Do Stated Racial Preferences Match Residential Mobility Behavior?

Benjamin Jarvis, University of California, Los Angeles
Robert D. Mare, University of California, Los Angeles

Individuals’ preferences for neighborhoods with varying racial characteristics affect patterns of residential mobility and segregation. “Stated” racial preferences may be elicited in surveys, typically via reactions to neighborhood vignettes (SP). Alternatively, preferences may be “revealed” by individuals’ residential histories (RP). Whereas SP data directly measure preferences, they may suffer from social desirability biases and oversimplify how individuals view neighborhoods. Whereas RP data directly measure behavior, actual moves result from a mixture of individuals’ preferences and economic and social constraints on their choices. Wave 2 of the Los Angeles Survey of Families and Neighborhoods uniquely obtains SP and RP data from the same individuals. Using these data, we show how SP and RP are associated and how this association varies across race-ethnic and socioeconomic groups and over time. Additionally, we show how to combine SP and RP data in individual-level models of residential choice.