European Survey Research AssociationEuropean Survey Research Association
 
Home About us Membership Conferences Journal Courses Minutes Contact

Login to your account:

Sign up | Reset password

Conferences

Conferences


ESRA2009: Conference main page | Overview of sessions | Time table

Warsaw 2009: Presentations and short courses


Empirical test of a generic cognitive dual-process model to explain attitude-behavior relations and occurrence of response effects

Session: Cognition in survey research (I)

Author:

  • Jochen Mayerl; University of Stuttgart, Germany

Abstract:

According to common dual-mode models of social cognition, two modes of information processing are distinguished: answering to survey questions in an automatic-spontaneous mode in contrast to answering in a deliberative-controlled manner. Based on a generic cognitive dual process model of attitude-behaviour research, it is argued according to the MODE model that attitude accessibility acts as a moderator of attitude-behavior relations in the spontaneous mode only.

Several hypotheses are derived regarding a) the predictive power of attitudes towards behavioral intentions and actual behavior and b) the occurrence of response effects while answering to attitude questions in surveys. The statistical test is conducted on money donations to charities using data of a 2005 nationwide two-wave German CATI-survey with 2002 respondents. The included measurement of response latencies allows the operationalization of the mode of information processing while answering to survey questions.

As a result of several structural equation models, the theorems of the generic dual process model are confirmed empirically. Therefore, attitudes are most predictive towards information processing and behavioural intentions when processing takes place in the spontaneous mode with high chronic attitude accessibility. Additionally, direct effects of attitudes on reported behaviour appear in the spontaneous mode with high accessible attitudes even if controlling the intention as a mediator variable. In contrast, no moderator effects of the chronic accessibility are found in the deliberative mode of information processing. As expected, simple judgement heuristics – e.g. acquiescence – only have a meaningful influence on answering to attitude questions in surveys when information is processed spontaneously with low attitude accessibility. Furthermore, deliberative judgments (i.e. judgments with longer response latencies) are based more on beliefs than spontaneous judgements.