A Principal Component Logistic Analysis of the Mobility Decision in Transitional Neighborhoods

This study represents an extension of the literature that examines the factors affecting a household's decision to make an intrametropoli-tan move by emphasizing the importance of neighborhood characteristics in addition to the traditional life-cycle variables on the decision to move. Because of the dichotomous nature of the dependent variable and the collinearity among the independent variables, we employ both principal component and logit analysis. The primary data base is drawn from a sample of households moving into and out of two areas in St. Louis.
The results indicate that many of the traditional life-cycle variables are significant, although some behave differently than in the usual case because of the transitional nature of the neighborhoods we examine. The results also suggest that a household's perceptions regarding the quality of the neighborhood are also important determinants of mobility. Finally, when the maximum likelihood logit analysis results are compared with the principal component logit results, we find that the principal component logit technique performs better in almost all instances.