Helmut W. Crott

Choice shift and group polarization: An analysis of the status of arguments and social decision schemes.

Two Choice Dilemma Questionnaire items were used to investigate the influence of persuasive arguments and social decision schemes on group decisions. Furthermore, the predictions of the persuasive arguments theory on the polarization of individual preferences were tested. Ss were given lists of persuasive arguments and a second individual decision was requested before the group discussion. The reduced paired comparison median model showed the best fit and the highest hit rate in predicting the group decisions.

Group decision, choice shift, and polarization in consulting, political, and local political scenarios: An experimental investigation and theoretical analysis☆

Abstract Five-person groups discussed for 15- or 30-min one of three different decision tasks: a personal consulting, a political attitudinal, or a local political problem. Groups were instructed to consider themselves committees who were to achieve joint decisions on the matter. Eleven Social Decision Schemes (SDS), resulting in different predictions of group decisions, have been tested for systematic bias and precision. The Black-Median model turned out to be best fitting with regard to both criteria, indicating that the group decision can be described as the result of a voting over alternative courses of action. Additionally, individual shift and group choice shift were analyzed with regard to Persuasive Argument Theory (PAT) and the Black-Median Decision Scheme. In a correlational analysis no indications could be found for a PAT process, whereas the SDS Black-Median covaried with individual shift as well as with group shift.

Shifting students’ to experts’ complex systems knowledge

This paper includes three analyses concerning: expert support in the selection of impact variables for scientific models relevant to environmental planning, the quality of students’ individual estimates of corresponding impacts before and after a group discussion, and the accuracy of artificially‐aggregated judgments of independent groups. Participants were students of environmental sciences at ETH Zurich. The first analysis revealed that during participation in an environmental case study, students’ individual estimates of impacts of variables which have been suggested by experts increased, as compared to the estimates of impacts of additional variables, which have been selected by the students. The remaining analyses consider group discussions on the strength of particular environmental impacts. The quality of the estimates was analyzed referring to expert estimates of the impacts.

Informative intervention to improve normative functioning and output of groups

Formal probability models of group decision making imply that the normative influence of large opinion factions tends to have a negative impact on the quality of decisions in difficult tasks and a positive impact in easy tasks that many people solve correctly. Consequently, an informative intervention (INFO) was developed that advises group members to evaluate task difficulty and to react correspondingly to normative majority or plurality influence. In the study, groups of 5 persons had to solve 2 types of intellective tasks: knowledge questions and logic problems. Compared to control groups, group members in the INFO condition were less overconfident, and achieved more correct group decisions.

Biases in Group Decision Making

Publisher Summary This chapter explains polarization in group decision making by group decision rules, which take into account the rank order of individual preferences over alternatives. Decision rules usually predict the outcome of a group decision on the basis of the individual preferences of the participants by integrating the individual preferences into a collective decision according to the given rule. Most decision rules take into account only the first preferences of the individuals. Some decision rules predict a distribution of group decisions. Others make exact point predictions or specify the range in which the group decision is most likely to fall. Research on shift and polarization has implicitly taken the arithmetic mean of the individual responses as the natural candidate for the group decision. The fact that the actual group decision deviates from the arithmetic mean have been interpreted by psychological effects, such as the effects of group atmosphere, leader characteristics, information exchange effects, and societal values. However, it is easy to verify that most decision rules leads to a shift in relation to the arithmetic mean, when the distribution of the original individual preferences within the group is not completely symmetric.

Momentum Effects in Discussions on Intellective Tasks: Comparing Informed and Non-Informed Groups

A probabilistic model of opinion change (PCD model, Crott, Werner, & Hoffmann, 1996) was applied to analyze the opinion formation process in five-person group discussions concerning intellective tasks. Control groups performed conventional face-to-face discussions, whereas an informational intervention (INFO) was applied in the experimental condition. Analyses performed in line with the PCD model examined the attraction of the correct answer, the size of the subgroups favoring each answer alternative, and a dynamic factor accounting for the possibility that, immediately subsequent to an individual’s opinion change towards a particular alternative, the probability of additional opinion changes to that alternative is dynamically increased. In both conditions the model variants including the dynamic factor were able to adequately describe opinion formation. A stronger attracting influence of the correct answer was observed in the INFO condition compared to the control condition.

Prozeßanalyse von Gruppenentscheidungen zu Aspekten ökologischer Stadtplanung

Zusammenfassung: Die Studie untersucht auf der Grundlage eines stochastischen Modells (Probabilistic Model of Opinion Change Including Distances; PCD-Modell) den Prozes der Meinungsanderungen in Gruppendiskussionen. 90 Studenten/innen der Umweltnaturwissenschaften hatten die Aufgabe, im Rahmen einer Fallstudie zur Neugestaltung eines Stadtteils von Zurich Einflusbeziehungen zwischen relevanten Situationsmerkmalen einzuschatzen. Im Laufe der Diskussion stieg die Hohe der Einschatzungen an und ihre Varianz innerhalb der Gruppen nahm ab. Zur Analyse des Meinungsbildungsprozesses wurden 16 Varianten eines PCD-Grundmodells getestet. Von den beiden postulierten, prozesbestimmenden Faktoren Grose der fur eine Alternative votierenden Untergruppe und durchschnittliche Attraktivitat der Alternativen erwies sich die Untergruppengrose als der bedeutsamere. Daruber hinaus fuhrte die Berucksichtigung der Distanzen zwischen den Alternativen zu einer deutlichen Verbesserung der Beschreibung des Prozesses, da die Meinungs...