No. There are several reasons that questions may be asked of only some GSS respondents. First, some questions are not applicable to some respondents. For example, the question on marital happiness (HAPMAR) is only asked of currently married respondents. Second, since 1988 many items in the replicating core of the GSS have been measured for a random two-thirds of each sample. This enables the GSS to monitor more trends than would be possible if each core item was asked of each respondent. There are three overlapping “ballots”, so that the bivariate association of every core item with every other core item can be measured each year. For more detail about the ballot design, see Appendix Q: Rotation and Double Sample Designs. To see which items are on which ballot in any given year, see Appendix U: Variable Use by Year/Ballot. Third, since 1994 the GSS has been administered to two samples in even-numbered years, rather than to a single sample each year. Questions in ISSP modules are usually asked of respondents in only one of these two samples. Questions in some topical modules are asked of respondents in only one of the two samples, but other topical modules cover both samples. In 2002 and 2004, many replicating core items were asked of respondents in only one of the two samples. For more detail about the two-sample design, see Appendix Q: Rotation and Double Sample Designs. To see which items are in which sample in any given year, see Appendix U: Variable Use by Year/Ballot.
