Passing the Test
- 1 December 1991
- journal article
- research article
- Published by SAGE Publications in Evaluation Review
- Vol. 15 (6) , 770-799
- https://doi.org/10.1177/0193841x9101500607
Abstract
Ecological regression analysis has successfully estimated minority and White voting behavior in jurisdictions throughout the United States. Critics have imposed assumptions on the technique that are far too restrictive and have misinterpreted relevant data from Los Angeles County. Ecological regression is more reliable than the "neighborhood model, " devised by defendants' experts for the Los Angeles litigation. Unlike ecological regression, the neighborhood model generates results that are an arbitrary function of how individuals are grouped into precincts: the classic form of the "ecological fallacy. " Ecological regression, not the neighborhood model, accounts for the actual results of elections in Los Angeles County.Keywords
This publication has 5 references indexed in Scilit:
- Ecological Regression and Voting RightsEvaluation Review, 1991
- Estimating the Extent of Racially Polarized Voting in Multicandidate ContestsSociological Methods & Research, 1988
- Ecological InferencePublished by SAGE Publications ,1978
- Correlation, Regression, and the Ecological Fallacy: A CritiqueJournal of Interdisciplinary History, 1974
- An Alternative to Ecological CorrelationAmerican Sociological Review, 1953