ANES

1992 TIME SERIES STUDY

How to Cite
DATA
Return to Data Center
Errata
QUESTIONNAIRES
Pre-election pdf
Post-election pdf
Coversheets pdf
CODEBOOK
Introduction pdf   text
Variables pdf   text
Appendix pdf   text
FIELD MATERIAL
Pre-election Respondent Booklet
Post-election Respondent Booklet
  ABOUT THE DATASET:
Time Series Study
Includes: 1990 panel variables
Completions: 2,485 pre-election; 2,255 post-election
Sample: fresh cross-section, panel
Modes used:face-to-face, telephone
Weights: V923008,V923009,V927000
Auxiliary files: nonresponse 'bias' file

STUDY CONTENT HIGHLIGHTS:
In addition to content on electoral participation, voting behavior, and public opinion, the 1992 ANES contains questions in other areas such as civic engagement, social altruism, and media exposure. Special-interest and topical content includes a new battery on feminism and the role of women in society and families, a comprehensive retrospective battery on the 1991 Persian Gulf War, and new questions on sexual harassment, homosexuals in the military, immigration policy, attitudes toward immigrants, and the first ANES time series battery on racial stereotypes.

STUDY DESIGN HIGHLIGHTS:
Besides being a stand-alone time series data collection, in which the full sample comprises a representative cross-section, the 1992 ANES is also, for the empaneled portion of the sample, the final wave in a three-wave study that first began with the 1990 ANES and subsequently continued in the 1991 Pilot, and which featured special inclusion of questions on developments in the Persian Gulf in all 3 waves. Two versions or 'forms' of the 1992 instrument were produced. The 'long-form' questionnaire was intended for face-to-face administration with cross-section respondents and with panel respondents who were not "out of range" in 1992. The 'short-form' questionnaire was intended for administration by telephone with "out of range" panel respondents only, although complications during implementation resulted in many long-form interviews also administered by phone. The short-form questionnaire excluded most demographic items and, additionally, some Congressional content due to the difficulty of assigning telephone respondents to congressional districts which were newly drawn in 1992. With one exception, question wording was undifferentiated by form. Among specific design features was the coding of probes used by interviewers for a subset of questions.

Beginning September 1, interviews were conducted prior to election day, November 3, using a pre-election instrument averaging 69 minutes (71 minutes 'long' version and 49 minutes 'short' version). Post-election interviews that averaged 64 minutes (65 minutes 'long' version and 44 minutes 'short'-version) were administered November 4,1992-January 13, 1993. The sample consisted of 1769 respondents who had previously been interviewed in the 1990 ANES (1992 panel respondents), together with fresh cross-section cases including 1515 eligible persons, for an eligible total of 3,284. Of these, 2,485 completed pre-election interviews (1,359 panel, 1126 fresh cross), and 2,255 were re-interviewed in the post (1,250 panel, 1005 fresh cross).

WEIGHTS SUMMARY:
V923008 is a full-sample weight that compensates for unequal probabilities of selection for different size households, and for differential nonresponse across geographic sample design categories and between panel and cross-section respondents. It is post-stratified by age and sex. V923009 is a corresponding weight for use with the panel alone. V927000, the "time series" weight, adjusts only for panel aging and attrition, and should be used when comparing the panel cases ro the total sample to previous unweighted studies.

AUXILARY FILES DESCRIPTION:
V923008 is a full-sample weight that compensates for unequal probabilities of selection for different size households, and for differential nonresponse across geographic sample design categories and between panel and cross-section respondents. It is post-stratified by age and sex. V923009 is a corresponding weight for use with the panel alone. V927000, the "time series" weight, adjusts only for panel aging and attrition, and should be used when comparing the panel cases or the total sample to previous unweighted studies.