Groups Stay Poles Apart

Charles Gibb (an educational psychologist in Humberside). Social Work Today – 31 August 1989.

It is probably impossible nowadays to attend any course that doesn’t have at least one small-group session where half-a-dozen or so students are required to tackle collectively a given issue: “Size of income is inversely related to size of dog. Discuss.”

This can be an effective way of learning if groups have a remit related to what the course is about and if tutors do enough groundwork to avoid the tendency of group discussions to meander aimlessly. Besides being an effective learning method, groupwork can have some secondary benefits. Successful groupwork gives participants an insight into how groups work and how to get the best out of them; useful for running workshops on parenting skills, for example.

Also, much professional decision making from establishing tea-money contributions to deciding to offer formal counselling services to a local hospital is carried out within groups of colleagues.

The more adept you are with these groups, the more you might expect to get out of them.

Of course, there are books on group dynamics – whatever that might be.

Getting the most out of the groups doesn’t depend on how much you know; it depends on what you know.

Early studies of group decisions were based on the assumption that consensus reflected the average of individual judgements and therefore tended towards moderation. If half a group thought a 14 year old mugger should be birched and half thought he should be given presents, the whole group would decide to give him a suspended sentence.

However, the assumption of moderation was blown apart in 1961 by a man called Stoner. He discovered that consensus tended towards extreme rather than moderation decisions. Imagine a group of eight people. Individually each is given a dilemma in which they are offered a cut-price video in a pub. The video has clearly fallen off the back of a lorry. What would you do?

Choices range from 1: would definitely take the video to 7: would refuse to have anything to do with it. Stoner – who had set out to investigate decision making in business – calculated an average of these individual responses then put the eight people together as a group and asked them to come up with a single group decision on the same dilemma. He found groups reached riskier decisions than individuals.

This phenomenon is called, “The Risky Shift”. As you can imagine, the practical implications attracted much debate. But so did the notion of Risky Shift itself. Ten years after Stoner’s original work, Fraser, Gouge & Billing, among others, began to find evidence that Stoner’s results were due partly to artefact. Stoner’s dilemmas tended to attract individual decisions that tended towards risk. So although the group decision was more risky, the individuals were already tending that way.

But what if the dilemmas had been constructed so as to encourage caution in individual decisions? It turned out that if they were, the subsequent group decision tended ever more towards caution. Thus you have not a group tendency toward risk, but towards extremity which could be either caution or risk – a group polarisation effect.

For instance: where the average of individual decisions on an issue leans towards risk, the group decision will be even more risky; where the average leans towards caution, the group decision will be even more cautious. The evidence for this is plentiful and easy to replicate. I tried it in 1983 with three groups of five people and, sure enough, polarisation of group decisions was significant in 14 out of 15 decisions made.

It is possible to speculate on why groups seem to polarise decisions like this. Perhaps individuals are aware that when they make decisions they are solely accountable and are therefore less extreme than they might otherwise be. Possibly group discussion among people who tend broadly in the same direction makes more arguments available to each individual so causing firmer resolve and conviction. Maybe the crucial point is that experimental decision making groups and teaching groups are not real and so have nothing to lose by testing extremities. In fact there is evidence that where real decisions with consequences – like over financial affairs – are concerned there is a shift towards moderation. On the other hand, there is also evidence that judges sitting as a group tend to make more extreme judgements than when sitting alone.

What are the implications?

The first that comes to mind is that group decisions will tend to be a mismatch for individual ones. This is a key point since it is almost always individuals who, one way or another, have to implement group decisions. Practising social workers implement group decisions which, if the polarisation evidence is correct, may cause them some conflict. On the other hand, there is nothing unusual about employees being less than wholeheartedly committed to doing what they are told.

But what about the groups themselves? If you are a tutor using small group methods, what use can you make of the polarisation phenomenon? Suppose you were running a workshop for teenagers on the dangers of drug abuse and at one point you split into groups each with a remit to decide: “You are offered heroin. There is no chance of being caught. What would you do?” Presumably, one of the main purposes of such a task is to get the participants to start thinking about the issues rather than acting blindly. The aim is to foster rational thought by providing opportunities for it. You can make use of the polarisation phenomenon to expand these opportunities.

Before doing each piece of group work ask for individual decisions. Then get the group decision. Now if there is polarisation the new group remit is to discuss each individual shift. This adds opportunity for a real discussion on why individuals changed and focuses these individuals more precisely on their views and why they have them. This technique doesn’t apply only to drugs; it can be used for any topic.

What about professional decision making groups of which you are a member? These groups don’t always polarise. But if they do you can see it happening. It’s almost tangible. So, how to manage it?

It all depends. You wouldn’t want to resist if the group discussion was really changing your views. After all you can’t be rigid just because you’ve heard about group polarisation. Knowledge is supposed to make one more, not less, flexible. But if the group doesn’t change your views and slips towards an extreme, then what? If you resist, the evidence suggests the group will pressurise you up to a point then give you up as a bad job and go ahead and polarise anyway.

Having clout in groups doesn’t so much rely on knowing about things like polarisation. It depends on power relationships within the group. It is rare to see a group where the person with the most clout – be it because of knowledge, personality, income, articulateness, job status or whatever – didn’t get his or her own way. Perhaps this is what group polarisation is all about. Perhaps the more powerful individuals make more extreme decisions (because they’re better able to get away with them) and then drag the group along with them.

In this way group polarisation becomes no more than a fancy name for group follow-my-leader. If people will run around blindly extolling universal virtues of group decision-making, they must also be prepared to pay its price; the illusion of agreement.