Abstract

Ninetysix university students (48 males, 48 females) were randomly assigned a partner (whom they did not know well), forming two dyad conditions: (a) samesex, and (b) mixedsex. The 48 dyads were audiotaperecorded in 20minute problem solving interactions, from which 300word language samples were transcribed for analysis. In Study 1, 9 trained observers coded 12 language variables previously shown to distinguish male from female language use. Discriminant analysis results demonstrated that a weighted combination of 8 variables could differentiate male from female interactants: male indicatorsinterruptions, directives, and conjunctions/ fillers begin sentence; female indicatorsquestions, justifiers, intensive adverbs, personal pronouns, and adverbials begin sentence. An analysis of variance of individuals gender discriminant function scores showed greater differences in genderlinked language behavior in samesex than in mixedsex dyads. In Study 2, 231 naive observers rated the 96 interactants, using the Speech Dialect Attitudinal Scale. MANOVA results showed that in samesex dyads, female interactants were rated higher on SocioIntellectual Status and Aesthetic Quality, but no gender difference was found on Dynamism. However, in mixedsex dyads, men were rated higher on Aesthetic Quality, whereas women were rated higher on Dynamism. Taken together, the analyses of the objective language data and the subjective attributional data provide partial support for the GenderLinked Language Effect in samesex dyads and for the attenuation of that effect in mixedsex dyads.

Cite this document ^(BETA)