Alfred J. Tuchfarber

Political Information Processing: Question Order and Context Effects

Data from two independent field experiments indicate that changes in question order and context may well account for an apparently precipitous decline of interest in politics at the time of the CPS 1978 American National Election Study. Evidence from a question order experiment with the SRC/CPS “feeling thermometers” also suggests that such contextual artifacts may not be atypical. Indeed, because of the many changes in the content and organization of the election studies over the years, context effects represent plausible rival hypotheses for a number of inexplicable shifts and trends in the time-series. In testing these hypotheses the authors derive and validate an information-processing model of how respondents infer their political “states of mind” from observations of their own question-answering behavior in the survey interview. In addition, the authors illustrate the wide applicability of the model tosubstantive problems in the discipline and its implications for the survey-based paradigm in political behavior research.

Experiments in filtering political opinions

Changes in the wording of “core” measures of political attitudes in the American National Election Studies have generated a good deal of controversy about artifactual sources of change in the U.S. electorate. This research, based on several field experiments and replications, investigates the effects of using or not using various types of opinion filter questions that have appeared in the SRC/CPS series over the years. The analysis shows that the use of a filter interacts significantly with a respondents level of education and interest in politics, particularly the latter, in determining whether a respondent will offer an opinion on a given public policy issue. But the study also demonstrates that such an interaction occurs primarily when the question about interest in politics is askedafter, rather thanbefore, a set of political issue items. In interpreting these order-and-context effects the authors develop a self-perception model of how respondents infer their interest in politics from information that isavailable in memory about their own behavior in the survey interview.

Interest in political campaigns: The influence of question order and electoral context

A split-ballot experiment shows that, when people are asked how interested they are in following political campaigns, their response depends not only on the order in which the question is asked, but also on the broader electoral context in which it is posed. When asked how interested they were in following the political campaigns immediatelyafter a question about whether or not they voted in the (1982) election, people were more likely to think they were interested in the campaign, especially if they claimed to have voted, than if they were asked about it immediatelybefore the question on whether or not they voted. This order effect, however, appears to depend onwhen the questions are asked. If asked within a few weeks after the election, there is little or no order effect. But later, as the memory of the campaign fades, the order of the questions makes a sizable difference in the results. This order effect also seems to be more pronounced among better-educated respondents, suggesting that they are more likely to feel pressured by a social norm to vote and to express an interest in political affairs, not only in “real life,” but in the survey interview as well. The paper concludes with a brief discussion of the implications for the design of the interview schedule used in the American National Election Studies.

Education and mass belief systems: An extension and some new questions

Using the SRC/CPSs national election surveys from 1956 to 1976, this paper investigates the effect of education on consistency among the publics domestic policy opinions. Evidence from both gamma correlations and factor analysis indicates that education has neither a strong nor a linear effect on issue constraint over the 20 years covered by the data, for the lowest and the highest education strata consistently show the highest levels of constraint. We do not conclude, however, that education is unrelated to recognition of ideological concepts, for almost one-half of the lowest education stratum do not use “liberal” and “conservative” terms. We conclude that issue constraint does not directly translate into ideology and suggest some new directions that future research should take if we are to evaluate effectively the effect of education on opinion structuring.

The changing nature of urban affirmative action programs: the search for race-neutral alternatives

In January, 1989, the U.S. Supreme Court cast in doubt the legality of minority set-aside programs. The Court required government set-aside programs to meet a “strict-scrutiny” standard of review and required cities to first explore race-neutral means for increasing minority firm participation prior to utilizing any type of rigid quota system. This paper examines how four cities have reacted quite differently, with some cities continuing old practices, and others developing a mixture of new policies and implementation structures. It is argued that the Supreme Courts decision will have only a minor impact on minority contracting programs, with cities seeking to combine affirmative action contracting programs with those that stress economic development.