News Release 

Speech and social class perception

Proceedings of the National Academy of Sciences

Researchers examine how speech patterns influence social class perception. Speech patterns are among the most significant means of social class perception. Using audio of speakers collected from the International Dialects of English Archive, Michael W. Kraus and colleagues explored whether hearing only speech allowed strangers to identify social class accurately. When 229 participants listened to 27 speakers uttering 7 words, the listeners accurately predicted the speaker's age, gender, race, and social class more than 50% of the time, which is the random guess rate. However, social class judgments were less accurate than judgments of the other categories. The authors also asked 302 participants to predict the social class of 35 speakers after listening to or reading a transcript of each speaker's self-description. Speaker social class was correlated positively with predicted social class in the speech condition, and participants made more accurate predictions after hearing speech than reading an identical transcript. In a separate experiment, 274 participants with hiring experience listened to answers to pre-interview questions from 20 prospective job candidates who varied in social class and were recruited from the broader New Haven, Connecticut, community. Participants judged the competence, fit, starting salary, and signing bonus of the candidates in favor of those from higher social class backgrounds, compared with those from lower social class backgrounds. The findings suggest that speech patterns may perpetuate social inequality, according to the authors.

Article #19-00500: "Evidence for the reproduction of social class in brief speech," by Michael W. Kraus, Brittany Torrez, Jun Won Park, and Fariba Ghayebi.

MEDIA CONTACT: Michael W. Kraus, Yale University, New Haven, CT; tel: 203-432-6034; email: michael.kraus@yale.edu

###

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.