Elsevier

Aggression and Violent Behavior

Volume 37, November 2017, Pages 91-101
Aggression and Violent Behavior

The effect of academic achievement on aggression and violent behavior: A meta-analysis

https://doi.org/10.1016/j.avb.2017.08.002Get rights and content

Highlights

  • Meta-analysis of studies examining the association between academic achievement and violent behavior

  • Low academic achievement is consistently associated with violent offending

  • Low academic achievement is associated with violent offending, controlling for nonviolent antisocial behavior

  • Findings support the differential etiology of violence hypothesis

Introduction

In this paper we present a meta-analysis of studies on the association between academic achievement and physically aggressive or violent behavior. The paper builds on the literature by testing the differential etiology of violence hypothesis proposed by Savage and Wozniak (2016), whereby academic achievement is among their “good prospects” for predicting violent, as opposed to nonviolent, criminal behavior. The current paper also provides estimates of effect sizes. An interest in the association between academic achievement and antisocial behavior has been seen in the published literature for decades (e.g., Jensen, 1976). Academic achievement has been inversely associated with conduct problems in children (e.g., Murray & Farrington, 2010), delinquency in adolescents (e.g., Lipsey and Derzon, 1998, Murray and Farrington, 2010), and even criminal activity in adults (e.g., Le Blanc, 1994). Some findings have implied that academic achievement may be related to violent antisocial behavior in particular (e.g., Katsiyannis & Archwamety, 1997). Rebellon and Van Gundy (2005) report that educational success was negatively associated with violent offending but not property offending in the National Youth Survey (NYS) data set. Lewis and colleagues found that incarcerated boys rated as “more violent” (having committed serious violent crimes) had lower scores for almost all tests, with significantly lower arithmetic scores (Lewis, Shanok, Pincus, & Glaser, 1979).

The differential etiology thesis suggests that the etiology of violent behavior is likely to be distinguishable from the etiology of nonviolent but antisocial behavior, an idea that is largely ignored in modern criminological theory. Savage and Wozniak (2016) include in their review of studies of violence those that measure violent criminal offenses, but also those that employ indicators of interpersonal violence where physical harm and intent to harm are present (such as studies of early physical aggression or self-report physical fighting). Academic achievement is equivalent to academic performance, and is usually operationalized with grades in U.S. studies, but might also be reflected in other indicators of academic success such as ratings of school performance (as was done in a study of Icelandic children, Bernburg & Thorlindsson, 1999), direct tests of reading or arithmetic, or given a “low secondary allocation” (as was used in a British study, Farrington, 1989).

Savage and Wozniak (2016) emphasize two main reasons that negative correlations between academic achievement and violence are theoretically sound, which also point to a stronger association between academic achievement and violence than between academic achievement and nonviolent offending (differential etiology). First, there is a close association between intelligence and academic achievement (Steinmayr, Ziegler, & Träuble, 2010; p. 14) and intelligence deficits have been associated with violent behavior in adolescents and adults in many studies (e.g., Ayduk et al., 2007, Barker et al., 2007, Cohen et al., 2003, Giancola, 2000). To the extent that low academic achievement is indicative of low intelligence, or specific cognitive deficits for some children, it may be associated with violence. Importantly, cognitive deficits and poor executive functioning in children have been associated with physically aggressive and antisocial behavior in many studies (see Savage & Wozniak's, 2016 chapter on intelligence and executive functioning). In several studies of offenders, violent offenders had significantly lower IQ scores than nonviolent ones, with recidivistic violent offenders having the lowest IQ of all (e.g., Holland et al., 1981, ⁎Kennedy, 2006, ⁎Lewis et al., 1979, ⁎Loeber et al., 2005).

Some of the reasons that intelligence is thought to be negatively associated with violence include the following. Those with low intelligence may also have weaker skills in complex problem solving (e.g., Stadler, Becker, Godker, Leutner, & Greiff, 2015), making it difficult to choose nonviolent options in a complex encounter. Verbal impairments (intelligence scores are often partly comprised of verbal tests) have been cited as among the “most well-established neurocognitive impairments associated with conduct behavior problems” (Barker et al., 2007, p. 593). In some studies, those with low intelligence have been less able to understand the inner workings of other persons (theory of mind) (e.g., Ibanez et al., 2013, Qualter et al., 2011) which is associated with empathy (Qualter et al., 2011). Some authors have reported significant correlations between intelligence and cognitive empathy (e.g., Schwenck et al., 2014). None of these in-and-of-themselves is inherent in academic underachievement, but low academic achievement may be indicative of hidden intellectual and cognitive deficits of this nature making intelligence an important potential confound in the association between academic achievement and violence.

Further, though the association between violence and intelligence is consistent across many studies, a review of the body of literature calls into question the association between intelligence and nonviolent criminality. Barker and colleagues have reported that a series of executive functions and verbal ability are negatively associated with physical aggression trajectories in their sample, but not with theft trajectories. In fact, controlling for violent offending, they report that measures of executive function and verbal intelligence were positively associated with frequent theft (Barker et al., 2007). Walsh (1987) found a negative association between IQ and violence, but a positive association between property crime and IQ. Bernat, Hall, Steffen, and Patrick (2007) report a negative correlation between WAIS-R scores and violence but not between WAIS-R scores and nonviolent offenses.

There is more to the theoretical foundation for the hypothesis that academic achievement is associated with violent behavior. School problems of many types are likely to engender frustration and negative emotionality (strain) and negative emotions clearly play a role in much aggressive and violent behavior. Although criminologists have largely ignored its role, the role of emotion has been featured in psychological theory and empirical research on aggression for a long time (Loeber & Hay, 1997). Seminal work on frustration and aggression (Dollard, Doob, Miller, Mowrer, & Sears, 1939) has been credited with initiating modern empirical work in this area (Baumeister & Bushman, 2007). Many authors in recent decades continue to emphasize the role of emotions in the etiology of aggression (e.g., Baumeister and Bushman, 2007, Beck, 1999, Bernard, 1990, Huesmann and Eron, 1992). When psychologists discuss causes of aggression, foremost on their lists of factors are items such as “unpleasant events,” (e.g., Bushman & Huesmann, 2010), frustration (e.g., Dollard et al., 1939), and anger (e.g., Scheff & Retzinger, 1991). Savage and Wozniak (2016) posit that negative emotionality is likely to have a special relationship with physically aggressive externalizing behaviors in young children and violence in adolescents because it is associated with “lashing out” (e.g., Dutton, 2011, Dutton et al., 1996, Eisenberg et al., 2001). While negative emotionality is associated with numerous forms of offending (e.g., Ferguson, 2011), some forms of negativity are likely to have a special relationship with physically aggressive behavior in children and violent behavior in adolescents and adults (e.g., Dutton, 2011, Dutton et al., 1996, Eisenberg et al., 2001).

The two emotions featured in studies of and theories about aggression are anger and shame (Baumeister & Bushman, 2007). In work by Kaplan (e.g., Kaplan et al., 1982), deviance springs from negative feelings that arise from self-derogation. According to Kaplan, self-derogation commonly occurs in the course of “normative participation” in various activities. While Kaplan did not emphasize school, it is clear that school experiences would fit neatly into his framework for deviance, and in one paper he reports that being “afraid of getting a bad report card” is associated with self-derogation (Kaplan & Pokorny, 1970). Doing poorly in school is highly likely to cause shame, frustration, and anger in many students.

In addition, low academic achievement may be caused by or correlated with other school problems, which are also likely to engender negative emotionality. Some authors have proposed that bonding to school (also referred to as school climate) might be protective against individual experiences of violence and community (e.g., Brookmeyer, Fanti, & Henrich, 2006). Other school problems include suspension, expulsion and, ultimately, low school attainment. Magnuson, Duncan, and Kalil (2006) point out that a sense of school connectedness and relationships with teachers play a particularly important role in the emotional and academic adjustment of middle school children. Thornberry draws our attention to the reciprocal nature of the relationship between academics and delinquency. Adolescents who become involved in delinquency “tend to have lower subsequent grades, develop weaker school bonds, and are less likely to graduate from high school or attend college” (Hoffmann, Erickson, & Spence, 2013, p. 631). Savage and Wozniak (2016) included school attachment in their review and concluded that measures of school attachment were consistently, negatively associated with violent behavior, and these relationships tended to withstand controls for academic achievement, in the few studies where such controls were applied.

Finally, Savage and Wozniak (2016) submit that the intensity of the experience, the grind of all-day every day attendance requirements “enhances the potential for dramatic positive impacts for those who benefit, and dramatic adverse impacts for those whose school experience is unhappy” (p. 41). To them, the “cumulative continuity” of negative interactions at school may snowball into larger problems (see also Payne & Welch, 2013). Their larger point is that the quantity and intense quality of the school experience magnifies whatever effect they exact on the developing child.

Emerging evidence suggests that, like intelligence, academic achievement may also be differentially related to violent over nonviolent offending. As an example, Hart, O'Toole, Price-Sharps, and Shaffer (2007) found that GPAs were significantly lower in violent compared to nonviolent delinquent adolescents. Loeber et al. (2005) reported that low academic achievement was significantly more common among violent compared to nonviolent offenders. There is no consensus on this point, however; other studies have reported significant negative associations between property delinquency and grades (e.g., ⁎Bernburg and Thorlindsson, 1999, ⁎Owens-Sabir, 2007, ⁎Rebellon and Van Gundy, 2005) and the difference may depend on the measure of academic achievement (e.g., reading vs. math; Marcus & Gray, 1998). Savage and Wozniak (2016) concluded from their comprehensive review of the evidence that academic achievement is a “good prospect” as a differential predictor of violence.

Thus, we examine what is known about empirical associations between academic achievement and violent behavior. Our summary will address the general research question about the consistency and size of that association. We will also examine the evidence to formally test whether the extant literature suggests that academic achievement is differentially associated with violent, compared to nonviolent antisocial behavior.

Section snippets

Selection of studies

The studies included in this review were derived from those acquired for a larger project on academics, intelligence, and executive functioning. An attempt was made to acquire all published studies relevant for understanding the effect of academic achievement on aggression and violent behavior. To that end, we conducted extensive searches of Criminal Justice Abstracts and PsycINFO, supplemented by a search in ERIC. We combined search terms related to potential independent variables with a list

Overview

Forty-three studies met the criteria outlined in our introduction and also reported adequate information to code a summary measure of significance (See Appendix A). We were able to convert coefficients to a common effect size for 36 studies. In some cases, the authors did not provide coefficients or adequate information for us to estimate them; in other cases the authors provided coefficients for which we do not have a conversion formula.

Overall effect of academic achievement on aggression and violence and the differential etiology of violence thesis

The weighted overall estimate of the association between

Discussion

In the present paper we have summarized findings from a body of literature on the effect of academic achievement on physical aggression and violence. While some studies report nonsignificant findings, our estimates are very consistent; in all of the weighted estimates, the negative association between indicators of violence and academic achievement was statistically significant. This was true for males and females, for children adolescents and adults, for simple and multivariate analyses. This

First page preview

First page preview
Click to open first page preview

References (104)

  • E.D. Barker et al.

    Developmental trajectories of male physical violence and theft

    Archives of General Psychiatry

    (2007)
  • R.F. Baumeister et al.

    Angry emotions and aggressive behaviors

  • A.T. Beck

    Prisoners of hate: The cognitive basis of anger, hostility and violence

    (1999)
  • P.E. *Bellair et al.

    Beyond the bell curve: Community disadvantage and the explanation of black-white differences in adolescent violence

    Criminology

    (2005)
  • P.E. *Bellair et al.

    Linking local labor market opportunity to violent adolescent delinquency

    Journal of Research in Crime and Delinquency

    (2003)
  • T.J. Bernard

    Angry aggression among the ‘truly disadvantaged’

    Criminology

    (1990)
  • J.G. *Bernburg et al.

    Adolescent violence, social control, and the subculture of delinquency: Factors related to violent behavior and nonviolent delinquency

    Youth and Society

    (1999)
  • K.A. Brookmeyer et al.

    Schools, parents, and youth violence: A multilevel, ecological analysis

    Journal of Clinical Child and Adolescnt Psychology

    (2006)
  • E.B. *Brownlie et al.

    Early language impairment and young adult delinquent and aggressive behavior

    Journal of Abnormal Child Psychology

    (2004)
  • E.T. *Bryant et al.

    Neuropsychological deficits, learning disability, and violent behavior

    Journal of Consulting and Clinical Psychology

    (1984)
  • B.J. Bushman et al.

    Aggression

  • S.B. *Campbell et al.

    Predictors and sequelae of trajectories of physical aggression in school-age boys and girls

    Development and Psychopathology

    (2010)
  • Y. *Choi

    Academic achievement and problem behaviors among Asian Pacific Islander American adolescents

    Journal of Youth and Adolescence

    (2007)
  • J. Cohen

    Statistical power analysis for the behavioral sciences

    (1988)
  • R.A. Cohen et al.

    Impulsivity and verbal deficits associated with domestic violence

    Journal of the International Neuropsychological Society

    (2003)
  • R.A. *Cohen et al.

    Neuropsychological correlates of domestic violence

    Violence and Victims

    (1999)
  • H. Cooper et al.

    The handbook of research synthesis

    (1994)
  • J. DeFronzo et al.

    Welfare assistance levels and homicide rates

    Homicide Studies

    (1998)
  • J. Dollard et al.

    Frustration and aggression

    (1939)
  • D.G. Dutton

    Attachment and violence: An anger born of fear

  • D.G. Dutton et al.

    Antecedents of abusive personality and abusive behavior in wife assaulters

    Journal of Family Violence

    (1996)
  • N. Eisenberg et al.

    The relations of regulation and emotionality to children's externalizing and internalizing problem behavior

    Child Development

    (2001)
  • P. *Ellickson et al.

    Early predictors of adolescent violence

    American Journal of Public Health

    (2000)
  • P. Fajnzylber et al.

    Inequality and violent crime

    Journal of Law and Economics

    (2002)
  • D.P. *Farrington

    Early predictors of adolescent aggression and adult violence

    Violence and Victims

    (1989)
  • D.P. *Farrington et al.

    Young men who kill: A prospective longitudinal examination from childhood

    Homicide Studies

    (2012)
  • C.J. Ferguson

    Video games and youth violence: A prospective analysis in adolescents

    Journal of Youth and Adolescence

    (2011)
  • C.J. Ferguson et al.

    Publication bias in psychological science: Prevalence, methods for identifying and controlling and implications for the use of meta-analyses

    Psychological Methods

    (2012)
  • S. *Feshbach et al.

    Cognitive competencies and aggressive behavior: A developmental study

    Aggressive Behavior

    (1984)
  • P.J. *Fite et al.

    Explaining discrepancies in arrest rates between black and white male juveniles

    Journal of Consulting and Clinical Psychology

    (2009)
  • P.R. Giancola

    Neuropsychological functioning and antisocial behavior–Implications for etiology and prevention

  • D.M. Gorman

    The bad practice of best practices

    (2004, February)
  • G.C.N. Hall

    Sexual offender recidivism revisited: A meta-analysis of recent treatment studies

    Journal of Consulting and Clinical Psychology

    (1995)
  • L. Hannon et al.

    The truly disadvantaged, public assistance, and crime

    Social Problems

    (1998)
  • L. Hannon et al.

    Moral commitment and the effects of social influences on violent delinquency

    Violence and Victims

    (2001)
  • J.L. *Hart et al.

    The risk and protective factors of violent juvenile offending

    Youth Violence & Juvenile Justice

    (2007)
  • L.V. Hedges et al.

    Statistical methods for meta-analysis

    (1985)
  • T.I. *Herrenkohl et al.

    Early adolescent predictors of youth violence as mediators of childhood risks

    Journal of Early Adolescence

    (2001)
  • T.I. *Herrenkohl et al.

    Protective factors against serious violent behavior in adolescence: A prospective study of aggressive children

    Social Work Research

    (2003)
  • J.P. Hoffmann et al.

    Modeling the association between academic achievement and delinquency: An application of interactional theory

    Criminology

    (2013)
  • Cited by (24)

    • Two perspectives on gaming social capital among Korean adolescents’ gamers

      2022, Entertainment Computing
      Citation Excerpt :

      These studies constructed research models in which the short-term effect of game playing was measured using highly violent or addictive games. Moreover, Ferguson [43] and Savage et al. [32] suggested a publication bias in existing gaming research since studies that fail to reject the null hypothesis regarding the negative effect of gaming were less likely to be published in academic journals. This calls for the long-term effects of game playing.

    • Suspended in context: School discipline, STEM course taking, and school racial/ethnic composition

      2021, Social Science Research
      Citation Excerpt :

      The practice of suspension removes students from classroom instruction and research has highlighted that students who experience school discipline have lower academic grades and achievement (Arcia 2006; Morris and Perry 2017), and take lower levels of coursework (Watts and Erevelles 2004). Similarly, students who demonstrate academic achievement are less likely to be seen as having behavioral problems (Savage et al. 2017; Schneider et al., 1998), evaluated based on their behavior (Oakes 2005), or commit delinquent offenses (Maguin and Loeber 1996; Yun et al. 2014). Other researchers, however, have shown that the negative link between school discipline and academic achievement may actually be driven by unconsidered academic achievement measures earlier in the life course (Anderson et al. 2019; Kinsler 2013).

    • Life course persistent antisocial behavior silver anniversary

      2020, Aggression and Violent Behavior
      Citation Excerpt :

      Since Moffitt's initial formulation, abundant research has validated Moffitt's hypothesis that subtle cognitive deficits in the form of low IQ, attention problems, and learning disabilities, all of which have substantial genetic influence (Fletcher, Lyon, Fuchs, & Barnes, 2018; Hinshaw, 2018; Knopik et al., 2017), adversely affects educational achievement (de Zeeuw, van Beijsterveld, Ehli, de Geus, & Boomsma, 2017; DuPaul & Langberg, 2015; Farrington, 2015; Fletcher et al., 2018), and thereby increases risk for antisocial behavior. Educational failure has been linked to increased risk for criminal offending through three mechanisms (Savage, Ferguson, & Flores, 2017; Wertz et al., 2018). First, school problems are likely to engender frustration and negative emotions which play a role in much aggressive and violent behavior.

    • Impulsivity and aggression: A meta-analysis using the UPPS model of impulsivity

      2019, Aggression and Violent Behavior
      Citation Excerpt :

      In comparison to other meta-analytic studies on correlates of aggression, the results for negative urgency, positive urgency, and lack of premeditation are larger. The effects for these three facets are larger than the effects for neighborhood disadvantage, academic achievement, 2d:4d digit ratio, and trait empathy (Chang, Wang, & Tsai, 2016; Savage, Ferguson, & Flores, 2017; Turanovic, Pratt, & Piquero, 2017; Vachon, Lynam, & Johnson, 2014) and are in line with personality traits like agreeableness and conscientiousness (Jones, Miller, & Lynam, 2011), suggesting that although the effects may be small-to-medium in size in comparison to effects in psychology, they large for individual differences in aggression research. Even though the effect sizes may be larger than other factors for aggression, there is still a large amount of variance unexplained by trait impulsivity.

    View all citing articles on Scopus

    Studies denoted with an asterisk were used in the meta-analysis.

    View full text