Virtuous opinion change in structured groups

Abstract Although the individual has been focus of most research into judgment and decision-making (JDM), important decisions in real world are often made collectively rather than individually, a tendency that increased recent times with opportunities for easy information exchange through Internet. From this perspective, JDM factors social context generalizability mundane realism relative to which ignores it. We delineate problem-space within we locate protocols used study or support collective JDM, identify common question posed by all these protocols—‘ What leading opinion change better (‘virtuous change’) agents? ’—and propose modeling approach paradigm using structured groups (i.e., some constraints on their interaction), answering question. This paradigm, based studies judge-adviser systems, avoids need interacting attendant logistical problems, lack power, poor experimental control. report an experiment our effects group size diversity judgmental forecasting performance illustrate approach. The found U-shaped effect probability change, but no amount virtuous change. Implications development more externally valid empirical theories design structured-group techniques discussed.