British Journal of Social Psychology

Turner, J. C.; Hogg, M. A.; Turner, P. J.; Smith, P. M.

doi: 10.1111/j.2044-8309.1984.tb00619.xpmid: N/A

Two experiments were carried out to test the hypothesis that failure or defeat increases group cohesion where members feel personally responsible for acting on the basis of group membership. Experiment 1 manipulated success or failure on a cooperative task where subjects (40 male and 40 female undergraduates) experienced high or low choice about doing the group task. Experiment 2 manipulated victory or defeat in intergroup competition under conditions of high or low commitment to the group (the subjects were 64 13‐ and 14‐year‐old schoolgirls). It was predicted and found that negative outcomes produced more cohesiveness than positive outcomes with high choice or high commitment, but less with low choice or low commitment. In addition, counter‐intuitive interaction effects were obtained on measures of self‐esteem and the causal attribution of group performance: in Expt 2 under high commitment, defeat increased cohesion and self‐esteem at the same time that it was attributed to internal factors. It is argued that increased cohesion following failure or defeat is a product of identification with a group to justify and explain behaviour which incurs costs.

Yardley, Krysia M.

doi: 10.1111/j.2044-8309.1984.tb00620.xpmid: N/A

Recently a great deal of attention has been paid to the task of conceptualizing and describing role play techniques. Most commonly, heavy reliance is placed on role theory terminology to achieve this task. In the present paper the author argues that such reliance, particularly on the part of the humanistic proponents of role play, is inappropriate given certain theoretical limitations of role theory. Further this leads to inadequate and misleading notions of what role play actually involves and also to unacceptable implications with respect to the ecological validity of role play experiments. New descriptive conceptualizations of role play are offered, pertaining to concepts of as‐ifness. Lastly the theatrical process is also examined in some detail to elucidate the processes of role play, and to illustrate the distinction between ecological and functional validity.

Gaskell, George; Smith, Patten

doi: 10.1111/j.2044-8309.1984.tb00621.xpmid: 6733400

A model of relative deprivation is proposed drawing on the writings of Gurr (1970) and Runciman (1966) and is discussed in relation to existing studies. Those aspects of the model concerning the intensity of affective reaction to relative deprivation are subjected to empirical test using samples of employed and unemployed black and white youth. The prediction that the degree of relative deprivation is negatively associated with attitudes to the perceived cause of deprivation receives only limited support. Various indices of relative deprivation are compared and it is concluded that present relative deprivation is the most appropriate operational measure. It is concluded that in understanding the attitudes of black and white youth to societal institutions other social characteristics are more important then relative deprivation.

Kemp, N. J.; Rutter, D. R.; Dewey, M. E.; Harding, A. G.; Stephenson, G. M.

doi: 10.1111/j.2044-8309.1984.tb00622.xpmid: N/A

Although it is widely believed that visual cues play an important part in person perception, the experimental literature is inconclusive. Two issues in particular remain unresolved: whether visual information affects participants in conversations in the same way it affects observers; and how the accuracy, quality, and confidence of impressions are related to one another. The purpose of the present experiment was to explore both issues, by incorporating participants and observers in one design, and examining all three aspects of impressions at once. Participants were video‐recorded in pairs, either face‐to‐face or over an audio‐only link, and the recordings were subsequently played to observers under conditions of sound and vision, vision only, or sound only. There were several differences in confidence estimates between participants and observers, but there was no evidence for either accuracy or quality effects. In addition, there was no evidence that observers were affected by whether or not they had visual information, or that face‐to‐face and audio participants differed in the impressions they created. It is concluded that visual cues play a smaller part in impression formation than has generally been supposed.

Bradley, Clare

doi: 10.1111/j.2044-8309.1984.tb00623.xpmid: N/A

Sex differences in academic achievement have often been observed but have usually been attributed to inherent differences between men and women students rather than to differential marking by examiners. The present study indicates that this may not be the case. Sex bias was shown to occur in the evaluation of students' projects. These results cannot be attributed to actual differences in achievement of men and women students. The implications of these findings for marking in general are discussed and recommendations are made for examination procedures which would help to eliminate such bias.

Gunter, Barrie; Furnham, Adrian

doi: 10.1111/j.2044-8309.1984.tb00624.xpmid: 6733401

This paper reports two studies which examined the mediating effects of programme genre and physical form of violence on viewers' perceptions of violent TV portrayals. In Expt 1, a panel of British viewers saw portrayals from five programme genres: British crime‐drama series. US crime‐drama series, westerns, science‐fiction series and cartoons which feature either fights or shootings. In Expt. 2, the same viewers rated portrayals from British crime‐drama and westerns which featured four types of violence, fist‐fights, shootings, stabbings and explosions. All scenes were rated along eight unipolar scales. Panel members also completed four subscales of a personal hostility inventory. Results showed that both fictional setting and physical form had signficant effects on viewers' perceptions of televised violence. British crime‐drama portrayals, and portrayals that featured shootings and stabbings, were rated as most violent and disturbing. Also, there were strong differences between viewers with different self‐reported propensities towards either verbal or physical aggression. More physically aggressive individuals tended to perceive physical unarmed violence as less violent than did more verbally aggressive types.

Ng, Sik Hung

doi: 10.1111/j.2044-8309.1984.tb00625.xpmid: N/A

The effects of equity and social categorization on reward allocation were tested by inserting a veridical input factor into the minimal categorization paradigm unconfounded by any need for inequity prevention. Forty‐seven Hong Kong Chinese students divided a monetary reward between two recipients who had contributed unequally to the reward. The more productive recipient was either an in‐group or out‐group member, or there was no group distinction between them (i.e. non‐categorization). Significant main effects of equity (favouritism to the more productive recipient) and social categorization (in‐group bias) were found on intergroup allocation, as was the main effect of equity on interpersonal allocation. When the forces of social categorization and equity were in harmony, equity effect was accentuated and nearly all the subjects behaved equitably; when in conflict, equity effect was neutralized and subjects were evenly split into equitable and inequitable allocators. No sex effect was found. The need for integrating equity and social categorization is highlighted.

Keyes, Susan

doi: 10.1111/j.2044-8309.1984.tb00626.xpmid: N/A

A pair of studies assessed the validity of two American instruments for use with British samples: the Personal Attributes Questionnaire (PAQ) as a measure of masculine/instrumental and feminine/expressive traits, and the Texas Social Behavior Inventory (TSBI) as a measure of social self‐esteem. Relationships also were explored between gender stereotypes and psychological malaise, as measured by the General Health Questionnaire (GHQ). Data from both studies verified the suitability of the PAQ and the TSBI for use with British adolescent samples. In Study 2, relationships between gender stereotypes and measures of personal adjustment were obtained, providing a partial replication of results previously reported for American samples (Spence & Helmreich, 1978).

Arcuri, Luciano

doi: 10.1111/j.2044-8309.1984.tb00627.xpmid: N/A

Davies, J.; Turner, I. F.

doi: 10.1111/j.2044-8309.1984.tb00628.xpmid: N/A

Friendship choices of 100 children between 7 and 11 years of age were examined. Little religious bias emerged from their choices of friends in class and in the playground. Out of school, however, Protestants tended to play with co‐religionists. This may reflect demographic factors rather than religious in‐group bias.