Does Anti-Drug Advertising Work?
) (General) Abstract This paper examines whether a national anti-drug advertising campaign is associated with a change in adolescents’ drug use behavior. Specifically, the primary objective of the paper is to investigate the relationship between adolescents’ recall of exposure to anti-drug advertising and their probability of trying marijuana, cocaine and crack, as well as their volume of drug use given trial. We use a version of an attribute-based discrete choice framework based upon work by Hanemann (1984) to generate a model of drug use behavior, and estimate the model using a modified sequential probit framework. We estimate the model using data from the Partnership for a Drug-Free America (PDFA) Partnership Attitude Tracking Survey (PATS) over the years 1987 through 1990. Since PDFA advertising did not begin until after the first set of data were collected, we had a form of “natural experiment” where we could first address the factors influencing drug use on a control group (respo
