Search

Using a theoretical framework not previously tested in environmental research, this study investigates the indirect relationship between ecological concern and voting behavior. Results indicate that the effects of ecological concern are mediated by attitudinal, normative, and behavioral intention variables. The results also provide a partial test of the validity of Ajzen and Fishbein's Theory of Reasoned Action. Key environmental, public policy, and communications implications are discussed. James D. Gill is Assistant Professor of Marketing and Lawrence A. Crosby is Associate Professor of Marketing, in the College of Business, Arizona State University, Tempe, Arizona 85287. James R. Taylor is the S.S. Kresge Professor of Marketing, Graduate School of Business Administration, University of Michigan, Ann Arbor, Michigan 48109. The authors gratefully acknowledge financial support provided by the College of Business, Arizona State University and by S.S. Kresge Research Funds. The authors also wish to thank an anonymous reviewer for technical contributions made to the manuscript. uiblic Opinion Quarterly Vol. 50:537-554 ? 1986 by the American Association for Public Opinion Research Published by The University of Chicago Press 0033-362X/86/0050-537/$2.50 This content downloaded from 157.55.39.27 on Wed, 07 Sep 2016 04:25:40 UTC All use subject to http://about.jstor.org/terms 538 JAMES D. GILL, LAWRENCE A. CROSBY, AND JAMES R. TAYLOR First, there have been numerous attempts to determine the antecedents of ecological concern, especially as they relate to the bases of market segmentation (e.g., Crosby and Gill, 1981; Murphy, 1978; Brooker, 1976; Henion and Wilson, 1976; Webster, 1975; Kinnear et al., 1974). Second, efforts have been made to develop or improve ecological concern measures (e.g., Antil and Bennett, 1979; Henion, 1976; Kinnear and Taylor, 1973). Third, several assessments of the impact of ecological concern on consumption and voting behavior have been conducted (Crosby and Taylor, 1982, 1983; Crosby et al., 1981; Lepisto, 1979; Henion, 1976; Kinnear and Taylor, 1973; Herberger and Buchanan, 1971; Kassarjian, 1971). Generally, the research indicates that ecological concern is related to, but is not highly correlated with, consumption behavior. Although often a good explanatory variable of group (market segment) behavior, prediction at the individual level proves difficult. This conclusion parallels that of social and behavioral scientists who have expressed reservations about the value of attitude data in environmental research (e.g., Endo and Nielsen, 1974). The weak relationship between ecological concern and behavior may mean that it is not an important explanatory variable. A more interesting possibility, however, is that ecological concern has an indirect rather than direct influence on behavior, with its influence being gradually diminished through successive levels of intervening variables. This view is consistent with Dunlap (1975) and Mazmanian and Sabatier (1981), who argue that a person's willingness to act upon his environmental attitude is subject to certain mediating influences (e.g., environmental issue salience, situational circumstances, individual differences). The researcher's theoretical perspective, of course, determines the hypothesized intervening variables and their relationships. Van Liere and Dunlap (1982) state that "since progress toward the solution of environmental problems is likely to depend more on proenvironmental behaviors than proenvironmental attitudes, the reasons for the weak relationship between environmental attitudes and behaviors and the conditions under which it can be strengthened clearly deserve examination."

High psychological commitment to a vote decision produces halo effects in the subsequent evaluation of election outcomes and is associated with greater preference stability. Among low commitment voters, preferences are more unstable and subject to less cognitive mediation. Social motivation moderates the influence of social versus personal outcomes in post-decision evaluation for high but not for low commitment voters.

Experimental results showed that education can increase the number of salient dimensions. However, the importance of key dimensions appeared unaffected by education. Information was found to influence product performance expectations and preferences, although sex role differences may exist.