L. Richard Hoffman

Group Problem Solving1

Publisher Summary This chapter provides an overview of the experimental evidence on group problem solving, factors that inhibit and promote the effective problem solving, and the present state of group problem solving research. The major barriers to effective problem solving are those conditions, which prevent the free expression of ideas in a group. Factors both obvious and subtle, such as pressures toward uniformity may be detrimental to the groups effectiveness if it prevents search for and discussion of alternative possibilities. The sources of pressures could be personnel characteristics, group members perceived ability, or prior experience. The larger the group the more inhibited the introverted people are, unless they have a strong stake in the outcome; and power structures in groups, while possibly mitigating the negative effects of personality factors, may also create new barriers to effective problem solving. Another contributor to ineffective problem solving is the failure of most groups to organize or plan their attack on the problem. There are several factors that promote the effective problem solving, such as intellectual ability of the group members, the members motivation to work cooperatively on the problem, and the diversity of the points of view and information relevant to the problem within the group. A formally appointed leader can also enhance a groups effectiveness by ensuring the free expression of ideas; however, he hinders the groups operation, if he tends to dominate the discussion. The efforts directed to inventing and testing new ways of encouraging creative group problem solving should advance our understanding of the problem-solving process and, when successful, would have practical value for society as well.

Acceptance and Quality of Solutions as Related to Leaders' Attitudes toward Disagreement in Group Problem Solving

An experiment was designed to test the hypotheses that (1) disagreement in a discussion can lead either to hard feelings or to innovation depending upon the attitude of the discussion leader; and (2) acceptance will depend upon the leaders perception of disagreement as well as upon the decision reached. The test situation utilized a role-playing format in which a foreman attempted to induce three subordinates to accept a change in work methods. Measures of the negative vs. positive effects of disagreement were obtained by the frequency with which the discussion leader reported having problem employees vs. idea men in his group. The results showed that when the foreman most often reported having problem employees, solutions were least likely to be innovative and acceptance was relatively low. When the foreman perceived his subordinates as idea men, innovative solutions increased and satisfaction with the solution was greatest for all concerned, despite the fact that all persons involved had to change their initial positions. Although previous research has indicated that satisfaction is a function of influence over the decision, regardless of its quality, this experiment supplies evidence that the quality of the solution can introduce a further source of satisfaction. This occurs when a solution is one that resolves differences and integrates differing viewpoints to form a new and better product.

DOES TRAINING WITH DIFFERENTIAL REINFORCEMENT ON SIMILAR PROBLEMS HELP IN SOLVING A NEW PROBLEM

An experiment to determine, for achieving correct solutions to the Hatrack Problem whether prior experience on a simpler version of the problem helps and the effects of positive and negative reinforcement. After 30 min., 50.0% of the 30 Ss with no experience but only 25.0% of the 60 Ss with experience on the simpler version solved the problem (p < .05). Positive and negative reinforcement showed no differential effects for 30 Ss receiving each. Previous experience may inhibit productive problem solving by providing too many false directions. Principles of reinforcement developed for learning are limited in their applicability to problem-solving behavior.