Factors that predict freshmen college students' preference to drink alcohol.

Dear Editor:

Students who decide to drink do so because of many environmental and social factors. Many students begin to drink because they feel that their peers also drink or find it socially acceptable to drink. Other factors that affect student drinking are the living arrangement of the student; if the college has a presence of Greek system; the presence of advertising and price; and parental and peer attitudes about drinking (National Advisory Council on Alcohol Abuse and Alcoholism, 2002). Peers affect drinking habits more often if the student aligns himself with close peer networks that are accepting of drinking or youth drinking. "Fitting in" was reported as being a maj or motivating factor of beginning heavy episodic drinking for college students (Weitzman, Nelson, & Wechsler, H., 2003). Additionally, if the community has a lax enforcement effort students find that they will suffer no consequences for drinking and therefore will feel no negative legal effects. (Wechsler, Lee, Kuo, Seibring, Nelson, & Lee, 2002; Toomey, & Wagenaar, 1999) Research has been conducted on State alcohol control policy and the rate of heavy episodic drinking of college students compared to local residents. These variables were related. (Nelson, Naimi, Brewer, & Wechsler, 2005) The researchers did control for instate/out-of-state attending of college compared to where they attended high school during the prediction, but the variable was not examined by itself for prediction of drinking in college. Students also have more of a tendency to drink alcohol and drink to excessive rates if the student lives on campus. If a student lives in sorority or fraternity housing, the drinking rate of a student is highest. On campus dormitory living has the second highest rate of drinking. Students who live with their families usually drink less then their on-campus peers. (Wechsler, et al., 2002; Wechsler, Lee, Nelson, & Kuo, 2002)

The competitiveness of finding a good college or university and attending college away from home could adversely affect students. Students who attend college in a state other than their home state may face different factors than students who attend college in their own hometown. Students who go away to college face a new environment full of new faces. The student has not had the opportunity to evaluate these potential friends in a safe environment. The prospect of "fitting in" may have the student more susceptible to peer pressure than he or she would be at home. A previous study found that college students who lived away from home had a greater than average rate of alcohol use and conducted heavy episodic drinking. Those that lived at home while attending college had a lower rate of heavy episodic drinking and alcohol use than the average. If the student left home to attend college immediately after high school, the increase in alcohol use developed quicker (Schulenberg, O'Malley, Bachman, & Johnson, 2000).

Students who drank in high school are three times more likely to begin heavy episodic drinking in college (Weitzman, et al., 2003). The question remains if location of attending college and high school drinking behaviors are significant predictors of college drinking behaviors. The current study will try to determine that question as well as determine which factors motivate students to begin drinking in college.

The data for this study was complied from the Harvard School of Public Health College Alcohol Study (CAS) survey of 2001 (Wechsler, 2001). The CAS is a survey compiled from approximately 10,904 college-students from 120 four-year universities and colleges within 38 US states and the District of Columbia, who were randomly selected and filled out anonymous questionnaires. The total sample was 64.1% female and 35.7% male. Of the respondents, 74% were Caucasian. The response rate for the 2001 questionnaire was 52%. (Wechsler, Lee, Kuo, et al., 2002) The mean age of the respondents was 20.82 with a standard deviation of 2.045. Of the respondents, 50.2% were under the legal drinking age of 21. For the current study, only the freshmen class was used. This reduced the sample to 2497 cases, which consisted of under 18 (0.8%), 18 year olds (41.5%), 19 year olds (49.6%), 20 year olds (3.0%), and 21 and older (5.2%). There were 65.6% female and 34.4 % male freshmen in the sub-sample. Of this freshman sample, approximately one-fourth (25.7%) were attending college out of state. This is representative of the total sample where most of the students (60.8%) remained in-state status.

To determine the student's use of alcohol in high school, the researcher looked at the response to the question "within the last year of high school, how often did you drink in a typical month". To study the location of attending college, the researcher interpreted the question instrument's item "Where was the high school that you attended during your senior year located?" The answers "different state" was interpreted to mean out-of-state and from the high school or far from the guardians' residence and locus of control; the answer "same state" would be instate and close to the guardians' locus of control. These items were compared to the question, "when did you last have a drink".

The students who did not drink in high school (answered never drank in high school) and began drinking within their freshman year (answered more than 30 days ago but in the past year or more on when they had their last drink) were singled out to find demographics and similarities. Thirteen importance-scale (Very important to not at all important) survey measures were used to observe the drinking factors of these students: (a) to get away from my problems and troubles; (b) to relax or relieve tension; (c) to get drunk; (d) to have a good time with my friends; (e) there is nothing else to do; (f) to celebrate; (g) to help me get my work done; (h) I like the taste; (i) as a reward for working hard; (j) to fit in with my friends; (k) to feel more comfortable when I'm with the opposite sex; (l) everyone else is drinking; and (m) because it's cheap.

The analyses were performed using the Statistical Package for Social Scientists software (SPSS Inc., version 12.0, Chicago, IL). A multi-linear regression was conducted to see if high school drinking habits (How often in your last year of high school did you drink alcohol in a typical month) and location of college attended (high school you attended in your senior year of high school located - in state or out of state) could predict college drinking (when did you last have a drink). Since high school drinking has been shown in other studies to predict college drinking, the researcher ran the output again controlling for high school drinking (How often in your last year of high school did you drink alcohol in a typical month) to find location of attending college (high school you attended in your senior year of high school located--in state or out of state) statistically predicted college drinking (when did you last have a drink). Stepwise regressions were then conducted to see which, if any, factors (from the thirteen reasons mentioned above) predicted freshmen drinking behaviors. Confidence intervals (CI) of 95% were used to find significance.

A simultaneous multiple regression analysis was conducted to assess the prediction of college drinking based on high school drinking behaviors and location of college in relation to guardian's home. Evaluations of linearity, normality, homoscedasticity, and multicollinearity showed that the assumptions were met within acceptable limits. Regression results showed that the linear combination of amount of drinks in high school and location of college significantly predicted college drinking, [R.sup.2] = .387, [R.sup.2.sub.adj] = .385, F(7, 2149) = 193.543, p <.001. This model accounted for 38.7% of the variance in college drinking.

A partial correlation was then conducted to assess the relationship between college drinking and attending a college instate or out-of-state, controlling for high school drinking. The Pearson correlations among the variables were assessed first. Using the Bonferroni approach to control for Type 1 error across the three correlations, a p [less than or equal to] .02 was required for significance. The results of the correlational analyses presented in Table 1 show that all three correlations were statistically significant and were greater than or equal to .063. The results showed that college drinking behaviors are positively correlated to location of college, but they are more positively correlated to the students' drinking behavior in high school.

To control for the students' high school drinking behavior, a partial correlation was run between location of college and drinking behaviors in college. The results of the partial correlation are presented in Table 2. These results show that where the student attends college remained statistically significant when high school drinking behaviors were removed. The significance level did diminish a small amount, so high school drinking does have influence on the correlation but not as much as believed. The correlation between locations of college, although weak, is still significant.

The data was analyzed to see how many of the freshmen reported drinking within the last year of high school. Of the freshmen polled 44.6 percent did not drink before attending college. Of these same students, 25.3 percent were still abstainers within their first year. The students, who started drinking within this year, were then studied to decide what factors may have led them to begin drinking. Of the original 10,904 students, only 2498 were freshmen. Of these freshmen, 523 started drinking within their first year of college. More female students (70.4%) than male (29.6) students were represented in this number. Most of these students were either 18 years old (41.5 %) or 19 years old (50.1%). Of these freshmen 74.8 % attended college in the same state and 25.2 % attended college out-of-state. A stepwise multiple regression was used to evaluate which, if any, of the drinking factors could significantly predict how many times they drank within the last thirty days. The data was collected from the 319 freshmen who began drinking within their first year and answered these Likert scale questions. Four of the factors (please see Table 3) entered into the overall model, [R.sup.2] = .216, [R.sup.2.sub.adj] = .216, F (4, 315) = 21.715, p<.001. These predictors accounted for 21.6 % of the variance in how many times they drank within the last 30 days. The results indicated that students who start drinking in their first year of school are likely to do so more often if they perceive that drinking is important to have a good time with friends, to be comfortable with the opposite sex, to reward themselves for hard work, and if drinking is important to fit in with friends.

Since 70 percent of the freshmen, who began drinking within their freshman year, were female, this population was examined to see which if any of the same drinking factors could predict how many times they drank within the last 30 days. The model changed slightly. Of the 320 freshmen that answered these questions, 220 were female. The group of 220 were entered into a stepwise multiple regression and five (please see table 4) of the Likert scale predictors entered into the overall model, [R.sup.2]= .292, [R.sup.2.sub.adj] = .275, F (5, 214) = 17.640, p<.001. These predictors accounted for 29.2 % of the variance in how many times they drank within the last 30 days. The results indicated that for female freshmen who start drinking in their first year of college are likely to do so more often if they perceive that drinking is important to relax or relieve tension, celebrate, feel comfortable with the opposite sex, reward themselves for working hard, and if drinking is important to get away from troubles.

To examine the out-of-state factor of why the out of state freshmen decide to begin drinking, a stepwise multiple regression was conducted. The analysis consisted only of the out-of-state freshmen, who did not drink in high school, but did pick up drinking within their first year of school. Only 147 cases fit these criteria. The females (70.1%) still outweighed the male (29.9) population. Of these only 82 answered all the factor items, and only these were allowed into this analysis. The analysis was to evaluate which, if any, of the drinking factors could significantly predict how many times they drank within the last 30 days. Only two (please see Table 5) of the thirteen factors entered into the final model, [R.sup.2]= .241, [R.sup.2.sub.adj] = .221, F (2, 79) = 12.51, p<.001. These two predictors accounted for 24.1% of the variance in how many times the out-of-state freshmen drank within the last 30 days. The results indicated that out-of-state freshmen who begin drinking during their first year do so if they are more likely to believe that drinking is important when having a good time with friends and when they are rewarding themselves for working hard.

Only one-fifth of the freshmen class actually started drinking within their first year. This could be because they had already started drinking before college, or they were still abstaining. Only one-fourth of the students were still abstaining within their first year. This study shows that freshmen who begin drinking during their first year do so to "fit in" this correlates with the past studies in literature (Weitzman, et al., 2003).

Of all the freshmen that begin drinking within their first year of college, 70% were female. This percentage seems to be very unbalanced. A reason that the percentage may be unbalanced could be within the study itself. The study was answered by more females (approximately 64.1%) than males. This high percentage may also be related to more males began drinking before college or have not picked up drinking yet. Females who begin to drink during this time of their lives are susceptible to date rape, harassment, assault, unwanted sex and legal repercussions. When the reasoning behind female freshmen drinking in college was examined, most of the significant factors: to relax or relieve tension, to celebrate, to feel comfortable with the opposite sex, as a reward for working hard, and to get away from troubles seemed to be stress related. If this were true, the way to combat female freshmen drinking would be to find other avenues that they can relieve their

stress. Instead of teaching the students the horrors of underage drinking, teach these students how to handle stress better.

A significant correlation between college drinking behaviors and location of college was found. This finding shows that change of scenery or change of normal patterns could lead to destructive behaviors in college freshmen. High school drinking influenced the college drinking behaviors in this study as well as with the literature (Wechsler, Dowdall, Davenport, & Castillo, 1995). However, when high school drinking behaviors were factored out of the study, location of college continued to be significant. The correlation between college drinking behaviors and location were slight, but they show if a student was predisposed to other mediating factors going to school out of state and farther away from old influences may help create a college drinking behavior. Factors that lead these freshmen to be susceptible to drinking if they live out of state are of social influence. These factors include if the student feels that he or she has to have alcohol to have fun with friends and to reward his or herself. These factors seem to be confusing because one is a social factor and the other is a stress-relieving factor. This may have occurred because gender was not factored into the equation because of such a small sample number (82).

The results of the study are limited. The College Alcohol Study itself is limited because it is a self-reported and voluntary survey. A true demographic of freshman class may not exist because of the high ratio of Caucasians and females in the freshman class responses. The study was also limited because the researcher was not able to examine the distance away from guardian. In future studies, mileage away from guardian's permanent residence or living arrangements of the student in relation to parental living arrangement should be examined in the context of beginning college-drinking behaviors.

CONCLUSION

One-fifth of college freshmen begin drinking within their first year of college. The focus of college administration should be the college freshmen who begin drinking in college. These students are more likely to drink because stress and social factors influence them to pick up alcohol. Theories that will stop these students from beginning to drink alcohol in college are zero tolerance alcohol control policies to make students more accountable for their actions and the implementation of stress reduction in a mandatory health classes. This implementation will include stress reduction techniques used extensively in anxiety classes such as deep breathing, exercise, verbalization of frustration, and instruction in proper socialization skills.

The correlation between attending school out-of-state and drinking within the student's first year is not highly correlated enough to warrant any parental concerns with sending their children out of state to attend college. The parents should be more concerned with the social factors of drinking particularly if the child is worried about fitting in and having a good time with friends while attending an out-of-state college. Parents are encouraged to continue to monitor their children when they are out-of-state to know what they are doing and with whom they are doing it.

The author thankfully acknowledges Dr. Henry Wechsler for the use of the 2001 College Alcohol Study data set, Jeff Hansen for preparation of the data, Dr. Mary McNeese for consultation on the statistics and manuscript, and Carol Lee for manuscript consultation.

REFERENCES

Nelson, T.F., Naimi, T.S., Brewer, R.D., & Wechsler, H. (2005). The state sets the rate: The relationship among state-specific college binge drinking, state binge drinking rates, and selected state alcohol control policies. American Journal of Public Health, 95, 441-446.

Schulenberg, J., O'Malley, P.M., Bachman, J.G., & Johnson, L.D. (2000) "Spread your wings and fly": The course of well-being and substance use during the transition to young adulthood. In: Crockett, L.J. & Silbereisen (Eds.) Negotiating adolescence in times of social change (pp. 224-255). New York, NY: Cambridge University Press.

Task Force of the National Advisory Council on Alcohol Abuse and Alcoholism. (2002). A Call to Action: Changing the Culture of Drinking at U.S. Colleges: Final Report of the Task Force on College Drinking. (NIH Publication No. 025010). Bethesda, MD: National Institute of Health.

Toomey, T.L. & Wagenaar, A.C. (1999). Policy options for prevention: The case of alcohol. Journal of Public Health Policy, 20, 192-213.

Wechsler, H. (2001). Harvard School of Public Health College Alcohol Study, 2001 [Computer File]. Boston, MA: Harvard School of Public Health.

Wechsler, H. & Dowdall, G.W., Davenport, A., & Castillo, S. ( 1995). Correlates of College student binge drinking. American Journal of Public Health, 85, 921-926.

Wechsler, H., Lee, J.E., Kuo, M., Seibring, M., Nelson, T.F., Lee, H. (2002). Trends in college binge drinking during a period of increased prevention efforts: Findings from 4 Harvard school of public health college alcohol study surveys: 1993-2001. Journal of American College Health, 50, 203-217.

Wechsler, H., Lee, J.E., Nelson, T.F. & Kuo, M. (2002b). Underage college students' drinking behavior, access to alcohol, and the influence of deterrence policies: Findings from the Harvard School of Public Health college alcohol study. Journal of American College Health, 50, 223-236.

Weitzman, E.R., Nelson, T.F., & Wechsler, H. (2003). Taking up binge drinking in college: The influences of person, social group, and environment. Journal of Adolescent Health, 32, 26-35.

Vicki Lindsay, Administration of Justice University of Southern Mississippi

Correspondence concerning this article should be addressed to Vicki Lindsay, Department of Administration of Justice, University of Southern Mississippi, 118 College Hall # 5127, Hattiesburg, Mississippi, 39406. E-mail: Vicki.lindsay@usm. edu

TABLE 1
Correlations among Drinking Predictors

                       College drinking
                       behavior               Location of College

Location of College    .083 ***
High School drinking   .516 ***               .063 **

* p<.05

** p<.01

*** P<.001

TABLE 2
Partial Correlations among Drinking Predictors

                       College drinking behavior

Location of College    .059 ***

* p<.05

** p<.01

*** p [less than or equal to] .006

TABLE 3
Stepwise Regression Analyses to assess which, if any, drinking
factor could predict how many times freshmen who started
drinking drank within the last thirty days

                                                 Bivariate  Partial
                     B    [beta]    t       p        r         r

Have a good time
With friends       -.245  -.283   -4.872  <.001    -.393     -.265

Be comfortable
With the opposite
Sex                -.237  -.190   -3.434   .001    -.299     -.190

Reward for
Working hard       -.202  -.169   -3.073   .002    -.314     -.171

Fit in with
friends             .159   .111    2.087   .038    -.047      .117

TABLE 4
Stepwise Regression Analyses to assess which, if any, drinking factor
could predict how many times freshmen females who began drinking drank
within the last thirty days

                                                 Bivariate  Partial
                     B    [beta]    t       p        r         r

Relax or
Relieve tension    -.326  -.342   -4.840  <.001    -.432     -.314

Celebrate          -.133  -.141   -2.084   .038    -.347     -.141

Feel comfortable
With the opposite
Sex                -.201  -.171   -2.791   .006    -.317     -.187

Reward for
Working hard       -.193  -.173   -2.571   .011    -.328     -.173

To get away
From troubles       .179    .135   2.054   .041    -.116      .139

TABLE 5
Stepwise Regression Analyses to assess which, if any, drinking factor
could predict how many times out-of-state freshmen who began drinking
drank within the last thirty days

                                                    Bivariate   Partial
                     B     [beta]     t       p         r          r

Have a Good time
With friends       -.268   -.295    -2.754   .007     -.412      -.296

Reward self for
Working hard       -.327   -.290    -2.710   .008     -.409      -.292