This case study describes the process used to obtain a representative sample in an immigrant community. A mixed-methods approach was utilized, including a geographic boundary study, abstraction of census data, and key informant and brief random interviews. This information was integrated to construct a sampling frame from which a two-stage, random cluster sampling procedure was implemented. This case study illustrates the challenges of selecting a representative sample in a dynamic community to elicit valid and reliable information about the impact of a multi-level intervention on residents. In addressing this challenge, we utilized a variety of methods to inform the development of the sampling frame to ensure the broadest representation of community respondents and thus the systematic delivery of the intervention.