Effect of Group Means on the Probability of Consensus

In this study, groups who could not reach a consensus were investigated using the group polarization paradigm. The purpose was to explore the conditions leading to intragroup disagreement and attitude change following disagreement among 269 participants. Analysis indicated that the probability of consensus was low when the group means differed from the grand mean of the entire sample. When small differences among group members were found, depolarization (reverse direction of polarization) followed disagreement. These results suggested the groups which deviated most from the population tendency were the most likely to cause within-group disagreement, while within-group variances determined the direction of attitude change following disagreement within the group.

💡 Research Summary

The paper investigates why some groups fail to reach consensus by applying the classic group‑polarization paradigm to a sample of 269 participants. Participants were first surveyed on a socially relevant issue, providing individual attitude scores that were used to compute each group’s mean and the overall sample (grand) mean. The participants were then randomly assigned to small discussion groups (approximately eight to ten members) and asked to deliberate on the same issue for a fixed period. After the discussion, researchers recorded whether the group achieved consensus and, for groups that did not, measured post‑discussion attitude changes.

Statistical analysis centered on logistic regression models predicting the binary outcome of consensus. The primary predictors were (1) the absolute difference between a group’s mean and the grand mean, reflecting how far the group’s initial stance deviated from the population tendency, and (2) the within‑group standard deviation, indicating the homogeneity of members’ pre‑discussion attitudes. Control variables included group size, gender composition, and the overall strength of pre‑discussion attitudes.

The results revealed two robust patterns. First, the probability of reaching consensus declined sharply as the distance between a group’s mean and the grand mean increased. In other words, groups whose initial average opinion was far from the overall sample’s average were the most likely to experience internal disagreement that prevented a collective decision. This effect held regardless of whether the group’s mean was higher or lower than the grand mean, suggesting that relative deviation—not the direction of deviation—is the critical factor.

Second, the direction of attitude change after a failed consensus depended on within‑group variability. When groups were relatively homogeneous (small standard deviation), participants tended to shift their post‑discussion attitudes in the opposite direction of the original group mean—a phenomenon the authors label “de‑polarization.” Conversely, groups with larger internal variability displayed the classic polarization effect, with attitudes moving further toward the initial group mean.

These findings extend existing theories of group polarization, which have traditionally emphasized that agreement leads to more extreme positions. The current study demonstrates that disagreement can produce the opposite effect, especially when members start out with similar views. Moreover, the research highlights the structural role of the group’s position relative to the broader population: groups that are outliers in the distribution of opinions are more prone to internal conflict, while the degree of internal consensus determines whether that conflict results in further extremism or a moderating backlash.

The authors discuss practical implications for organizational decision‑making, public policy deliberations, and any context where collective judgments are required. They suggest that managers and facilitators should monitor both the “distance” of a team’s average stance from the organizational norm and the homogeneity of its members. Teams that are too divergent from the norm may need additional integrative mechanisms (e.g., structured deliberation protocols) to avoid stalemate, whereas maintaining a modest level of internal diversity can foster constructive de‑polarization after disagreement.

Limitations include the use of a student sample, focus on a single issue, and reliance on external observers to code consensus, which may introduce subjectivity. Future research is recommended to replicate the design with more diverse populations, multiple topics, and objective consensus‑measurement techniques.

In sum, the study provides empirical evidence that (1) the farther a group’s mean deviates from the grand mean, the lower its likelihood of reaching consensus, and (2) within‑group variance determines whether post‑disagreement attitude change amplifies or reverses the original polarization. These dual mechanisms offer a nuanced understanding of how group composition influences collective decision outcomes.