Geography, Statistics, and Ecological Inference

Annals of the Association of American Geographers, Vol. 90, No. 3, pp. p. 601-606, September 2000

6 Pages Posted: 17 Jan 2008

Abstract

I am grateful for such thoughtful review from these three distinguished geographers. Fotheringham provides an excellent summary of the approach offered, including how it combines the two methods that have dominated applications (and methodological analysis) for nearly half a century - the method of bounds (Duncan and Davis, 1953) and Goodman's (1953) least squares regression. Since Goodman's regression is the only method of ecological inference widely used in Geography (O'Loughlin), adding information that is known to be true from the method of bounds (for each observation) would seem to have the chance to improve a lot of research in this field. The other addition that EI provides is estimates at the lowest level of geography available, making it possible to map results, instead of giving only single summary numbers for the entire geographic region. Whether one considers the combined method offered the solution (as some reviewers and commentators have portrayed it), a solution (as I tried to describe it), or, perhaps better and more simply, as an improved method of ecological inference, is not importatnt. The point is that more data are better, and this method incorporates more. I am gratified that all three reviewers seem to support these basic points. In this response, I clarify a few points, correct some misunderstandings, and present additional evidence. I conclude with some possible directions for future research.

Suggested Citation

King, Gary, Geography, Statistics, and Ecological Inference. Annals of the Association of American Geographers, Vol. 90, No. 3, pp. p. 601-606, September 2000, Available at SSRN: https://ssrn.com/abstract=1083740

Gary King (Contact Author)

Harvard University ( email )

1737 Cambridge St.
Institute for Quantitative Social Science
Cambridge, MA 02138
United States
617-500-7570 (Phone)

HOME PAGE: http://gking.harvard.edu

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
63
Abstract Views
867
Rank
632,837
PlumX Metrics