This sample essay on Group Polarisation reveals arguments and important aspects of this topic. Read this essay’s introduction, body paragraphs and the conclusion below.
Group polarisation refers to the tendency for groups to make more extreme decisions or judgements after discussion when compared to the initial views of group members. This phenomenon has interested psychologists since 1961, when Stoner discovered the risky shift, and is central to understanding social influence in groups; therefore the major theories of social influence have been invoked to explain it, so far with little conclusive success.
This essay will cover the origins of group polarisation and examples of it in everyday life, and will then proceed to discuss a number of theories proposed to explain it.
It will then be attempted to ascertain which of these best explains group polarisation and why. Until the 1960’s it was generally accepted that a groups attitudes were the mean or average of its comprising individuals attitudes. In 1961 Stoner asked some business students to make judgements on a choice dilemma questionnaire, this was done individually and then in groups.
Group decisions were found to be, on average, more risky than average pre-discussion, individual decisions. This was known as the “shift to risk”, and was quickly replicated by Wallach et al. (1962).
Wallach et al. evised twelve hypothetical choice dilemmas, and asked participants to rate the lowest level of risk that they found acceptable for somebody to take a risky course of action. They found that 92% of group decisions were riskier than the average individual decision; and importantly, that it resulted in permanent cognitive change, with 39% taking riskier actions after the group decisions (Wallach and Kogan, 1965).
Early explanations for this shift were based around the idea that groups have a diffusion of responsibility; with no-one individually held responsible, they can afford to take risks.
However, the “shift to risk” was soon found to be a “shift to extremity”, with several experiments finding that groups are not always risky. Stoner (1968) found reliable shifts to caution on issues of well being and safety; e. g. prisoner to attempt escape, if the prisoner escapes he lives, but if not he will die. It was concluded that groups polarise, they shift to extremities, ‘the more extreme a group is to begin with, the more extreme it seems to become’ (Brown, 2000, p. 199).
This was a very important discovery which challenged the classic conformity theory experiments which typically showed convergence on the group norms. Sherif’s 1935 auto kinetic experiment showed groups converging on a range of judgements close to the mean or median of preliminary individual judgements. Asch’s line experiments produced similar results (1952, 1955 & 1956). The “risky shift” experiments show simultaneous convergence and shift to a more extreme group norm be it negative or positive.
Polarisation doesn’t just occur in theoretical choice dilemmas, it is a general phenomenon that may be observed in everyday life. Kalven and Ziesel (1966) studied jury decisions in America and found that the initial majority were predictive of 90% of decisions. Blacovich et al. (1975) found that groups bet more than individuals in Blackjack; while McCauley et al. found groups to be more cautious than individuals when betting on horses. Moscovici and Zavalloni (1969) asked French students their attitudes towards President de Gaulle and the USA, the students then formed groups and gave group attitude responses.
These group responses were more extreme than their individual responses with de Gaulle being seen more positively and the USA more negatively than previously. However, almost all the studies conducted into group polarisation have been within laboratories using ad-hoc groups and with no realistic outcome for the participants. Therefore it could be argued that these results cannot be generalised Of the experiments that have been conducted with real-experiment groups, the results have been less reliable (e. g. Semin and Glendon, 1973) though it is a phenomenon that is occurring regularly.
The first theory that tried to account for group polarisation was by Wallach and Kogan (1965); based on diffusion of responsibility, they held that discussion reduced any anxiety felt about the negative consequences of making a choice, because the responsibility is shared. This didn’t explain why shifts to caution or shifts on decisions which resulted in no consequences occurred. The group decision schemes theory (Davis, 1979), was one of the first influential theories in the field; Davis held that groups have implicit rules for combining individual positions on decisions; the most common rule being majority rule.
This means the group decision will be more extreme or “skewed” whenever the majority is biased towards a particular extreme. There were many criticisms levelled at the theory which didn’t adequately resolve why polarisation actually occurred. It was rejected, as skewness does not account for shifts in dyads where no majority can exist; it doesn’t explain why there are shifts in the group median as well as the mean and the shift may also runs counter to the initial majority. Also it didn’t justify why the group shift was internalised and even affected non-participating observers (Lamm, 1967).
Lastly, as Graesser (1975) points out, the theory is not falsifiable because group decision schemes theory is based upon infinite maths calculations and therefore it is always possible to claim that the correct scheme has yet to be discovered. A number of other theories were proposed, including familiarisation theory (Bateson, 1965) and various leadership theories, none proved to be watertight and were dismissed. Only two theories have survived and both have tended to rework one of the two dual processes recognised by the social dependency paradigm: normative and informational influence.
First we shall consider normative influence and social comparison theory; an extension of Festingers (1954) social comparison theory, Sanders and Baron (1977) have modified it to explain polarisation. Normative influence is being liked; fulfilling and conforming to the groups expectations so as to boost positive feeling about oneself. Sanders and Baron tried to prove that people often value more extreme values than they actually hold; however they are generally scared of being too extremist.
However when in a group it is actually found that they hold only moderate views, and so shift them further in the group direction to be seen more positively. ‘The key factor in this social comparison explanation is peoples knowledge of other group members positions relative to the dominant social values in question’ (Brown, 2000, p. 202). Therefore, as Teger and Pruitt (1967) found, there is no need for any discussion, as long as group members positions can be inferred, their will be no need for a verbal exchanging of views.
As the values become more explicit so people polarise more extremely; Baron and Roper (1976) adapted Sherif’s (1936) auto kinetic effect paradigm by effectively informing participants that light distance was correlated with high intelligence. This inevitably led to participants estimating higher distances depending upon the claim made by the participant beforehand. However Burnstein and Vinokur (1977) claim that by associating light movement with intelligence, they have provided the participants with a persuasive argument to estimate larger distances (Burnstein and Vinokur, 1977, p. 27).
Burnstein and Vinokur have taken a totally opposing stance towards solving the polarisation phenomenon. They hold that it is a result of informational influence, of persuasive arguments which cause people to accept information as the truth. Vinokur and Burnsteins persuasive arguments theory (1974) holds that a culturally given pool of arguments exist for and against the decision in question, which participants can delve into and exchange during discussion. For a shift to occur it depends upon the persuasiveness of the new arguments generated in discussions.
However, availability, direction and persuasiveness of these arguments vary, which may allow for experiments whereabouts the arguments do not result in group polarisation. A persuasive argument is defined ‘as a statement judged to have a certain cogency’ (Burnstein and Vinokur, 1977, p. 326), but the arguments impact can only be effective within certain social contexts; for otherwise it wouldn’t be possible to determine its impact. Sanders and Baron have conceded that a persuasive argument also has an effect on group polarisation, and that the two theories work in conjunction with one another.
This was rejected on the grounds that it was too complicated and that persuasive arguments can explain polarisation and convergence. There is a huge wealth of evidence lending credence to the persuasive arguments theory; including polarisation when exposed to arguments without exposure to others positions and polarisation when just thinking about an issue. Burnstein and Vinokur found that (i) if an individual could argue but not compare, then polarisation occurred (1973), (ii) if he could compare but not argue, polarisation vanishes or is greatly reduced (1973, 1975).
Burnstein et al. conceded that social comparison may play an indirect role in influencing polarisation as ‘information about others may guide the person in generating arguments’ (Burnstein and Vinokur, 1977, p. 318). Myers and Lamm also disagree with the persuasive arguments theory, they concluded that ‘(although) the evidence for (informational theories such as persuasive arguments) is compelling… it also appears that group polarisation is not fully explained by a passive process of cognitive learning.
More dynamic processes of cognitive rehearsal and verbal commitment are also likely contributors’ (Myers and Lamm, 1976 cited in Burnstein and Vinokur, 1977, p. 316). Burnstein and Vinokur argue that this is invalid because persuasive arguments involves a dynamic process whereabouts the persons brain acts like a central processing unit. Another criticism of the persuasive arguments theory is that their research generally involves the use of choice dilemma questionnaires introduced by Kogan and Wallach (1964); as these only measure shifts involving a risk-caution dimension they can’t really be used to generalise to choice shifts in general.
Many debates have taken place between exponents of the social comparison theory and the persuasive arguments theory ultimately leading to an inconclusive end. An alternative to Turners (1987) self categorisation theory, formulated by Wetherell (1987), proposes that for convergence to lead to polarization rather than just conformity; arguments have to be persuasive and extreme positions desirable only to the degree that they represent the norms of a group with which the person is psychologically identified.
What group members converge upon is not the ingroup average, but the ingroup norm. This idea was tested on students by Mackie and Cooper (1984) who confirmed that participants attitudes were significantly more affected when they knew the arguments were coming from an ingroup rather than an outgroup. Sanders and Baron admit that there are ‘probably several processes responsible for group induced choice shifts’ (Sanders & Baron, 1977, p. 303).
Burnstein and Vinokur still remain convinced that persuasive arguments can adequately explain group polarisation; but in conclusion (and as Lamm and Myers (1978) deduce) in a naturalistic setting there is no neat dichotomy. There is no clear cut separation between the two main theories, they operate unconsciously within each other and therefore both are necessary in daily life, probably along with several other factors, in order to produce the phenomenon known as group polarisation.