Alcohol consumption and positive study practices among African American college students.

Abstract

This investigation focused on the relation between college student drinking behavior and study skills, behaviors, habits, and attitudes among undergraduate students at a predominantly African American university. Students (N = 492) were administered a multimedia alcohol survey with an embedded measure of study practices. The negative and generally statistically significant correlations between study practices and drinking outcomes signify that lower scores on these study practices subscales were associated with greater alcohol use and abuse. Multiple regression analyses evaluated the contribution of study habits and attitudes to the prediction of alcohol consumption outcomes, beyond sex, age, and grades. The inclusion of measures of study practices enhanced the prediction of African American student alcohol use significantly and substantially.

INTRODUCTION

There is a growing literature on college student alcohol use and abuse, and there is a well-established body of research that examines the important role of student study practices in academic success and student feelings of self-confidence, self-efficacy, and self-competence. While we know that alcohol use and abuse are negatively correlated with academic performance (Engs, Diebold, & Hanson, 1996; Schulenberg, Bachman, O'Malley, & Johnston, 1994), there is no research that directly examines the patterns between study practices and alcohol use and abuse. It is essential that educational policymakers recognize the associations that may exist between college student alcohol consumption and study habits and attitudes in order to understand the needs of students and develop innovative programs specifically designed to reduce alcohol abuse. If important associations do exist between alcohol consumption and study practices, then academic policymakers need a complete and precise understanding of those study habits and attitudes that best predict reduced alcohol abuse. For example, is it study methods and habits, or is it student attitudes toward their education or their instructors that are associated with alcohol consumption? Answers to such questions will help policymakers, researchers, and clinicians devise more effective prevention and intervention strategies.

Student alcohol use and abuse

Research demonstrates that a host of problems--physical, educational, social, disciplinary, psychological, and legal problems--are associated with college student alcohol consumption (Gonzalez & Wiles, 1981; Walfish, Wentz, Benzing, Brennan, & Champs, 1981). For example, student alcohol consumption has been linked to depression and suicide in both high school and college students (DeSimone, Murray, & Lester, 1994) and to high-risk sexual activity (Gainey, 1993). Alcohol use and abuse rates are high, especially at predominantly white universities (Wechsler, Davenport, Dowdall, Moeykens, & Castillo, 1994); binge drinking is also a problem (Wechsler, 1996; Presley, Meilman, & Cashin 1997). Furthermore, students who drink frequently are not typically concerned about their drinking behavior (Burrell, 1992).

African American student alcohol consumption rates are significantly lower than rates for white students--1/3 the rate, in fact--and African American students do not endorse alcohol expectancies as positively as white students (Meilman, Presley, & Lyerla, 1994). Even so, a study at a predominantly African American southern university classified 13% of the student body as high-risk drinkers (Grenier, Borskey, & Fosle, 1998). More importantly, at similar levels of alcohol consumption as whites, African American adolescents (Welte & Barnes, 1987) and African American adults (Herd, 1988) experience more alcohol-related problems.

Student study skills, behaviors, habits and attitudes

According to Gadzella, Goldston, and Zimmerman (1976) and Pace (1990), the best predictor of grade-point average (GPA) is student study habits and practices. Indeed, there is extensive research that demonstrates that positive study practices are associated with academic success (Fontana, 1986; Howard, 1993; Silverman & Riordan, 1974; Trapey & Harris, 1979). This pattern holds for African American students (Tucker, 1973), as well as for students internationally (de K. Monteith & de Wet, 1984; Eikeland & Manger, 1992). Rau and Durand (2000) have recently reaffirmed these associations with their finding of a relationship between an "academic ethic"--a student worldview that emphasizes daily, methodical, disciplined, and sober study--and academic performance. However, African American students seem particularly prone to drop out because of poor study habits: At an Indiana university, 65% of the African American dropouts blamed poor study habits (Sailes, 1990). Significantly, for all students low GPA has been associated with low levels of perceived competence, security, and self-esteem (Howard, 1993).

Policy Implications

Most predisposing and protective factors for student alcohol abuse and alcohol-related problems are beyond the purview of universities and colleges--many of these factors are already in place before the student leaves home to attend college, for example, family history of alcohol abuse (Weitzman & Wechsler, 2000), peer relationships (Curran, 2000), and personality" characteristics such as social conformity and sensation-seeking (Windle, 1994). Still, there is increased effort to provide the findings of current student alcohol abuse research to academic policymakers so they can design effective programs to address student alcohol abuse (Wechsler, Nelson, & Weitzman, 2000). There is also an increased effort to familiarize faculty, and academic departments with alcohol abuse prevention concepts and approaches (Ryan & DeJong, 1998). Additionally, university programs and substance abuse prevention campaigns are now being examined for their effectiveness (Maney, Theororou, & Vasey, 2001; Higher Education Center for Alcohol and Other Drug Prevention, 2004; Wechsler, Kelly, Weizman, San Giovanni, & Seibring, 2000). Importantly, an environmental management approach is increasingly being emphasized. This approach recognizes that alcohol behavior is influenced at multiple levels--including personal, peer, institutional, community, and policy--and it suggests that a variety of environmental management strategies, including seemingly indirect academic policies such as longer library hours, recreational options, and higher academic requirements, are being developed to address student alcohol abuse (Higher Education Center for Alcohol and Other Drug Prevention, 2004).

Because study practices and study attitudes (and the feelings of competence and self-esteem they may promote) are within the reach of college officials, their role in relation to student alcohol consumption is critical to understand, especially at predominantly African American universities.

METHOD

Sample

Undergraduate students at a predominantly African American southern university were administered a confidential multimedia student alcohol survey. The analysis sample (N = 492) is a subset of a stratified random sample of students who could be located and who agreed to participate. Students ranged in age from 18 to 45 years, with an average age of 19.9 years (s = 3.5), with 165 (34%) freshmen, 122 (25%) sophomores, 126 (26%) juniors, and 79 (16%) seniors; 51% of the sample was female (n = 252). With respect to marital status, six respondents were married, five were divorced, and 481 were single; 34 had children. The majority of the students lived in on-campus dormitories (n = 442, 91%), and 36 (7%) students were members of a fraternity or sorority. Average survey administration time was 50 minutes (s = 16.0) and each student was compensated $10 after participating in the study.

Measures

The student alcohol survey is a multimedia survey administered in a laboratory setting. The respondent uses headphones connected to a computer to listen to digitally recorded introductions and instructions, which are simultaneously viewed on the computer screen. The respondent enters his or her answers directly into the computer, which selects the next questions based on a predetermined branching pattern programmed into the survey. The student alcohol survey included a number of pre-existing questionnaires and scales, thereby assuring some degree of comparability with other large national surveys.

Alcohol consumption. Questions measuring alcohol consumption were adapted from Windle (1999), the College Alcohol Study (Wechsler, 1997), and the Core Alcohol and Drug Survey (Core Institute, 1994). For each of four different beverage types (beer, malt liquor, wine and wine coolers, and liquor), survey participants were asked about the quantity and frequency of their alcohol consumption in the past 12 months. From these items, quantity-frequency indices (QFIs) were computed by multiplying the quantity by the frequency for each beverage type, as well as an aggregated Overall QFI representing consumption of all beverages combined. For example, a student who reported drinking two beers each week would have a QFI-Beer value equal to 104 and an Overall QFI of 104. Another outcome variable was developed to describe binge-drinking, operationalized as the reported number of times six or more drinks (all beverages combined) were consumed in one sitting in the past year.

Study practices. In order to measure study practices and attitudes, a subset of the 100-item Brown and Holtzman Survey of Study Habits and Attitudes (SSHA, 1984) was included in the multimedia student alcohol survey. Four dimensions of study habits and attitudes are measured by the SSHA: Delay Avoidance (DA) assesses a student's promptness in completing academic assignments, lack of procrastination, and freedom from distraction (i.e., "I put off writing reports, term papers, etc., until the last minute"); Work Methods (WM) measures a student's use of effective study procedures and skills, and efficiency in doing assignments (i.e., "I skip over the figures, graphs, and tables in a reading assignment"); Teacher Approval (TA) assesses a student's opinion of teachers and their classroom behavior and pedagogical methods (i.e., "I think that instructors expect students to do too much studying outside of class"); and Education Acceptance (EA) assesses a student's approval of educational objectives, requirements, and practices (i.e., "Unless I really like a course, I believe in doing only enough to get a passing grade"). A five-point response scale is employed for all SSHA items, where 0 = "Rarely," 1 = "Sometimes," 2 = "Frequently," 3 = "Generally," and 4 = "Almost always."

In order for the entire student alcohol survey to remain within a time limit of approximately one hour, it was decided that the length of the SSHA would be reduced by approximately 50%. Because it was important that all items be evaluated from the perspective of target respondents, a small group of undergraduate psychology students (four females and one male) was convened to select the "best" SSHA items based on (a) item appropriateness and relevance, (b) comprehension of terminology and theme, and (c) repetitiveness of theme. Item selection was an iterative process with the primary goal of reducing the length of the SSHA. Each item was considered in turn, with discussions revolving around the three general criteria, and discussions were directed toward the elimination, revision, re-wording, and streamlining of items.

Based on the three general criteria, 46 questions were selected for use: 18 DA items, 11 EA items, 10 WM items, and 7 TA items. Several items were eliminated because they were too simplistic. An example of an item revision was the DA item "I like to have a radio, record-player, or television set turned on while I am studying," which was re-worded to "I like to have a radio, stereo, or TV turned on while I am studying." Anachronistic language was reworded throughout.

The internal consistency reliabilities of the abbreviated SSHA subscales were evaluated using coefficient alpha. A coefficient alpha of 1.0 indicates a perfectly reliable scale; for research purposes, coefficient alpha should exceed 0.70. For the 18-item DA subscale, alpha was 0.79 in this sample; Brown and Holtzman (1984) reported a somewhat more acceptable internal consistency of 0.89 in a sample of freshman college students. The 11-item EA subscale yielded a coefficient alpha of 0.69, compared to Brown and Holtzman's report of 0.87. Alpha for the ten-item WM subscale was 0.74 and 0.68 for the seven items of the TA in this sample; Brown and Holtzman reported coefficient alphas of 0.87 for both of these subscales. These internal consistencies were all acceptable.

Test-retest reliability, which examines whether a behavior or trait is stable across time, was assessed by computing the zero-order Pearson correlation between the same measure on the same individuals collected at two different times. For the test-retest study, when students finished the survey, they were asked to return to the laboratory the next day for a second survey administration. This use of a 24-hour time interval departs from the typical test-retest time interval of two weeks or longer; however, the literature regarding adolescent and young adult drinking patterns indicates that college student alcohol consumption is highly variable (Bailey, Flewelling, & Rachal, 1992; Harris, Wilsnack, & Klassen, 1994)--the use of a two-week time interval would include an unignorable element of opportunistic, time-based variability, whereas the 24-hour time interval better reflects the measurement error of interest associated with subject response processes such as memory and recall.

Forty-eight students surveyed in the larger study voluntarily participated in the test-retest study; the sample was 40% female and 60% male, and the average age was 20 years old. The test-retest reliabilities for all of the abbreviated SSHA subscales were satisfactory: DA: 0.86; EA: 0.80; WM: 0.70; and TA: 0.70. Brown and Holtzman (1984) report somewhat higher four-week test-retest reliabilities ranging from 0.88 to 0.93 for the four subscales and 14-week reliabilities ranging from 0.83 to 0.88.

Statistical Analyses

Zero-order Pearson correlations were computed to assess the relationships between the student drinking variables and the abbreviated SSHA subscales, and t-tests examined mean differences between subgroups of interest (e.g., males and females, drinkers and abstainers). Standard hierarchical regression procedures (Cohen & Cohen, 1983) were used to test the contributions of individual differences and study practices variables to the set of alcohol-related outcomes. Model 1 always takes the simplest form in the hierarchical or stepwise analyses, representing the prediction of a single outcome variable from the set of covariates:

Alcohol outcome (i.e., Overall QFI) = sex + age + GPA

Model 2 is always:

Alcohol outcome = sex + age + GPA + DA + EA + WM + TA

Because Model 1 is hierarchically nested within Model 2, the fit of the two models can be directly compared using a function of the models' coefficients of determination ([R.sup.2]s). A statistically significant increase in the [R.sup.2] indicates that the inclusion of the second set of variables (here, the SSHA subscale scores) enhances the prediction of the outcome variable; a nonsignificant change in the [R.sup.2] is evidence that the inclusion of the second set of variables does not significantly improve the prediction of the outcome.

For all analyses, each alcohol outcome was modeled separately, and each SSHA subscale appeared separately in the regression model. An alpha of 0.01 was used as the criterion for statistical significance of correlations, t-tests, and regression models, but a more liberal alpha of 0.05 was used as the criterion for statistical significance of the regression parameter estimates.

RESULTS

In order to describe the distributions of the alcohol outcomes and the subscales of the SSHA, means and standard deviations were computed and are displayed in Table 1. The SSHA subscales were scored in such a way that greater numbers indicate better study habits and attitudes, with a 0 indicating the worst study practices. The maximum values obtainable were 72, 44, 40, and 28, for the Delay Avoidance (DA), Education Acceptance (EA), Work Methods (WM), and Teacher Approval (TA) subscales, respectively. When the means in Table 1 are converted into the SSHA metric and roughly compared to the most recent SSHA norms (Brown & Holtzman, 1984), they indicate very generally that this sample of students can be characterized as possessing somewhat above-average study habits and attitudes.

Sex differences

Also displayed in Table 1 are the subgroup averages for males and females on each of the alcohol outcomes and predictor variables. Statistical tests indicate that, with respect to the alcohol outcomes, males reported consuming more beer and malt liquor than females, while females reported drinking more wine than males. Only the difference in liquor consumption failed to reach statistical significance. On average, males reported drinking a total of about 280 alcoholic beverages per year (approximately 5.4 drinks per week), whereas females drank about 94 drinks per year (approximately 1.8 drinks per week). For binge-drinking, too, there was a statistically significant difference between males and females (t = -2.65, p < 0.05): Males reported binge-drinking 7.4 times a year--three times more often than females, who reported binge-drinking 2.4 times per year. With respect to study habits and attitudes, females achieved better scores on all SSHA subscales. When drinkers and non-drinkers were examined for differences in average SSHA subscale scores, both DA (t = 3.24, p < 0.05) and EA (t = 2.47, p < 0.05) obtained statistical significance, indicating that non-drinkers have a higher approval of education and are less likely to be distracted from school work or procrastinate.

Correlations

Pearson correlations between the student drinking variables and the SSHA subscales are displayed in Table 2. Although the correlations were generally small, the fact that most of the signs were negative is noteworthy, and indicates that better study practices are generally associated with less alcohol consumption. All correlations involving beer and malt liquor consumption were statistically significant, with a pattern of larger correlations between both DA and EA and the alcohol outcomes. DA and EA also correlated significantly with liquor consumption and overall alcohol consumption; however, correlations between the SSHA subscales and both wine consumption and binge drinking were small in size and not statistically significant.

Regression analyses

The multiple regression results indicated that the addition of the four SSHA subscales to the regression models enhanced the prediction of alcohol outcomes, meaning that study practices accounted for substantial variation in all dependent variables. In some cases the increase in the [R.sup.2]s was substantial; in all cases, the increase was statistically significant. Table 3 summarizes the results of the multiple regression analyses and lists the statistically significant predictor variables for each analysis. (For completeness, the two-way interactions of each SSHA subscale by sex were analyzed in order to investigate the possible joint effects of these predictors on alcohol consumption; however, no two-way interactions achieved significance and so these results are not described further.)

In the prediction of the overall alcohol consumption for all beverages combined, [R.sup.2] = 0.04 (p < 0.05) using only sex, age, and self-reported GPA (Model 1); however, the predictability increased to [R.sup.2] = 0.10 (p < 0.05) with the inclusion of the SSHA subscales (Model 2). This increase in model R-square was statistically significant ([F.sub.479] = 8.85, p < 0.0001). Although the parameter estimates for sex and GPA were statistically significant in Model 1, sex (b = 5.20, t = 3.24, p < 0.01), GPA (b = 2.13, t = 3.56, p < 0.001), DA (b = 0.38, t = 3.35, p < 0.001) and EA (b = 0.40, t = 2.24, p < 0.05) were statistically significant predictors in Model 2.

For the prediction of each alcoholic beverage type, the Model 1 R-squares were statistically significant, as were the Model 2 R-squares and the statistical tests for incremental change. It is noteworthy that the regression coefficients for EA obtained statistical significance for the prediction of beer (b = 0.12, t = 2.93, p < 0.01), wine (b = 0.10, t = 2.53, p < 0.05), and liquor (b = 0.12, t = 2.76, p < 0.01), in addition to overall alcohol consumption. Additionally, TA (b = 0.16, t = 2.34, p < 0.05) contributed to the prediction of wine consumption and WM (b = 0.12, t = 2.06, p < 0.05) to the prediction of liquor consumption.

For the prediction of binge drinking, [R.sup.2] = 0.05 (p < 0.01) using only sex, age, and self-reported GPA (Model 1); with the inclusion of the SSHA subscales in Model 2, the predictability increased to [R.sup.2] = 0.10 (p < 0.001), a statistically significant increase ([F.sub.4,167] = 3.62, p = 0.01). The parameter estimates for sex and age were statistically significant in Model 1 and in Model 2, sex (b = 0.42, t = 2.18, p < 0.05), age (b = 0.08, t = 2.53, p < 0.05), DA (b = 0.04, t = 2.68, p < 0.01), and TA (b = 0.05, t = 2.15, p < 0.05) were statistically significant.

DISCUSSION

The current study found that the SSHA subscales are generally inversely associated with student alcohol outcomes. Specifically, Delay Avoidance and Education Acceptance are negatively correlated with beer, malt liquor, liquor, and overall alcohol consumption, and Work Methods and Teacher Approval are negatively correlated with beer and malt liquor consumption. The lower a student scores on these study practices subscales, the worse are his or her self-reported study skills, habits, and attitudes, and the more alcohol is consumed. With respect to the binge-drinking measure, correlations with Teacher Approval and Education Acceptance only approached statistical significance.

In terms of predicting student alcohol consumption from the SSHA subscales and the set of background variables (student sex, age, and grade-point average), the addition of the SSHA subscales to the models significantly enhanced the prediction of alcohol outcomes. Delay Avoidance was found to be a statistically significant predictor of beer, wine, and liquor consumption, as well as overall alcohol consumption. Education Acceptance was found to be a statistically significant predictor of overall alcohol consumption but no specific beverage type. It should be noted that two of the 11 EA items had particular relevance to beliefs and behaviors often associated with alcohol consumption: "I believe that having fun is more important than studying" and to a lesser extent "I don't go to classes whenever there is something I'd rather do." The SSHA Work Methods subscale also contributed to the prediction of liquor consumption, while Teacher Approval contributed to the prediction of wine consumption and binge drinking.

The negative correlation between the SSHA subscales and student alcohol consumption, and the predictive capability, is an important empirical finding for educational policymakers to figure into comprehensive strategies designed to address student alcohol abuse. While correlations and regression models in no way denote causation, understanding this inverse association will help educational policymakers appreciate a potential protective influence for student alcohol abuse and alcohol-related problems. The fact that SSHA scores independently contributed to the prediction of alcohol outcomes beyond grade-point average, a common measure of academic success or aptitude, highlights the unique importance of study habits and attitudes and confirms that these variables are not redundant with academic achievement. Most importantly, we have here confirmatory evidence of a potential protective factor that is within the purview of academic policymakers. Attention to study practices could be implemented within the emerging environmental management approach, which recognizes that alcohol behavior is influenced by diverse factors at multiple levels (Higher Education Center for Alcohol and Other Drug Prevention, 2004).

These findings are especially pertinent for policymakers at institutions serving African American students who have suffered from the social structural disadvantages specific to the American racial experience. These social structural disadvantages have shaped a cluster of problems experienced by many African American college students, including poor study practices and study attitudes (Sailes, 1990) and an increased level of problems experienced with lower levels of alcohol consumption compared to whites (Welte & Barnes, 1987).

One direction for future research may be to examine the connection between study practices and attitudes and feelings of self-competence, self-confidence, and self-efficacy Howard's (1993) hypothesis that "Students' perception of self is highly correlated between study skills and academic performance" (p. 6) is pertinent here. Systematic attention to effective study practices and positive study attitudes may--by the indirect route of strengthening students' feelings of confidence, self-competence, and self-efficacy--help to provide an effective defense against student alcohol use and abuse. Another direction for future research might be to examine existing college programs that work to improve study practices and attitudes. Through the use of longitudinal models examining the relation between changes in specific study practices and attitudes and changes in drinking behavior, special attention could be focused on students who are successful in these interventions and whether or not their post-intervention alcohol consumption is decreased.

Author Note

Bruce dePyssler, Department of English and Mass Communication, North Carolina Central University; Valerie S. L. Williams, RTI Health Solutions, RTI International Research, Triangle Park, North Carolina; Michael Windle, Department of Psychology, University of Alabama at Birmingham.

Research supported by National Institute on Alcohol Abuse and Alcoholism and the National Center for Minority Health and Health Disparities, NIH.

Correspondence concerning this article should be addressed to Bruce dePyssler, North Carolina Central University, Department of English, 1801 Fayetteville Street, Durham NC 27707. E-mail: [e-mail].

References

Bailey, S. L., Flewelling, R. L., & Rachal, J. V. (1992). The characterization of inconsistencies in self-reports of alcohol and marijuana use in a longitudinal study of adolescents. Journal of Studies on Alcohol, 53(6), 636-647.

Brown, W. F., & Holtzman, W. H. (1984). Manual." Survey of Study Habits and Attitudes. New York: Psychological Corporation.

Burrell, L.F. (1992). Student perceptions of alcohol consumption. Journal of Alcohol and Drug Education, 37, 107-113.

Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences, 2nd ed. Hillsdale, NJ: Erlbaum.

Core Institute. (1994). Core Alcohol and Drug Survey--Long Form. Carbondale IL: Author, Southern Illinois University.

Curran, P. J. (2000). A latent curve framework for the study of developmental trajectories in adolescent substance use. In J. S. Rose, L. Chassin, C.C. Pressman, & S. J. Sherman (Eds.), Multivariate applications in substance use research: New methods for new questions (pp. 1-42). Mahwah, NJ: Erlbaum.

de K. Monteith, J. L. & de Wet, J. j., (1984) Personality and other characteristics of the underachieving child. Paper presented at the International Conference: Education for the Gifted "Ingenium 2000," Stellenbosch, Republic of South Africa.

DeSimone, A., Murray, P., & Lester, D. (1994). Alcohol use, self-esteem, depression, and suicidality in high school students. Adolescence, 29, 939-42.

Eikeland, O. J., & Manger, T. (1992). Why students fail during their first university semesters. International Review of Education, 38, 489-503.

Engs, R., Diebold, B., & Hanson, D. (1996). The drinking patterns and problems of a national sample of college students, 1994. Journal of Alcohol and Drug Education, 41, 13-33.

Fontana, D. (1986). Students' attributes of academic success. Assessment and Evaluation in Higher Education, 20, 345-365.

Gadzella, B., Goldston, J., & Zimmerman, M. (1976, April). Interrelationships of study habits and attitudes, locus of control, motivation achievement and academic achievement. (ERIC Document Reproduction Service No. ED143928)

Gainey, R. (1993). A study of undergraduate college students' alcohol consumption related to high-risk sexual activity. (ERIC Document Reproduction Service No. ED361636)

Gonzalez, G., & Wiles, W. (1981). The incidence of alcohol usage as a factor in disciplinary cases. National Association of Student Personnel Administrators Journal, 19, 33-38.

Grenier, C., Borskey, E., & Fosle, D. (1998). A survey analysis of alcohol use at a black university in the Deep South. Journal of Child and Adolescent Substance Abuse, 7, 79-92.

Harris, T. R., Wilsnack, R. W., & Klassen, A. D. (1994). Reliability of retrospective self-reports of alcohol consumption among women: Data from a U.S. national sample. Journal of Studies on Alcohol, 55(3), 309-314.

Herd, D. (1988). Drinking by black and white women: Results from a national survey. Social Problems, 35, 493-505.

Higher Educational Center for Alcohol and Other Drug Prevention, U.S. Department of Education (2004). What campuses and communities are doing. Retrieved August 19, 2005 from www.edc.org/hec/ ideasamplers/.

Howard, J. R. (1993). Study behaviors of undergraduate and graduate students. (ERIC Document Reproduction Service No. ED360581)

Maney, D. W., Theororou, E., & Vasey, J. J. (2001, March). A pilot study of an alcohol education intervention targeting the frequently binge-drinking college student. (ERIC Document Reproduction Service No. ED451150)

Meilman, P., Presley, C., & Lyerla, R. (1994). Black college students and binge drinking. Journal of Blacks in Higher Education, 4, 70-71.

Pace, R. C. (1990). The undergraduates: A report of their activities and progress in college in the 1980s. Los Angeles: Center for the Study of Evaluation, University, of California.

Presley, C. A., Meilman, P. W, & Cashin, J. R. (1997). Weapon carrying and substance abuse among college students. Journal of American College Health, 46, 3-8.

Rau, W., & Durand, A. (2000). The academic ethic and college grades: Does hard work help students to "make the grade"? Sociology, of Education, 73, 19-38.

Ryan, B. E., & DeJong, W. (1998). Making the link: Faculty and prevention. Retrieved August 19, 2005 from www.edc.org/hec/pubs/facultyprevention.html

Sailes, G. A. (1990). An investigation of black student attrition at Indiana University. (ERIC Document Reproduction Service No. ED338747)

Schulenberg, J., Bachman, J., O'Malley, P., & Johnston, L. (1994). High school educational success and subsequent substance abuse: A panel analysis following adolescents into young adulthood. Journal of Health and Social Behavior, 35, 45-62.

Silverman, M. S., & Riordan, K. (1974). Study skills training for "high risk" freshmen. Journal of College Student Personnel, 13, 63.

Trapey, E. R., & Harris, J. B. (1979). Study skills courses make a difference. Journal of College Student Personnel, 20, 62-66.

Tucker, S. J. (1973). Multiplepredictors of academic success of the black male college student. (ERIC Document Reproduction Service No. ED076167)

Walfish, S., Wentz, D., Benzing, P., Brennan, F., & Champs, S. (1981). Alcohol abuse on a college campus: A needs assessment. Evaluation and Program Planning." An International Journal, 3, 163-168.

Wechsler, H. (1996). Alcohol and the American college campus: A report from the Harvard School of Public Health. Change, 28, 20-25.

Wechsler, H. (1997). College Alcohol Survey. Boston, MA: Harvard School of Public Health.

Wechsler, H., Davenport, A., Dowdall, G., Moeykens, B., & Castillo, S. (1994). Health and behavioral consequences of binge drinking in college. Journal of the American Medical Association, 272, 1672-1677.

Wechsler, H., Nelson, T., & Weitzman, E. (2000). From knowledge to action: How Harvard's College Alcohol Study can help your campus design a campaign against student alcohol abuse. Change, 32, 38-43.

Wechsler, H., Kelly, K., Weizman, E., San Giovanni, J.P., & Seibring, M. (2000). What colleges are doing about student binge drinking: A survey of college administrators. Journal of American College Health, 48, 219-226.

Weitzman, E. R., & Wechsler, H. (2000). Alcohol use, abuse, and related problems among children of problem drinkers: Findings from a national survey of college alcohol use. Journal of Nervous and Mental Disorders, 188, 148-154.

Welte, J. W. & Barnes, G. M. (1987). Alcohol use among adolescent minority groups. Journal of Studies on Alcohol, 48, 329-336.

Windle, M. (1994). A study of friendship characteristics and problems behaviors among middle adolescents. Child Development, 65, 1764-1777.

Windle, M. (1999). Alcohol use among adolescents. Thousand Oaks, CA: Sage.

Bruce dePyssler, North Carolina Central University

Valerie S.L. Williams, RTI Health Solutions

Michael Windle, University of Alabama at Birmingham

Table 1

Descriptive statistics for alcohol outcomes and the SSHA subscales
(* p < 0.01)

                               Mean (s)      Males   Females      t

Overall QFI                 160.5 (1237.8)   280.2    94.2     -3.92 *
QFI-Beer                     45.1 (521.5)     98.1    15.7     -7.68 *
QFI-Malt liquor              40.5 (588.6)     85.6    15.4     -5.54 *
QFI-Wine                     18.4 (186.3)     15.0    20.3      3.50 *
QFI-Liquor                   56.5 (532.1)     81.4    42.7     -1.95
Binge-drinking                4.2 (43.1)       7.4     2.4     -2.65 *
Delay Avoidance (DA)         46.7 (26.5)      44.3    48.0      4.08 *
Education Acceptance (EA)    31.6 (15.9)      29.4    32.9      6.62 *
Work Methods (WM)            29.5 (15.9)      27.7    30.5      4.96 *
Teacher Approval (TA)        18.8 (12.1)      17.6    19.4      3.93 *

Table 2

Correlations among alcohol outcomes and the SSHA subscales (* p < 0.01)

      Overall     QFI-         QFI-         QFI-      QFI-       Binge
        QFI       Beer      Malt liquor     Wine     Liquor    drinking

DA    -0.23 *    -0.24 *      -0.19 *      -0.06     -0.18 *      0.03
EA    -0.23 *    -0.26 *      -0.22 *      -0.02     -0.16 *     -0.15
WM    -0.10      -0.14 *      -0.14 *       0.04     -0.05       -0.06
TA    -0.11      -0.14 *      -0.14 *       0.07     -0.10       -0.18

Table 3

Adjusted R's for multiple regression analyses (* p < 0.01)

                    Model 1      Model 2
Alcohol outcome    [R.sup.2]    [R.sup.2]

Overall QFI         0.04 *       0.10 *
QFI-Beer            0.11 *       0.15 *
QFI-Malt liquor     0.06 *       0.08 *
QFI-Wine            0.04 *       0.06 *
QFI-Liquor          0.04 *       0.08 *
Binge-drinking      0.05 *       0.10 *

                        Statistically significant
Alcohol outcome           predictors (p < 0.05)

Overall QFI        1: sex GPA;        2: sex GPA DA EA
QFI-Beer           1: sex;            2: sex DA
QFI-Malt liquor    1: sex;            2: sex
QFI-Wine           1: sex age GPA;    2: sex age GPA DA TA
QFI-Liquor         1: sex age GPA;    2: age GPA DA WM
Binge-drinking     1: sex age;        2: sex age DA TA

Note: For each alcohol outcome, the increase in predictability
from Model 1 to Model 2 was statistically significant.