The Development of an Indirect Aggression Scale

This report presents the development of a psychometric scale, measuring indirect aggression as a personality trait. Twenty items were generated after researching some previously developed indirect aggression scales (Forrest et al. 2005) and the ‘International Personality Item Pool’ website (IPIP 2013).

The scale consisted of 20 items, each of which belonged to one of three different dimensions of indirect aggression – social exclusion, malicious humour and guilt induction. The scale was administered to 20 undergraduate Psychology students from Coventry University. After analysis were conducted it was found that the scale had a high internal consistency and good face validity. Future validation in larger samples may be necessary for the better establishment of the scale as an assessment measuring indirect aggression.

The aim of this report is to present the development of a scale, which can access indirect aggression as a personality trait. Indirect aggression was first introduced by Feshbach (1969) who referred to it as a behaviour that aims to harm an individual’s social relations. The nature of indirect aggression usually enables the aggressor to remain unknown, which makes it a preferable aggressive strategy (Brendgen 2012). There are not many studies that researched indirect aggression in terms of the different theories of personality.

However, the most influential approaches to sex differences in aggression Bandura’s (1977) social learning models, referred to indirect aggression as a trait that is learned over time. For example, according to the social script model children tend to adopt aggressive schemas that initiate and guide their aggressive behaviour (Huesmann and Guerra 1997). Another way of learning to use indirect aggression is through modeling behaviour, which implies that the environment that children are raised in is crucial to the establishment of the specific behaviours that children adopt (McLaughlin 2006).

According to Bjorkqvist et al. (1992) and their developmental theory of aggression, different types of aggression develop at different stages of an individual’s life. Thus, with the development of sociocognitive skills, children begin to employ more indirect forms of aggressive behaviour. Theory of mind is a typical example of a sociocognitive skill, which enables people to explain and predict behaviour on the basis of their own mental representations (Wellman et al. 2001). Children with good Theory of mind skills were found to be better at manipulating other people’s mental representations.

Therefore, good Theory of mind skills were possitively associated with the use of indirect aggressive strategies (Sutton et al 1999). In terms of evolution, for a long time, psychologists argued that men are more aggressive than women because they have to fight off competition in order to have a better access to female mates and pass their genes successfully. However, Campbell et al. (2001) suggested that females have just as much interest to find high quality mates as men and that female competition do exist. Campbell et al. 2001) also suggested that females compete with each other for quality mates using mainly indirect aggressive strategies.

Many studies discovered that direct physical aggression is more typical for men and that females tend to engage in more indirect aggressive behaviour than men (Bjorkqvist et al. 1992; Crick et al. 1997). Some evolutionary strategic models (Archer 2001; Hess and Hagen 2006) present the idea that sex differences in aggression result from the different selective environments of males and females.

Based on this statement, Hess (2006) concluded that indirect aggression for women and physical aggression for men are simply adaptations. Goldstein (1999) argued that our behaviour is a result of our experience and our genetic makeup. In a study exploring aggression reduction strategies, he discovered that if a particular behaviour, such as aggression is interrupted at an early age, the chances of that behaviour to progress into a serious problem are very small. Goldtein (1999) also concluded that people raised in different environments process things in a different way.

This cognitive approach to learned behaviour could be applied to indirect aggression as well. For example if an individual was raised in an environment where applying psychological and emotional abuse is a regular practice, their responses to indirect aggression would differ from those of a person who’s never experienced such abuse. The most popular scale for measuring Indirect Aggression in adults is Forrest et al. ’s (2005)

Scale, which was developed in two separate versions – an Aggressor version and a Target version. Items in the scale were generated on the basis of a few qualitative interviews that were conducted. The scale measured indirect aggression on three levels – social exclusion, guilt induction and malicious humour. This variety of subscales allowed Forrest et al. (2005) to create a scale that can measure indirect aggression in greater depth. The current study will consider all three dimensions of indirect aggression, measured in Forrest et al. ’s (2005)

Scale in order to develop a personality trait assessment, which includes a sufficient amount of various behaviours related to indirect aggression. However, as a scale measuring covert aggression, self-reports of indirect aggression are less likely to be truthful, because in general indirect means of aggressive behaviour are used exactly in order to hide someone’s harmful intentions (Bjorkqvist 1994).

The scale development in this report requires a detailed research of the previously developed indirect aggression scales in order to be determined the types of items that will be most appropriate to include. Additionally, due to the nature of the personality trait being tested, selfreporting may result in socially-desirable, rather than honest answers.

The design used in the study was scale development. The items from the scale were tested using a mixed correlation design. All items were intercorrelated with each participant’s response. Participants The scale was tested on 20 undergraduate Psychology students (5 males and 15 females) at Coventry University. The mean age of the participants was 21. 15 (SD=1. 46) Materials An Indirect Aggression Questionnaire was created (see Appendix 6) in order to measure the extend to which people employ indirect aggressive strategies in their life.

The questionnaire consisted of 20 items and for each item participants could indicate the extend to which they relate to it, using a Likert Scale. The format of the four-level Likert Scale was the following: Never; Sometimes; Often; Very Often. Some examples of the items in the scale are: ‘I have purposefully ignored people’, ‘I have intentionally stopped talking to a friend in order to make them feel bad’ and ‘I have intentionally made other people look stupid’. A debrief form was distributed to all participants after they completed the scale (see appendix 5).

The first step towards designing a new scale was to do some research on Indirect Aggression in general and all the personality theories that this particular trait relates to. The examples of indirect aggressive strategies found in some journal articles were such as ‘rumor spreading’, ‘social exclusion’, ‘rejection’ etc (Archer 2004; Card et al. 2008). In order to provide further ideas of the types of behaviours that could be included in the new Indirect Aggression Scale, the ‘International Personality Item Pool’ website (IPIP 2013) was researched as well.

Based on Forrest et al. ‘s (2005) study, three different dimensions measuring indirect aggression were included in the scale – social exclusion (7 items), malicious humour (9 items) and guilt induction (4 items). As a result, 20 items that indicate indirect aggressive behaviour were generated. It was made sure that all items included in the scale did not repeat, because repetition of items would have probably faked the internal consistency of the results. After the questionnaire was ready, it was distributed in a lecture to 20 third level undergraduate Psychology students.

Before completing the scale, each participant gave their verbal consent to take part. When all scales were completed, participants were asked to give their feedback about the scale in terms of whether the scale seemed to measure what it was intentioned to measure. After all scales were collected participants were debriefed and all data was analyzed. Cronbach’s Alpha was performed in order to check internal reliability. Results The mean age of participants was 21. 15 years (SD=1. 46). The total number of participants was 20(5 males and 15 females).

The descriptive statistics for all tested items showed the mean value (M=41. 00) and the standard deviation (SD=14. 056). The median score (38. 00) and the mode (23. 00) were also revealed. The frequency distribution showed a skewness of 0. 566 (SE=0. 512) and kurtosis of -0. 872 (SE=0. 992). The z scores were calculated and suggested that the scale items were normally distributed.

In order to check the internal reliability of the items, a Cronbach’s Alpha was conducted. The main purpose of Cronbach’s Alpha is to give an average correlation between items, by analyzing the inter-correlations of all items (Cronbach 1951). The 20 items gave a score of 0. 952, which could be interpreted as high internal consistency . However, after reviewing the Item-Total Statistics it was determined that if item 9 was deleted, the internal consistency would increase even more. Therefore, item 9 was deleted. and as a result the Cronbach’s Alpha of the scale increased to 0. 55. On the basis of the received feedback about the scale, the face validity of all items was determined to be good enough to measure behaviours associated with indirect aggression.

Discussion

This study reported the development of a psychometric scale, measuring indirect aggression as a personality trait. The scale measured three dimensions of behaviours related to indirect aggression – social exclusion , malicious humour and guilt induction. The final scale consisted of 19 items  and showed great internal consistency, which could also be interpreted as high validity.

The face validity checks suggest that the main limitation of the scale was that the four-level Likert scale for all items was heavily weighted to positive. A possible solution for this would be to add one more level, which will be negatively keyed, such as ‘Not Often’. Furthermore, the scale was tested on a small group of undergraduate students within the age range of 18-25 years and with a prevalence of the female participants. Therefore, the scale can not be considered as valid for other populations.

However, the total scores were normally distributed, which indicates that there was enough variance within the participants sample to properly test the scale. Due to the nature of the personality trait being tested and the fact that the scale was administered in a lecture (decreased perceived anonymity), self-report measures are likely to be more socially- desirable, rather than honest. Therefore, in order to avoid the possibility of participants responding in a socially desirable manner, some rational techniques could be used such as to build in control features into the self-report instrument (Paulhus 1991)

In the future it would be interesting to distribute the scale to a larger sample of participants with an equal number of males and females. In this way further validity of the scale will be conducted, which can contribute to the better establishment of the scale as an assessment measuring indirect aggression. Furthermore, in order to test Bandura’s (1977) social learning theory the scale could be tested on people that participate in the same social group. According to Bandura (1977) people tend to adopt behaviours that they observe in their social environment.

Since, indirect aggression is considered to be a learned behaviour (Huesmann and Guerra 1997), the prediction will be that people participating in the same social group will show similar results on the scale. An indirect aggression scale was developed and indicated to be reliable and valid for the population that it was tested on. The scale contained less items than Forrest et al. ’s (2005) Indirect Aggression Scale, which could be considered as a benefit as it does not bore the participant and holds their attention better.

After being validated to larger and various populations, the scale could be used as a measure, which can correctly indicate the levels of indirect aggression employed by individuals. Additionally, the scale could be used as an assessment of indirect aggression as a learned behaviour, which can contribute to the better understanding of the learning perspective of the different theories of personality.

References

1. Archer, J. (2004) ‘Sex Differences in Aggression in Real-World Settings: A Meta-Analytic Review’. Review of General Psychology 8, 291-322
2. Bandura, A. (1977) Social Learning Theory. Englewood Cliffs, NJ: Prentice-Hall Bjorkqvist, K. (1994) ‘Sex Differences in Physical, Verbal and Indirect Aggression: A Review of Recent Research’. Sex Roles 30, 177-188
3. Bjorkqvist, K. , Osterman, K. and LKaukiainen, A. (1992) ‘The Development of Direct and Indirect Aggressive Strategies in Males and Females’.
4. Bjorkqvist, K. (1994) ‘Sex Differences in Physical, Verbal and Indirect Aggression: A Review of Recent Research’. Sex Roles 30, 177-188
5. Brendgen, M. (2012) ‘Development of Indirect Aggression Before School Entry’. Encyclopedia on Early Childhood Development Campbell
6. Muncer, S and Bibel, D. (2001) ‘Women and Crime: An Evolutionary Approach’. Aggression and Violent Behaviour 23, 229-238
7. Card, N. , Stucky, B. , Sawalani, G. and Little, T. (2008) ‘Direct and Indirect Aggression During Childhood and Adolescence: A Meta-Analytic Review of Gender Differences, Intercorrelations, and Relations to Maladjustmen’. Child Development 79(5), 1185-12229
8. Crick, N. , Casas, J. and Mosher, M. (1997) ‘Relational and Overt Aggression in Preschool’. Psychology 33, 579-588
9. Crombach, L. (1951) ‘Coefficient Alpha and the Internal Structure of Tests’. Psychometrika 16(3), 297-334
10. Feshbach, N. (1969) ‘Sex Differences in Children’s Models of Aggressive Responses Towards Outsiders’. Merrill-Palmer Quarterly 15, 249-258
11. Forrest, S, Eatough, V and Shevlin, M. (2005) ‘Measuring Adult Indirect Aggression: The Development and Psychometric Assessment of the Indirect Aggression Scales’. Aggressive Behaviour 31, 84-97
12. Goldstein, A. , Syracuse, U. (1999) ‘Aggression REduction Strategies: Effective and Ineffective’. School Psychology Quarterly 14(1), 40-58
13. Hess, N. and Hagen, E. (2006) ‘Sex Differences in Indirect Aggression. Psychological Evidence From Young Adults’. Evolution and Human Behaviour 27, 231-245