Monitoring the AIDS epidemic in the United States: a network approach

Respondents in the 1988 General Social Survey (GSS) were asked to scan their acquaintance networks to identify all those who had been a victim of a homicide or had acquired immunodeficiency syndrome (AIDS). Estimates of the sex, race, age, and regional breakdowns for homicides in the last year and for people with AIDS were compared with official statistics. The GSS estimates for the distribution of homicide victims replicate the official statistics quite well. The GSS estimates for AIDS cases suggest that the data provided to the Centers for Disease Control may underestimate by a substantial margin the prevalence of AIDS in the white population of higher socioeconomic status, overstate the relative prevalence of the disease in the minority populations, underestimate the prevalence of the disease in the Midwest, and overstate it for the East.