You are not currently logged in.
Access JSTOR through your library or other institution:
If You Use a Screen ReaderThis content is available through Read Online (Free) program, which relies on page scans. Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
The Reliability of Aggregated Public Opinion Measures
Bradford S. Jones and Barbara Norrander
American Journal of Political Science
Vol. 40, No. 1 (Feb., 1996), pp. 295-309
Published by: Midwest Political Science Association
Stable URL: http://www.jstor.org/stable/2111703
Page Count: 15
Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
Preview not available
Using generalizability theory, we discuss the O'Brien (1990) method of aggregate-level reliability assessment. We contrast this theoretical conception of reliability with the theory of individual-level internal consistency. The methodology used is the O'Brien (1990) aggregate-level reliability technique for the R:A design, where "R" denotes that survey respondents reside or are nested within aggregate units (A). The aggregate unit in this analysis is the state. One-way analysis of variance is used to estimate reliabilities. Individual-level measures of reliability are inappropriate techniques when one is evaluating the reliability of aggregated measures of public opinion. Estimating aggregate-level reliability is methodologically simple and should be a common practice among the many researchers using aggregate-level variables created from individual-level data.
American Journal of Political Science © 1996 Midwest Political Science Association