Master of Arts

thesisThis study explores the roles that gender socialization and the acoustics of [s] play in the sociophonetics of perceived gay and straight speech. There have been two main approaches to the study of why some men sound gay and some sound straight: acoustic and social reasons. Many sociolinguistic studies have focused only on the phonetic cues utilized in sounding gay or straight, or on cues listeners use to judge someone as sounding as such (Jacobs et al. 2000; Piccolo 2008; Zimman 2010). In previous studies (e.g., Jacobs et al. 2000; Munson et al. 2006; Zimman 2010), the phonetic variable [s] has been found to be the best acoustic cue for differentiating between gay and straight speech. Studies (Gaudio 1994; Linville 1998; Munson et al. 2006) that have explored social reasons as to why men are perceived as sounding gay or straight have argued that self- identified sexual orientation as well as membership in a larger gay or straight community may influence perceptions of speech. Only one study (Renn 2002) has focused on how gender socialization influences judgments of men's voices as sounding gay or straight. Fourteen self-identified gay and straight Latter-day Saint returned missionaries were recruited as participants in a sociolinguistic interview, which resulted in the collection of demographic information as well as approximately one-hour audio recordings of three differ- ent speech styles (two reading styles and spontaneous speech). These speaker-participants also completed the Kinsey Scale (Kinsey et al. 1948) to determine sexual identity and a psychometric recalled childhood gender socialization questionnaire (Zucker et al. 2006) used to determine childhood activities and habits. Short samples from the interviews were used in a subsequent matched guise study where listener-participants were asked to judge the speech samples as sounding gay or straight along a five-point Likert scale. Four acoustic properties of [s], highest peak frequency, highest amplitude, spectral center of gravity, and spectral skew; actual orientation; style; and the results from the childhood gender socialization questionnaire were analyzed using a mixed effect model of statistics to determine their effects in the judgment of a voice sample being perceived as sounding gay or straight. The results showed that those speaker-participants who had lower scores on the recalled childhood gender socialization questionnaire were more likely to be judged as sounding gay while those who had higher scores were more likely to be judged as sounding straight. Speaker-participants were judged as sounding significantly ""straighter"" when reading the scientific reading sample as opposed to the dramatic reading or when speaking spontaneously. The highest amplitude and spectral skew measures yielded significant results but in the opposite direction that was hypothesized. Those with a lower average skew and average higher amplitude in [s] were judged as sounding straight, those with a higher average skew and average lower amplitude were judged as sounding gay. Results counter to Jacobs et al. (2000); Munson et al. (2006); Zimman (2010, 2013) may be the result of participant outliers or may actually be a phenomenon that is unique to the Mormon male population