Predictors of accent-based prejudice

This thesis investigated accent-based stereotyping and prejudice – a line of research originating in Lambert et al. (1960) – by studying perceptions of four accented groups. Participants recruited from Amazon’s Mechanical Turk listened to audio clips where the speakers had native accents from either Toronto, Latin America, Arabic countries, or India. They then evaluated the speakers on several dimensions based on the Stereotype Content Model (SCM) and the solidarity-status-dynamism model (SSD), and completed direct measures of prejudice. Speakers were not evaluated differently on measures of prejudice but were stereotyped differently. Participants higher in right-wing ideologies held more negative stereotypes of speakers and demonstrated greater prejudice. Comparing theoretical models indicated that the more commonly-used SCM provides a suitable alternative to the SSD model. Implications for research on accent-based prejudice are discussed.