College Binge Drinking: Deviant Versus
School of Business Administration at Widener University, Chester,
Abstract: College binge drinking is examined from the perspectives of two cul-
tures. The traditional culture views binging as deviant; the second culture pro-
motes it. In this context, logit regression is used to explore the effects of
various factors, including student employment and parental education. Employed
students are less likely to binge than are students who are not employed. Also,
students whose mother is a college graduate, but whose father is not, are more
likely to binge than other students. The prescriptions for reducing binge drinking
are different when the behavior is perceived as mainstream rather than deviant.
The research calls for the development of a process for promoting cultural change
in an environment of continually changing student leadership.
Keywords: Binge drinking, college students
Over 40% of college students have been found to engage in heavy epi-
sodic alcoholic use or binge drinking (1, 2). This heavy alcohol use poses
a serious health threat. Binge drinkers are more likely than non-binge
drinkers to engage in unplanned sexual activity, to damage property,
or to get injured. Furthermore, students at schools with higher binge
levels are more likely than students at schools with lower binge levels
to experience problems as a result of the alcohol-related behaviors of
Address correspondence to Karen Leppel, School of Business Administration
at Widener University, One University Place, Chester, PA 19013, USA. E-mail:
The American Journal of Drug and Alcohol Abuse, 32: 519–525, 2006
Copyright Q Informa Healthcare
ISSN: 0095-2990 print/1097-9891 online
Considering alcohol use from 1980 to 2004, Johnston et al. (1)
remarked that while there were significant fluctuations in binging by high
school students and young adults not enrolled in college, college binging
remained stable and at a high level. They suggest that entrenchment
of binging in the college culture may be responsible for the observed
Durkin, Wolfe, and Clark (3) expressed their analysis of binge drink-
ing in terms of the social bond, the connection between the individual and
society. Deviant behavior occurs when the social bond is weak or lacking.
The social bond has four elements: attachment, involvement, commit-
ment, and belief. In their analysis, Durkin et al. (3) consider only the tra-
ditional culture, not the secondary college partying culture. The current
article examines binge drinking from both the traditional perspective pre-
sented by Durkin et al. (3) and also from the perspective of the secondary
II. SOCIAL BOND THEORY
The first element, attachment, refers to an individual’s ties to others. In
the traditional culture, attachment would usually be to family members
who have encouraged the student to attend college and avoid deviant
behaviors such as binge drinking. According to the theory, individuals
with strong attachments would be less likely to engage in the deviant
behavior. However, in the secondary culture, the attachment might be
to a fraternity brother or sorority sister. And the student might be
encouraged by the attachment to pursue a culturally promoted activity
such as binge drinking. This analysis is consistent with the findings
of a positive relation between binging and fraternity=sorority house
residence (4, 5).
The second, element, involvement, consists of the amount of time
spent in behavior promoted by the society. The traditional culture
expects the student to spend a lot of time studying. The secondary
culture, on the other hand, perceives binging as an activity that builds
comradery but reduces time available for studying. This discussion is con-
sistent with the finding of Wechsler, Dowdall, Davenport, and Castillo (6)
that students who felt that ‘‘academic work is not very important’’ were
more likely to binge than other students. Employed students, having less
time for schoolwork and the social life of the campus, are less likely to
binge (4, 6).
The third element, commitment, represents the time, energy, and
other resources already invested in behavior promoted by the society;
the individual would not want to jeopardize this investment. In the
traditional culture, commitment could be to getting a good education and
graduating with high grades. Binge drinking could interfere with school
Insofar as higher parental education reflects higher income, parents’
education may have an income effect in which the amount of money the
student has available to spend on commitments is increased. If the stu-
dent has a commitment to the secondary partying culture, then binging
would increase. This analysis agrees with the findings of Wechsler et al.
(6) and Chaloupka and Wechsler (4). In order to determine if the effects
of mother’s and father’s education differ, the current analysis uses a four-
category parental education variable: neither parent graduated from col-
lege, only the father graduated from college, only the mother graduated
from college, and both parents graduated from college.
Also operating through the commitment element are two opposing
effects of student employment on binge drinking. First, earnings from
a job could be devoted to the social life, increasing binging. Second,
the job may generate a commitment to things besides the campus social
scene, reducing binging. Which effect dominates is not clear and may
vary with the number of hours worked.
Students at four-year institutions have a greater investment in the
college culture. They are therefore more likely to binge than students at
two-year schools (Centers for Disease Control and Prevention (7)).
The fourth element, belief, concerns the acceptance of the social
value system. Weakening the individual’s beliefs increases the likelihood
that a person will engage in deviant behavior. The beliefs of the tra-
ditional college culture include a high value on diligence and hard work.
Durkin et al. (3) reported that two measures of the belief component,
respect for authority and acceptance of conventional beliefs, were inver-
sely related to the frequency of binging. If having a mother who has a col-
lege education but a father who does not, reflects an unconventional
background and less acceptance of conventional beliefs, binging may
be greater in these cases.
Some groups of students, such as married and older students, are
less likely than single and younger students to be a part of the secondary
college culture. Married students have a separate and different life
away from the campus. Older students have generally spent some time
away from school and their younger peers and have developed a differ-
ent set of values. Thus, married and older students have been found
to binge less than single and younger students (6, 7). In addition,
Wechsler et al. (6) and the Center for Disease Control and Presentation
(CDC) (7) have reported that whites binge more than other racial=
ethnic groups. The current study therefore includes control variables
College Binge Drinking521
III. DATA AND ESTIMATION
This study used data collected by the CDC via the National College
Health Risk Behavior Survey (NCHRBS). Unfortunately, the survey
was only administered in 1995. If college binge drinking experienced dra-
matic changes over the last decade, results based on this survey would be
merely of historical interest. However, given the stability of college binge
drinking (1), the NCHRBS can shed light on this continuing problem.
Among data collected was information on whether the student in the
past 30 days consumed five or more alcoholic drinks within a couple of
hours. Other data included gender, marital status, age, race=ethnicity,
hours worked per week, parental education, whether the student lived
in a fraternity=sorority house, and whether the institution was a two-year
or four-year school. Complete information was available for these vari-
ables for 2139 women and 1314 men. Frequencies of explanatory vari-
ables are given in Table 1.
The SAS LOGISTIC procedure was used to estimate equations for
the log of the odds of binge drinking for men and women. Results of
the logit estimation are presented in Table 2. Women who worked less
than 40 hours per week were less likely to binge than women who were
not employed. Men who worked 1 to 19 hours or 40 hours or more were
less likely to binge than men who were not employed. Opposing effects
Table 1. Frequencies of variables by gender (%)
25 Years or Older
Live in Fraternity=Sorority House
Work Zero Hours
Work 1–19 Hours
Work 20–39 Hours
Work 40 or More Hours
Neither Parent College Graduate
Dad College Graduate, Mom Not
Mom College Graduate, Dad Not
Both Parents College Graduates
Number of Observations
seem to be operating here. Job involvement and work commitment push
students to binge less. However, employment brings in money and makes
binging more affordable. Why the relative sizes of the effects vary with
gender and employment level requires further research.
Concerning parental education, the lowest predicted probability of
binging was for students for whom neither parent is a college graduate.
The probability of binging was significantly higher for students for
whom the mother was a college graduate, but the father was not. The
other parental education variables had no significant impact. Less accept-
ance of conventional beliefs by students whose mother has a college
Table 2. Logit estimation results
25 Years or Older
Live in Fraternity=
Hours Work: 1–19
Hours Work: 20–39
Hours Work 40þ
Mom College Grad,
Dad College Grad,
Both Parents College
.0194.1534 1.020.0239 .1620 1.024
Note: The left-hand variable is ln[p=(1?p)], where p is the probability of bin-
ging at least once in the past 30 days.
aThis likelihood ratio test statistic compares separate specifications for males
and females to a common specification for both.
?Significant at the 10% level.
??Significant at the 5% level.
???Significant at the 1% level.
College Binge Drinking523