A Comparison of Block Group and Census Tract Data in a Hedonic Housing Price Model

Abstract

The consideration of neighborhood effects in urban and regional models has been constrained by data inadequacies. Analysts have customarily used data aggregated at the census tract level to characterize areas differentiated by public service provision or socioeconomic composition. This paper introduces use of the block group, an aggregation unit typically 20 to 25 percent of the size of a census tract, formulated for general use in the 1970 censuses for all urbanized areas. 1