The Journal of Psychology, 2004, 138(2), 101–114

Self-Control, Perceived Opportunity,

and Attitudes as Predictors of
Academic Dishonesty
AARON U. BOLIN
Department of Psychology and Counseling
Arkansas State University

ABSTRACT. Academic dishonesty is a persistent and pervasive problem on college cam-

puses. Researchers have suggested a variety of factors that influence academic dishonesty.
The present study is an examination of the roles of self-control, attitude toward academic
dishonesty, and perceived opportunity in predicting academic dishonesty. The dataset con-
sisted of 853 survey responses from university students across the United States. The
results showed that attitude toward academic dishonesty mediated the relationship
between self-control and academic dishonesty and also between perceived opportunity
and academic dishonesty. Implications of these findings are briefly discussed.
Key words: academic dishonesty, attitudes, cheating, general theory of crime

ACADEMIC DISHONESTY is a persistent and pervasive problem on college

campuses in the United States. Over 60 years ago, Drake (1941) reported that
23% of undergraduate students had engaged in some form of academic dis-
honesty or cheating. Although there is some debate as to whether the problem
has grown since then (for a review see Brown & Emmett, 2001), recent esti-
mates of the incidence of cheating on college campuses suggest that the
majority of all students cheat at some point in their college careers (Davis,
Grover, Becker, & McGregor, 1992; Sierles, Hendrickx, & Circle, 1980; Stern
& Havlicek, 1986).
In an effort to understand this problem, researchers have identified a variety
of factors that appear to be correlated with academic dishonesty. For example,
most researchers agree that men, students at large state-supported institutions,

I wish to thank two anonymous reviewers for several useful suggestions that were incor-
porated into this manuscript.
Address correspondence to Aaron U. Bolin, Department of Psychology and Coun-
seling, Box 1560, Arkansas State University, State University, AR 72467;
abolin@[Link] (e-mail).

101
102 The Journal of Psychology

and students with lower academic abilities cheat more often than women, stu-
dents at small private colleges, and students with higher academic abilities do
(Brown & Emmett, 2001; Davis et al., 1992). McCabe and Trevino (1997) found
that age, fraternity or sorority membership, peer approval of dishonesty, and peer
cheating were also associated with higher rates of cheating by college students.
Unfortunately, research into the causes of academic dishonesty has been largely
focused on describing relationships between variables without regard to theoret-
ical integration or explanation of the phenomenon.
My purpose in the present study was to examine the topic of academic
dishonesty within the theoretically rich and broader context of deviant behav-
ior and delinquency. Like academic dishonesty, deviant behavior has been
associated with a variety of factors including age (Gottfredson & Hirschi,
1990), gender (Mears & Ploeger, 1998), association with delinquent peers
(Empey & Stafford, 1991), self-control and perceived opportunity (Grasmick
& Tittle, 1993), and organizational identification (Eve & Bromley, 1981).
However, research into deviant behavior (unlike research on academic dishon-
esty) is often placed in the context of empirically supported theories (Bolin &
Heatherly, 2001).
One theory in particular seems to offer the hope of clarifying the nature of
academic dishonesty. According to the general theory of crime (Gottfredson &
Hirschi, 1990), lack of self-control, perceived opportunity, and the interaction
between them are the major causes of all deviant behavior, including academic
dishonesty. People who lack self-control have personalities that predispose them
to commit deviant acts (Arneklev, Grasmick, Tittle, & Bursik, 1993). When
opportunities for deviance present themselves, people who lack self-control are
unable to resist the temptation.
If previous findings on academic dishonesty are reinterpreted in the con-
text of the general theory of crime, a clearer picture of the phenomena begins
to emerge. The opportunity for academic dishonesty on a college campus is
omnipresent. This opportunity is increased further by (a) joining a fraternity
or sorority, (b) associating with peers that cheat or approve of cheating, and
(c) attending a large state-supported school (McCabe & Trevino, 1997). An
individual without self-control is very likely to cheat in such a tempting envi-
ronment.
Although the absence of self-control appears to be a valid explanation for
cheating that takes place impulsively in response to a perceived opportunity,
the general theory of crime does not explain why students with self-control do
not cheat (Gottfredson & Hirschi, 1990). Having self-control might lead stu-
dents to deliberately (rather than impulsively) cheat in circumstances that are
common on college campuses today: detection is unlikely, opportunity is high,
norms favor cheating, and cheaters have an advantage in the race for a high
GPA (Graham, Monday, O’Brien, & Steffen, 1994; Wood, Pfefferbaum, &
Arneklev, 1993). In short, lack of self-control may be sufficient to explain
 Bolin 103

cheating in an opportunity-rich environment, but having self-control does not

seem sufficient to explain why some students do not cheat when cheating may
be in their best interest (Vazsonyi, Pickering, Junger, & Hessing, 2001). In
support of this assertion, Grasmick and Tittle (1993) found that a substantial
proportion of variance in deviant behavior is left unexplained by the variables
of self-control and opportunity alone; the general theory of crime could not
fully explain the phenomenon. If Grasmick and Tittle’s results hold for all
deviant behavior, then an additional variable or variables may also be needed
to explain academic dishonesty.
Some evidence suggests that this additional variable may be attitude toward
academic dishonesty. For example, Piquero and Tibbets (1996) found that the
effect of self-control on deviance was mediated by attitudinal variables such as
perceived pleasure and perceived shame for the act. Bolin and Heatherly (2001)
found that attitudes toward deviant behavior were good predictors of actual
behavior in two large samples. In addition, Davis et al. (1992) showed that atti-
tudes toward academic dishonesty and perceived opportunity for academic dis-
honesty both had an impact on cheating among college students.
I based the present study on the premise that the general theory of crime, as
proposed by Gottfredson and Hirschi (1990), is an inadequate explanation of
cheating among college students unless attitude toward academic dishonesty is
added to the model. Haines, Diekhoff, LaBeff, and Clark (1986) found prelimi-
nary support for this revision to the general theory of crime model. Using step-
wise regression, they found that age (a correlate of self-control), attitudes, and
perceived opportunity all made significant and independent contributions to the
prediction of academic dishonesty. However, the use of stepwise regression did
not allow Haines et al. to test for mediation.
On the basis of prior research and the preceding discussion, I hypothesized
that the relationship between self-control and academic dishonesty would be
mediated both by perceived opportunity and attitude toward academic dishon-
esty. The proposed causal chain that flows from self-control to perceived oppor-
tunity to academic dishonesty is consistent with the explanation for deviant acts
offered by the general theory of crime. The proposed causal chain that flows
from self-control to attitudes toward academic dishonesty is the suggested mod-
ification to the general theory of crime, especially with regard to academically
dishonest behavior.
Figure 1 summarizes the proposed relationships among these four variables.
A nondirectional path between the error terms of perceived opportunity and atti-
tude toward academic dishonesty is included in this model to represent the cor-
relation between these two variables after controlling for the effects of self-con-
trol. Because there appears to be no empirical evidence or theoretical rationale
that suggests which of these two variables should come first in the causal chain,
this relationship is left free to vary.
104 The Journal of Psychology

Perceived
opportunity

Low Academic
self-control dishonesty

Attitude towards
academic dishonesty

FIGURE 1. Proposed model of relationships among variables.

Method

Participants

Participants were recruited from colleges and universities around the United
States through various internet sites and psychology instructors. The initial data
set consisted of 853 student responses to an internet survey but was reduced by
deleting incomplete (n = 10) and duplicate responses (n = 44). A response was
considered incomplete if the participant did not respond to 20% or more of the
items. A response was considered duplicate if two or more responses originated
from the same internet protocol address and all items, including the participants’
birthdays, were identical. The final dataset consisted of 799 responses from col-
lege students around the United States. The median age of participants was 20
years; 62.1% reported a birth date that would make them a traditional-age col-
lege student (ages 18–22). Nearly 70% (n = 554) were women, and 64% (n =
508) were freshmen or sophomores.

Measures

The Academic Dishonesty Scale consists of nine behavioral items adapted

from McCabe and Trevino (1997). Participants are asked to indicate how often
they had engaged in each academically dishonest behavior since beginning their
college careers using a 5-point Likert-type scale ranging from not even one time
(1) to many times (5). The internal consistency reliability estimate based on the
current sample for the Academic Dishonesty Scale suggested an adequate level
of reliability (α = .90; see Table 1).
 Bolin 105

The Perceived Opportunity Scale consists of eight items adapted from

McCabe and Trevino (1997) that deal with the participants’ perceptions of the
frequency and acceptability of academically dishonest behaviors at their home
institution and the likelihood of academic dishonesty being detected. Participants
used a 5-point Likert-type scale ranging from strongly agree (1) to strongly dis-
agree (5). Previous research has shown that the items on the Perceived Opportu-
nity Scale were valid predictors of academic dishonesty (McCabe & Trevino).
The internal consistency reliability estimate based on the current sample for the
Perceived Opportunity Scale suggested an adequate level of reliability (α = .73;
see Table 1).
The Attitude Toward Academic Dishonesty Scale contains four items adapt-
ed from Davis et al. (1992) that deal with participants’ moral evaluations of
cheating. Participants used a 5-point Likert-type scale ranging from strongly
agree (1) to strongly disagree (5). Previous research has shown that items on the
Attitude Toward Academic Dishonesty Scale are valid predictors of academic
dishonesty (Davis et al.). The internal consistency reliability estimate based on
the current sample for the Attitude Toward Academic Dishonesty Scale suggest-
ed an adequate level of reliability (α = .75; see Table 1). Items on these three
scales are contained in the Appendix.
I also used the Self-Control Scale, which consists of 24 items (for a list of
items see Grasmick & Tittle, 1993). The items measure the six facets of self-con-
trol first proposed by Gottfredson and Hirschi (1990): impulsivity, preference for
physical activity, risk taking, self-centered, preference for simple tasks, and tem-
per. Participants used a 5-point Likert-type scale ranging from strongly agree (1)
to strongly disagree (5) to rate each item. Previous research has shown that the
Self-Control Scale is a valid predictor of many types of deviant behavior (see
Grasmick & Tittle). The internal consistency reliability estimate based on the
current sample suggested an adequate level of reliability (α = .83; see Table 1).

Design and Procedure

Survey order and distribution. I combined items from each scale into a single sur-
vey and posted it on the internet. Participants’ responses to the survey were auto-
matically appended to a database set up for that purpose, and notices were then
sent to several Web sites that advertise online studies. I sent additional notices to
psychology instructors around the country via e-mail suggesting the survey as an
extra credit assignment. Many instructors indicated, via e-mail, that their stu-
dents would be allowed to complete the survey for extra credit.
Participants completed the scales in the following order: Perceived Oppor-
tunity, Attitude Toward Academic Dishonesty, Self-Control, and Academic Dis-
honesty. I computed composite scores for each scale as a unit-weighted sum of
all items. If the participants left any survey responses blank, they were encour-
aged, but not required, to provide responses to all items. All responses were com-
106 The Journal of Psychology

TABLE 1. Descriptive Statistics (N = 661)

Variable α SD 1 2 3 4

1. Academic dishonesty .897 .133 — .351 .506 .287

2. Perceived opportunity .730 4.813 — .439 .135
3. Attitude toward
academic dishonesty .753 .176 — .436
4. Self-control .834 11.179 —

Note. All correlations are significant at p < .01.

pletely anonymous, but participants had the option of printing a generic receipt
after their responses had been logged for the purpose of obtaining course credit.

Estimation method and fit criteria. I tested the hypothesis with a path analysis
and the LISREL software package (Jöreskog & Sörbom, 1989). As input for this
program, I computed a variance–covariance matrix by using listwise deletion of
missing data. Measurement error was accounted for in all models by using the
standard practice of fixing the error variance of observed variables to [(1 – relia-
bility) × the variance] and fixing the path between each latent construct and its
observed indicator to the square root of the indicator’s reliability. All parameters
were estimated using maximum likelihood estimation.
An assessment of the overall model fit was based on both absolute and incre-
mental fit indices. Absolute indices include the chi-square likelihood ratio test, the
standardized root mean residual (SRMR), and the root mean square error of
approximation (RMSEA; Cudeck & Browne, 1983; Mulaik et al., 1989; Steiger,
1988). A good fit of the model was indicated by a nonsignificant chi-square, a
SRMR of less than .05 and a RMSEA of less than .05 (Browne, 1982). I used the
Non-Normed Fit Index (NNFI; Tucker & Lewis, 1973) to compare alternative
models and the Comparative Fit Index (CFI; Bentler, 1990) to compare the non-
central chi-square with the null model. Given the relatively low number of indi-
cators in the models being tested, fairly conservative cutoffs of .95 were used for
both indices. In addition to the proposed model, I also computed parameter esti-
mates for several alternative models (Medsker, Williams, & Holahan, 1994).

Results

Descriptive Statistics

Table 1 shows the means, standard deviations, internal consistencies, and

intercorrelations for the variables under investigation. A visual inspection of fre-
 Bolin 107

quency histograms and computation of skew and kurtosis statistics for each
variable revealed that academic dishonesty and attitude toward academic dis-
honesty were both negatively skewed. As a result, both variables were subject-
ed to logarithmic transformation before any analysis in an effort to normalize
their distributions. The distributions of self-control and perceived opportunity
were both approximately normal.

Model Fit

A summary of fit indices for the saturated model, the proposed model, the next-
most-likely alternative model, and the null model is given in Table 2. The saturated
model and null model are included only as a point of comparison: The saturated
model provided the best possible fit to the data and the null model provided the
worst possible fit to the data. The proposed model differed from the saturated model
by constraining one relationship to zero, the direct path from self-control to acade-
mic dishonesty. (If there is no direct path between self-control and academic dis-
honesty, then the relationship between these two variables in the current sample is
mediated by perceived opportunity and attitude toward academic dishonesty.) The
increment in the fit function associated with this change was nonsignificant, χ2(1, N
= 661) = 0.041, p = .84, which provided empirical support for the decision to con-
strain this path. Self-control did not have a direct effect on academic dishonesty. All
of the fit indices for the proposed model also met the a priori standards for good fit;
the proposed model provided a very good fit to the actual data. In addition, all struc-
tural paths in the proposed model were statistically significant except the path from
perceived opportunity to academic dishonesty.
Parameter estimates and fit indices were also calculated for the next-most-
likely alternative model. The next-most-likely alternative model differed from
the proposed model by constraining one relationship to zero, the direct path from
perceived opportunity to academic dishonesty. The increment in the fit function
associated with this change was nonsignificant, χ2(1, N = 661) = 3.209, p = .20,
which provided empirical support for the decision to constrain this path. All of
the other fit functions also met the a priori standards for good fit; the next-most-
likely alternative model provided a very good fit to the actual data. In addition,
all structural paths in this model were statistically significant.
Because the proposed model and the next-most-likely alternative model both
provided an acceptable fit to the data, selecting the most appropriate model was
somewhat more complicated. The choice between models must be based on both
empirical and theoretical considerations. On theoretical grounds, it makes sense
for the relationship between perceived opportunity and academic dishonesty to
be dependent on each individual’s attitude toward academic dishonesty; noticing
an opportunity to cheat is unlikely to lead to cheating behavior unless an indi-
vidual also has a favorable attitude toward cheating. Although the difference in
empirical fit between the proposed model and the next-most-likely alternative
 108

TABLE 2. Descriptive Statistics (N = 661)

Fit indices
Model Description χ2 df RMSEA SRMR NNFI CFI
The Journal of Psychology

Saturated model All predictors have direct and indirect 0.00 0 .000 .000
paths to academic dishonesty; best
possible fit
Proposed mediated
model No direct path from self-control to 0.04 1 .000 .002 1.000 1.000
academic dishonesty
Next-most-likely
alternative model No direct path from perceived oppor- 3.21 2 .031 .014 .994 .998
tunity to academic dishonesty
Null model No relationships between any of the 571.20** 6 .378 .294 .122 .122
latent variables; worst possible fit

Note. RMSEA = root mean square error of approximation; SRMR = standardized root mean residual; NNFI = Non-Normed Fit Index; CFI = Com-
parative Fit Index.
*p < .05. **p < .01.
 Bolin 109

model was relatively small, the greater parsimony, uniformly significant paths,
and overall good fit of the next-most-likely alternative model made it preferable
to the proposed model. This result suggested that the proposed direct relationship
between perceived opportunity and academic dishonesty could be removed from
the model with little consequence, empirically or theoretically (see Figure 2).

Discussion

Interpretation

The proposed relationships depicted in Figure 1 were generally confirmed in

the current study; there was no direct relationship between self-control and aca-
demic dishonesty. However, the proposed model was not selected as the preferred
model for the data in the current sample. In particular, the relationship between
perceived opportunity and academic dishonesty in the proposed model was not
necessary. The next-most-likely-alternative model (Figure 2), a revision of the
proposed model that deleted this unnecessary path, was selected as the preferred
model for the data in the current sample. Overall, the current results suggest that
attitude toward academic dishonesty plays a critical role in the explanation of
academic dishonesty; nearly 40% of the variation in academic dishonesty is
explained by its relationship with attitude toward academic dishonesty in the cur-
rent sample.

.966**

Perceived
.507**

opportunity
1.00** *
6* R2 = .04 .606**
.18

Low Academic
self-control dishonesty
.54 .699** R2 = .39
8**

9*
*
.62

Attitude towards
academic dishonesty
R2 = .30

FIGURE 2. Parameter estimates for the next-most-likely alternative model.

The standard solution is shown here. Values on each directional path are stan-
dardized beta weights, χ2(2, N = 661) = 3.25, p = .197, RMSEA = 0.031.
*p < .05. ** p < .01.
110 The Journal of Psychology

Mounting empirical evidence (Arneklev et al., 1993; Piquero & Tibbets,

1996; Vazonyi et al., 2001) and the current results suggest that the general theo-
ry of crime is an inadequate explanation of deviant acts such as academic dis-
honesty. However, the current results demonstrate that the theory is viable with
the addition of attitudes as an intervening variable. From a logical standpoint, the
addition of attitudes to the model makes sense. Having a favorable attitude
toward academic dishonesty or any deviant behavior should make it easier to take
advantage of opportunities for the behavior (Haines et al., 1986). Having a favor-
able attitude toward academic dishonesty should also clear the way for individu-
als with or without self-control to act, some impulsively and others deliberately.
Of course, it is also possible that attitudes toward cheating are the result of cheat-
ing behavior rather than the cause, or even that attitudes toward cheating and
cheating behavior are reciprocally determined.

Implications

These findings have important implications for interventions aimed at reduc-

ing academic dishonesty. Self-control is a relatively stable personality trait that
is shaped in childhood (Gottfredson & Hirschi, 1990). If the general theory of
crime was accurate, and self-control really was the primary cause of academic
dishonesty and other deviant acts, then intervention would need to take place
well before students enter college. According to the general theory of crime,
interventions aimed at increasing self-control would probably not be very effec-
tive in reducing academic dishonesty once students reach college age.
Perceived opportunity for academic dishonesty is also difficult to change.
Interventions aimed at reducing the perceived opportunity for academic dishon-
esty might include an increase in surveillance, an increase in sanctions, and the
disruption of deviant social networks. However, determined cheaters would like-
ly create new opportunities. In addition, reducing the perceived opportunity for
academic dishonesty would require eternal vigilance; dishonesty rates would be
expected to return to previous levels if the intervention was removed. If the gen-
eral theory of crime was accurate and perceived opportunity mediated the rela-
tionship between self-control and academic dishonesty, then increasing sanctions
and supervision in an effort to deter academic dishonesty would be successful
only as long as the intervention remained in place.
However, the current results suggest that the general theory of crime is not
accurate unless attitudes are added to the model. Because attitudes are less endur-
ing than personality traits like self-control and require less frequent intervention
than the eternal vigilance needed to reduce opportunity, interventions aimed at
influencing student attitudes toward academic dishonesty would seem to have a
higher likelihood of success at a much lower cost. In fact, the recent trend toward
the development of honor codes to deter academic dishonesty may be effective,
in part, by influencing student attitudes toward academic dishonesty (McCabe &
 Bolin 111

Trevino, 1993). Student attitudes also seem amendable to change through inter-
ventions such as education (Ames & Eskridge, 1992). Furthermore, Uhlig and
Howes (1967) found that students were less likely to take advantage of opportu-
nities to cheat if they had negative attitudes toward academic dishonesty.

Limitations

Several limitations of the current study are worth noting. First, the use of
self-report data, especially for sensitive topics such as academic dishonesty, rais-
es several questions about the accuracy of the data. Also, because participation
was anonymous and recruitment efforts were not uniform throughout the coun-
try, it seems unlikely that the current sample is representative of the total popu-
lation of college students in the United States. In addition, most of the students
in the current sample were enrolled in a psychology course. Students who take
psychology courses and participate in extra credit assignments are a highly selec-
tive sample. However, this limitation may be more imagined than real consider-
ing that Eskridge and Ames (1993) showed similar results for a sample of mixed
academic majors.
A closely related problem lies in the analysis decisions to use composite
measures, to transform skewed variables, and to compute the variance–covari-
ance matrix using listwise deletion of missing data. It is possible that repeating
the analyses using other analysis strategies may slightly alter some path esti-
mates in the final model. To guard against this possibility, the analyses were
repeated using a variety of different analysis strategies. Because there were no
appreciable differences in the results, this limitation is probably not a serious
threat to the validity of these findings.
Finally, many potentially confounding variables were left unexplored or
unmeasured by the current design. In particular, group differences in the inter-
relationships among variables may exist by gender, ethnicity, and type of institu-
tion. As a result of these limitations, these findings may not generalize to other
samples.

Future Directions

Future researchers in this area should continue to probe the limits of the gen-
eral theory of crime and the explanation of deviant acts such as academic dis-
honesty. One need in this area is for methods that allow a more direct measure of
deviant behavior. Although the current study made it possible for students to give
anonymous reports of cheating, the data still suffers from the limitations of all
self-reported data. In addition, the current study should be expanded to include
noted demographic predictors of academic dishonesty. It would be interesting to
see whether demographic predictors of academic dishonesty, such as gender and
age, predict incremental variance in academic dishonesty after controlling for
112 The Journal of Psychology

attitudes. It may also be important to determine whether these relationships are

similar for other populations such as high school students and students at uni-
versities in other countries.
In addition, a logical next step in this line of research is to design an interven-
tion that targets student attitudes toward academic dishonesty and then evaluate the
effectiveness of this intervention in reducing academic dishonesty. As a prelude to
this initiative, it may be necessary to first examine how attitudes toward cheating
develop and change over the course of an academic career. Finally, these findings
should be replicated using a larger and more representative sample.

REFERENCES
Ames, G., & Eskridge, C. (1992). The impact of ethics courses on student attitudes and
behavior regarding cheating. Journal of College Student Development, 33(6), 556.
Arneklev, B. J., Grasmick, H. G., Tittle, C. R., & Bursik, R. J. (1993). Low self-control
and imprudent behavior. Journal of Quantitative Criminology, 15, 307–331.
Bentler, P. M. (1990). Comparative fit indices in structural models. Psychological Bul-
letin, 10, 238–246.
Bolin, A., & Heatherly, L. (2001). Predictors of employee deviance: The relationship
between bad attitudes and bad behavior. Journal of Business and Psychology, 15(3),
405–418.
Brown, B. S., & Emmett, D. (2001). Explaining variations in the level of academic dis-
honesty in studies of college students: Some new evidence. College Student Journal,
35(4), 529–539.
Browne, M. W. (1982). Covariance structures. In D. M. Hawkins (Ed.), Topics in multi-
variate analysis. Cambridge, England: Cambridge University Press.
Cudeck, R., & Browne, M. W. (1983). Cross-validation of covariance structures. Multi-
variate Behavioral Research, 18, 147–167.
Davis, S. F., Grover, C. A., Becker, A. H., & McGregor, L. N. (1992). Academic dishon-
esty: Prevalence, determinants, techniques, and punishment. Teaching of Psychology,
19(1), 16–20.
Drake, C. A. (1941). Why students cheat. Journal of Higher Education, 12, 418–420.
Empey, L. T., & Stafford, M. C. (1991). American delinquency: Its meaning and con-
struction. Belmont, CA: Wadsworth.
Eskridge, C., & Ames, G. A. (1993). Attitudes about cheating and self-reported cheating
behaviors of criminal justice majors and noncriminal justice majors: A research note.
Journal of Criminal Justice Education, 4(1), 65–78.
Eve, R., & Bromley, D. G. (1981). Scholastic dishonesty among college undergraduates:
Parallel test of two sociological explanations. Youth and Society, 13, 3–22.
Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stan-
ford University Press.
Graham, M. A., Monday, J., O’Brien, K., & Steffen, S. (1994). Cheating at small colleges:
An examination of student and faculty attitudes and behaviors. Journal of College Stu-
dent Development, 35, 255–260.
Grasmick, H. G., & Tittle, C. R. (1993). Testing the core empirical implications of Gott-
fredson and Hirschi’s general theory of crime. Journal of Research in Crime and Delin-
quency, 30(1), 5–30.
Haines, V. J., Diekhoff, G. M., LaBeff, E. E., & Clark, R. E. (1986). College cheating:
Immaturity, lack of commitment and the neutralizing attitude. Research in Higher Edu-
 Bolin 113

cation, 25(4), 342–354.

Jöreskog, K. G., & Sörbom, D. (1989). LISREL: A guide to the program and applications
(2nd ed.). Chicago: SPSS.
McCabe, D. L., & Trevino, L. K. (1993). Academic dishonesty: Honor codes and other
contextual influences. Journal of Higher Education, 64(5), 522–537.
McCabe, D. L., & Trevino, L. K. (1997). Individual and contextual influences on acade-
mic dishonesty: A multicampus investigation. Research in Higher Education, 38(3),
379–397.
Mears, D. P., & Ploeger, M. (1998). Explaining the gender gap in delinquency: Peer influ-
ence and moral evaluations of behavior. Journal of Research in Crime and Delinquen-
cy, 35(3), 251–267.
Medsker, G., Williams, L. J., & Holahan, P. J. (1994). A review of current practices for
evaluating causal models in organizational behavior and human resources management.
Journal of Management, 20, 439–464.
Mulaik, S. A., James, W. S., Van Alstine, J., Bennett, N., Lind, S., & Stilwell, C. D. (1989).
Evaluation of goodness-of-fit indices for structural equation models. Psychological
Bulletin, 105, 430–445.
Piquero, A., & Tibbets, S. (1996). Specifying the direct and indirect effects of low self-
control and situational factors in offenders’ decision making: Toward a more complete
model of rational offending. Justice Quarterly, 13, 481–510.
Sierles, F., Hendrickx, I., & Circle, S. (1980). Cheating in medical school. Journal of
Medical Education, 55, 124–125.
Steiger, J. H. (1988). Aspects of person–machine communication in structural modeling
of correlations and covariances. Multivariate Behavioral Research, 23, 281–290.
Stern, E. B., & Havlicek, L. (1986). Academic misconduct: Results of faculty and under-
graduate student surveys. Journal of Allied Health, 5, 129–142.
Tucker, L. R., & Lewis, C. (1973). A reliability coefficient for maximum likelihood fac-
tor analysis. Psychometrika, 38, 1–10.
Uhlig, G. E., & Howes, B. (1967). Attitude toward cheating and opportunistic behavior.
The Journal of Education Research, 60(9), 411–412.
Vazsonyi, A. T., Pickering, L. E., Junger, M., & Hessing, D. (2001). An empirical test of
a general theory of crime: A four-nation comparative study of self-control and the pre-
diction of deviance. Journal of Research in Crime and Delinquency, 38(2), 91–132.
Wood, P. B., Pfefferbaum, B., & Arneklev, B. J. (1993) Risk-taking and self-control: Psy-
chological correlates of delinquency. Journal of Criminal Justice, 16, 111–130.

APPENDIX
Scale Items

Perceived Opportunity (adapted from McCabe & Trevino, 1997)

Plagiarism and cheating on tests occur frequently at this school.
I have personally observed another student cheating on a test many times at this
school.
My closest friend would strongly disapprove if he/she found out I had cheated in a
course.
A typical student at this school would strongly disapprove if he/she found out I had
cheated in a course.
A typical student at this school would report someone who had cheated on a test.
The penalties for academic dishonesty at this school are severe.
The faculty understand the policies on academic dishonesty.
The faculty support the policies on academic dishonesty.
114 The Journal of Psychology

Attitude Toward Academic Dishonesty (adapted from Davis et al., 1992)

It is “wrong” to cheat.
Students should go ahead and cheat if they know they can get away with it.
Students should try to cheat even if their chances of getting away with it are very
slim.
I would let another student cheat off my test if he/she asked.
Academic Dishonesty (adapted from McCabe & Trevino, 1997)
Copied material and turned it in as your own work.
Used unfair methods to learn what was on a test before it was given.
Copied a few sentences of material from a published source without giving the author
credit.
Helped someone else to cheat on a test.
Collaborated on an assignment when the instructor asked for individual work.
Copied from another student during a test.
Turned in work done by someone else.
Received substantial help on an individual assignment without the instructor’s
permission.
Cheated on a test in any way.
Used a textbook or notes on a test without the instructor’s permission.

Original manuscript received April 18, 2003

Final revision accepted October 6, 2003