Group Dynamics: Forces Inside Groups

1Eight people, less than eight times the effort

Around 1913 the French agricultural engineer Max Ringelmann asked men to pull on a rope, alone or in groups, while he measured the force. Groups pulled harder than individuals, but not in proportion: eight men together pulled with much less than eight times the force of one man.

Part of the loss could be poor coordination. But in 1979 Bibb Latané, Kipling Williams and Stephen Harkins asked students to shout and clap as loudly as they could, alone or in groups, and then blindfolded them and played noise through headphones so they only believed others were shouting with them. Each person still produced less sound when he thought he was part of a group. People were not only less coordinated in groups: they tried less hard.

The study of such forces inside groups is what Kurt Lewin called group dynamics.

21. The structure of a group

• Norms. Muzafer Sherif (1935) placed people in a dark room with a single point of light, which seems to move although it is still (the autokinetic effect). Alone, people gave very different estimates of the movement; in groups, their estimates converged on a shared norm, which they kept when later tested alone. Norms form quickly, even about ambiguous things, and outlast the group that made them.
• Roles. Kenneth Benne and Paul Sheats (1948) described task roles (initiator, information-giver, summariser), maintenance roles (encourager, harmoniser, gatekeeper who lets quiet members speak) and individual roles that serve the person, not the group (blocker, dominator, joker).
• Status. Members' rank in the group, formal or informal. High-status members speak more and are listened to more, whether or not their ideas are better.
• Cohesion. The forces holding the group together. Cohesive groups usually perform better, especially when their norms favour high performance, but cohesion can also feed groupthink.

32. Social loafing

Steven Karau and Kipling Williams (1993) combined 78 studies and found social loafing to be a reliable effect. Their collective effort model says people work hard when they expect their effort to make a difference to an outcome they value. Loafing is reduced when:

• each person's contribution can be identified and evaluated;
• the task is meaningful or the group matters to members;
• the group is small and members know one another;
• members expect others to work hard too.

Their review also found less loafing among women and in collectivist cultures than among men and in individualist cultures. A teacher who gives one mark to a group report invites loafing; one who also assesses each member's part reduces it.

43. Group polarisation

Group discussion tends to push members' views to a more extreme position in the direction they already leaned (Serge Moscovici and Marisa Zavalloni, 1969). People hear new arguments that support their side, and they want to appear at least as committed as others. Polarisation can make cautious groups more cautious, and angry groups angrier, which is one reason why closed online groups can harden opinions.

54. Groupthink

Irving Janis (1972) studied disastrous decisions by expert government teams, such as the United States' failed invasion of Cuba at the Bay of Pigs in 1961, and proposed groupthink (Figure 1).

Janis recommended remedies: the leader should hold back his or her own view at first; one member should play "devil's advocate"; outside experts should be invited; the group should split into sub-groups; and a "second-chance" meeting should be held before a final decision.

How strong is the evidence? Janis built the theory mainly from historical case studies chosen because they ended badly, so it is easy to see groupthink in hindsight. Laboratory tests have given mixed results: a review by James Esser (1998) found that cohesion alone does not reliably produce poor decisions, while directive leadership and closed procedures matter more. The remedies are widely used and sensible, but the full model has not been confirmed.