The effects of semantic and stylistic variations in language on perceptions of social influence characteristics

This study investigated the effects of variations in certain semantic and stylistic components of a counselor's language on perceptions of that counselor's social influence as measured through the dimensions of perceived expertness, trustworthiness and attractiveness. The language variables used in this study were derived from a linguistic theory called case grammar which is also the basis for the Computer Assisted Language Analysis System used in this study to verify the exact variations in the independent variables. The subjects were 132 undergraduate students divided into four groups. Each subject listened to one of four 20-minute counseling audiotapes that systematically differed only in the semantic and stylistic properties of the counselor's language. Subject perceptions of the counselor were measured via the Counselor Rating Form (Barak & LaCrosse, 1975). A multivariate analysis of variance was performed to determine the effects of the language variations, with one way analyses of variance used as follow-up tests. The results showed that: (a) semantic variations had no significant effect on perceptions of the counselor's social influence, (b) stylistically complex language yielded significantly better ratings of the counselor's expertness than did use of a simple style, (c) there is no interaction effect for these semantic and stylistic measures on the three social influence dimensions. The implications of these findings for counselor training programs, for research on conceptual level and for research on language as a mediator of the social influence process are discussed. Methodological considerations for future research are included.