree Slightly Neutral Slightly Agree Strongly
Disagree Disagree Agree Agree

1 2 3 4 5 6 7

 

 

 

 

 

 

 

l. I have high standards for my performance at work or at school.

2. I am an orderly person.

3. I often feel frustrated because I can't meet my goals.

4. Neatness is important to me.

5. If you don't expect much out of yourself you will never be succeed.
6. My best just never seems to be good enough for me.

_7. I think things should be put away in their place.

_8. I have high expectations for myself.

_9. I rarely live up to my high standards.

_10. I like to always be organized and disciplined.
11. Doing my best never seems to be enough.

12. I set very high standards for myself.

13. I am never satisﬁed with my accomplishments.

l4. Iexpect the best from myself.

151

 

Yes

Yes

Yes

Yes

15. I often worry about not measuring up to my own expectations.

16. My performance rarely measures up to my standards.

17. I am not satisﬁed even when 1 know I have done my best.

18. I am seldom able to meet my own standards for performance.

19. I try to do my best at everything I do.

20. I am hardly ever satisﬁed with my performance.

21. I hardly ever feel that what I've done is good enough.

22. I have a strong need to strive for excellence.

23. I often feel disappointment after completing a task because I know I could
have done it better.

24. Using the scale above, please rate the degree to which you agree that you
are perfectionistic.

No 25. Do you think of yourself as perfectionistic? Please circle.
(Note that this may apply only to areas of your life that are of
importance to you. It need not apply to all areas of your life).

No 26. Have signiﬁcant others (persons you know well) said that you
are perfectionistic or referred to you as perfectionistic?

No 27. Have you ever been referred to as perfectionistic by a
counselor or a therapist?

No 28. Have you ever entered counseling or psychotherapy in an

attempt to deal with or reduce your perfectionism?

152

 

“2.... 02.1 "'1!

 

 

llleljjl]Mlll‘rﬁlllﬁl]SI