FINDINGS

Mixed Types of Abuse

In many studies the authors combine abuse types. In most cases, based on prevalence rates reported in other studies, we can assume that neglect and physical abuse are most common, and sexual abuse is much less common in these sets of combined cases. Findings from our limited set of studies employing combined measures of abuse suggest a strong positive association between exposure to abuse and violent behavior as other reviewers have reported. The findings also strongly suggest that maltreated individuals are more likely to commit nonviolent offenses as well.

There is also limited evidence that the levels of victimization may distinguish between property crime and violent crime offenders (e.g., Cuevas, Finkelhor, Turner, & Ormrod, 2007). In the 6 studies where violent and nonviolent offenders were compared, violent offenders were more likely to have been abused in 5, and the difference was statistically significant in a preponderance of comparisons (PoC3) in 3 of those studies (see Tables 8.1 and 8.2). A close consideration of

Summary of STUDY Results

Independent

Variable

Number of Studies (k)

rn

о

©

X

Mixed Types of Abuse

Violent

Offender Violence Violent vs. Nonviolent Offenders

Nonviolent Offending Offender Nonviolent

  • 9
  • 6
  • 6
  • 8
  • 3
  • 0
  • 0
  • 0

1

0

  • 0
  • 0
  • 0
  • 0
  • 0
  • 0
  • 1
  • 1
  • 0
  • 1
  • 2
  • 0
  • 2

3

1

  • 7
  • 5
  • 3
  • 4 1

Physical Abuse

Violent

Offender Violence Violent vs. Nonviolent Offenders

Nonviolent Offending Offender Nonviolent

  • 15
  • 7
  • 11
  • 12
  • 4
  • 1
  • 0
  • 0
  • 0
  • 0

0

  • 1
  • 0
  • 1

0

  • 0
  • 1
  • 2
  • 0
  • 1
  • 5
  • 3
  • 4
  • 5 3
  • 9
  • 2
  • 5
  • 6 0

Neglect

Violent

Offender Violence Violent vs. Nonviolent Offenders

Nonviolent Offending Offender Nonviolent

  • 4
  • 2
  • 4
  • 4
  • 2
  • 0
  • 0
  • 0
  • 0
  • 0

1

  • 1
  • 0
  • 1

1

  • 0
  • 1
  • 2
  • 0
  • 1
  • 2
  • 0
  • 2

2

0

  • 1
  • 0
  • 0

1

0

Sexual Abuse

Violent

Offender Violence Violent vs. Nonviolent Offenders

Nonviolent Offending Offender Nonviolent

  • 9
  • 5
  • 6
  • 8
  • 3
  • 0
  • 0
  • 0
  • 0
  • 0

0

  • 1
  • 0
  • 1

1

  • 1
  • 1
  • 1

1

1

  • 2
  • 3
  • 4
  • 2
  • 1
  • 6
  • 0
  • 1

4

0

Trauma

Violent

Offender Violent Violent vs. Nonviolent Offender

Nonviolent Offending Offender Nonviolent

  • 3
  • 5
  • 2
  • 2
  • 2

0

2

  • 0
  • 0
  • 0
  • 0
  • 0
  • 0
  • 0
  • 1
  • 0
  • 0
  • 0
  • 0
  • 0

0

  • 2
  • 0
  • 1

0

3

1

  • 2
  • 1
  • 1

Щ Findings for the study overall are ambiguous

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

© Relationship is reported as 0 or simply “not significant” without associated direction

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

Summary of COMPARISON Results

Independent

Variable

Number of Comparisons

О

©

X

Mixed Types of Abuse

Violent

Offender Violence Violent vs. Nonviolent Offenders Nonviolent Offending

Offender Nonviolent

  • 41
  • 14
  • 18
  • 50
  • 9

3

2

  • 1
  • 0
  • 0
  • 0
  • 3 9
  • 1
  • 4
  • 9
  • 1
  • 2
  • 20
  • 1
  • 27
  • 8
  • 6
  • 29
  • 4

Physical Abuse

Violent

Offender Violence Violent vs. Nonviolent Offenders

Nonviolent Offending Offender Nonviolent

  • 55
  • 21
  • 40
  • 40
  • 18
  • 3
  • 2
  • 2

8

1

  • 0
  • 5
  • 12
  • 0
  • 5
  • 20
  • 6
  • 11
  • 13
  • 10
  • 32
  • 8
  • 15
  • 19
  • 2

Neglect

Violent

Offender Violence Violent vs. Nonviolent Offenders

Nonviolent Offending Offender Nonviolent

  • 11
  • 10
  • 11
  • 19
  • 7
  • 1
  • 2
  • 0
  • 3
  • 3

0

5

  • 6
  • 0
  • 3
  • 6
  • 1
  • 3
  • 8
  • 0
  • 4
  • 2
  • 2

8

1

Sexual Abuse

Violent

Offender Violence Violent vs. Nonviolent Offenders

Nonviolent Offending Offender Nonviolent

  • 25
  • 18
  • 16
  • 33
  • 15
  • 3
  • 3
  • 4
  • 8
  • 6
  • 3
  • 5
  • 6
  • 4 4
  • 8
  • 10
  • 4
  • 7
  • 4
  • 11
  • 0
  • 3
  • 14
  • 1

Trauma

Violent

Offender Violent Violent vs. Nonviolent Offender

Nonviolent Offending Offender Nonviolent

  • 6
  • 22
  • 3
  • 4 3

0

6

  • 0
  • 0
  • 2
  • 0
  • 0
  • 0
  • 0
  • 0

0

  • 8
  • 0
  • 1

0

  • 6
  • 8
  • 3
  • 3
  • 1

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

© Relationship is reported as 0 or simply “not significant” without associated direction

• Findings are in the expected direction of the attachment hypothesis but are not statistically significant studies that attempt to eliminate possible sources of spuriousness suggests that this difference is real. Famularo et al. looked at juveniles presenting to court and found a significant difference between violent and nonviolent juvenile offenders in their experience of abuse, controlling for age, sex, and ethnicity (Famularo, Kinscherff, Fenton, & Bolduc, 1990). Loeber et al. (2005) also found a significant difference; they did not employ any control variables, but they used a sample originally drawn from the community, so we have fewer concerns about the kinds of sampling bias that may crop up when violent and nonviolent offenders from a correctional sample are compared. Pollock, Mullings, and Crouch (2006) found a significant difference in abuse history between newly admitted violent and nonviolent female offenders, and they made a significant effort to ensure that the nonviolent group was truly nonviolent.

Studies in which the PoC was not statistically significant do not comprise a strong statement against our differential etiology thesis. McClintic (2003) found null effects almost across the board leading to concerns about his methodology. Hamalainen and Haapasalo (1996) provide one simple association comparing violent to nonviolent offenders, and the difference between them was not statistically significant. However, their sample size was very small (n = 34), and our estimated difference between groups, based on the authors’ reported data, was close to the tail of the distribution (our estimated z = 1.48, p = .069) and would have been statistically significant in a larger sample. Hamilton, Falshaw, and Browne (2002) find numerous associations in the predicted direction, and the failure of the data to provide a “preponderance” of comparisons is mainly due, in our opinion, to some of the very narrowly defined measures of abuse (such as “intrafamily repeat victimization only”).

Thus, a close consideration of the findings clearly suggests that violent offenders are more likely to have endured some undifferentiated “abuse” than nonviolent offenders.

If we look at studies that report findings for violent and nonviolent offending separately, we found evidence suggesting that measures of mixed forms of abuse are associated with both (Harrell, 2007; Heck & Walsh, 2000; Stouthamer-Loeber, Loeber, & Homish, 2001). Findings reported by Widom (1989a) in one of her early articles suggest that nonviolent but not violent offending was associated with earlier abuse. Her later studies (e.g., Widom & Maxfield, 2001; Widom & White, 1997), which had somewhat more sophisticated analyses and a longer follow-up period, suggest that the association between a mixed measure of abuse and violence was statistically significant. The association was less consistent for violent than nonviolent offending, and consistent for females but not usually significant for males.

There is some limited evidence that a mixed measure of abuse is more consistently associated with violent than nonviolent offending. Findings reported by Mersky, Topitzes, and Reynolds (2012) are informative. While most of the associations between any abuse and property crime were significant, most of the associations between abuse and “nonviolent-only” conviction were in the right direction but not significant. Since many of those convicted of property crime may also have had convictions for violence, it could be that the violent behavior among the property offenders was actually driving the stronger association estimated for them. A similar pattern was reported in an earlier analysis of the same sample (Mersky & Reynolds, 2007) and findings reported by Siegel and Williams (2003) and Steiner and Wooldredge (2009) (in an offender sample) lean toward this interpretation.

We did not do a separate analysis for running away as a form of delinquency. Running away may play an important role among abused children in perpetuating delinquency (and also victimization) (e.g., Kaufman & Widom, 1999; Kim, Tajima, Herrenkohl, & Huang, 2009), and we do not elucidate that possibility here.

 
Source
< Prev   CONTENTS   Source   Next >