FORUM on Corrections Research

Academic Contributions

Gender Research

Traditional and new perspectives for understanding and researching gender and aggression¹

Kelly Taylor²
Department of Psychology, University of Ottawa

This article will introduce literature related to the study and understanding of gender differences in the manifestation of aggression among adults, while discussing how and why this is critical within a correctional environment.

Evidence suggests that aggression manifests itself differently in men and women and that existing instruments and methodological approaches fail to capture the true complexity of aggression or the varying manifestations of aggression in men and women. It is argued that although the measurement of physical or verbal aggression is adequately captured by existing instruments, more covert forms of aggression, such as relational aggression, can be better understood by a gender-informed approach. The incorporation and consideration of such a framework may assist in advancing one's understanding of the factors that affect aggressive behaviour in men and women.

Associations between anger and aggression, aggression and violence and, in turn, the sequence of aggression, violence and crime are evident. Furthermore, Correctional Service Canada's Mental Health Strategy, Intensive Intervention Strategy for women and a variety of specialized treatment/programming approaches underline the organization's understanding that managing emotions and aggression among offenders is important.

Traditional perspectives

Predominant theories of aggression include the frustration-aggression hypothesis (Dollard, Doob, Miller, Mowrer & Sears, 1939) and the AHA! Syndrome (Anger, Hostility, Aggression) (Spielberger, Johnson, Russell, Crane, Jacobs & Worden, 1985). These two theories have dominated past research on aggression, including the way we view and measure this construct. In short, they suggest that aggression is directly related to frustration, anger and hostility, characteristics described as precursors to aggression.

The causes of aggression are complex, however, including biological, emotional, social and cognitive factors. While the interconnectedness of these antecedents to aggressive behaviour is clear, predominant theories and ensuing methodology fail to consider aggression's complexity.

Recent research examined the major meta-analyses conducted on psychological gender differences, including those examining gender differences in aggression (Hyde, 2005). Findings revealed strong evidence for the gender similarities hypothesis³; that is, 78% of effect sizes⁴ related to gender differences are small or close to zero. Hyde (2005) maintained, however, that evidence suggesting that there are only small gender differences in aggression is typically refuted when researchers consider the type of aggression being examined.

The missing piece in meta-analytic findings to date relates to how traditional views of aggression have led researchers to ignore non-traditional forms of aggressive behaviour which are more difficult to observe and measure - specifically, indirect and covert forms of aggression. Also of interest are variables identified within these meta-analytic findings that have the potential to act as mediator variables in the study of gender differences in aggression.

Indirect forms of aggression

Developmental psychologists⁵ suggest that traditional measurement instruments for aggression are most effective in measuring direct aggression. They may therefore underestimate or provide inaccurate findings about the level of aggression among women, for whom direct aggression may not be the preferred mode of expression.

Subtle, social, relational, psychological and a variety of additional indirect forms of aggression allow aggression to take place in a very covert as opposed to overt manner. It is this covertness that makes indirect forms of aggression elusive and difficult to measure in a valid and reliable way.

Crick and Grotpeter (1995) defined relational aggression as behaviour that ". . . harms others through damage to their peer relationships or to the threat of such damage" (p. 313). Damage to a relationship generally occurs through manipulation or control, threatening withdrawal of acceptance or friendship, or using social exclusion or rumour spreading as forms of retaliation.

Most research examining relational aggression has been conducted with children and adolescents. Much of the research with children states that girls are more likely to use relational aggression than are boys. In contrast, most of the studies with adults indicate that the use of relational aggression is higher among men than women. To date, five published studies examine relational aggression among adults (Linder, Crick & Collins, 2002; Loudin, Loukas & Robinson, 2003; Storch, Bagner, Geffken & Baumeister, 2004; Storch, Werner & Storch, 2003; Werner & Crick, 1999). General characteristics found to be associated with relational aggression for adults include peer-rejection, anti-social personality, low pro-social skills, depression, lack of perspective taking, social anxiety and loneliness. Further, there is evidence to suggest gender differences in some of these areas.

. . . indirect and covert forms of aggression, not routinely identified as aggression, are successfully exploited in prison settings.

Another related area of research, directly implicating the offender population and the prison environment, is the work of Jane Ireland (2005). Ireland focuses on bullying among prisoners, and she broadly defines bullying as follows:

An individual is being bullied when they are the victim of direct and/or indirect aggression happening on a weekly basis, by the same perpetrator or different perpetrators. Single incidences of aggression can be viewed as bullying, particularly where they are severe and when the individual either believes or fears that they are at risk of future victimization by the same perpetrator or others. An incident can be considered bullying if the victim believes that they have been aggressed towards, regardless of the actual intention of the bully. It can also be bullying when the imbalance of power between the bully and his/her victim is implied and not immediately evident (2005, p. 5).

Ireland's research identifies four categories of prisoners: ‘pure bullies,' ‘pure victims,' ‘bully/victims,' and ‘those not involved.' In researching these four groups, Ireland underlines that indirect and covert forms of aggression, not routinely identified as aggression, are successfully exploited in prison settings. Here, victims are gravely affected while the likelihood of the perpetrator being identified is vastly reduced. These forms of aggression take a substantial toll on victims, emotionally and psychologically, often resulting in suicide, and yet the incidents often go unrecognized and unchallenged.

Mediating variables

There is also clear evidence that, within research on gender differences in aggression, there are several mediating variables that affect the research outcome (Knight, Fabes & Higgins, 1996). Study characteristics, including type of data collection (direct observations vs. peer-parent-teacher reports vs. self-report), type of setting (field vs. laboratory), type of aggression (e.g., psychological vs. physical), type of surveillance (private vs. semi-private vs. public), and/or freedom of choice to aggress have all been identified as mediating variables.

In addition, concepts such as provocation (Bettencourt & Miller, 1996; Bettencourt & Kernahan, 1997), gender-role identity (Milovchevich, Howells, Drew & Day, 2001; Walker, Richardson & Green, 2000) and even the gender of all those involved in research experiments (Harris, 1992; Harris, 1995) have been particularly emphasized for their relevance in measurement and methodological approaches.

Within a correctional environment, gender-role identity and provocation are two constructs that have been identified as relevant for research findings pertaining to gender differences in aggression. Gender-role identity is distinct from gender in that measuring gender differences through one's selected identity as opposed to their physiological sex characteristics provides unique information. Provocation, traditionally viewed as important in the elicitation of aggression, is more complex than once believed. That is, different forms of provocation affect men and women differently, and the gender of the provoker is influential to the outcome. Gender-role identity and provocation should be given a prominent role in the study of gender differences in aggression.

Linear perspectives

The predominant and perpetuating method used in research on gender differences in aggression seems to be an inexorable linear approach. A linear approach assumes a proportional, sequential development of events. To date, most research has focused on a sequence of aggression beginning with provocation, in turn leading to anger and hostility, inevitably manifesting in some form of aggression. The core assumption suggests a direct relationship between these constructs without considering other mediating variables. Furthermore, this body of knowledge has focused almost exclusively on physical and verbal forms of aggression, which are argued to be expressed primarily by men.

Researchers have identified the difficulties with using linear approaches in the study of emotions (Vallacher & Nowak, 1994). Linear arguments are likely to suggest that there is simply a direct relation between anger and aggression for both men and women, without considering the additional gender-relevant components that contribute to differences in aggression (i.e., mediating variables and form of aggression).

According to Lewis and Granic (2000), linear approaches may impede theoretical progress and hinder one's ability to model the relationship between goals, emotions and emotion regulation in a convincing way, in turn supporting a reductionist approach to personality processes. Furthermore, linear thinking tends to be very static in nature, not allowing for the emergence of more dynamic features of the manifestation of aggression in men and women.

Non-linear approaches to psychological research have the capacity to capture the dynamism and complexity of a variety of social psychological phenomena (Vallacher & Nowak, 1994). Lewis and Granic (2000) describe non-linear approaches as being particularly sensitive to the interactions among the relevant variables.

Non-linear perspectives

Because of the complexity of aggression, approaches that can best capture and conceptually represent this complexity are preferred. Aggression needs to be viewed as a dynamic process as opposed to a fixed and static concept. Consider for example covert, often unobservable, forms of aggression such as relational aggression (explained in the section on indirect forms of aggression). Unlike overt, observable aggression, which appears at one point in time, relational aggression is more likely to occur over a period of time. Static and linear methodological approaches preclude researchers from gathering valid data on this form of aggression; a one-time measure of this construct may actually conceal its occurrence. On the other hand, a methodology that includes a time-series analysis has the potential to reveal this temporal form of aggression. The focus is on observation of phenomena with the aim of clarifying across-time relationships among variables, temporal trends and cycles (Hokanson, Tate, Niu, Stader & Flynn, 1994).

Furthermore, by examining the dynamic relationship and interactions between gender-identity, provocation and aggression, for example, we have the potential to reveal more complex gender differences in aggression, thus providing us with greater explanatory power. Examining independent variables in isolation fails to reveal the relevance of other independent variables, and their combinations, in the varying manifestations of aggressive behaviour in men and women.

A non-linear gender-informed model of aggression

A non-linear gender-informed model of aggression has the potential to accomplish two goals: 1) to encourage researchers to consider this area by means of non-linear statistical techniques and non-linear methodologies; and 2) to place gender as a central component of the model, as opposed to introducing a paradigm that perpetuates an approach to research where gender is either ‘noise' in a system or where female responses are relegated to the category of being an exception to the general rule.

Figure 1 presents a non-linear gender-informed model of aggression. Placing gender differences at the centre of this model and avoiding linear thinking in its development, this model begins to clarify the non-linear (complex) nature of aggression. Within this model, links without arrows are not proposed as causal. Links with arrows are causal, and solid lines represent certainty in the outcome while dashed lines represent that the outcome is feasible but may not always come to fruition.

At the outset, a non-linear design may not appear to differ dramatically from past efforts. Indeed, many of the elements required for a non-linear strategy have been used in traditional linear approaches. Two factors argue for the development of a non-linear framework, however: 1) the addition of gender-sensitive and gender-relevant variables; and 2) the importance of allowing for the complexity of gender differences in aggression to be represented in research designs. With multiple variables interacting in complex manners, implementing a non-linear framework may allow previously undetected gender differences in aggression to emerge. Such a framework is not projected as a panacea, but rather as inspiration for future innovative research and methodology.

For correctional workers, such a framework may inspire a different viewpoint and perhaps a different understanding of the complexity, causes and manifestation of aggression in male and female offenders. Most Correctional Service Canada core correctional programs target aggression to some extent. For programs such as anger and emotions management, family violence and violence prevention, and treatment approaches such as dialectical behaviour therapy, targeting aggression is integral. Furthermore, current organizational initiatives such as gender-specific violence prevention programming for women, and the initiation of research on bullying in our women's facilities, emphasize the significance of this area and how it is directly implicated in the organization's priorities.

All in all, a better understanding of the nature, causes and manifestation of aggression is essential for optimal treatment success and positive correctional outcomes.  

---

References:

Bettencourt, B., & Miller, N. (1996). Gender differences in aggression as a function of provocation: A meta-analysis. Psychological Bulletin, 199, 422-447.

Bettencourt, B., & Kernahan, C. (1997). A meta-analysis of aggression in the presence of violent cues: Effects of gender differences and aversive provocation. Aggressive Behavior, 23, 447-456.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155-159.

Crick, N., & Grotpeter, J. (1995). Relational aggression, gender, and social-psychological adjustment. Child Development, 66, 710-722.

Dollard, J., Doob, L., Miller, N., Mowrer, O., & Sears, R. (1939). Frustration and Aggression. New Haven: Yale University Freer.

Harris, M. (1992). Sex, race, and experiences of aggression. Aggressive Behavior, 18, 201-217.

Harris, M. (1995). Ethnicity, gender, and evaluations of aggression. Aggressive Behavior, 21, 343-357.

Hokanson, J., Tate, R., Niu, X., Stader, S., & Flynn, H. (1994). Illustration of concomitant time series analyses in a case of somatoform disorder. Cognitive Therapy and Research, 18, 413-437.

Hyde, J. (2005). The gender similarities hypothesis. American Psychologist, 60, 581-592.

Ireland, J. (2005). Bullying among prisoners: Innovations in theory and research. Oregon: Willan Publishing.

Knight, G., Fabes, R., & Higgins, F. (1996). Concerns about drawing causal inferences from meta-analyses: An example in the study of gender differences in aggression. Psychological Bulletin, 119, 410-421.

Lewis, M.D., & Granic, I. (Eds.). (2000). Emotion, development, and self-organization: Dynamic systems approaches to emotional development. New York: Cambridge University Press.

Linder, J., Crick, N., & Collins, W. (2002). Relational aggression and victimization in young adults' romantic relationships: Associations with perceptions of parent, peer, and romantic relationship quality. Social Development, 11, 69-86.

Loudin, J., Loukas, A., & Robinson, S. (2003). Relational aggression in college students: Examining the roles of social anxiety and empathy. Aggressive Behavior, 29, 430-439.

Milovchevich, D., Howells, K., Drew, N., & Day, A. (2001). Sex and gender role differences in anger: An Australian community study. Personality and Individual Differences, 31, 117-127.

Spielberger, C., Johnson, E., Russell, S., Crane, R., Jacobs, G., & Worden, T. (1985). The experience and expression of anger: Construction and validation of an anger expression scale. In M.A. Chesney & R.H.

Rosenman (Eds.), Anger and hostility in cardiovascular and behavioral disorders (pp. 5-30). Hemisphere: McGraw-Hill.

Storch, E., Bagner, D., Geffken, G., & Baumeister, A. (2004). Association between overt and relational aggression and psychological adjustment in undergraduate college students. Violence and Victims, 19, 689-700.

Storch, E., Werner, N., & Storch, J. (2003). Relational aggression and psychological adjustment in intercollegiate athletes. Journal of Sport Behavior, 26, 155-167.

Vallacher, R., & Nowak, A. (1994). Dynamical systems in social psychology. San Diego, CA: Academic.

Walker, S., Richardson, D., & Green, L. (2000). Aggression among older adults: The relationship of interaction networks and gender role to direct and indirect response. Aggressive Behavior, 26, 145-154.

Werner, N., & Crick, N. (1999). Relational aggression and social-psychological adjustment in a college sample. Journal of Abnormal Psychology, 108, 615-623.