Intercept studies, clinical trials, and cluster experiments: to whom can we extrapolate?

While sample surveys rely on an explicit theory of the relationship between a sample and its reference population, many nonsurvey-based designs lack a clear relationship to the population to which inferences are to be drawn. For example, in order to be applicable to large populations, intercept studies--which include observational studies of migrant populations and randomized studies such as mall intercepts--must make implicit or explicit assumptions about the probability of finding members of the population. In conducting clinical trials, constraints imposed by the research often lead large segments of the population to have zero probability of observation. Although such experiments generally conduct research at only a few sites, investigators usually intend to apply their results quite broadly. This paper compares intercept studies and clinical trials with particular emphasis on inference to a population for which the probabilities of being observed are unknown, unreliable, or very small.