关键词: MAU决策程序, 自由讨论, 信息不分享, 群体决策质量

Abstract: In this study, laboratory experiments to compare the decision making p rocedure of MAU (multi attribute utility) with unstructured discussion on the e f fect of decision quality in the 33AbstractIn this study, laboratory experiments to compare the decision making p rocedure of MAU (multi attribute utility) with unstructured discussion on the e f fect of decision quality in the 33% information shared condition we re conducted, and the factors that affected decision quality were also preliminarily explored . Two kinds of materials for experiments were used in this study. One was the pr ofiles of three hypothetical candidates for student body president (less difficu lt decision level) and the other was the profiles of six hypothetical faculty ca ndidates (high difficult decision level). Subjects met in three person groups ( s mall size) or six person groups (large size) to decide which of the candidates was b est suited for the position of student body president or would be the best perso n to teach an English course. Each profile for candidate contained five positiv e items, three neutral items and one negative item. The actual items member in a group received were not identical. 33% of the total items was share d (each memb er received the same items) and the remaining items were unshared (only one memb er re ceived). There were 288 undergraduates of non psychological speciality particip a ting in these experiments who were divided according to three experimental cond i tions, i.e., size of the groups (3 person or 6 person), decision making proce dur es (MAU or unstructured discussion) and the levels of task difficulty (3 candid a tes or 6 candidates). So there were 8 groups of subjects under each condition. A ssignment to experimental conditions and to decision making groups within each condition was random. The results showed: (1) MAU procedure used by groups was more helpful to reduce some negative effects of group dynamic processes. The number of attributes (espe cially, unshared information) used in the MAU was more than that of unstructured condition. (2) Compared to groups under conditions of unstructured discussion, MAU groups did not show their superiority on attitudinal and interaction criteri a (such as self confidence, satisfaction with the final group decision). Under s ome conditions, even lower levels of those criteria were shown in MAU groups tha n those under conditions of unstructured discussion. (3) Much longer time of mak ing a decision could be seen in MAU groups than that in groups under conditions of unstructured discussion. (4) Under the MAU condition, group size had no e ffect on the pooling of shared and unshared information, while the effect of the complexity of the task was mediated by the group size. MAU procedure was suitab le to the tasks with comparatively lower level of complexity as the size of grou p increased (in this study it is 6 person groups). At the end of this paper, so m e suggestions were put forward. It might be a better strategy for a group to poo l information by using MAU procedure first and then to discuss and reach consens us by using unstructured discussion, for example.

Key words: MAU decision making procedure, unstructured discussio n, unshared information, quality of decision making of groups