I agree that increased political diversity in social psychology, like many other forms of diversity, would be a plus because it would foster diversity of thought on social issues. However, Duarte et al. have put forward this enlightened idea in an accusatory manner that mischaracterizes research and theory in the field.

To reveal the target article's biased perspective, I note the authors' analysis of the presumed undermining of social psychology by its political liberalism. Their first example is their claim that social psychologists are in denial about stereotype accuracy. Not so. Gordon Allport, a founder of research on stereotyping, argued for stereotypes' “kernel of truth” (Allport Reference Allport1954/1979, p. 190), and his nuanced theorizing discouraged the notion that stereotypes are mere fictions. Consistent with Allport, understanding of accuracy requires differentiating between accuracy at the group and the individual levels (e.g., Ryan Reference Ryan, Stroebe and Hewstone2003). Sheer logic dictates that group stereotypes, as mental averages of group members, wrongly describe atypical individuals even while they may convey considerable accuracy at the group level. For example, in arguing that that gender stereotypes are “data-driven representations of social reality,” Wood and Eagly (Reference Wood, Eagly, Olson and Zanna2012, p. 91) reviewed numerous studies showing that beliefs about sex differences and similarities are moderately to highly correlated with empirical data on the personality traits, abilities, social behaviors, and occupational distributions of women and men (e.g., Hall & Carter Reference Hall and Carter1999). These same authors also reviewed research spelling out the considerable potential of group stereotypes to mischaracterize individual group members.

Concerning a wide range of other stereotypes, Koenig and Eagly (Reference Koenig and Eagly2014) provided strong evidence of their grounding in observations of group members' behaviors. Their studies tested the proposition that stereotypes of group members derive from people's observations of their behaviors in the social roles in which group members are overrepresented relative to their numbers in the population. Yet, neither Koenig and Eagly nor the advocates of gender stereotype accuracy appear to have raised the ire of their social psychological colleagues.

Despite many social psychologists' considerable open-mindedness concerning group-level stereotype accuracy, most stereotype research addresses, not accuracy, but the negative consequences of stereotypes for individuals. One theme is that stereotypes disadvantage strivers from lower-status groups who attempt to take on new roles. It is stereotypes' descriptive accuracy that lends them the power to suppress the aspirations of those individuals who strive to break the strictures of stereotypes. Such individuals can face backlash (Rudman et al. Reference Rudman, Moss-Racusin, Phelan and Nauts2012) and depressed performance of stereotype-relevant tasks (Steele & Aronson Reference Steele and Aronson1995). Other research emphasizes the many ways that stereotypes legitimize the societal status quo (e.g., Cuddy et al. Reference Cuddy, Fiske, Glick and Zanna2008). Yet, stereotypes' group-level accuracy and their support of the societal status quo are two sides of the same coin.

Another example that Duarte et al. offer of the presumed undermining of social psychology is their claim that social psychology is dominated by the view that prejudice and intolerance are limited to the political right. To support this claim, they feature a small number of studies that yielded one-sided characterizations of conservatives but soon faced contrary evidence produced by other researchers. On this point, Duarte and colleagues appear to be unfamiliar with the massive amount of research in social psychology on the effects of attitudes and ideology on information processing (see Eagly & Chaiken Reference Eagly, Chaiken, Gilbert, Fiske and Lindzey1998). A fundamental proposition of attitude theory is that attitudes exert selective effects at all stages of information processing. Hundreds of studies have tested the proposition that people's attitudes bias information processing in favor of material that is congruent with their attitudes. Such congeniality effects are common in research on exposure and attention to attitude-relevant information and the perception, judgment, and evaluation of such information. Despite complexities arising from competition between pressures toward congeniality and pressures toward accuracy (e.g., Hart et al. Reference Hart, Albarracín, Eagly, Brechan, Lindberg and Merrill2009), neither attitude theory nor its typical findings yield support for the idea that congeniality biases are limited to or stronger among persons on the political right.

Duarte et al. correctly describe social psychology as populated mainly by political liberals. The phenomenon stems from liberals' attraction to a field that they believe produces knowledge that can facilitate social change. Following from the social movements of the last 50 years, adherents of increasing equality on the basis of gender, race and ethnicity, and sexual orientation have flocked to social psychology. However, their preferences for progressive social change do not invariably produce biased science, given that liberal, like conservative, psychological scientists are constrained by the shared rules of post-positivist science. When bias is present, it tends to be corrected over time, as illustrated by the aftermath of Jost et al.'s (Reference Jost, Glaser, Kruglanski and Sulloway2003) article.

Duarte et al. have stigmatized the entire field of social psychology based in large part on their exaggeration of social psychologists' hostility to group-level stereotype accuracy and their overemphasis on a few studies that negatively characterized conservatives. Their article thus displays their lack of broad knowledge of theory and research in this discipline. Also, they have unwittingly illustrated one of social psychology's oldest principles – that attitudes bias information processing, in this case by fostering their highly selective and one-sided characterization of social psychology. It is fortunate that they have published in a journal that allows others to correct their misjudgments.