Study
Dietz (2016) conducted a randomized controlled trial at two public high schools within the Orange County Public Schools district in central Florida. Participants were juniors and seniors in high school, and the intervention was delivered during their regularly scheduled government or economics classes. The treatment group received the ALC; the control group received a similarly structured presentation about body image. The presentations were both 45 minutes.
A total of 180 high school students participated in the evaluation. The sample was 57 percent female, and 62 percent identified as white, 27 percent as Hispanic, 18.9 percent as African American, 4.4 percent as Asian or Southeast Asian, 2.7 percent as Hawaiian/Pacific Islander, 1.7 percent as American Indian or Alaska Native, and 10 percent as biracial or other. No statistically significant differences were found between groups on demographic variables, baseline expectancies, or baseline alcohol consumption.
Alcohol consumption was measured using the timeline follow-back procedure to recall drinking behavior over the last 30 days. Alcohol expectancies were measured using the CEOA. A mixed-model application of GLM, using baseline alcohol expectancies or baseline alcohol consumption as a covariate, was used to analyze between-group differences. The study author was independent of the program. The study author conducted subgroup analyses on differences between male and female students.
Study
Fried and Dunn (2012) conducted a randomized controlled trial at a large state university in Orlando, Florida. Participants were members of fraternities at the university. The treatment group received the ALC intervention, and the control group received a media literacy presentation on advertisements for personal appearance products. The presentations were both 50 minutes.
Four fraternity chapters were included in the study and were randomly assigned to the ALC condition (two chapters, n = 148) or the control condition (two chapters, n = 102). Of the 235 participants who completed baseline data, 209 completed the 4-week follow-up data. The mean age of participants was 19.95 years. Approximately 78.5 percent identified as white, 15.3 percent as Hispanic, 1.4 percent as Asian American, 1 percent as African American, and 3.3 percent as other. There were no statistically significant differences between the groups on age, class standing, or ethnicity at baseline. There was a statistically significant main effect for peak BAC, mean drinks per sitting, and peak drinks per sitting between groups. There was also a statistically significant main effect for the self-perception subscale of the CEOA. Baseline differences between groups were taken into account during later analyses.
Daily alcohol consumption for the 4 weeks prior to baseline and follow up was measured using the retrospective, self-report, timeline follow-back procedure. Alcohol expectancies were measured using the CEOA. Between-group differences in alcohol expectancies and alcohol consumption were analyzed using ANCOVAs with baseline alcohol expectancies or baseline alcohol consumption as covariates. The study authors did not conduct subgroup analyses.
Study
Fried (2010) conducted a quasi-experimental study at a large state university in Orlando, Florida to study the effectiveness of the Alcohol Literacy Challenge (ALC). Participants were eligible for the study if they were members of one of four fraternities or four sororities. The study author administered either the expectancy challenge presentation to the treatment group or the educational presentation to the control group during a prescheduled chapter meeting. Those in the treatment group received an Alcohol Expectancy Challenge presentation, which presented pharmacological effects of alcohol and challenged students’ positive expectancies regarding alcohol. The control group received an educational presentation of similar length.
The mean age of the 525 participants was 19.76 years. Forty-nine percent were male, and approximately 83 percent were white, 12.9 percent were Hispanic, 1.3 percent were Asian American, 0.3 percent were African American, and 2.5 percent were of other race/ethnicity. At baseline, no statistically significant differences were found between groups based on age, class standing, ethnicity, average weekly peak blood alcohol content (BAC), average drinks per sitting, or average drinking days per week. There were also no statistically significant differences in alcohol expectancies on the Comprehensive Effects of Alcohol Scale (CEOA) for six of the seven subscales (sociability, tension reduction, liquid courage, sexuality, cognitive/behavioral impairment, and risk and aggression). There was a statistically significant difference between the two groups on the self-perception subscale. This difference was taken into account in interpreting overall results.
Alcohol consumption was measured using the retrospective, self-report, timeline follow-back procedure. Alcohol expectancies were measured using the CEOA, a factor model-based expectancy measure. Follow-up data was collected 4 weeks after the initial presentation. Of the 525 participants who completed baseline measures, 354 (67.4 percent) completed follow-up measures. The study author used a 2 (experimental, control) x 2 (pretest, posttest) x 2 (male, female) mixed analysis of variance on the seven subscales of the CEOA and on drinking behavior, and conducted subgroup analyses by looking at differences between male and female students.
