Ecological Regression and the Analysis of past Politics

Bibliographic Data

ID3336752
AuthorsJ Morgan Kousser (0000-0001-5261-2859, California Institute of Technology, corresponding author)
Year1973
Volume4
Issue2
Pages237
Publication date1973-01-01
Peer ReviewedYes
Open AccessNo
TypeARTICLE
VenueThe Journal of Interdisciplinary History (JOURNAL)
Journal identifiersISSN: 0022-1953 • E-ISSN: 1530-9169
PublisherJSTOR (PUBLISHER)
DOI10.2307/202265
OpenAlexW2324218386
LanguageEN
Citations received27

Regression estimation of \ncell entries in contingency tables is among the most useful statistical \ntechniques for political historians. Developed in the 1950s by statisticians \nwho were attempting to circumvent the so-called "ecological \nfallacy," regression estimation has received a good deal of attention in \nother social scientific disciplines, but surprisingly little in history. In a \nrecent article in this journal, Jones provided a short introduction to the \ntechnique pioneered by Leo A. Goodman. In addition, Jones tested its \naccuracy by comparing survey results with estimates of voting behavior \nfrom the 1960 presidential election, and urged the historical profession \nto utilize the Goodman procedure. Although helpful as far as it goes, \nJones's paper does not treat the theory, mathematical background, and \nassumptions of the method in sufficient detail to enable historians to \nemploy it creatively. Nor does he offer the researcher advice on how to deal with typical difficulties which arise in an actual analysis of past \ndata-for example, the problems of nonlinearity and logically impossible \nestimates

Biology · Geography · Political science · Politics · Regression · Regression analysis · Statistics · Census and Population Estimation · Ecology · Environmental Science · Historical Economic and Social Studies · Law · Mathematics · Social and Cultural Dynamics