The role of discussion in changing opinion regarding a matter of fact.

I N a previous article 2 the writer mentioned a series of experiments carried on at Syracuse University, the purpose of which was to provide data which might aid in answering such questions as the following: Under controlled conditions, what per cent of individuals alter their expressions of opinion during discussion? Is the accuracy of individual judgments or opinions increased by discussion? Is the average opinion of the group more accurate after discussion than before? Is the typicality of the group increased during discussion? Are men more or less likely to change their opinions than are women? If there are changes, are they due to discussion itself or to other factors? Is there a relationship between willingness to alter opinion during discussion or to take the leadership in discussion groups, and standing in ascendance-submission, aggressiveness and mental tests? It was also suggested that a topic to be discussed under experimental conditions should allow for (1) interest and comprehension on the part of the subjects, (2) diversity of opinion, (3) quantitative standards of measurement, (4) central point from which to measure dispersion, (5) sufficient range for expression of opinion so that no one might hold a more extreme opinion or judgment than is provided for in the scale of measurement. The present article will describe several experiments which were designed to furnish data bearing on the above-mentioned problems.