Testing contrasting interaction models for discriminating between consensual and dissentient decision-making groups

This study tested three models of group interaction for their ability to discriminate between groups that reach consensus and those that do not. Fifteen mock juries (seven hung and eight conviction) constituted the sample. Model 1, a “functional action model, ” successfully dis-criminated between consensus and hung juries based on the relative number of simple dis-agreements. Model 2, an “interact pattern model, ” successfully discriminated between con-sensus and hung juries based on the sequential redundancy of interaction patterns that perpetuate or resolve ambiguity, that digress from the work at hand, and that clarify or resolve conflict. Model 3, Fisher’s four phases of development, failed to discriminate between consensus and hung juries. Researchers studying small group decision making have a clear mandate to investigate relationships between the structure of social interaction and group performance outcomes. A growing body of research evidence demonstrates that both the processual forms and the thematic content of group communication explain a significant proportion of variance in models linking individual and group-level task interaction and group performance outcomes (Fisher, 1970b