Group Decision Making: Two Heads Over One?

Published: 17 December 2014

Could groupthink be harming decision making?

False premises about weapons of mass destruction led to the decision to invade Iraq without any apparent consideration of the risks involved and the failure to develop contingency plans to deal with these risks.

The tendency for behavioural scientists to focus on individual aspects of decision making neglects the fact that many important decisions are taken by groups. In organisations for example, boards of directors or designated senior groups usually make important strategic decisions. In households important decisions are usually taken following discussions between family members.

Involving others is thought to lead to better decision-making for three reasons:

  • groups use more extensive information by drawing on the diversity of group members’ knowledge and experience;
  • groups are more likely to detect errors and bias in reasoning given that each individual’s contribution is appraised by other group members; and
  • each individual is more motivated to work hard given that others are evaluating the time and effort they devote to the task.

However, research challenges these assumptions and shows there are circumstances when group performance is actually much worse than individual decision making. In the November issue of the Harvard Business Review, Reid Hastie and Cass Sunstein outline a broad range of group biases that debunk assumptions about the superiority of groups.

  1. The common knowledge bias debunks the first reason that groups are thought to be better. Research shows a strong tendency for groups to share and discuss the common information held by all members with little or no discussion of the unique information that each individual brings to the situation.
  2. The second reason for group superiority is challenged by research showing that common errors in individual decision making, such as the confirmation bias, are often enhanced in group settings i.e. a stronger tendency to focus on and weight more heavily information that supports initial views and to neglect or discount information that does not. People’s own propensity to follow this mode of thought is reinforced when they see others using it as well. This leads to polarisation of group judgements e.g. the group attitude to risk is much more extreme than the attitudes held by each individual.
  3. The third potential for superiority of groups is challenged by research showing ‘social loafing’ – a tendency for individuals to sit out meetings by making minimal or no contribution. This is particularly likely when individuals are not given space to make their contribution, they feel that their contribution is not listened to or acted on or when the most powerful group member dominates proceedings.

These group related biases, often subsumed under the title ‘groupthink’, have been shown to lead to disastrous outcomes.

For example, these biases were at the heart US Senate Intelligence Committee report that explained how false premises about weapons of mass destruction led to the decision to invade Iraq without any apparent consideration of the risks involved and the failure to develop contingency plans to deal with these risks.

All groups, be it board members, senior management teams or a household, need to recognise and manage these biases if they are to achieve the potential advantages they have over individuals when making decisions.

Professor John Maule 

More from linstock

Should insights from behavioural science about how people really behave direct and determine B2B marketing strategies in the immediate future?

Change in our emotional state affects the way we make judgements and reduce the effectiveness of decisions. There are four important issues to consider.

John has spent the last 35 years looking at how individuals and groups take decisions, publishing books and academic articles on the subject.

Good consultancy often gets lost in the pitch process, where shiny ideas and guesswork are more appealing than grappling with difficult questions.

Leave a Reply

Name *

Mail (not published) *


* Required field