This presentation is an informal discussion of a questionnaire context effect which led to an erroneous conclusion in analysis of data from the Monitoring the Future project; an annual survey of high school seniors that has occurred since 1975, and that focuses heavily on their drug use and attitudes about drugs. In 1980, the researchers reported distinct shifts in peer norms about the regular use of cigarettes, alcohol, or marijuana. Anomalous trends in subsequent years, however, revealed that this shift was due primarily to the placement of the questions about friends' disapproval immediately after the questions about parents' disapproval. A revised format, omitting the parent questions, revealed that the supposed shift in peer norms about drug use was a contextual effect of the placement of peer approval questions after parent approval questions. The paper concludes with speculations about how the context effect can influence or distort correlation findings. The presentation includes references along with seven visual displays of data. (TE)