Hobart and William Smith Colleges Professor Notes Serious Flaws in Recent Harvard Study on Alcohol U – Hobart and William Smith Colleges \
The HWS Update

Hobart and William Smith Colleges Professor Notes Serious Flaws in Recent Harvard Study on Alcohol U

Wesley Perkins, developer of the social norms approach, defends its effectiveness and blasts the Harvard study


(July 25, 2003) Geneva, N.Y. – H. Wesley Perkins, dubbed the “father of social norms” by the Los Angeles Times, argues that the recently released study of college alcohol use and abuse conducted by Harvard University’s Henry Wechsler is seriously flawed.

“In Henry Wechsler’s latest article and press release he attempts to challenge the growing evidence in support of the social norms approach to substance abuse prevention with a so-called ‘national evaluation of social norms marketing’ on college campuses,” said Perkins. “But he failed dramatically.”

Perkins noted that Wechsler first provided a biased and limited review of the research literature. He ignored controlled experiments in which providing normative feedback about the majority of students exhibiting moderate or healthy behavior and holding responsible attitudes has led to reductions in problem drinking. He also ignored the most recent compendium of case studies on college campuses that fully adopted the model with great success, edited by Perkins, titled “The Social Norms Approach to Preventing Substance Abuse in School and College Age Substance Abuse” (Jossey-Bass, 2003).

Interestingly, notes Perkins, Wechsler also fails to mention his own previous research (Perkins and Wechsler, Journal of Drug Issues, 1996) with his College Alcohol Survey data showing perceived norms as the strongest predictor of alcohol abuse.

“The study most recently cited that reputedly refutes the effectiveness of social norms approaches contains fundamental flaws that make the results meaningless,” noted Perkins.

Perkins noted several flaws:

· First, the survey that Wechsler’s research was not designed for this level of analysis: it has very small samples at each school (as few as 50 or 60 representing schools with tens of thousands of students).

· Second, and a more fundamental problem, the study has not identified and separated out for comparison the schools that have actually been conducting intensive social norms interventions.

Perkins explains, “Wechsler simply asks a single administrator on each campus whether or not the school had conducted a ‘social norms’ campaign on their campus between 1997 and 2000–with no definition or control for what that phrase meant. In the late 1990s, although the social norms approach to substance abuse prevention was gaining attention and a couple dozen projects had been initiated with U.S. Dept. of Education and other government funding, there was still much ignorance about what this approach entailed. Back in that time period one would have been hard pressed to identify more than a hundred schools nationwide implementing an actual model of normative feedback about the true positive norms that exist among the majority of students and not compromised by including the traditional model of scare tactics.

“Yet the Wechsler et al. study of 98 schools found almost 40 percent of administrators saying they had done some sort of social norms campaign during that period. If correctly identified, that would translate to an unbelievable 1,500 schools nationwide doing such campaigns in the 1990s. This is a truly absurd figure for anyone like myself who was working in the field conducting research and working with schools who were interested in the social norms approach at the time.”

Perkins notes that today, although interest has grown substantially, the approach still stands as a minority alternative to traditional prevention efforts. He said that only about 200 schools nationwide have actively engaged in any kind of significant social norms effort, and even then the projects are occasionally compromised by scare campaigns on campus that inflate misperceptions of what is normal behavior and the attitudes of most students.

· Third, beyond the single administrator saying his or her campus had done some type of social norms campaign, Wechsler and team examine the student survey data for corroboration looking at whether or not students said they had been given information about drinking rates as part of the school’s educational campaign. But here again, there is no specification of what rates were provided.

“If the campaign simply reported the rates of high risk drinking or the number of drinking consequences that occurred at a student’s school, the student could answer affirmatively—the exact opposite the of type of information that would be given in a social norms approach,” noted Perkins. “If students had simply seen material about the ‘binge’ rate at their school—the popular but controversial Wechsler measure often reported by media—then they would have also said yes to this question.

· Fourth, the other measures ostensibly identifying schools with social norms interventions simply ask whether students saw alcohol education material on posters, handouts, and mailings or in school newspaper articles. Again the type of content, whether traditional health education or social norms, was unspecified. So in the end, a large number of schools doing all types of media campaigns of varying content and intensity levels are grouped together and labeled as social norms efforts. Then the study concludes by finding this large group of what includes mostly traditional educational interventions has had no impact just like the rest of the schools—hardly surprising.

“As my recent book makes clear, one must look at programs that intensively apply the model with fidelity and good evaluation measures to see the positive results coming from this approach,” concluded Perkins.

# # #

For more information on Perkins’ work: http://alcohol.hws.edu/alcohol/index.html.