Beyond dummy variables and sample selection: what health services researchers ought to know about race as a variable

Health Serv Res. 1994 Apr;29(1):1-16.


Objective: This article addresses the definitional and methodological problems associated with the most common uses of race as a variable in health research.

Principal findings: Although often used in health services research, race is a poorly understood concept because it lacks conceptual clarity. Moreover, the measurement problems with race have not yet been adequately addressed. As a result, many quantitative models that attempt to explain race differences are inadequate to inform health or social policy.

Conclusions: Researchers should treat the race variable with the same degree of caution and skepticism with which it treats any other variable.

MeSH terms

  • Bias
  • Birth Certificates
  • Brazil
  • Continental Population Groups* / classification
  • Cultural Characteristics
  • Data Interpretation, Statistical*
  • Ethnic Groups* / classification
  • Female
  • Health Behavior
  • Health Policy
  • Health Services Research / methods*
  • Humans
  • Japan
  • Male
  • Models, Theoretical
  • Organizational Policy
  • Risk Factors
  • Socioeconomic Factors
  • United States