Social Psychology
Page 20
a Head Teacher at a junior school
b Chief negotiator for a union
c The ‘joker’ in a group of friends
d They are all formal roles
11
Group processes
There are not too many of us who have never either worked in a group, or in front of a group. As a student, we may be required to work in a group to collect data, or to present the findings of our individual work in front of a group; as an employee, we may be tasked by our boss to work in a group to find a solution to a problem, and then to present our solution to him, or our colleagues, or the entire company. Having looked at the nuts and bolts of groups previously (Chapter 10), this chapter starts to look at the more intricate and applied workings of a group. It will cover the effects that a group has on an individual’s performance, whether we can predict if a given task would be more effectively completed by an individual or a group, and whether working as a part of a group brings out the optimal performance in an individual. It will conclude by looking at whether groups always make better decisions than individuals.
Audience effects and social facilitation
It has been noted on many occasions that an individual carrying out a task alone performs very differently to when they have to carry out the same task in front of an audience. For example, we might practise our presentation the night before in front of a mirror and think that it went pretty well, but when we deliver it in front of our colleagues the next day, it might be a totally different story.
One of the first people to notice this so-called ‘audience effect’ was Norman Triplett, and he is sometimes credited with conducting the first social psychology experiment. Triplett was very interested in cycling and regularly studied magazines listing the race times of cyclists. He noted with interest that individuals always seemed to cycle faster when others were present (e.g. they were paced) than when they cycled alone. Triplett (1898) speculated that competition between people had the effect of energizing and improving performance on motor tasks. He decided to put this to the test and created a ‘competition machine’; this was basically two fishing reels attached to a board. He then measured the time it took children to reel-in the fishing twine, either when they performed alone, or when they performed side by side with another individual (e.g. in competition). He found that the latter condition promoted faster times, and he concluded that competition energized performance. It should be noted that Triplett did not carry out any statistical analysis on his data (it was the 1890s after all) but merely reached his conclusions by looking at the trends in the data (in fact, a more recent analysis of his data using modern statistical techniques concluded that most of his findings were non-significant!).
However, this idea laid dormant for a number of years until Floyd Allport in the 1920s revived the idea and referred to it as social facilitation. He thought that this idea of performance enhancement would extend to circumstances beyond competition, and believed that the mere presence of others would suffice. A typical experiment to test this assumption would be to ask participants to write down as many associations as they could for a word printed at the top of an otherwise blank piece of paper. It was found that more than 90 per cent of participants were able to generate more word associations when they were performing the task in the presence of others than when they were performing it alone.
Key idea: Social facilitation
The idea that an individual’s performance is affected by the presence of other people.
However, other research conducted using different, and sometimes more complex, tasks didn’t always find the same results. It seemed that sometimes, as well as having an enhancing effect, performance could be adversely affected by the presence of others. This thorny issue remained a problem until the 1960s when a researcher named Robert Zajonc (his surname rhymes with ‘science’) came along with his ‘drive theory of social facilitation’.
According to Zajonc’s theory, the presence of others causes an instinctive arousal which motivates individuals to perform ‘habitual behaviour patterns’ (dominant responses). He claimed that this arousal acted as a ‘drive’, and, based on his theory, made two predictions:
1 The presence of others will facilitate (e.g. improve) performance when an individual’s dominant response is the correct one.
2 The presence of others will impair (e.g. worsen) performance when an individual’s dominant response is an incorrect one.
Using cockroaches (whose dominant response apparently is to seek shade by running in a straight line when a light is pointed at them), and an interesting piece of apparatus (a kind of stadium) that allowed other cockroaches to act as an audience, Zajonc found evidence to support his theory. When in front of an audience, an individual cockroach would perform its dominant response quicker than if alone; however, when required to perform a non-dominant response (in this case, to turn a corner to find the shade), it was performed slower when other cockroaches were present than when alone.
Zajonc concluded that the mere presence of an audience can facilitate or impair an individual’s performance dependent on their dominant response. When our dominant response is the correct one, we are likely to perform it even better in front of others (for example, a well-practised saxophonist will perform their favourite piece even more brilliantly in front of an audience – all other things being equal); when our dominant response is an incorrect one, we are likely to perform worse in front of others. Generally, these findings have been replicated on humans in a number of different domains. For example, above average, or below average, pool players were asked to perform either alone or in front of four passive observers. Results found that, for the good players, performance increased from 69 per cent (alone) to 80 per cent (in front of others), whereas it decreased from 36 per cent (alone) to 25 per cent (in front of others) for the poorer players.
However, competing theories argue that ‘mere presence’ isn’t enough. Cottrell (1972) advanced the ‘evaluation apprehension theory’ which, whilst using the idea of Zajonc’s drive, claimed that increased arousal is caused by fear of evaluation, and not simply the mere presence of others. Its premise is that we have an instinctive need to look good in front of others. Cottrell conducted a simple experiment where participants were asked to carry out a well-learned task in front of either an inattentive audience (they were blindfolded) or an attentive audience who were paying close attention to the participant’s performance. A third condition had other people in the room (mere presence) but only because they were waiting to participate in a different experiment. Cottrell found that only when the audience was attentive was social facilitation found. However, other research has not supported this theory, and it’s also hard to believe that the cockroaches in Zajonc’s original experiment were worried about being evaluated by other cockroaches!
A third more convincing explanation for social facilitation is Baron’s (1986) ‘distraction-conflict theory’. This theory claims that increased arousal is caused by attentional conflict between paying attention to the audience and also to the completion of the task. Together, these two distractions will lead to an impaired performance on difficult tasks, but on easier tasks, the drive is able to overcome the distractions and performance improves. In an experiment to test the theory, participants were asked to perform a hard or easy task either alone, alongside someone completing the same task (which is more likely to be distracting as there is the tendency to compare performance with the other), or alongside someone completing a different task (minimal distraction as they are not a relevant source of comparison). Results showed that the hard task was performed worse in the distraction condition, but the easy task was performed more successfully. Distraction-conflict theory has two main strengths: it can account for the findings from animal studies, such as Zajonc’s cockroaches, and it can allow any form of distraction (e.g. noise) to produce social facilitation effects.
There are some general problems with ‘drive’ theories of social facilitation. For
example, it’s not clear as to what the ‘drive’ actually is. Zajonc’s drive is a bit nebulous and hard to define (and measure), and therefore it is difficult to confirm or refute the existence of it as a mediating mechanism. Although physiological measures of arousal may access the ‘drive’, the absence of physiological arousal is no guarantee that the drive is not operating.
Having looked at the effect that a group (an audience) can have on an individual’s performance, it is now time to look at what happens when a group tries to perform a task; that is, the task requires some form of interaction and co-ordination between individuals in the group. It is likely that the presence of all the group members will have an effect on one another in terms of their performance, and it is also likely that how well the group performs may also depend on the nature of the task itself. This leads to the question of whether it is possible to predict whether any given task can be performed better by a group than by an individual – which is a useful prediction to be able to make when trying to determine the best use of one’s resources.
Steiner’s task taxonomy
In order to make this prediction, Steiner (1972, 1976) produced a ‘task taxonomy’ to tackle this question. The taxonomy enables group tasks to be classified dependent on whether or not it is possible to divide up labour to complete a task, whether there is a predetermined standard for the task that needs to be met, and how the input of individuals contributes to the overall group performance. Each of the three dimensions is best represented by asking three questions:
1 Is the task divisible or unitary?
2 Is it a maximizing or optimizing task?
3 How are individual outputs related to group outputs?
A divisible task benefits from the division of labour, that is different people perform different tasks (such as producing a newspaper or building a house). A unitary task cannot be broken down into sub-tasks (such as pulling a rope).
A maximizing task is an open-ended task that stresses quantity, that is, an individual does as much as they can (such as selling raffle tickets for charity, or generating ideas). An optimizing task has a pre-determined standard that needs to be met, and the goal should neither be under-achieved or exceeded (such as being asked to recruit 50 participants for a study, or bake 12 cupcakes for a charity bake sale).
The final question regarding individual inputs can be broken down into five further types of task:
1 Additive: the output of the group is the sum of all individual inputs. For example, selling raffle tickets for charity or collecting data for a study. The group performs better than the best individual member.
2 Compensatory: the output of the group is the average of all the individuals’ inputs. For example, a team time trial over an obstacle course. The group performs better than many of its members.
3 Disjunctive: the output of the group is that of one selected individual’s input, and it’s not always guaranteed to be the best one. For example, a group has to select one of its members to complete a task, answer a question or solve a problem. The group performs equal to, or less than, the best member.
4 Conjunctive: the output of the group is that of the slowest or least able individual. For example, in a team race, the team’s time is that of their last member to cross the line. The group performs equal to its worst member.
5 Discretionary: the output of the group in relation to its members’ input is not predicted by the features of the task. Instead, it is possible for the group to decide on the best way to complete the task.
Steiner’s task taxonomy makes it possible to classify a task with reference to these three sets of parameters. For example, making a car is a divisible, optimizing, conjunctive task. It should therefore be possible to determine whether a specific task is more suitable for a group than an individual.
Another thing that Steiner predicted was that a group’s actual output would never be equal to a group’s potential output. He believed that this was due to some form of process loss. This could be due to problems in co-ordinating the behaviour of the various individuals (co-ordination loss) or due to psychological factors (motivation loss).
Key idea: Process loss
Deterioration in group performance, compared to individual performance, due to some form of interference between group members.
Ringelmann, in 1913, was one of the first people to investigate this phenomenon, which gave rise to the so-called Ringelmann effect. He asked young men to pull on a rope linked to a measuring device, and found that the more individuals there were pulling, the less the average force was exerted per person. A later experiment attempted to determine the exact nature of this process loss by using real groups and pseudo groups of differing sizes. The real groups had everyone attempting to pull a rope as hard as they could, whereas the only person pulling in the pseudo group was the first individual; the other individuals pretended to be pulling but were actually confederates. In this way, it was possible to distinguish how much of the Ringelmann effect was due to process loss (i.e. the problem of everyone pulling together at the same time) and how much was due to motivation loss (i.e. no need to pull as hard because others in the group would ‘take up the slack’). This deterioration in performance due to motivation loss was christened ‘social loafing’.
Key idea: Ringelmann effect
The amount of force exerted by an individual decreased for each additional individual added to the group effort.
’Social loafing is the tendency for group members to expend less individual effort when working together than when working alone. In most studies of social loafing, this motivation loss is due not to a particular pattern of interaction among group members, but to the absence of interaction coupled with the belief that individual task contributions are pooled.’
(Levine & Moreland, 1998, p. 440)
Social loafing
Social loafing has been found to occur across a number of different experimental tasks. For instance, when participants were asked to clap, cheer and shout, the amount of noise generated by two people together fell by 29 per cent compared to the amount generated by one individual on their own. When there were four people in the group, the output was reduced by 49 per cent, and by 60 per cent when there were six in a group. Generally it has been found that after the group size increases beyond about eight, there is a levelling-off effect of the amount by which the group output is affected. It has also been found that social loafing may be a cultural effect (see Spotlight below).
Spotlight: Is social loafing a cultural phenomenon?
A study by Earley (1993) compared the group and individual outputs of participants (full-time managers) from the USA (an individualistic culture), Israel, and China (a collectivistic culture). They were asked to carry out a range of tasks, over the space of an hour, which approximated those carried out by managers (such as form completion and memo writing – bear in mind these were the days before e-mail existed). In the ‘individual condition’, participants were asked to put their name on each completed item and told that their individual completion rate would be calculated; in the ‘group condition’, they believed they were working as part of a team and were told to put each completed item in a team basket so that the team’s output could be calculated at the end. It was found that participants with a more individualistic outlook (i.e. from the USA) completed more items when they believed they were performing as an individual compared to when they believed they were performing as a team member. Conversely, individuals with a more collectivistic outlook (i.e. from China) completed more items when they worked as part of a team than when they performed as an individual. These results lead to the conclusion that culture is a factor in social loafing.
A number of explanations have been put forward to explain social loafing. These include:
• Output equity: individuals loaf in groups because they expect other people to do the same. They loaf in order to maintain equity.
• Evaluation apprehension: the presence of others in a group allows a
greater degree of anonymity for an individual. They become less identifiable and if they are unmotivated to perform a task, it provides cover for an individual to hide. Performing alone would not provide such cover, thus the individual would be more apprehensive about being evaluated and they would be more likely to overcome their unmotivated state.
• Matching to standard: people loaf because they have no clear idea of the level of performance that is expected of them. They are therefore more likely to set themselves a lower threshold of achievement.
• Social impact theory: the number of other people in a group affects an individual’s attitudes and behaviour. If there are two people in a group, the impact of being asked to do something can be diffused, hence the feeling that the level of effort required is reduced. The more people in a group, the less and less responsibility each individual might feel for the task. As a cynic once claimed: “Teamwork means never having to take all the blame!”
As social loafing can be damaging to the productivity of a group, a number of ways to reduce it have been suggested (see Spotlight below). Of course, social loafing is not an inevitable consequence of working in a group, and people may work harder as part of a team when they consider that a task is of great importance and where they feel that their colleagues may not be working quite as hard as they should. Also, where competition exists with another group, individuals are less likely to loaf; the presence of an ‘outgroup’, and potential intergroup competition, has the potential to transform a group’s behaviour (see Chapter 14).