Language reflects sociocultural structures, such as gender, and affects individuals' perceptions and cognitions. In gendered languages, male-inflected parts of speech are generally used for both sexes (i.e., masculine generics), thus proliferating stereotypes, inequality, and misattributions. We hypothesized that masculine-generic inflection in a questionnaire would bias women's reports compared with a gender-neutral inflection (e.g., "he or she"). We tested our prediction using an academic motivation questionnaire. We found that women reported lower task value and intrinsic goal orientation in the masculine-generic form compared with the gender-neutral form, and lower self-efficacy than men in the masculine-generic form. These findings suggest that questionnaires and surveys written in gendered languages or translated into them may contain construct-irrelevant variance that may undermine the validity of their scores' interpretations, thus risking the possibility of false conclusions.

People lie more when they work as a group rather than alone. However, do people suspect and morally evaluate groups and individuals differently when they are suspiciously successful? In four experiments, we examine whether (a) suspiciously successful individuals and groups are judged and punished differently and (b) individual group members are judged differently from the group as one unit. Results suggest that people suspect successful groups and individuals to the same extent. However, group members are less likely to be suspected, judged negatively, punished, and reported on, when they are judged as separate individuals compared with as a group. The findings demonstrate a bias in judgment of group members, stemming from the method of evaluation—holistic or separate. We suggest that in order to minimize bias when judging misconduct by a group, the moral evaluation and punishment of all group members should be considered simultaneously.

We examine whether people seek information that might help them make sense of others’ dishonest behavior. Participants were told that a hypothetical partner (either a friend or a stranger) had engaged in a task in which the partner could lie to boost their earnings at the expense of the participant’s earnings. Participants were less likely to search for information that can justify potential dishonest behavior conducted by a friend than by a stranger (Experiment 1). When participants knew for certain that their partners had lied to them, they were less likely to assume that that the lie was justified when told that the partner was a friend rather than a stranger (Experiment 2). The results imply that people are more likely to search for information that may reduce the severity of possible dishonest behavior when a stranger, rather than a friend, is responsible for the behavior.

No abstract