Excessive alcohol consumption is a widespread concern among many college campuses. Most of the data on the prevalence and adverse consequences associated with college drinking are gathered from self-report surveys, which require respondents to have a knowledge of standard drink sizes. Unfortunately, the reliability and validity of these data are questionable because college students are typically unable to define or pour standard drinks. Efforts to improve college students’ self-reported alcohol consumption are warranted. Some researchers suggest that we can improve self-reports of alcohol consumption by providing college students with feedback on the accuracy of their standard drink free-pours (White et al., 2005). However, the evidence supporting the use of feedback to improve the validity of self-report is limited by aggregate data, lack of repeated measures, evaluation of only one type of beer, and no evaluation of the effects of feedback on observable behavior (e.g., free-pour). The current study replicated White et al. (2005) using a single-case design and repeated measures to examine the effects of feedback on college students’ self-report as well as their free-pours. Results showed feedback improved the accuracy of college students’ free-pours of standard servings of beer containing 5% alcohol by volume (ABV); however, this skill did not generalize to pouring standard servings of a higher ABV (8%) beer. Unlike White et al. (2005), the feedback had little effect on college students’ self-reported alcohol consumption. Future researchers should use a similar single-case design to evaluate if feedback on various types of alcohol (such as beer, wine, and liquor) systematically affects college students’ self-reports. If so, this feedback could potentially be used as a method to improve the reliability or validity of college students’ self-reported alcohol consumption, and potentially lead to more accurate evaluations of alcohol reduction interventions.

This exploratory study ( n =55) informs efforts aimed at addressing the problem of collegiate high-risk drinking by evaluating the effectiveness of three methods for teaching standard drink measures to first-year college students. These methods include the use of numeric measures in ounces, two-dimensional (2-D) images, and three-dimensional (3-D) models. All of these represent the appropriate volume for four standard drinks—beer, wine, hard alcohol, and alcohol contained in one mixed drink—as defined by the National Institute on Alcohol Abuse and Alcoholism (NIAAA). By increasing students’ knowledge of standard drink measures and their own accurate pouring ability, they may improve their efficacious use of protective strategies, including tracking of the number of drinks they consume. In this study, when the distinction between over-pouring and under-pouring was eliminated and the absolute values of the percentages of inaccurate pouring became the focus, all three interventions were found to improve the accuracy of the ability to pour standard-size mixed drinks. Although violations to some of the statistical assumptions underlying ANOVA precluded use of this inferential technique to directly compare the three interventions, the data analytic approaches used did suggest that both the 2-D and the 3-D interventions may be more beneficial, overall, than the NIAAA numerical intervention. Associations between knowledge of standard drink measures, self-efficacy in pouring standard drinks, self-efficacy in using protective strategies, use of protective strategies, pouring accuracy, alcohol consumption, and negative consequences associated with alcohol usage were explored. Use of 3-D standard drink models offers prevention educators a simple yet potentially more effective method for teaching students and can be conducted easily in a variety of educational settings. Additionally, this type of education may ultimately improve the accuracy of data collection on student drinking. Suggestions for both administrators and further research are presented.

Self-report methods are often used as the main or only measure of college students' alcohol consumption (e.g., Alfonso, Hall, & Dunn, 2013; White et al., 2003). However, these data are limited because students are rarely able to accurately state or pour the standard serving size for various types of alcohol, rendering the validity of self-report questionable (Midanik, 1982; White et al., 2003). Moreover, students are often required to participate in university alcohol training programs; these programs are generally information-based and rarely contain a skills training component, in which students practice identifying and pouring standard servings of alcohol. Researchers have evaluated the effectiveness of nutrition and alcoholic beverage portion training procedures that make use of practice, measurement aids, and verbal feedback; however, to our knowledge, the effect of stimulus fading on a free pour procedure has not been assessed even though research suggests stimulus fading is an effective training procedure. In the current study we assessed the effects of individualized 1) verbal feedback, 2) superimposition training (i.e., measurement aids), and 3) stimulus fading training on college students' ability to accurately free pour a standard serving of beer into 18 oz training and generalization cups using an ABA and in some cases, an ABACA design. Results suggest that all three types of training are effective in teaching college students to free-pour the standard serving size of beer. Further research is required to evaluate effective individual and group trainings, use of different alcohol types and various sized containers, and maintenance and generalization of the skill over time and across settings.