Development and evaluation of theory-based alcohol education programs.
Excessive alcohol consumption leads to more than 1,400 unintentional deaths and 500,000 unintentional injuries every year on college campuses. It also contributes to myriad other high-risk behaviors such as drunk driving, unprotected sex, and physical fighting (Hingson, Heeren, Zakocs, Kopstein, & Weschler, 2002). The primary behavioral risk factor for alcohol-related problems is heavy episodic alcohol use or binge drinking, defined as consuming more than five drinks in one sitting for men or more than four drinks for women (Wechsler et al., 2002). According to recent national data, approximately 2 in 5 college students engage in binge drinking (Wechsler et al., 2002). In an effort to curb what many describe as an epidemic of binge drinking among students, the Centers for Disease Controls (CDC) has established goals under the auspices of "Healthy People 2010" for reducing binge drinking among high school and college students to 11% and 20%, respectively (U.S. Department of Health and Human Services, 2000).
In their review of college drinking prevention programs, Ziemelis, Bucknam, and Elfessi (2002) point out the paucity of research on theory-based interventions. This is surprising given the growing body of literature that collectively applies change theories to health behaviors. For example, the Theory of Planned Behavior (TPB) has been used to predict several health behaviors, including alcohol consumption. The TPB suggests that behavior is a function of behavioral intentions (the perceived likelihood of performing a given behavior). Three constructs directly influence behavioral intentions including: 1) attitude (positive or negative evaluations of performing a behavior), 2) subjective norms (perception of social expectations regarding adoption of a given behavior), and 3) perceived behavioral control (personal beliefs regarding the ease of performing the target behavior) (Godin & Kok, 1996). Attitudes toward a particular behavior are based on beliefs about the benefits or drawbacks that will result from engaging in such a behavior. For example, one's attitude toward drinking may be formed out of beliefs that drinking will facilitate social assimilation. On the other hand, one may also believe that a negative consequence of drinking will involve suffering a hangover the next day.
Subjective norms also play a role in forming a set of beliefs and subsequent intention to act. The degree to which a student feels drinking is accepted or even expected among peers also influences decisions associated with drinking. Finally, the individual must see the behavior as something that can be controlled. If drinking is perceived as somehow beyond ones ability to refuse, social norms and attitudinal beliefs will not come into play.
In research using the TPB, Marcoux and Shope (1997) successfully predicted use, rates of use, and abuse of alcohol among a large sample of middle school students accounting for high amounts of variance in intention to use alcohol. Schlegel, D'avernas, Zanna, Decourville, and Manske (1992) later examined the TPB constructs as predictors of problem drinking among high school students in a 12-year longitudinal study. Results showed that the TPB was effective in predicting both frequency of drinking behavior and quantity of alcohol consumed during each episode. The authors conclude that primary prevention of problem drinking should be aimed at changing beliefs and attitudes to prevent excessive drinking, as well as improving perceived behavioral control (Schlegel et al., 1992). The development and delivery of TPB based curriculum for college students is described below.
Prior to developing the curriculum, a series of six focus groups were conducted by a trained doctoral student. Focus group sessions were recorded, discussions were transcribed, and responses were coded. The first group was asked to discuss general campus issues that impacted quality of life. The results included several health-related issues, which became the starting point for the subsequent group. Each group was provided key points from the preceding group as a basis for discussion. After the sixth focus group, it was determined that no new information had been obtained and that no additional sessions were needed. Demographic data for all six groups indicate that the representation was approximately proportional to the student population. The gender distribution was 25 females and 28 males. The racial/ethnic distribution was as follows: two African Americans, four Asian Americans, one American Indian, 32 Caucasians, four Hispanics, and two International students. In addition, there was one disabled student represented in the sample.
The themes that emerged from the focus groups included: opportunities and resources to be physically active, knowledge about and availability of healthy food, financial concerns, personal safety, sexually transmitted diseases and unintended pregnancy, and alcohol and drug abuse by others. Specific suggestions for alcohol and drug use included: counseling available in residence halls, shuttle system from bars to campus, required classes on health issues for all students, and more social opportunities and resources as alternatives to parties.
Based on data from these focus groups and from the National College Health Risk Behavior Survey (CDC, 1997), a survey was developed and administered to 400 randomly selected undergraduate students. This survey emphasized quality of life concerns 1) around campus and 2) in "my own situation."
Concerns around campus included: availability of entertainment choices, availability of physical fitness facilities, drug abuse by others, alcohol abuse by others, people that smoke, safety of cyclists and pedestrians, safety of being on campus after dark, availability of healthy food, and affordability of medical care. Concerns about "my own situation" included: how my body looks, my current physical activity level, current debt, debt I will have when I graduate, feeling safe where I live and when I drive, having a trusted person to answer important health questions, and information about the food I eat.
The results of this survey indicated that students were more concerned about how consumption of alcohol by others might impact them. Campus specific behavioral and enforcement data suggest these concerns were well founded because episodic drinking was problematic on this campus. For example, data from the Office of Student Affairs indicated that the binge-drinking rate during the 1999 academic year was higher compared to regional or national binge drinking rates. Other notable markers were an increase (over a two year period) in the number of times students reported drinking in the last 30 days, and the number of students who reported being drunk three or more times in the past month. Also, many binge drinking associated behaviors increased during the same two-year period including missing classes, having unprotected sex, and driving after drinking. Finally, enforcement data indicated an increase in the number of alcohol-related violations (193 to 201) and arrests (44 to 64) from 1998 to 1999.
The curriculum was comprised of two components, an Internet Assignment and Interactive Education session. The purpose of the Internet Assignment was to provide students with core knowledge by asking them to search for and provide answers to 10 questions to facilitate participation and discussion during the classroom presentation. All assignments were completed on-line before being sent to a central database for grading. After the assignments were completed and graded, scores were automatically sent via e-mail to the student and to the instructor.
The theoretical basis of the 50-minute Interactive Education session was the Theory of Planned Behavior (TPB). A substantial portion of the curriculum was designed to change attitudes toward binge drinking for three reasons; first, students indicated that they were concerned about the impact of alcohol on quality of life, second, there was evidence that alcohol related problems had worsened, and third, a review by God in and Kok (1996) indicated that of the constructs included in the TPB, attitudes had the biggest effect size. The curriculum was presented by a trained undergraduate Peer Health Educator and consisted of an "impaired vision" activity, a presentation that described short and long term effects of alcohol consumption, alcohol laws and violations, a trivia game that addressed the misconceptions about alcohol consumption on campus, and video testimonial of a student whose life had been impacted by a drunk driving accident. The student who presented the curriculum had completed a one-year training program and was subsequently recruited to deliver the curriculum to Freshman Orientation classes.
In the current study, the TPB was used to develop a one-hour, peer-lead alcohol education curriculum designed for college freshman. The purpose of this quasi-experimental study was to examine the effects of a TPB-based, peer education intervention for binge drinking at a college campus on behavioral attitudes, subjective norms, perceived behavioral control, and behavioral intentions.
Participants (n = 34, 85% response rate) were enrolled in two sections of Freshman Orientation at a large Midwestern university. The instructors of these sections requested that the alcohol curriculum be delivered in their classes. A research assistant contacted the instructors for permission to conduct a study associated with the delivery of the alcohol curriculum. Once instructor approval was obtained; the research assistant scheduled the classes.
One week prior to delivery of the educational content, the research assistant administered an informed consent and a survey instrument (described below) using procedures approved by a University Institutional Review Board. Immediately following the curriculum delivery, participants completed the survey instrument again.
The TPB instrument developed for this study was based on Wall, Hinson, and McKee (1998) and Conner, Graham, and Moore (1999), and included 4 demographic items (gender, age, place of residence and marital status) and 31 items that were modeled after a TPB instrument designed to measure alcohol behavior. The scale used in this study had 5 subscales that corresponded to the constructs of the TPB: Attitudes, Subjective Norms and Motivation to Comply, Perceived Behavioral Control, and Behavioral Intention.
The Attitude scale was comprised of five items that measured attitudes toward binge drinking, (e.g., Binge drinking helps me: fit in, have fun, relax, forget about problems, interact with people). The response options were on a 5-point Likert scale ranging from "1--Strongly agree" to "5--Strongly disagree". The Subjective Norm scale included seven items that measured the extent to which the respondent believed seven key referents (parents, friends, significant others, doctors, professors, bosses, coworkers) felt that they should or should not binge drink. Response options were on a 7-point Likert scale ranging from "1--Strongly believe I should NOT" to "5--Strongly believe I should". Two additional response options, "6--Don't know" and " 7--Don't have" were added for this scale only. No respondent selected the latter two options so no special treatment of these responses was necessary. Smaller scores on this scale were more desirable.
The Motivation to comply scale included seven items that measured the extent to which respondents were motivated to behave consistently with the perceived beliefs of the key referents above. Response options were on a 5-point scale ranging from "1--Very important" to "5--Very unimportant". Smaller scores on this scale were more desirable.
The behavioral intention item measured intention to drink four or more drinks per sitting on three or more occasions during the subsequent 14 days. The response option ranged from "1--Very likely" to "5--Very unlikely". Smaller scores on this item were more desirable.
Data were entered into SPSS 11.5 for PC (SPSS, Chicago, IL) then checked for accuracy by verifying that the computed range for each variable was within the expected range and by visually rechecking approximately 5% of the surveys. To determine whether the survey instrument was reliable, internal reliability coefficients were computed for all scales except the one-item Behavioral Intention scales at both pre--and post-test. Coefficients ranged from .81 to .95 indicating that all the sub-scales were of good to excellent internal consistency. Accordingly, all scales were employed in subsequent analyses.
In order to check whether any of the demographic variables were associated with the variables of interest, Pearson correlation coefficients were computed between age and each of the 10 scale scores. Similarly, Spearmans Rho coefficients were computed between the scale scores and both gender and place of residence. No students in the sample were married, thus this variable was omitted from all subsequent analyses.
In order to determine effect from pre--to post-intervention, and to perform a power analysis, MANOVA was performed using all five sets of pre- and post-intervention scale scores. Use of MANOVA was indicated by checking for both the normality and sphericity assumptions. The skewness statistic, which was less than an absolute value of 1.0 for all variables, indicated that the data were normally distributed. Mauchly's test of sphericity indicated that in this sample, sphericity could not be assumed. Thus, a Greenhouse-Geisser adjustment was made prior to computing F values.
Because the MANOVA yielded significant effects, comparisons were computed using paired t-tests. The use of the multiple t-tests is sometimes accompanied by a Bonferroni adjustment that functionally makes it more difficult to find significance particularly in small samples. The rationale for this adjustment is to reduce the chance of Type I error (finding a difference when none exists). However, the usefulness of this approach has been criticized (Curran, 2000) and might possibly lead to an inflated rate of Type II error (finding no difference when one does exist). Because there were only five comparisons planned a priori, a more prudent approach that minimized error and maximized power in this analysis was to use p < .05 as the indicator of statistical significance. Finally, a stepwise regression model was run for both pre--and post-data using Attitude, Subjective Norms, Motivation to Comply, and Perceived Behavioral Control as predictors of Behavioral Intention.
Neither age nor place of residence was associated with any of the scales at either time. Gender was associated with subjective norms in the post-test condition only. Because there was only one association and because the sample was already small, the main analyses were conducted with both genders in one group. However, to determine the functional significance of this difference, variable means by gender were examined. Mean subjective norm score at post-intervention decreased (desirable) in women but increased in men. It is possible that because the class facilitator was a female, the females were more normatively influenced than the men but this is speculative.
The MANOVA was significant (Greenhouse-Geisser F = 56.5, df = 4, p = .00). The computed power for the MANOVA was 1.0 with an effect size, of .63. Paired t-tests indicated that Attitude was the only variable that changed from pre--to post-intervention (t = 2.13, p = .04). The regression model for pre-test data indicated that of all TPB constructs, only Attitude predicted Behavioral Intentions (Adj [R.sup.2] = .21, p = .004). The results for post-test data were similar. The first variable included in the second regression model was Attitude (Adj [R.sup.2] = .38, p = .000). The addition of Perceived Behavioral Control created a significant change in the variance accounted for in the model (Adj [R.sup.2] = .44, p = .037).
Binge drinking on college campuses continues to be a significant problem that causes immediate and long-term damage to student health and wellbeing. Among behavior change models the Theory of Planned Behavior has demonstrated success in predicting a variety of health behaviors, including drinking behavior. Past research indicates that of the three constructs in the TPB (Attitude, Subjective Norms, and Perceived Behavioral Control), Attitude has the largest effect size, suggesting that influencing attitudes may have the largest impact on behavioral intentions (Godin & Kok, 1996). Therefore, the present study sought to alter college students' intentions toward binge drinking by designing and testing a peer-led alcohol education curriculum emphasizing the TPB construct of attitude. Statistical analyses revealed that attitude significantly changed from the pre-test to the post-test, suggesting that the intervention may have had an impact. Additionally, it appears that in addition to Attitude, Perceived Behavioral Control may have been enhanced by the alcohol education program.
There are several strengths of the current study that are worth noting. First, the intervention was based on an initial needs assessment that confirmed a widespread student concern with alcohol abuse. Second, the curriculum and aim of the intervention were constructed based on a theoretical model, namely the Theory of Planned Behavior. Many of the past attempts by college administrators and other educators to alter drinking behavior on campuses have been poorly developed due to a lack of theory in the formation of the programs. Finally, the curriculum was delivered by a trained undergraduate peer health educator who would be more likely to engage freshmen around alcohol behavior than would a faculty member or other authority figure.
Among the limitations of this study was the non-randomized and relatively small sample. Also, there was not a comparison group for this study. Furthermore, the intervention was of a relatively low intensity. While a more intensive intervention may have created a greater impact, this is a question that must be answered by future research.
This study supports the use of a peer-led educational intervention based on the Theory of Planned Behavior to change attitudes toward binge drinking and suggestively, perceived behavioral control as well. However, further studies with better designs and larger samples are needed to confirm these findings as well as evaluate the impact on actual drinking behaviors.
Centers for Disease Control (1997). Youth risk behavior surveillance: National college health risk behavior survey--United States, 1995. Morbidity and Mortality Weekly Report, 46(SS-6), 1-54.
Connor, M., Graham, S., & Moore, B. (1999). Alcohol and intentions to use condoms: Applying the Theory of Planned Behaviour. Psychology and Health, 14, 795-812.
Curran, E. D. (2000). Multiple comparisons: Philosophies and illustrations. American Journal of Physiology: Regulatory, integrative and comparative physiology, 279(1), R1-8.
Godin G., & Kok, G. (1996). The Theory of Planned Behavior: A review of its applications to health-related behaviors. American Journal of Health Promotion, 11(2), 87-98.
Hingson, R. W., Heeren, T., Zakocs, R. C, Kopstein, A., & Weschler, H. (2002). Magnitude of alcohol-related mortality and morbidity among U. S. college students ages 18-24. Journal of Studies on Alcohol, 63(2), 136-144.
Marcoux, B., & Shope, J. (1997). Application of the Theory of Planned Behavior to adolescent use and misuse of alcohol. Health Education Research, 12(3), 323-331.
Schlegel, R. P., D'avernas, J. R., Zanna, M. P., Decourville, N. H., & Manske, S. R. (1992). Problem drinking: A problem for the Theory of Reasoned Action? Journal of Applied Social Psychology, 22, 358-385.
U.S. Department of Health and Human Services (2000). Healthy People 2010: Understanding and improving health. (2nd ed.) Washington, DC: U.S. Government Printing Office.
Wall, A. M., Hinson, R. E., & McKee, S. A. (1998). Alcohol outcome expectancies, attitudes toward drinking, and the Theory of Planned Behavior. Journal of Studies on Alcohol, 59, 409-419.
Wechsler, H. Lee, J. E., Kuo, M., Seibring, M., Nelson, T., Lee, H. (2002). Trends in college binge drinking during a period of increased prevention efforts: Findings from the 4 Harvard School of Public Health College Alcohol Study Surveys: 1993-2001. Journal of American College Health, 50(5), 203-217.
Ziemelis, A., Bucknam, R., & Elfessi, A. (2002). Prevention efforts underlying decreases in binge drinking at institutions of higher education. Journal of American College Health, 50(5), 238-252.
Troy B. Adams and Daniel R. Evans are affiliated with Arizona State University, Department of Exercise and Wellness
Rachel M. Shreffler is affiliated with Oklahoma State University, University Health Services
Katheran J. Beam is a graduate of Oklahoma State University
Correspondence concerning this article should be addressed to Troy Adams, Arizona State University East, 7350 E. Unity, Mesa, AZ 85212. Phone: 480-727-1958
|Printer friendly Cite/link Email Feedback|
|Author:||Beam, Katheran J.|
|Publication:||Journal of Alcohol & Drug Education|
|Date:||Sep 1, 2006|
|Previous Article:||Enhancing prevention programs' credibility through the use of a logic model.|
|Next Article:||Use of a virtual reality driving simulator as an alcohol abuse prevention approach with college students.|