Self-categorization theory

The self-categorization theory presented by Turner, Hogg, Oakes, Reicher and Wetherell (1987) represents a further development of the theory of social identity . It was published in 1987 under the title “Rediscovering the social group. A Self-Categorization Theory ”published. Turner et al. understand this theory not as a substitute for the theory of social identity, but as a more general theory that includes the theory of social identity. The goal of self-categorization theory is not to explain a certain behavior between groups, but to describe the cognitive requirements and mechanisms that make it possible for people to show special behaviors as group members in the first place.

The basic requirements of self-categorization theory

Self-concept

An important concept in self-categorization theory is self-concept, which can be defined as “a set of cognitive representations of the self accessible by a person” (Turner et al., 1987, 44). The self-concept consists of various components that become salient depending on the situation . For example, it is conceivable that someone has the two self-concepts of himself that he is a good chess player but a bad soccer player. At a chess tournament, the self-concept that this person has about their ability to play football will understandably not be as prominent as the concept of their ability to play chess.

Hierarchical organization of categorizations

Another requirement that Turner et al. (1987) assume that the cognitive representations mentioned above are organized in the form of categorizations that are derived from similarities within a class and differences between the classes. These self-categorizations are structured as a hierarchical system, as described by Rosch (1978). This system is similar to a pyramid in that a category that is at the top of the hierarchy includes several subordinate categories but can never be fully described by a single subordinate category. For example, a particular dog might be part of the Pekingese category while another dog might be part of the German Shepherd category. At the next level, Pekingese could be part of the "small dogs" category, while sheepdogs could be part of the "big dogs" category. However, both belong to the "dog" category. They can be assigned to this category via characteristics that they both have (they may both bark, they both come from the wolf, etc.) and via differences, for example to a cat that clearly does not belong to the “dog” category. Neither the German Shepherd nor the Pekingese, for example, make noises that could be described as “meow”; neither the Pekingese nor the German Shepherd have a habit of bringing home dead animals, etc.

With regard to self-concept, Turner assumes three levels of self-categorization. The top level is the concept of oneself as a human being as opposed to animals or plants. The middle, the "social" level is that of the in-group and out-group categorizations. At this level, one classifies oneself as a member of certain groups by looking for similarities to certain groups or making distinctions to members of other groups. The lowest level that Turner describes is the "individual level". This level is based on similarities and differences between oneself and other in-group members.

example

An example should clarify this system: For example, Mr. Meier sees himself as a human being and therefore clearly differs from a horse. Mr. Meier lives in a small German town, is a member of the rabbit breeders' association and an employee of a bank. The rabbit breeders' association usually meets in good company, there is usually a lot of drinking and laughter - so Mr Meier sees himself as a sociable person and is therefore (in his eyes) significantly different from a member of the local reading group. At work, Mr. Meier is very tidy and conscientious, which, in his eyes, sets him apart from someone who belongs to the professional group of artists. A colleague of Mr. Meier, who is also a bank employee, has a family and is happily married. Mr. Meier, on the other hand, has not yet found a suitable woman despite great efforts, which sometimes makes him very melancholy. This is where Mr. Meier differs individually from his colleague.

The metacontrast value

Another prerequisite of the theory is that a categorization takes place on the basis of comparisons between stimuli which are both part of the next higher level of the hierarchical system mentioned above. This means that categorizations and comparisons are mutually dependent. One cannot exist without the other. The creation of a category follows the so-called principle of metacontrast. This means that a collection of stimuli is categorized into a unit in such a way that the differences on a relevant comparison dimension between them are minimal and the differences to other stimuli are maximal. This directly follows the metacontrast value, which Turner defines as follows: The MCR ( meta contrast ratio ) is equal to the perceived difference between the members of a category and other stimuli, divided by the perceived difference within the members of a category.

${\displaystyle MCR={\frac {(der-durchschnittlich-wahrgenommene-Unterschied-zwischen-den-Kategorien)}{(der-durchschnittlich-wahrgenommene-Unterschied-innerhalb-einer-Kategorie)}}}$

This metacontrast value can now be calculated for each member of a category. The member of the category with the highest MCR is, according to Turner et al. also the most prototypical member of the category. It follows logically that the prototypicality of a member of a category is a dynamic value that depends on the comparison category. For example, someone who compares a group of Germans with French might come to a different conclusion about the most prototypical German than someone who compares a group of Germans with a group of English. Therefore, one speaks of the relative prototypicality of a group member. This contradicts Rosch's (1978) view that a category is defined by its most prototypical member, since, according to Turner, the prototypicality of a category member is interrelated with the category itself. Another logical conclusion would be that recategorization occurs if the MCR is less than 1. However, Turner et al. (1987, 47ff.) Not on this case.

A condition for a comparison is that both stimuli are similar on a more abstract level. Ideally, this comparison is made on the least abstract level, which just includes both stimuli. Paradoxically, it follows that a difference between two stimuli can only be determined if they are similar on a more abstract level. An example should clarify this: Dogs and cats can be compared with one another on the “pets” level, so they are similar on this level. A comparison on the “living being” level would also be possible, but not very useful, as this level is too abstract. A comparison on the “canids” level would not be possible, however, since the cat is not included in this category.

Applied to self-categorizations, this means that a comparison of two individuals takes place at the next higher level of the hierarchical system presented above and is therefore a comparison within groups. A comparison between groups is therefore a comparison within the level “human being” and so on.

The salience of a self-category varies with the frame of reference, i.e. the individual self-categories become salient if comparisons are made only within one's own group. The social self-categories become salient when comparisons only take place within the level of “human being” and the self-categories with regard to the conception of oneself as a human being again become salient when comparisons take place between life forms.

Some hypotheses of self-categorization theory

From the above and other requirements, Turner et al. (1987) a large number of hypotheses regarding the formation and function of groups and their phenomena. Only a few selected ones should be presented here.

Relationship between individual and social level

The first by Turner et al. (1987) is in principle equivalent to the first continuum of the theory of social identity . Tajfel and Turner (1986) differentiate between intergroupal and individual behavior, Turner et al. predict an inverse relationship between the individual (the lowest) level and the social (middle) level of self-categorization, with the difference that Turner et al. already relate to perception and not just to behavior. This means that, depending on the situation, people categorize themselves primarily as a member of a certain group or as an individual person. Situations in which membership of a group becomes salient reduce the perception of inter-individual differences within the group and vice versa.

Depersonalization

Another hypothesis states that factors that increase the salience of the comparison dimension at the in-group– out-group level lead to an increase in the perception of the similarity with the in-group members. In this way, individual self-perception is depersonalized, i.e. stereotypes are used that describe the character of in-group membership. According to the third hypothesis, this depersonalization is the cause of all known group phenomena. Turner points out that depersonalization in this context is not to be understood as a loss of individual identity, but rather a change from the individual level of self-categorizations to the social level of self-categorizations. The next of the presented hypotheses revolve around this central concept of depersonalization.

Basic conditions for depersonalization

Membership in one or more groups is a prerequisite for the depersonalization effect described above. Psychological groups in self-categorization theory come into being as soon as two or more people perceive and define themselves via in-group– out-group categories. This idea is thus similar to the definition of social groups by Tajfel and Turner (see theory of social identity ).

Self-categorization theory is equivalent to the assumption that you categorize yourself based on similarities and that you are more likely to assign yourself to a category with which you have a high degree of similarity than a category with which the perceived similarities are less assume that the higher the perceived similarity between them, the more likely a collection of individuals will form a group. If, on the other hand, a greater resemblance to other people is perceived, the probability that the collection of individuals will form a group decreases.

example

An example should explain this: Messrs. Schröder, Fischer, Clement, Koch, Stoiber and Merz are snowed in in a mountain hut. Mr. Schröder, Mr. Fischer and Mr. Stoiber like to play chess, while the other three men prefer the game of Skat. The perceived similarity on the comparative dimension “preferred leisure activity” is therefore greater between Mr Schröder and Mr Fischer, for example, than between Mr Schröder and Mr Clement. So according to the theory it is likely that both the chess players and the skat players will each form a group.

But why do the men named in the example form categories based on the dimension “preferred leisure activity”? Obviously, the associated categories must be very salient. When asked which category is salient, Turner et al. Say that salience is a product of cognitive accessibility to a category and suitability to the situation. In the example above, the gentlemen have a long time ahead of them in which there is nothing to do. Therefore, both the accessibility and the fit to the situation are relatively high. But suppose all six gentlemen are politicians, and Messrs. Schröder, Fischer and Clement are members of one party, while the other three gentlemen are members of another party. After they can leave the mountain hut, their preferred leisure activity plays a relatively minor role. There is a lot to do in everyday political life, so leisure time is out of the question, both cognitive accessibility and the fit of the situation are falling rapidly and it is relatively unlikely that the group constellation described above will continue to exist because other dimensions of comparison are now more accessible and fit more into the new situation.

Social influence

The statements of the self-categorization theory regarding social influence are essentially based on Festinger's theory of social comparison (1954). It is argued there that people need to evaluate their own opinions and abilities. This is done either through the examination of clearly objective, real criteria or through social criteria, that is, through comparison with other people whose opinions, attitudes and skills seem to represent the most accurate representation of reality. The possibility of reducing the uncertainty about one's own opinion, attitude or ability by means of social criteria is obviously preferred if an attractive comparison group is available (cf. e.g. Miller, 1977). Turner et al. present some hypotheses regarding social influence (i.e. the formation of opinions within groups).

Subjective validity

A central role is played by one's own security or uncertainty about the “correctness” of one's own opinions, attitudes, etc. This security is referred to as subjective validity and is directly related to the perceived difference between one's own opinion and the opinion of other similar people. The smaller this difference, the higher the subjective validity and vice versa. A low subjective validity can be reduced:

1. by attributing the difference of opinion to relevant differences between oneself and others,
2. by attributing the difference of opinion to relevant differences in the situation,
3. through mutual social influence to reach an agreement.

Uniformity pressure

The degree of uniformity pressure, i.e. the mutual need to agree, is a product

1. the perceived similarity between both sides,
2. the perceived similarity of the situation (for example, one group member may perceive a situation as threatening, while another group member cannot see a threat),
3. the expected or perceived different mutual opinion about the assessment of the situation,
4. the importance of subjective validity for the group.

This product can actually be understood as a product in a mathematical sense. If one of these factors is zero for a person, he or she will feel no pressure of uniformity. The difference between points two and three is to be understood in such a way that one person must perceive the situation as similar on the one hand, but on the other hand must also believe that the other person assesses the situation in the same way as the person in question. If there is pressure to be uniform, it determines the direction of influence within a group (who influences whom?) From the persuasiveness of the individual group members. The persuasiveness of each group member results from the perceived prototypicality of the group member, because a group member has more persuasive power, the greater the perceived support of the group for his opinion, attitude, etc. and each group member tries to "fit into the group as well as possible ".

Group polarization

Under group polarization refers to deflect the effect that opinions and attitudes of members of a group for a discussion within the group more in that direction that has emerged even before the discussion. For example, if a group of students were asked for their opinion on tuition fees, it could be determined that the students are more likely to be against it. If a second survey, which is carried out after the discussion of the topic within the group, shows that the group is very much against tuition fees, one would speak of group polarization.

This effect is relatively easy to explain using the hypotheses and assumptions presented above. A prerequisite for the effect is that individuals need to evaluate or validate their own opinions and attitudes. One way to do this, as mentioned above, is to make social comparisons. If individuals identify with a group, the most prototypical opinion of the group is also the most valid opinion. It can therefore be assumed that individuals try to adapt their opinion to the most prototypical opinion of the group, the more the pressure for uniformity within the group increases is. Using a few numerical examples, it can be shown that extreme opinions gain more prototypicality the more extreme the most prototypical opinion is. All of the following numerical examples are Turner et al. (1987, 82ff.) Taken.

The following values ​​for A, B and C are intended to reflect the perceived opinion of the individual group members, the values ​​for O the perceived opinion of the out-group members.

Table 1. Numerical example 1 for group polarization

Outgroup	Outgroup	Member A	Member B	Member C	Outgroup	Outgroup
−3	−2	−1	0	+1	+2	+3

The most prototypical member of the group (with regard to the opinion asked) can now be calculated using the meta contrast ratio described above:

${\displaystyle MCR_{x}={\frac {durchschnittlicherUnterschiedzuMitgliedernderout-group}{durchschnittlicherUnterschiedzuanderenGruppenmitgliedern}}}$

${\displaystyle MCR_{A}={\frac {(2+1+3+4)\div 4}{(1+2)\div 2}}={\frac {2.5}{1.5}}=1.{\bar {6}}}$

${\displaystyle MCR_{B}={\frac {(2+3+2+3)\div 4}{(1+1)\div 2}}={\frac {2.5}{1}}=2.5}$

${\displaystyle MCR_{C}={\frac {(4+3+1+2)\div 4}{(2+1)\div 2}}={\frac {2.5}{1.50}}=1.{\bar {6}}}$

In this example, B is the most prototypical opinion and also reflects the mean of the psychological scale. The distance from A and C to B is always the same. Group polarization would therefore also not be expected in this example. But if you choose an example in which the opinions are not distributed symmetrically around the mean value of the psychological scale, the numerical ratios change:

Table 2. Numerical example 2 for group polarization

Outgroup	Outgroup	Outgroup	Member A	Member B	Member C	Outgroup
−3	−2	−1	0	+1	+2	+3

Again the most prototypical opinion of the group is to be determined:

${\displaystyle MCR_{A}={\frac {(3+2+1+3)\div 4}{(1+2)\div 2}}={\frac {2.25}{1.5}}=1.5}$

${\displaystyle MCR_{B}={\frac {(4+3+2+2)\div 4}{(1+1)\div 2}}={\frac {2.75}{1}}=2.75}$

${\displaystyle MCR_{C}={\frac {(5+4+3+1)\div 4}{(2+1)\div 2}}={\frac {3.25}{1.5}}=2.17}$

In this example, too, B is the most prototypical member of the group, but C's opinion is more prototypical than A's, so A's motivation to change his mind towards the prototypical opinion of the group should be greater than C's motivation to do so to do.

It seems that one can accurately predict the polarization effect of a group if one had data on the opinions within the group before and after a discussion, as well as the data on the opinion of the out-group perceived by the members of the group. However, this presupposes that there is a high pressure of uniformity within the group. However, as explained above, the pressure of uniformity depends on the perceived similarity between the group members, the perceived similarity of the situation within the group, the expected or perceived different mutual understanding of the assessment of the situation and the importance of the subjective validity for the group. If only one of these factors is not optimal, the uniformity pressure in the group drops rapidly and a prediction of the group polarization would be much more difficult than in the numerical examples mentioned above.

literature

• Festinger, L. (1954): A theory of social comparison processes. Human Relations , 7, 117-140
• Miller, RL (1977): Preference for social vs non-social comparison as a means of self-evaluation. Journal of Personality , 45, 343-355.
• Rosch, E. (1978): Principles of categorization. In E. Rosch and BB Lloyd (Eds.), Cognition and Categorization (pp. 27-48), Hillsdale, NJ: Erlbaum.
• Tajfel, H. & Turner, JC (1986): The social identity theory of intergroup behavior. In: S. Worchel & WG Austin (eds.): Psychology of intergroup relations (pp. 7–24). Chicago, IL: Nelson Hall.
• Turner, JC, Hogg, MA, Oakes, PJ, Reicher, SD & Wetherell, MS (1987): Rediscovering the social group. A Self-Categorization Theory . New York, NY: Basil Blackwell