Findings: Parent Education

Parent education is another good prospect for an educational predictor of violence. It was negatively associated with violence in 12 out of 14 studies, and 6 of these reported a PoC that was statistically significant (see Tables 5.3 and 5.4).

In complex, multivariate models, the association between parent education and violence holds. Bellair and McNulty (2005) found that those reporting serious adolescent violence were less likely to have parents who were college graduates. In another of their analyses of Add Health data, McNulty and Bellair (2003) report that adolescent violence is consistently lower among those whose parents were college graduates. Low maternal education has been associated with being on a high and rising trajectory of physical aggression (Huijbregts, Seguin, Zoccolillo, Boivin, & Tremblay, 2008). In a set of structural models presented by Gottfredson et al. (1991), the total effect of parent education was negatively associated with violence for both boys and girls.

The evidence on maternal education and physical aggression has been consistent in younger samples (e.g., Koda, 1999; Nagin & Tremblay, 2001). In the National Institute of Child Health and Human Development (NICHD) sample (2004), looking at development from birth to 3rd grade, maternal education was much lower among the high physical aggression trajectory group compared to the other groups. The odds ratio (OR) comparing high to low groups reflected a large difference (OR = .37).

Some findings on the association between parent education and violence have been weak, inconsistent or unsupportive of an association between parent education and violence (e.g., Bernburg & Thorlindsson, 1999; Martino et al., 2008; Piquero, 2000; Ribeaud & Eisner, 2010).

Unfortunately, few studies have provided evidence to help us understand whether or not parent education has a special relationship with violence beyond

Table 5.3 Summary of Study-Level Findings Related to Other School- Related Factors and Violent and Nonviolent Offending

Summary of STUDY Results

Independent Variable Category

Number of Studies (k)

@

о

©

X

Parent Education

Violent Nonviolent Violent vs. Nonviolent

  • 14
  • 3
  • 2
  • 2
  • 1
  • 0
  • 0
  • 1
  • 1
  • 0
  • 0
  • 0
  • 6
  • 1
  • 0
  • 6
  • 0
  • 1

Learning

Violent Nonviolent Violent vs. Nonviolent

  • 7
  • 1
  • 4
  • 1
  • 0
  • 0
  • 1
  • 0
  • 0
  • 0
  • 0
  • 0
  • 3
  • 1
  • 2
  • 2
  • 0
  • 2

School Problems

Violent Nonviolent Violent vs. Nonviolent

  • 6
  • 0
  • 3
  • 0
  • 0
  • 1
  • 0
  • 0
  • 0
  • 0
  • 0
  • 0
  • 0
  • 0
  • 1
  • 6
  • 0
  • 1

Academic Attainment

Violent Nonviolent Violent vs. Nonviolent

  • 13
  • 5
  • 10
  • 0
  • 0
  • 1
  • 2
  • 3
  • 3
  • 0
  • 0
  • 0
  • 5
  • 0
  • 3
  • 6
  • 2
  • 3

Щ Findings are ambiguous

О Findings are in the opposite direction of the attachment hypothesis (not necessarily statistically significant)

© Relationship is reported as “null” or coefficient = 0

Findings are in the expected direction of the attachment hypothesis but are not statistically significant

X Findings are in the expected direction of the attachment hypothesis and are statistically significant

its association with antisocial behavior. While a strong preponderance of studies indicates a parent education-violence connection, the 3 studies reporting the association between nonviolent offending and parent education do not demonstrate that an association exists. In the 3 studies where authors provide coefficients for both violent and nonviolent offending, 2 report a stronger pattern of findings for violent than nonviolent offending (Gottfredson et al., 1991; Maschi, 2006). The other study reports ambiguous findings for both (Bernburg & Thorlindsson, 1999). The number of studies comparing violent to nonviolent offenders is also meager. Data from the Pittsburgh youth study strongly support a differential etiology hypothesis; violent offenders were significantly more likely to have a “poorly educated mother” than nonviolent offenders (Loeber et al., 2005). The authors did

Table 5.4 Summary of Comparison-Level Findings Related to Other School- Related Factors and Violent and Nonviolent Offending

Summary of COMPARISONS

Independent Variable Category

Number of Comparisons

О

©

X

Parent Education

Violent Nonviolent Violent vs. Nonviolent

  • 50
  • 11
  • 3
  • 4
  • 6
  • 2
  • 3
  • 0
  • 0
  • 13
  • 4
  • 0
  • 30
  • 1
  • 1

Learning

Violent Nonviolent Violent vs. Nonviolent

  • 17
  • 2
  • 5
  • 2
  • 0
  • 0
  • 1
  • 0
  • 0
  • 8
  • 2
  • 2
  • 6
  • 0
  • 3

School Problems

Violent Nonviolent Violent vs. Nonviolent

  • 13
  • 0
  • 8
  • 0
  • 0
  • 1
  • 0
  • 0
  • 0
  • 0
  • 0
  • 2
  • 13
  • 0
  • 5

Academic Attainment

Violent Nonviolent Violent vs. Nonviolent

  • 35
  • 17
  • 14
  • 2
  • 12
  • 5
  • 0
  • 0
  • 0
  • 13
  • 1
  • 5
  • 20
  • 4
  • 4

О Findings are in the opposite direction of the attachment hypothesis (not necessarily statistically significant)

© Relationship is reported as “null” or coefficient = 0

Findings are in the expected direction of the attachment hypothesis but are not statistically significant

X Findings are in the expected direction of the attachment hypothesis and are statistically significant

not test the association in a multivariate model. Piquero (2000) reports that years of maternal education were nearly identical among frequent nonviolent offenders and frequent violent offenders in the Philadelphia portion of the Collaborative Perinatal Project.

Another problem with drawing conclusions here is that these studies as a body were not designed to test the effects of parent education. In our own analysis of Add Health data, we see that simple correlations between measures of parent education and violence in wave 1 and 2, dummy indicators of violent vs. nonviolent offenders and even between violent compared to frequent nonviolent-only offenders are not large but are statistically significant at a very low level of probability (because of the large sample size).

Parent education (wave 1)

and frequency of violence, wave 1 r = - .100** and frequency of violence, wave 2 r = - .073** and violent vs. nonviolent dummy code wave 1 r = - .112** and violent vs nonviolent dummy code, wave 2 r = - .119** and violent vs. frequent nonviolent-only, wave 1 r = - .056*

  • * p < .05
  • ** p < .01

In wave 1, the parents of adolescents who reported violent behavior had an average "parent education" score of 5.54 compared to 6.17 among those who report nonviolent offending only. This difference is also evident in comparison to chronic nonviolent-only offenders (mean = 6.00). If we run a multivariate model, regressing frequency of violent offending on parent education, controlling for age, sex, disadvantaged minority status, family income, intact family, peer alcohol and drug use, community disorder, drug and alcohol use of the participant, and frequency of nonviolent offending in wave 2, we find that parent education is still significantly, negatively associated with violence in wave 2 (t = - 3.10, p < .001). If we run a logistic regression, predicting whether the subject is violent or nonviolent-only, using the same control variables plus a control for frequency of nonviolent offending, we find that parent education is significantly lower among violent than nonviolent- only offenders, with a very large Wald statistic (Wald = 10.125, p = .001). While these effects are not large in magnitude, they are compelling as a theory test.

We conclude that low parent education remains a good prospect in the differential etiology of violence though this conclusion does not yet have strong.

 
Source
< Prev   CONTENTS   Source   Next >