Some social mechanics of group decision making: The distribution of opinion, polling sequence, and implications for consensus.

The 1st study extrapolated earlier findings (J. H. Davis et al; see record 1989-15329-001) that the critical 4th voter in 6-person mock juries evenly divided between guilty- and not-guilty-inclined jurors (3,3) were significantly influenced by the preceding sequence (guilty or not-guilty faction voting first) and timing of a straw poll. Results implied that both procedural variables would have an effect on jury verdict distributions, but their magnitude was surprisingly low, especially for sequence. Exp 2 focused only on sequential voting in (4,2) groups in which sequential voting by majority-minority factions first was again observed to influence critical individuals, although the effect was sharply mediated by the "leniency bias." The additional empirical parameter estimates permitted a more comprehensive exploration of probable verdict consequences. Results showed that group-level decisions, under the particular conditions studied, remained counterintuitively robust against significant social influence pressures observed at the individual level, and familiar from numerous studies of conformity and minority influence. Discussion emphasized conceptual hazards associated with inferring group-level actions from individual member behavior. (PsycINFO Database Record (c) 2010 APA, all rights reserved)