Doctoral Dissertation Research: Conversational Turn-Taking in Network Context

The proposed doctoral dissertation research addresses the question of how social networks influence the mechanics of conversation, specifically turn-taking. The research methodology includes two components: (1) An observational component codes turn-taking in meetings of managers in a large corporation. Because preliminary research has identified it as important for turn-taking, also recorded is the identity of any person addressed by a given speaker. (2) A questionnaire component administers a social network questionnaire repeatedly at 3-month intervals. Included are sociometric questions pertaining to respect, attributions of influence, the age of a relationship, frequency of interaction, liking/affect, and mutual confiding. Data analysis entails transforming the conversation data into multiple 2-dimensional matrices, with each matrix summarizing the incidence of a particular 'transition' involving that pair of individuals, e.g., the frequency with which ego addressed alter immediately following a remark by alter to some third person. A variety of methods for the simultaneous analysis of multiple matrices will be employed to relate these conversational matrices to those deriving from the questionnaire. The research contributes most directly to network analysis, which tends to gloss over the mechanics of interaction, and conversation analysis, which has long maintained an anti-structural stance.