ANES

1990-1992 MERGED FILE

How to Cite
DATA
Return to Data Center
Errata
CODEBOOK
Introduction pdf   text
Variables pdf   text
Appendix pdf   text
  ABOUT THE DATASET:
Type of Study: ANES Merged File (Time Series and Pilot studies)
Sample universe: eligible voters in U.S. households
Sample composition: cross-sections (Time Series), Pilot subset
Sample selection: per component study
Number of waves: 4 total
Completions: per component study
Modes used: face-to-face, telephone
Instrument format: paper questionnaire, CATI
Weights: V923009, V927000
Auxiliary files: none

STUDY CONTENT HIGHLIGHTS:
This file is an ANES derivative file which comprises merged data from 2 Time Series studies and 1 Pilot study: the ANES 1990 Time Series Study, the ANES 1991 Pilot Study, and the ANES 1992 Time Series Study. All variables from the component studies are present, including 1990 vote validation variables. (Note: this release has sometimes been incorrectly referred to as the 'Gulf War file' or 'Gulf War study' due to the presence of Gulf-War-related questions in all its component studies; the official name of this release, however, is the ANES 1990-1992 Merged File).

STUDY DESIGN HIGHLIGHTS:
Respondents from the 1990 Time Series study were re-interviewed in the 1991 Pilot study; they were also re-interviewed in the 1992 Time Series study, for which a cross-section was constructed from these re-interview cases together with a set of 'fresh cross' cases (respondents first selected and interviewed in 1992). All cases from the 1990 Time Series study and 1991 Pilot study are present, however from the 1992 Time Series study only 'panel' cases are represented in this file: 'fresh cross' cases from 1992 are not included here.

Participation across the 3 studies is as follows.

Case Count1990 Post1991 Pilot 1992 Pre1992 Post
298X   
34X X 
265X XX
323XX  
75XXX 
985XXXX
Total1980138313591250

WEIGHTS SUMMARY:
V923009 and V927000 apply to the 1,359 panel cases from the 1992 Study. V923009 compensates for the unequal probabilities of selection for respondents in different size households, and for differential nonresponse across geographic sample design categories; it is post-stratified to 1990 Census proportions for 24 age by sex by Census region categories. V927000 corrects for panel attrition and aging among the 1992 panel cases.

AUXILIARY FILE DESCRIPTION:
No auxiliary file is present for this derivative file.