This study into the nature of “the floor” actually began as an open-ended inquiry into sex differences that might occur beyond the sentence level in the multi-party interaction of five informal committee meetings. Technical difficulties prompted the trying out of several different transcription displays, most of which failed to capture the “feel” of the interaction and each of which biased (in its own way) the perception of what had actually gone on. The type of unconventional display eventually used was intended to show the floor holder in the center of the page, flanked by co-occurring talk. Because there were many episodes for which a single floor holder could not be identified, the primary focus of the study shifted to the nature of the floor itself. Questions about sex differences became a secondary and succeeding focus.
In the analysis, “floor” and “turn” were distinguished on the basis of “participant-sense” rather than technical criteria. Two kinds of floors were subjectively identified: F1, a singly developed floor; and F2, a collaborative venture where several people seemed to be either operating on the same wavelength or engaging in a free-for-all. The two kinds of floors were differentiated objectively by such features as quantity and frequency participation, language functions, number of nonturn utterances, overlaps, and pauses. There were indeed sex/language differences, but these were related to the type of floor being developed. Men took more and longer turns and did more of the joking, arguing, directing, and soliciting of responses F1 's. Turn length and frequency differences were neutralized in F2's, and certain language functions were used by women to a greater extent in F2's than in F1 's. (Conversational analysis, gender and language, qualitative research methodology.)