Author's Keywords: racial block voting, ecological inference	Convention All Academic Convention is the premier solution for your association's abstract management solutions needs. Submission - Custom fields, multiple submission types, tracks, audio visual, multiple upload formats, automatic conversion to pdf. Review - Peer Review, Bulk reviewer assignment, bulk emails, ranking, z-score statistics, and multiple worksheets! Reports - Many standard and custom reports generated while you wait. Print programs with participant indexes, event grids, and more! Scheduling - Flexible and convenient grid scheduling within rooms and buildings. Conflict checking and advanced filtering. Communication - Bulk email tools to help your administrators send reminders and responses. Use form letters, a message center, and much more! Management - Search tools, duplicate people management, editing tools, submission transfers, many tools to manage a variety of conference management headaches! Click here for more information.
Association: Name: American Political Science Association URL: http://www.apsanet.org
Citation: URL: http://www.allacademic.com/meta/p59489_index.html Direct Link: HTML Code:

Grofman, Bernard. and Barreto, Matt. "Ecological Regression and Ecological Inference in the Presence of Systematic Bias in the Measurement of the Independent Variable" Paper presented at the annual meeting of the American Political Science Association, Hilton Chicago and the Palmer House Hilton, Chicago, IL, Sep 02, 2004 <Not Available>. 2009-05-26 <http://www.allacademic.com/meta/p59489_index.html>

Grofman, B. and Barreto, M. , 2004-09-02 "Ecological Regression and Ecological Inference in the Presence of Systematic Bias in the Measurement of the Independent Variable" Paper presented at the annual meeting of the American Political Science Association, Hilton Chicago and the Palmer House Hilton, Chicago, IL Online <.PDF>. 2009-05-26 from http://www.allacademic.com/meta/p59489_index.html

Publication Type: Conference Paper/Unpublished Manuscript Review Method: Peer Reviewed Abstract: In ecological inference involving estimating levels of racial bloc voting (RBV) it is often the case that data is missing (at the precinct level) on the racial composition of the electorate, but data is available (at the precinct level) on surrogate variables such as Spanish surname registration the racial proportions among the total population or among the voting age population. It has been a topic of concern for several decade how to adjust ecological techniques to compensate for such systematic measurement errors in the independent variable (see, e.g., discussion in Kousser, 1973; Grofman, Migalski, and Noviello, 1985; Grofman and Migalski, 1988; Loewen and Grofman, 1989). The availability of a computer program to implement King's (1997) ecological inference method has sparked a renaissance of ecological studies, especially in political science, leading to a renewed interest in this adjustment question (see esp. King, 1997: 71-72; Cho, King and Cain, 2002; Zax, 2002). After first reviewing the basic Goodman (1953, 1959) and King (1997) single equation techniques, we introduce different two equation methods which might be used to cope with the problem of measurement error in the independent variable. For the two single-equation methods and the first four of the double equation methods, we use simulations to evaluate the magnitude of the problem caused for reliable ecological inference of patterns of racial bloc voting when we have systematic and substantial measurement error in the independent variable. Here we show that, for our very simplified hypothetical data, all double equation approaches outperform both the single-equation techniques, and are essentially identical in their results. To further assess methods and compare resulting estimates of racial bloc voting parameters, we apply both the single-equation techniques and two of the double equation ecological techniques to estimate Latino and non-Latino turnout and bloc voting patterns in recent elections in California involving both Hispanic and non-Hispanic candidates -- using Spanish surname registration as our surrogate for the Latino composition of the actual electorate. In the elections we study, we find: (a) in estimating Latino and non-Latino levels of turnout, basic single-equation bivariate ecological inference and single-equation bivariate ecological regression produce essentially identical results; and (b) in estimating levels of racial bloc voting, double-equation bivariate ecological inference a la King and double equation bivariate ecological regression a la Goodman produce essentially identical results. Also, in the first two real world elections we report on, because reliable inferences about voting in the City of Los Angeles can be made-- either from Spanish surname sign-in data, in the case of precinct-specific estimates of turnout; or from exit poll data, in the case of specification of bloc voting patterns at the aggregate leve -- we can directly test the accuracy of ecological techniques in reproducing the parameters for behavior at the individual level. For these elections, we find: (c) that either of the double equation approaches yields estimats of levels of racial bloc voting that are more accurate than those from either of the single-equation techniques; and (d) that the double equation estimates correspond to a quite reasonable degree with the exit poll estimates of the true racial bloc voting parameters.

Heuristic and Systematic Biased Processing of Political Messages: Effects of Candidate Preference and the Level of Interest in Politics on Attitudes toward Issues

The Independent Variable Problem: Evaluating Alternative Measures of Attitudes towards Democracy

Ecological Footprints: Measuring Sustainability and Problem Solving