A multivariate analysis of changes in population from 2000 to 2010 at the county-level in the U.S.: Working paper series--13-03

The release of the 2010 Census data offers new opportunities to examine population changes occurring at the county level. We employ Ordinary Least Squares Regression to identify factors that potentially explain changes in population growth rates between 2000 and 2010 for all counties in the forty-eight contiguous states. A number of demographic and economic variables as well as noneconomic and amenity-based variables emerged as significant explanatory variables that underlie the changes in county populations. We extend the analysis using K-Means clustering as a technique for assigning counties into groups which share common properties with other counties in the same cluster. Based upon the results of our study we conclude that the set of economic variables is relatively more important in explaining population change when compared with demographic and other noneconomic variables in the model. However, this outcome may have, in part, been influenced by the economic recession in the latter years of the decade