Measuring and Modeling Neighborhoods

TL;DR

An easy-touse online survey instrument that allows respondents to draw their neighborhoods on a map is developed and a statistical model is proposed to analyze how the characteristics of respondents, relevant local areas, and their interactions shape subjective neighborhoods.

Abstract

Granular geographic data present new opportunities to understand how neighborhoods are formed, and how they influence politics. At the same time, the inherent subjectivity of neighborhoods creates methodological challenges in measuring and modeling them. We develop an open-source survey instrument that allows respondents to draw their neighborhoods on a map. We also propose a statistical model to analyze how the characteristics of respondents and local areas determine subjective neighborhoods. We conduct two surveys: collecting subjective neighborhoods from voters in Miami, New York City, and Phoenix, and asking New York City residents to draw a community of interest for inclusion in their city council district. Our analysis shows that, holding other factors constant, White respondents include census blocks with more White residents in their neighborhoods. Similarly, Democrats and Republicans are more likely to include co-partisan areas. Furthermore, our model provides more accurate out-of-sample predictions than standard neighborhood measures.

Figures (35)

• Figure 1: Map with brush tool used to draw neighborhoods.
• Figure 2: Descriptive statistics for respondent neighborhoods.
• Figure 3: Model schematic. The respondent's location is indicated by the black house in the center. Blocks are labeled in the order in which they are considered for inclusion in the neighborhood, with the number indicating the graph-theoretical distance and the letter the spatial distance tiebreaker. Blocks shaded purple have been included in the neighborhood, while blocks shaded light orange have been excluded. Parentheses around a block label indicate that it will not be considered for inclusion in the neighborhood because none of its neighboring blocks which are closer to the respondent belong to the neighborhood.
• Figure 4: Illustration of kernel function across a range of values of the $\alpha$ parameter, indicated by different colors. The length scale shown here is arbitrary; in the model, it is estimated as the $L$ parameter.
• Figure 5: Selected full model coefficient posteriors, scaled to show the percentage point change in probability of a block's inclusion for a baseline probability of 50%. Plotted are 90% and 50% credible intervals, with posterior medians displayed to the right of each interval. Section S5 of the SI contains the full results table for the other variables specified in the model specification.