Llectual talent, which in prior work was assessed with survey purchase TAK-385 questions about what is required for success [1, 5]. Moreover, we hypothesized that this linguistic measure of a field’s ability beliefs should also (inversely) predict whether women and African Americans pursue degrees in that field. Minimal linguistic measures similar to ours have been used successfully in past research to examine psychological variables on a large scale. For instance, several studies have used word counts from the language used in online forums (e.g., blogs, chat rooms) to track changes in psychological climate following XL880 site threatening events such as September 11 (e.g., [23, 24]). These word-count measures revealed the expected post-event increase in use of words indicating emotional negativity (e.g., “guilty”), cognitive processing (e.g., “think”), and orientation toward others and the community (e.g., “share”)–an increase that was followed by a gradual return to pre-event baselines [23]. Thus, simple word counts can be a powerful tool for studyingPLOS ONE | DOI:10.1371/journal.pone.0150194 March 3,3 /”Brilliant” “Genius” on RateMyProfessors Predict a Field’s Diversitymacro-level psychological phenomena that are difficult to capture adequately otherwise. Here, we used them as a measure of the extent to which academic fields value brilliance and genius, expecting to find an inverse relationship between the frequency of brilliance-related words on RateMyProfessors.com and the diversity of a field. Beyond testing this key prediction, we used our word-count data to explore several other questions that are relevant to the FAB hypothesis (e.g., is there bias in attributions of brilliance?). We now go on to outline the four research questions we sought to answer with these data. The background societal stereotypes that impugn the intelligence of groups such as women and African Americans are a core component of the FAB framework. There is already evidence for these stereotypes (e.g., [3, 25]), but our data allow us to document these stereotypes as well, at least with respect to gender. (Race information is not available for the instructors on RateMyProfessors.com.) We thus ask whether “brilliant” and “genius” are used more for male than for female instructors (Question #1). While we predict a gender bias in the attribution of intelligence-related superlatives, it shouldn’t be the case that any superlatives are used more often for male than for female instructors. We will thus also tally superlatives that speak fpsyg.2017.00209 more generally to instructors’ skill (such as “excellent” and “amazing”), expecting these to show a less genderbiased distribution. Our main goal, which is to provide a conceptual replication of Leslie, Cimpian, Meyer, and Freeland’s [1] findings, is captured by Question #2: Does use of “brilliant” and “genius” on RateMyProfessors.com predict diversity at the PhD level? We expect to replicate these prior findings: Fields with more brilliance-related language on RateMyProfessors.com (which may indicate a more brilliance-oriented general climate) should have fewer female and African American PhDs. In contrast, the educational and career choices of groups who are SART.S23503 not stereotyped as lacking brilliance, such as Asian Americans, should be unrelated to a field’s emphasis on brilliance. Rather than simply looking at the raw relationship between climate and diversity, we will also compare the predictive power of our linguistic measure of a field’s climate against s.Llectual talent, which in prior work was assessed with survey questions about what is required for success [1, 5]. Moreover, we hypothesized that this linguistic measure of a field’s ability beliefs should also (inversely) predict whether women and African Americans pursue degrees in that field. Minimal linguistic measures similar to ours have been used successfully in past research to examine psychological variables on a large scale. For instance, several studies have used word counts from the language used in online forums (e.g., blogs, chat rooms) to track changes in psychological climate following threatening events such as September 11 (e.g., [23, 24]). These word-count measures revealed the expected post-event increase in use of words indicating emotional negativity (e.g., “guilty”), cognitive processing (e.g., “think”), and orientation toward others and the community (e.g., “share”)–an increase that was followed by a gradual return to pre-event baselines [23]. Thus, simple word counts can be a powerful tool for studyingPLOS ONE | DOI:10.1371/journal.pone.0150194 March 3,3 /”Brilliant” “Genius” on RateMyProfessors Predict a Field’s Diversitymacro-level psychological phenomena that are difficult to capture adequately otherwise. Here, we used them as a measure of the extent to which academic fields value brilliance and genius, expecting to find an inverse relationship between the frequency of brilliance-related words on RateMyProfessors.com and the diversity of a field. Beyond testing this key prediction, we used our word-count data to explore several other questions that are relevant to the FAB hypothesis (e.g., is there bias in attributions of brilliance?). We now go on to outline the four research questions we sought to answer with these data. The background societal stereotypes that impugn the intelligence of groups such as women and African Americans are a core component of the FAB framework. There is already evidence for these stereotypes (e.g., [3, 25]), but our data allow us to document these stereotypes as well, at least with respect to gender. (Race information is not available for the instructors on RateMyProfessors.com.) We thus ask whether “brilliant” and “genius” are used more for male than for female instructors (Question #1). While we predict a gender bias in the attribution of intelligence-related superlatives, it shouldn’t be the case that any superlatives are used more often for male than for female instructors. We will thus also tally superlatives that speak fpsyg.2017.00209 more generally to instructors’ skill (such as “excellent” and “amazing”), expecting these to show a less genderbiased distribution. Our main goal, which is to provide a conceptual replication of Leslie, Cimpian, Meyer, and Freeland’s [1] findings, is captured by Question #2: Does use of “brilliant” and “genius” on RateMyProfessors.com predict diversity at the PhD level? We expect to replicate these prior findings: Fields with more brilliance-related language on RateMyProfessors.com (which may indicate a more brilliance-oriented general climate) should have fewer female and African American PhDs. In contrast, the educational and career choices of groups who are SART.S23503 not stereotyped as lacking brilliance, such as Asian Americans, should be unrelated to a field’s emphasis on brilliance. Rather than simply looking at the raw relationship between climate and diversity, we will also compare the predictive power of our linguistic measure of a field’s climate against s.