The effects of 157 university students’gender, attitudes toward women (Spence, Helmreich, & Stapp, 1973), and just‐world beliefs (Lerner, 1980) on their perceptions and attributions regarding the perpetrator and victim of an instance of wife abuse were examined. Hierarchical regression analyses revealed two patterns of results, each differentially associated with participants’gender. Consistent with Heider's (1958) balance theory, males blamed and derogated the wife/victim more as their attitudes toward women became less favorable. Among females, in contrast, those with positive attitudes toward women blamed, but did not derogate, the wife/victim more as their just‐world beliefs became stronger. The latter finding is interpreted in view of research which suggests that women may blame a victim of violence toward women in an effort to gain perceived control over the possibility of their own potential victimization. The implications of these findings for understanding and changing people's perceptions of the victims of wife abuse are discussed.

In a postal survey, 113 respondents completed the Rokeach (1967) Terminal Value Survey with an additional value, "health," and 15 questions about their preventive health behavior (PHB). Correlation and regression analyses showed that health and other values were related to overall PHB, suggesting that health researchers should also consider the role of values other than health. In consistency with Rokeach's theory, respondents who reported good PHB valued health more than did those who reported poor PHB. There were also significant differences in the general value orientation of extraversion versus introversion. Health educators might apply knowledge of such differences to a value confrontation program. As hypothesized, regression analyses for each of the 15 PHB items showed that health value was more likely to be predictive of behavior that involved a direct rather than indirect risk to health. For behaviorally consistent respondents, health value was correlated with both types of behavior, whereas for behaviorally inconsistent respondents, health value was only correlated with direct-risk behavior. This suggests that health researchers should ensure that subjects are aware of the relevance of a PHB to health before concluding that health value, or a related construct, cannot be predictive of PHB. Finally, the regression analyses suggest that it was possible to predict specific PHBs from the more general value ranks. This finding raised questions about the validity of Ajzen and Fishbein's (1977) notion of levels of specificity and provided further support to Rokeach's value theory.

In this article we review research relevant to Rokeach's (1973) suggestion that, by appealing to socially shared conceptions of what is good, people may use values to ego defensively rationalize or justify their attitudes. In line with this value justification hypothesis, research suggests that, although attitudes may originally stem from the relative importance that people ascribe to various values, once formed, attitudes may well produce self‐serving biases that affect both the values that people deem relevant to an issue and the complexity or open‐mindedness of their reasoning about an issue. In addition, just as people may appeal to values to justify their attitudes toward social issues such as nuclear weaponry or abortion, data suggest that people may exaggerate perceptions of intergroup value differences in an effort to rationalize prejudicial intergroup attitudes and justify discrimination. Aspects of the ego defensive use of values that merit elaboration and have yet to be addressed, as well as the more general implications of a functional approach to the study of values, are discussed.

No abstract