Health survey methods with minority populations: some lessons from recent experience.

  • 1 January 1992
    • journal article
    • Vol. 2  (3) , 273-87
Abstract
Until recently, minority populations have been inadequately or inaccurately represented in health research. Researchers are now recognizing the need to improve the validity and reliability of data on the health status and health-related behaviors of minorities. This paper discusses important methodological issues in conducting health survey research in minority communities: construction of an appropriate sampling frame, response rates, attrition from panel studies, and response patterns. These themes are illustrated with data from three field studies at the New England Research Institute. Two of these studies focus on inner-city Puerto Rican youth, a group rapidly increasing in size. The extent and multiplicity of problems experienced by this group affect the complexity of survey protocols. The third study is a random-digit-dial telephone survey on health care utilization for coronary heart disease by black and white adults from three inner-city neighborhoods in Boston. The conclusions drawn from the Institute's experience are corroborated by other scientific studies. First, the sociocultural characteristics of the community or group selected for study must be considered in planning and implementing any survey research on minority populations. Second, ensuring the quality of field work with minority groups may be expensive because of high residential mobility and lack of preexisting sampling frames. Third, there is no reason to expect any diminution of data quality with minority groups, provided the resources for data collection are adequate. The quality of data is undoubtedly proportional to the field efforts expended, but the costs of high-quality survey work are often not appreciated. The paper questions the utility of the term "minority research," for it disregards the considerable variation between and within minority groups and subcultures.

This publication has 0 references indexed in Scilit: