American National Election Studies, 2002

Data Archive > U.S. Surveys > General Population > National > National Election Studies > Summary


The NES/CPS American National Election Studies 2002 was conducted by the Center for Political Studies of the Institute for Social Research, under the general direction of Principal Investigators Nancy Burns and Donald R. Kinder. Dozens of substantive themes are covered including: interest in political campaigns, attentiveness to media coverage, and political participation. In this data file, variables 226 through 233 and variables 312 through 316 include indicators on religiosity, congregational affiliation, and church attendance. Respondents were also asked about financial contributions to their church (variable 685). For more information, go to: http://www.electionstudies.org/studypages/2002prepost/2002prepost.htm

Data File
Cases: 1,511
Variables: 727
Weight Variable: 4) WEIGHT1, 5) WEIGHT2
Data Collection
Date Collected: September to December, 2002
Funded By
These materials are based on work supported by the National Science Foundation under grant numbers SBR-9707741, SBR-9317631, SES-9209410, SES-9009379, SES-8808361, SES-8341310, SES-8207580, and SOC77-08885, as well as the Russell Sage Foundation under grant number 82-00-01, and the University of Michigan.
Collection Procedures
Survey data were taken utilizing telephone interviews.
Sampling Procedures
The 2002 National Election Study entailed both a pre-election interview and a post-election re-interview. The pre-election survey was conducted from September 18th, 2002 through November 4th, 2002. The post election survey was conducted from November 6th, 2002 through December 6th, 2002. The study is unusual in that it is the first midterm study in the NES time series to include both a pre-election and post-election survey. It is also the first NES biennial survey conducted entirely by telephone. The 2002 NES study sample was comprised of two components. The first was a "Panel" which contained the 1,807 respondents who had provided an interview in the 2000 study. Of these, 1187 were successfully re-interviewed in 2002. The second portion of the sample was a "Fresh Cross-Section" of respondents comprised of a newly pulled RDD sample of 1,175 phone numbers. Among these, 254 were determined to be non-sample and 324 resulted in successful interviews.
Principal Investigators
Nancy Burns
Donald R. Kinder
Additional Notes
2002 NES Analysis Weights:
There are two weight variables in NES 2002 data file. The variable, WEIGHT1, should be used for weighted analysis of the data for the 1,511 pre-election respondents. The second weight variable, WEIGHT2, should be used for weighted analysis of the data for the 1,346 post-election respondents.

These weights have been developed for combined cross-sectional analysis of the 2002 NES panel and new RDD cross-sectional samples. However, these weight variables may also be used by analysts who are interested in analyzing only the data for panel respondents who were interviewed in 2000 and reinterviewed in 2002. The final 2002 NES pre-election and post-election weights have been scaled so that the weights sum to the nominal sample sizes of the pre-election (n=1511) and post-election (n=1346) respondent samples.

The 2002 NES analysis weights are constructed as the product of three factors: a factor to reflect differential sampling or inclusion probabilities for survey respondents; a factor to compensate for differential nonresponse among sampled persons; and a post- stratification factor to adjust weighted demographic distributions for the sample to more precise population values estimated from the U.S. Current Population Survey (CPS). The following paragraphs provide a brief summary of the each of these three major components.
Citation
Burns, Nancy, Donald R. Kinder, and the National Election Studies.
NATIONAL ELECTION STUDIES, 2002: PRE-/POST-ELECTION STUDY [dataset].
Ann Arbor, MI: University of Michigan, Center for Political Studies [producer and distributor], 2003.

Note: some category labels in the data provided by the National Election Study were incomplete.