Replication Data for: Assessing Robustness of Findings About Racial Redistricting’s Effect on Southern House Delegations

We assess the robustness of Shotts’s (2003a) finding that racial redistricting increases the number of Southern representatives to the left of the U.S. House median. Our results, which are based on Monte Carlo simulations and an alternative measure of representatives’ preferences, are broadly consistent with Simons and Mallinson’s (2015) null findings. The data do not support the claim that racial redistricting promotes liberal policy outcomes; nor do they support the claim that it promotes conservative policy outcomes. At a methodological level, we suggest techniques that researchers can use to assess how robust their findings are to noise in variables based on estimated values like DW-Nominate scores.

本研究评估了肖茨（Shotts, 2003a）的一项核心发现的稳健性，该发现指出种族选区重划会增加政治立场位于美国众议院中位数左侧的南方众议员人数。本研究基于蒙特卡洛模拟与议员政治偏好的替代测度展开分析，所得结果总体上与西蒙斯与马林森（Simons and Mallinson, 2015）的零结果研究结论相一致。研究数据既不支持种族选区重划可推动自由派政策结果的主张，也不支持其可推动保守派政策结果的主张。在方法论层面，本文提出了可供研究者使用的分析技术，用于评估其研究结果对基于DW-Nominate得分（DW-Nominate scores）这类估计值所生成的变量噪声的稳健性。