Situational and Personality Factors as Direct or Personal Norm Mediated Predictors of Pro-environmental Behavior: Questions Derived From Norm-activation Theory

Abstract

Studies that use the norm activation theory (Schwartz, 1977) to explain pro-environmental behavior often focus on personal norms and on two situational activators, i.e., awareness of need and situational responsibility (e.g., Vining & Ebreo, 1992). The theory's other situational activators, efficacy and ability, and its personality trait activators, awareness of consequences and denial of responsibility, are generally ignored. The current article reports on two studies - a mail survey among the general public (N = 345) and a laboratory experiment among university freshmen (N = 166)–that found that (1) inclusion of additional activators improved the norm activation theory's potential to explain pro-environmental behavior and (2) personal norms significantly mediated the impact of activators on pro-environmental behavior. Theoretical issues and issues concerning environmental management evoked by these results are discussed.

ACKNOWLEDGEMENTS

This study was performed at the Centre for Energy and Environmental Research as partial fulfillment of the Ph.D research of Paul Harland who is now senior researcher at the The Hague Police Service, The Netherlands. Henk Staats and Henk Wilke are at the Centre for Energy and Environmental Research, Department of Social and Organizational Psychology, Leiden University, The Netherlands.

We are indebted to Peter de Heus and Mieneke Weenig for their helpful suggestions regarding statistical analyzes. We are extremely indebted to John Thøgersen for his valuable comments on an earlier version of the manuscript.

Notes

¹It should be noted that all environmental studies published so far label situational activators with names that were originally reserved for personality trait activators. In those studies, situational awareness of need, for instance, is usually termed ‘awareness of consequences’, a label that had originally been reserved in NAT for a personality trait activator, i.e., a person's receptivity for the consequences that his or her behavior might have for the welfare of others. Following this practice causes problems of discriminating the constructs when both situational and personality trait activators are included, as in the current article (Study 2). Therefore, we stick to Schwartz' original labels here and thus re-label activators included in studies we refer to when necessary.

²A copy of the full questionnaire is available on request from the first author.

³Separate analyses for each of these awareness of need items were performed in order to test whether these modest scale reliabilities influenced our results. Minor differences were observed that did not influence the final results or lead to different conclusions. Therefore, analyses that are based on the 2-item measure of awareness of need will be presented.

⁴Separate analyses lead to the same conclusions (see footnote 3).

Note. All measures range from 1 to 7, except the personal norm measures that range from 1 to 4. A higher score indicates a more positive stance on the construct. A number of respondents did not own cars, therefore the N for “Other transport forms” is lower.

**p < .01.

***p < .001 (two tailed).

Note. Entries are beta weights.

^a The z-score for mediation was calculated only for activators that were related to personal norms and to behavioral intention (Kenny et al., 1998).

*p < 0.05.

**p < 0.01.

***p < 0.001 (two tailed).

⁵The available number of participants did not permit to test a design that would include the four situational activators as well as the two personality trait variables. As our focus was on the newly introduced activators efficacy and ability as well as on the personality traits, we decided to exclude situational responsibility in favor of the other activators.

⁶We would like to thank Shalom Schwartz and John Ferrari for providing us with the measures for the personality traits.

⁷Whether modest alpha scores are usually obtained with the awareness of consequences scale could not be determined because scale reliabilities are not reported elsewhere. Dropping one of the stories did not produce a higher alpha. Acknowledging that a scale with a higher alpha would be preferable, we decided to use this less optimal one.

Note.

^a The z-score for mediation was calculated only for activators that were related to personal norms and to behavioral intention (Kenny et al., 1998).

^b Means are not reported for the experimentally manipulated factors (which have scores 0 or 1).

^c A higher score represents a higher intention to deny responsibility for one's behavior.

^d A higher score represents a higher awareness of the consequences of one's actions

‡p < .10.

*p < 0.05.

**p < 0.01.

***p < 0.001 (two tailed).