Publication type
Conference Paper
Series
American Association for Public Opinion Research Conference, Phoenix, Arizona 13-16 May, 2004
Authors
Publication date
June 1, 2004
Abstract:
Panel studies are of crucial importance to our understanding of the complex, interacting nature of causal processes in the social world. A limitation to valid and reliable inference from panel studies, however, derives from the reflexive nature of humans as research subjects; the act of observation often serves to transform the behaviour of those being observed (Kalton and Citro 2000). In panel studies this source of error falls under the general heading of ‘panel conditioning’ bias and relates to the fact that responses to questions in later rounds of the panel may be influenced by those given in earlier waves. Factors such as ‘satisficing’, cognitive consistency and self-presentation are likely to play important roles in influencing responses between waves. However, while conditioning effects are widely believed to be pervasive in panel survey research, the general consensus appears to be that their impact on marginal distributions is, at most, negligible (Holt 1989).
In this paper, we take a somewhat different approach and examine the effect of panel membership on the internal consistency or reliability of attitude items. Using data from the British Household Panel Study, we evaluate the hypothesis that simply administering political attitude questions serves to increase the apparent consistency or ‘constraint’ of respondents’ Political Belief Systems (Sturgis 2001). Such an effect is argued to arise because the interview prompts respondents to reflect on the subject matter of the questions and to think more deeply about how the issues covered may be inter-related (Jagodzinski et al 1987). Our results show that estimates of internal reliability increase between the first two administrations of the attitude items examined but not between subsequent waves. These increases in scale reliability are, furthermore, predominantly concentrated amongst the least politically engaged respondents in the sample.
Subject
Notes
not held in Res Lib - bibliographic reference only
#511693