Abstract
Studies using the self-report altruism scale (SRA) to measure trait altruism have found that altruism is uncorrelated with antisocial motives and positively correlated with being female and number of sex partners. However, it is unknown whether the SRA scale meets basic psychometric standards, such as unidimensionality. The act-frequency format of the SRA likely undermines its ability to isolate any one motive for behaving prosocially. In a pilot study (N =276), factor analysis indicated that the SRA contains dimensions related to the performance of favors (Factor 1), charitable giving (Factor 2), and politeness among strangers (Factor 3). Factor 2 predicted laboratory charitable giving, possibly due to having items related to charity. The present study (N = 814) replicated the three-factor structure of the SRA, with the majority of items loading on Factor 1. Factor 1 was negatively associated with criterion measures of altruism, and was the only factor that positively correlated with antisocial motives and number of sex partners. Factors 2 and 3, as well as other self-report measures of altruism, generally had the exact opposite correlates of Factor 1. Overall, treating the SRA as unidimensional has obscured trait altruism’s relationships to other constructs.