ABSTRACT

In this chapter we extend an approach to studying the role of housing and neighborhood amenities in residential sorting and neighborhood change, originally presented in Delmelle and Nilsson (2021). By incorporating a longer temporal perspective and a neighborhood classification approach that is more aimed at capturing change over time, namely latent class growth modeling, we show that this classification technique yields emore interpretable results compared to the corresponding text analysis in Delmelle and Nilsson (2021) and leads to better predictions in some cases but not all.