Assessing the Role of Parental Tolerance for Child Misbehavior in Predicting Child Physical Abuse Ji

1
Assessing the Role of Parental Tolerance for
Child Misbehavior in Predicting Child Physical
AbuseJill A. Star, Lori J. Klinger, Meena
Lambha, Ember Lee, Elizabeth V. Brestan -
Auburn University Linda Anne Valle, Ph.D.
Northern Illinois University (now at the Centers
for Disease Control and Prevention)
Results
Introduction

• Hypothesis 1 was partially supported in that the
  CRI and the Rigidity scale were significantly
  correlated in the nonabusive, nonmatched group (r
  .32, p lt .05), but
• not for the abuse sample or the matched non
  abuse sample.
• 2) Hypothesis 2 was supported in that the CRI
  significantly predicted group membership in both
  the nonmatched (? .52, p .000) and matched
  groups (? .74, p .000). Additionally, the CRI
  demonstrated greater predictive strength than the
  CAP Abuse total score or the CAP Rigidity score
  for both the nonmatched and matched groups (see
  Figure 2).
• 3)Hypothesis 3 was not supported in that the
  abuse sample had significantly lower scores on
  the CRI than the comparison sample in both the
  nonmatched and matched
• groups (nonmatched t -11.144, p lt .001
  matched t -6.082, p lt .001).

According to the National Clearinghouse on
Child Abuse and Neglect (2003), approximately
903,000 children were the victims of some form of
child maltreatment in 2001. Nineteen percent of
the 903,000 children suffered physical abuse
(NCCANCH, 2003). It is estimated that the direct
and indirect annual costs of child maltreatment
in the United States are approximately 94
billion (Fromm, 2001). Sedlak and Broadhurst
(1996) defined child physical abuse as occurring
when a child, under the age of 18, has been
injured or is at risk for injury by being hit
with a hand or other object, being kicked,
shaken, burned, thrown, stabbed, or choked by a
parent or a parent-surrogate. Although much
research has focused on the characteristics of
parents with a history of physical child abuse
such as being abused themselves, substance abuse,
and psychiatric problems, no study to date has
addressed the role of parental tolerance for
child misbehavior as a predictor of child
physical abuse. Trickett and Susman (1988) found
that abusive parents tend to be less satisfied
with their children than nonabusive parents. One
possible reason for this parental lack of
satisfaction may be due to their lack of
tolerance for their childs misbehavior.
Parental tolerance has been defined as the
level of annoyance a parent has for their childs
misbehavior and the Child Rearing Inventory (CRI)
was recently developed to measure the level of
tolerance parents have for their childs
misbehavior (Brestan, Eyberg, Algina, Johnson,
Boggs, 2003). Based on a sample of 236 mothers,
higher scores on the CRI were linked to parent
report of problematic child behavior on the
Eyberg Child Behavior Inventory (ECBI Eyberg
Pincus, 1999). The primary aims of the present
study were to establish the discriminative and
convergent validity of the CRI. The convergent
validity of the CRI was investigated by examining
its relation to the Rigidity scale of the Child
Abuse Potential Inventory (CAP Milner, 1986).
The discriminative validity of the CRI was
assessed by using CRI scores to predict child
abuse status among a sample of physically abusive
families and comparison families. Hypotheses This
study included 3 hypotheses 1) It was
hypothesized that the CRI and the Rigidity scale
would be significantly correlated, thus proving
evidence for the convergent validity of the CRI.
2) It was hypothesized that the CRI would
predict group membership in the abusive and the
normative groups. 3) Finally, it was
hypothesized that the sample with a history of
physical abuse would have significantly higher
scores than the comparison sample.
Discussion
The results of this study suggest that the CRI
may be used as part of a larger test battery to
assess child physical abuse risk. As
hypothesized, the CRI predicted group membership
in the abusive sample and normative sample, and
the scores on the CRI were significantly
different between the two groups. However, the
direction of the difference was the opposite of
what was expected in that the abusive sample had
significantly lower scores on the CRI, indicating
a higher tolerance for their childs misbehavior.
One possible explanation is that the abusive
sample may initially tolerate child misbehavior
until the intensity of problematic behaviors
increases. This increased intensity or frequency
of negative child behavior may then lead to a
change in the parents perception of the child, a
brief period of decreased tolerance, and abusive
parenting. Among a sample of non-clinic referred
mothers, the CRI was significantly correlated
with their report of their childs behavior
problems (ECBI Intensity Scale Brestan et al.,
2003). Another possibility is that the abusive
sample answered the CRI items in a more socially
acceptable manner to decrease negative
consequences for their family. The CRI only
contains 11 items and may have high face
validity, thus allowing parents the opportunity
to fake responses. Future research should
focus on the process through which parents with a
history of physical abuse report less tolerance
for child behavior relative to the normative
sample. In this study the CRI did predict group
membership, but it is unclear how the abuse and
nonabuse samples came to answer the items
differently. A revision of the CRI could include
more items and a lie scale to decrease the face
validity of the measure.
Method
Participants The participants in this study
consisted of 161 parents with children between
the ages of 4 and 12 years. Fifty-one parents
had no history of child physical abuse and were
recruited in Auburn, AL, and DeKalb, IL and 110
parents were referred to the University of
Oklahoma Health Sciences Center following
substantiated child physical abuse (see Figure
1). Two groups were used for analyses. One
group consisted of the full database (the
nonmatched group) and the other group consisted
of 40 parents (20 from each group) who were
matched on demographic characteristics such as
SES, IQ, age of child, etc. (the matched
group). Measures Three measures, in addition to
a demographic questionnaire, were used in this
study the CRI Total Score, the CAP Abuse
subscale, and the CAP Rigidity subscale. Child
Rearing Inventory (CRI). The CRI (Brestan,
Eyberg, Algina, Johnson, Boggs, 2003) consists
of 11 items that measure parental tolerance for
child misbehavior. Each item is scored on a 1-4
Likert-type scale, with higher scores indicating
lower levels of parental tolerance for child
misbehavior. A previous study provided
preliminary construct validity for the CRI by
using the Eyberg Child Behavior Inventory (ECBI)
Problem Scale and showed good internal
consistency and test-retest reliability for the
CRI (Brestan et al., 2003). Child Abuse
Potential Inventory (CAP). The CAP (Milner,
1986) consists of 160 items presented in an
agree/disagree format. This measure is
hypothesized to estimate the respondents
potential risk for committing child physical
abuse. The CAP is a standardized measure of
child abuse potential that indicates whether the
parent possesses an array of personal and
interpersonal characteristics that are similar to
the characteristics of known physical child
abusers. The CAP has shown high internal
consistency, test-retest stability, and future
predictive validity (Milner, 1986). CAP Abuse
Subscale. The CAP Abuse subscale consists of 77
items that yield a score predictive of child
physical abuse. Higher scores indicate a
greater likelihood of the occurrence of
child physical abuse. CAP Rigidity Subscale.
The CAP Rigidity subscale consists of 14 items.
It represents the rigidity of the parents
attitudes toward his or her childs
appearance and behavior. Higher
scores on this subscale represent a more rigid
parental attitude.
Figure 2
Figure 1