1972-1994 General Social Survey Cumulative File
The GSS contains three weight variable (ADULTS, OVERSAMP, FORMWT) that users
may wish to employ as well as a weight-related variable (ISSP). This section
briefly discusses these variables.
The full-probability GSS samples used since 1975 are designed to give each
household an equal probability of inclusion in the sample. (Call this
probability Ph.) Thus for household-level variables, the GSS sample is self-
weighting. In those households which are selected, selection procedures
within the household give each eligible individual equal probability of being
interviewed. In a household with n eligible respondents, each has probability
Ph of being in a selected household, and 1/n * Ph of actually being interviewed.
Persons living in large households are less likely to be interviewed, because
one and only one interview is completed at each preselected household. The
simplest way to compensate would be to weight each interview proportionally
to n, the number of eligible respondents in the household where the interview
was conducted. N is the number of persons over 18 (ADULTS) in the household.
A discussion of the weight as well and a post-stratification variant of
weighting by ADULTS appears in GSS Methodological Report No. 3.(9)
As described in the previous section, the 1982 survey included an oversample
of blacks. To make the 1982 survey a representative cross-section the user
can either exclude the black oversample cases by excluding codes 4 and 5
on SAMPLE or weight the file by OVERSAMP. To make the 1987 survey a
representative cross-section the user can either exclude the black oversample
by excluding code 7 on SAMPLE or weight the file by OVERSAMP. Users should
adopt one of these procedures in all cases except when analyzing only blacks
from the 1982 and/or 1987 cross-sections and oversamples.
Problems with form randomization procedures on the 1978, 1980, 1982-1985
surveys necessitate the use of FORMWT when variables appearing on only one
form are analyzed. A complete list of form-related variables appears in
Appendix P. Full details on the form randomization problem and of the weight
created to correct for it appear in GSS Methodological Report No. 36.(10)
The International Social Survey Program supplement was administered to Form 1
cases in 1985 and as such must be weighted for FORMWT as discussed above.
In addition because this was a self-administered supplement completed after
the main GSS questionnaire there is 10% non-response. Users may wish to use
ISSP to study supplement non-response bias and perhaps develop a weight to
compensate for same.(11) Similarly one may wish to adjust for supplement
non-response in other years.
In general the GSS samples closely resemble distributions reported in the
Census and other authoritative sources. Because of survey non-response,
sampling variation, and various other factors the GSS sample does deviate
from known population figures for some variables. The GSS does not calculate
any post-stratification weights to adjust for such differences. For relevant
discussion of distributional variation caused by non-response and other
factors see GSS Methodological Reports Nos. 3, 5, 9, 16, 21, 25, 79.(12)
Differences from the Census and other changes in distributions due to
alterations in sampling include the following:
1. In 1972 blacks were over-represented. The 1972 survey was the last
to utilize the 1960 NORC sample frame and it is believed to have
under covered rapidly growing suburban areas.
2. All full-probability samples under-represent males. This is discussed
in GSS Methodological Report No. 9.
3. Block quota samples under-represented men in full-time employment,
see GSS Methodological Report No. 7.
4. Coverage of Mormons increased significantly when the 1980 sample
frame was adopted. This was due to the addition of a primary sampling
unit in Utah. For more details see GSS Methodological Report No. 43.
5. People eighteen years old appear to be under-sampled although this
is actually not the case. Age is assigned based on year of birth and
the assumption that one's birthday has already occurred. However, to
be in the sample one must have actually reached his/her eighteenth
birthday and since the GSS is fielded in March every year only about
one-quarter of those born eighteen years prior to the current year
have reached majority by the interview dates. Thus nineteen year
olds as classified on the GSS consist of approximately one-quarter who
have turned nineteen since the first of the year and three-quarters
who will turn nineteen by the end of the calendar year. The same is
true for ages 20 and up. For eighteen year olds on the GSS only those
who have turned eighteen since the first of the year are included.
Thus the number of eighteen year olds in the GSS is approximately
one-quarter the number of nineteen year olds (See Appendix E). The
"missing" eighteen year olds are not under-represented in the sample,
but are merely counted as nineteen year olds.
If the merged GSS is thought of as designed to equally sample time, there are
numerous deviations due to such factors as 1) sample size variation across
surveys, 2) the absence of surveys in 1979 and 1981, 3) experiments (See App. 0)
4) switching of items from permanent to rotating status, 5) switching from
across survey rotation to sub-sample rotation, 6) late starting and terminated
time series, or 7) some combination of these. For more information on these
issues and possible adjustments see GSS Methodological Report No. 52. (13)
(1) For selection procedures, see Benjamin Ying and Carol Richards,
"The 1972 NORC National Probability Sample." Chicago: NORC, August 1972.
(3) Frederick Stephan and Philip McCarthy, SAMPLING OPINIONS.
(New York: John Wiley and Sons, 1958, Chapter 10.)
(4) Alaska and Hawaii are not included in this sample.
(5) The selection methods used are similar to those described in
standard textbooks, e.g., W. E. Deming, SAMPLE DESIGN IN BUSINESS RESEARCH
(New York: Wiley & Sons, 1960), and L. Kish, SURVEY SAMPLING (New York:
Wiley & Sons, 1965).
(6) In the actual implementation of the selection method, subsamples 1
and 4 resulted in 51 PSUs, whereas subsamples 2 and 3 produced only 50 PSUs.
The result was not unexpected and is due to a technical reason, details of
which will be provided on request. The inequality of subsample sizes does
not affect the equal probability characteristics of the sample.
(7) Steven G. Heeringa and Judith H. Connor, The 1980 SRC/NORC National
Sample Design and Development. Ann Arbor: Institute for Social Research, 1984.
(8) For further details on the 1990 National Sample, see Roger Tourangeau,
Robert A. Johnson, Jiahe Qian, and Hee-Choon Shin, SELECTION OF NORC's 1990
NATIONAL SAMPLE. Chicago: NORC, 1993.
(9) C. Bruce Stephenson, "Weighting the General Social Surveys for Bias
Related to Household Size," February 1978.
(10) Tom W. Smith and Bruce L. Peterson, "Problems in Form Randomization
on the General Social Surveys," July 1986.
(11) See Tom W. Smith, "Attrition and Bias on the International Social
Survey Program Supplement," GSS Methodological Report No. 642 February 1986.
(12) C. Bruce Stephenson, "Probability with Quotas: An Experiment,"
GSS Methodological Report No. 3, April 1979; Tom W. Smith, "Response Rates on
the 1975-1978 General Social Surveys with Comparisons to the Omnibus Surveys
of the Survey Research Center, 1972-1976," GSS Methodological Report No. 5,
June l968; Tom W. Smith, "Sex and the GSS: Nonresponse Differences," GSS
Methodological Report No. 9, August 1979; Tom W. Smith, "The Hidden 25%: An
Analysis of Nonresponse on the 1980 General Social Survey," GSS Methodological
Report No. 16, May 1981; Tom W. Smith, "Using Temporary Refusers to Estimate
Nonresponse Bias," GSS Methodological Report No. 21, February 1983; Tom W.
Smith, "Discrepancies in Past Presidential Vote," GSS Methodological Report
No. 25, July 1982; and Tom W. Smith, "Notes on John Brehm, THE PHANTOM
RESPONDENT: OPINION SURVEYS AND POLITICAL REPRESENTATION. GSS Methodological
Report No. 79, 1993.
(13) Tom W. Smith, "Rotation Designs of the GSS," February 1988.
1972-1994 General Social Survey Cumulative File