The availability of granular geographic data presents new opportunities to understand how neighborhoods are formed and how they influence attitudes and behavior. To facilitate such studies, we develop an online survey instrument for respondents to draw their neighborhoods on a map. We then propose a statistical model to analyze how the characteristics of respondents and geography, and their interactions, predict subjective neighborhoods. We illustrate the proposed methodology using a survey of 2,572 voters in Miami, New York City, and Phoenix. Holding other factors constant, White respondents tend to include census blocks with more White residents in their neighborhoods. Similarly, Democratic and Republican respondents are more likely to include co-partisan census blocks. Our model also provides more accurate out-of-sample predictions than standard distance-based neighborhood measures. Lastly, we use these methodological tools to test how demographic information shapes neighborhoods, but find limited effects from this experimental manipulation. Open-source software is available for implementing the methodology.
翻译:粒子地理数据的可用性为了解邻里如何形成以及它们如何影响态度和行为提供了新的机会。为了便利这些研究,我们开发了一个在线调查工具,供答卷人绘制自己的邻里地图。然后我们提出了一个统计模型,分析答卷人的特点和地理及其相互作用,预测主观邻里。我们用对迈阿密、纽约市和凤凰城2 572名选民的调查来说明拟议方法。保持其他因素不变,白人答卷人往往包括普查区,其邻里有更多的白人居民。同样,民主党和共和党的答卷人更可能包括共同党派的普查区。我们的模型还提供比标准的远程邻里措施更准确的抽样预测。最后,我们使用这些方法工具来测试人口信息如何塑造邻里,但发现试验性操作的效果有限。可以使用开放源软件实施方法。