The Recall of Past Partisanship: Feeble Memories or Frail Concepts?

Abstract

AbstractIt has been suggested that recall questions are unreliable measures of change in party identification. Data reported in this article confirm that 50 per cent of respondents in the 1974-1980 Canadian National Election Study three-wave panel had inconsistent patterns of recall. This finding urges caution in analysis; more importantly, such inconsistency raises questions about how to interpret recall behaviour in the light of party identification theory. The available evidence and unclear theory seem to point toward the possibility that at least one-half of the national sample lacked a meaningful federal party identification.