Question Wording and Partisanship: Change and Continuity in Party Loyalties During the 1992 Election Campaign

Abstract

Gallup macropartisanship varies more over time than aggregate measures of partisanship employing the standard Michigan Survey Research Center (SRC) party identification measure, but previous analyses do not provide direct evidence about why Gallup macropartisanship is more variable. Although these differences could result from the short-term focus of the Gallup party affiliation question, aggregate-level analyses cannot test the effects of question wording directly. Between March and October 1992, we conducted a series of question-wording experiments, employing six statewide computer-assisted telephone interview (CATI) surveys of Michigan adults, including a four-wave panel study. Our analyses strongly suggest that the Gallup measure responds more to short-term political conditions and clearly demonstrate that the Gallup measure is less stable over time. These individual-level results help explain why Gallup macropartisanship varies more over time than aggregate measures of partisanship employing the standard SRC measure and raise questions about the degree to which one can generalize from analyses using the Gallup data to the research literature on party identification.