A study is reported, illustrating the methodological and practical challenges of sampling a difficult-to-reach population and of studying a population defined by criteria other than the standard demographic categories. In order to help inform the development of a sampling plan and a research design for a larger scientific investigation of preadolescents who migrate to the United States from Puerto Rico, a study of 286 public and private schools in the United States (New Jersey) was conducted, ascertaining students' migratory movements. Examined are the distribution of arrivals (i.e., transfers-in from Puerto Rico) among districts and among schools in particular geographic areas, the stability of in-migration patterns and numbers over time, and the post-migration mobility. Although focused on this population, the study addresses issues likely to be encountered generally by researchers who study "nonstandard" or difficult-to-reach populations. Considered are both cross-section and longitudinal designs.