Child Abuse & Neglect 32 (2008) 1127–1137

Contents lists available at ScienceDirect

Child Abuse & Neglect

Cognitive and emotional differences between abusive and non-abusive

fathers夽
Karen J. Francis a,∗ , David A. Wolfe b
a
McMaster Children’s Hospital/McMaster University, Chedoke Site, Hamilton Health Sciences Centre, Hamilton, ON, Canada L9C 7N4
b
CAMH/University of Toronto, London, ON, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Objective: Abusive fathers perpetrate a substantial portion of child physical abuse. Despite
Received 30 July 2007 this, little is known about how they differ from non-abusive fathers. This study compared a
Received in revised form 21 May 2008
broad range of cognitive and affective factors between physically abusive and non-abusive
Accepted 23 May 2008
fathers.
Methods: Abusive (n = 24) and non-abusive (n = 25) fathers completed standard measures
Keywords:
assessing their experience and expression of anger, mental health, parenting stress, and
Fathers
Child abuse their empathy and perceptions of children’s socio-emotional signals.
Child maltreatment Results: Abusive fathers differed from comparisons on almost all constructs. They experi-
Abusive parents enced more anger and were more likely to express that anger aggressively. They reported
Abusive fathers more mental health concerns (such as depression, hostility, and paranoid ideation), more
stress in parenting, and significantly less empathy for their children. They were also more
likely to perceive children’s emotional expressions as depicting negative emotions, such as
anger and disgust.
Conclusions: Abusive fathers struggle with a myriad of difficulties that likely contribute to
their problematic parenting. These difficulties are both inter- and intra-personal in nature.
Practice implications: The findings suggest that abusive fathers require comprehensive
assessment that includes mental health screening. Interventions should be selected care-
fully to target abusive fathers’ high levels of negative affect and negative perceptions.
Treatment strategies should address problems related to parenting style (e.g., managing
stress and interpretation of children’s socioemotional signals) as well as their personal
adjustment (e.g., cognitive behavioral strategies for regulating affect and cognitive distor-
tions).
© 2008 Elsevier Ltd. All rights reserved.

Introduction

Fathers perpetrate a substantial proportion of child physical abuse in North America. They are the perpetrators in 67% of
physical abuse cases in Canada based on data from two-parent families (Trocme et al., 2005) and are also over-represented
among child physical abuse perpetrators in the United States (Guterman & Lee, 2005). The majority of fatalities resulting
from physical abuse are perpetrated by fathers or other male caregivers (Department of Health & Human Services, 2006).
Despite this, studies of maltreating parents typically have involved mothers only, and few data exist specific to abusive
fathers. Research that has considered abusive fathers has tended to focus primarily on the intimate partner relationship

夽 This study was supported by a doctoral research award to the first author from the Social Sciences and Humanities Research Council.
∗ Corresponding author.

0145-2134/$ – see front matter © 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chiabu.2008.05.007
1128 K.J. Francis, D.A. Wolfe / Child Abuse & Neglect 32 (2008) 1127–1137

rather than the father–child relationship. For this reason, researchers and clinicians are calling for greater understanding of
abusive fathers and appropriate intervention options for them (Coohey, 2006; Dubowitz, 2006; Scott & Crooks, 2004; Scott,
Francis, Crooks, & Kelly, 2006; Scott, Francis, Crooks, Paddon, & Wolfe, 2006).
Research on abusive mothers and spouse-abusive men may inform our understanding of fathers who abuse their children,
where no data on abusive fathers yet exists. Although theoretical models of child maltreatment have been based primarily
on information derived about child abusive mothers, these models serve as a framework for beginning to understand abusive
fathers. The developmental–ecological–transactional model of child abuse highlights that abuse is a function of individual,
family, community and social factors (National Research Council, 1993). Based on this model, psychological characteristics
identified among samples of abusive mothers include their own experiences of child maltreatment, mental health difficulties,
and high levels of negative emotion, particularly anger and depression, and relationship distress. In addition, cognitive
distortions and information-processing biases have been identified in abusive mothers, such as lack of empathy and negative
attribution bias when interpreting child behavior (Azar, 1997; Bugental, 1993; Milner, 2003). Similar emotional and cognitive
characteristics have also been observed in partner-abusive men (Bancroft & Silverman, 2002; Murphy & Eckhardt, 2005).
A history of child abuse and neglect in the family of origin has been associated with a greater likelihood of perpetrating
child maltreatment in the future (Wolfe, 1999) and has been identified as a risk factor for men’s intimate partner violence
(Delsol & Margolin, 2004). However, research with abusive fathers has not yielded consistent findings. Some studies have
found that, similar to abusive mothers, abusive and at-risk fathers are more likely than non-abusive fathers to report their
own childhood histories of abuse (e.g., Straus & Smith, 1990); Coohey’s (2006) study, in contrast, did not find a difference in
histories of abuse between non-abusive and abusive fathers.
Mental health concerns, particularly a history of substance abuse and violent behavior, have also been identified as a risk
factor in fathers’ and surrogate fathers’ perpetration of child abuse and neglect (De Bellis et al., 2001; Schaeffer, Alexander,
Bethke, & Kretz, 2005). De Bellis et al. (2001) conducted a study of the psychiatric comorbidity of mothers and fathers of
maltreated children. A history of substance abuse or dependence was identified in 85% of fathers or surrogate fathers, and
approximately three quarters had a history of police involvement due to violent behavior. In contrast, mothers presented
with mood and anxiety disorders, with only a third presenting with substance abuse difficulties. The authors’ conclusion was
that male caregiver mental illness and substance abuse was more serious and less amenable to conventional intervention
than female caregiver mental illness. Unfortunately, the study did not report which parent in each family was the identified
perpetrator. Similarly, substance use has been associated with fathers’ increased risk of child maltreatment as determined
by their scores on the Child Abuse Potential Inventory (Ammerman, Kolko, Kirisci, Blackson, & Dawes, 1999; Moss, Mezzich,
Yao, Gavaler, & Martin, 1995).
Parents’ perceived stress associated with their childrearing role has also been identified as a risk factor for the perpetration
of child abuse and a special intervention priority (Azar & Wolfe, 2006). Maltreating families may experience more stress as
a result of a number of factors, including unemployment, financial hardship, and child behavior (Guterman & Lee, 2005).
Further, distress and discord have been identified in the relationships of batterers, whose relationships are characterized by
high levels of conflict and low levels of satisfaction (Schumacher, Feldbau-Kohn, Slep, & Heyman, 2001).
The role of negative affect, particularly anger, has been supported in studies of abusive mothers (e.g., Bugental, 1993;
Mammen, Kolko, & Pilkonis, 2002) and maritally violent men (Eckhardt, Barbour, & Stuart, 1997; Jacobson, Gottman, Waltz,
Rushe, & Babcock, 2000; Norlander & Eckhardt, 2005). In conjunction with negative beliefs and interpretations, high levels
of anger and other negative arousal interfere with rational problem solving and information processing and may lead to
aggressive cognitions and impulses (Slep & O’Leary, 2001). Anger interferes with parents’ ability to respond to their child in
a proactive, responsive and balanced manner (Denham et al., 2000; Hakim-Larson, Dunham, Vellet, Murdaca, & Levenbach,
1999).
In addition to the emotional dimensions noted above, other elements are likely to play a role in men’s violence. Empathy,
particularly perspective-taking and empathic concern, has been identified as an important variable to consider in the study
of violence and aggression (Beven, O’Brien-Malone, & Hall, 2004). Empathy includes both affective and cognitive components
(Davis, 1994). Specifically, the affective component refers to an emotional involvement with the experience of others, while
the cognitive component refers to an intellectual appreciation and recognition of the experience of others. Spouse abusive
male partners and child abusive mothers both show lack of empathy for their victims, which interferes with parents’ ability
to meet the needs of their child and is associated with aggression (Bugental, 1993). Scott and Crooks (2004) have observed
clinically that abusive fathers focus on their own needs rather than those of their children. In perhaps the only study that has
examined fathers’ empathy, Perez-Albeniz and de Paul (2004) found that high-risk fathers reported less perspective-taking
compared with low-risk fathers. Fathers’ risk level was determined by their scores on the Child Abuse Potential Inventory
(CAPI; Milner, 1986), rather than identification by police or child protective services. Their study was conducted with parents
drawn from community schools and the sample size of fathers was very small (9 high-risk fathers and 12 low-risk fathers).
Researchers have also noted that maltreating mothers (Hildyard & Wolfe, 2007; Zahn-Waxler & Wagner, 1993) and woman-
abusing men (Holtzworth-Munroe & Hutchinson, 1993) perceive their victim’s emotions more negatively. However, abusive
fathers’ perceptions of children’s emotional expressions have not been investigated.
In summary, both maltreating mothers and male batterers experience more negative affect, particularly anger, than their
non-abusive counterparts. As well, both groups are more likely to have deficits in empathy and to misinterpret the behaviors
of others. Further, the experience of child maltreatment and mental health difficulties may also differentiate abusive from
non-abusive family members. However, few studies have examined cognitive and affective characteristics that differentiate
 K.J. Francis, D.A. Wolfe / Child Abuse & Neglect 32 (2008) 1127–1137 1129

men who physically abuse their children from those who do not. The present study involved fathers with a documented
record of physically abusing their child. We predicted that abusive fathers would differ from non-abusive fathers in terms
of reporting more mental health concerns, parenting stress, and anger. We further expected abusive fathers to show less
empathy toward their child and to perceive children’s emotional cues more negatively.

Method

Participants

A sample of abusive fathers was obtained based on referrals by Child Protective Services (CPS) to the study over a 28-
month period. Criteria for inclusion in the abuse sample were a record of perpetrating physical abuse of a biological or
step-child under the age of 12 within the past 3 years (thus, all fathers had been identified through CPS investigation as
having physically abused their child), access to this child, fluency in English, and at least a Grade 4 literacy level.
Exclusion criteria were major cognitive or psychological impairments or prior child sexual abuse perpetration. Of 31
abusive fathers referred, 7 were excluded based on these criteria (4 due to language or literacy problems and 1 due to each
of the other 3 criteria). Thus, data from 24 abusive fathers were available for analyses.
Thirty non-abusive (comparison) fathers were recruited from within the same community as the abusive fathers, using
the same inclusion and exclusion criteria noted above with the exception of abuse history. Comparison fathers were screened
by interview and questionnaire to ensure that they did not have a history of child maltreatment or woman abuse or prior
contact with child protective services. Two fathers were excluded due to elevated scores on the Abuse Scale of the CAPI;
three others were excluded due to past involvement with child protection services or history of abuse of the child’s mother.
Further information on recruitment follows below.

Measures

Mental health: The Brief Symptom Inventory (BSI; Derogatis, 1993) is a 53-item self-report questionnaire that assesses
current psychological symptomatology in clinical and non-patient populations. The BSI measures the experience of symptoms
in the past 7 days, and assesses somatization, obsessive-compulsive symptoms, interpersonal sensitivity, depression, anxiety,
hostility, phobic anxiety, paranoid ideation, and psychoticism. It also provides three global indices. Participants are asked
to respond to each item by endorsing the degree to which they were distressed by the symptom described on a scale from
0 (not at all) to 4 (extremely). The BSI has high internal consistency (Cronbach’s alpha: .71–.85), test retest reliability, and
convergent, discriminant, and construct validity (Derogatis, 1993).
Parenting stress: The Parenting Stress Index—Short Form (PSI/SF; Abidin, 1995) consists of 36 items from the full length, 120-
item PSI. The items are in a five-point response format, ranging from “strongly agree” to “strongly disagree.” The PSI measures
parental stress in the parent–child relationship through consideration of child, parent, and child–parent relationship factors.
The three scales (Parent Distress, Parent–Child Dysfunctional Interaction, and Difficult Child) yield a Total Parenting Stress
score. Internal reliability coefficients for the three scales range from .80 to .87 (.91 for the total score); acceptable test-retest
reliability and discriminant and convergent validity of the PSI have been demonstrated (Abidin, 1995).
Anger: The State-Trait Anger Expression Inventory-2 (STAXI-2; Spielberger, 1998) is a 57-item self-report questionnaire
that measures the experience and expression of anger. Each item has a four-point response format. The STAXI-2 assesses
state anger, trait anger, the expression of anger, and the control of anger. For the present study the State Anger, Trait Anger,
Anger Expression Out, and Anger Expression Index were used, as each of these scales relates to the experience and outward
expression of anger. The State Anger scale assesses the intensity of anger at a particular time on a four-point scale (1 = not at
all, 4 = very much so). The Trait Anger scale measures how often angry feelings are experienced over time (1 = almost never,
4 = almost always), with high scores suggesting the frequent experience of anger, frustration, and feelings of being treated
unfairly by others. The Anger Expression Out scale assesses the individuals’ outward expression of anger through verbally
or physically aggressive behavior toward other persons or objects in the environment. The STAXI-2 also includes an overall
Anger Expression Index; high scores on this scale indicate the experience of intense angry feelings and are associated with
difficulties in interpersonal relationships. The STAXI-2 is a conceptually and psychometrically strong measure of anger across
a wide variety of normative groups (Eckhardt, Norlander, & Deffenbacher, 2004).
Empathy and perceptions of child emotions: The Interpersonal Reactivity Index (IRI; Davis, 1983) is a 28-item self-report
questionnaire assessing cognitive and affective dimensions of empathy. The perspective-taking and empathic concern scales
have shown strong validity and reliability and were used in the current study. Some of the items were modified to reflect
a parent’s approach to his child rather than to a friend or others [e.g., I sometimes find it difficult to see things from (my
child’s) point of view]. Each scale contains seven items scored on a five-point Likert scale ranging from “does not describe
me very well” to “describes me very well.” The IRI has demonstrated convergent and discriminant validity (Davis, 1983) and
internal consistency (Atkins & Steitz, 2002), and discriminates high-risk violent offenders from non-offenders (Beven et al.,
2004).
The IFEEL Pictures (Butterfield & Ridgeway, 1993) were used to assess fathers’ interpretations of children’s socio-emotional
signals. These pictures are a series of 30 color photographs portraying head and shoulder views of 1-year-old children’s
emotional expressions. Although some pictures display “prototypical” emotions, most pictures were selected for their blends
1130 K.J. Francis, D.A. Wolfe / Child Abuse & Neglect 32 (2008) 1127–1137

of emotion, which allows for variability of interpretation. Further, the photographs vary by coloring, shadowing, and clarity.
Participants are instructed to “Please tell us, in one word if possible, the strongest and clearest feeling that each baby is
expressing. There is no right or wrong answer. Please respond with what first comes to your mind.” Each of the participants’
30 responses was categorized according to the comprehensive IFEEL Pictures Lexicon into one of the following clusters:
surprise, interest, joy, contentment, passive, sad, cautious–shy, shame–guilt, disgust–dislike, anger, distress, fear, and other.
The “other” category was comprised of unclear responses or responses that did not fit into any of the emotion categories (e.g.,
devious, moody, hungry). The number of responses fitting into each emotion category was computed for each participant,
and group averages were compared. Further, the total number of participants’ atypical responses was computed through
comparison to a frequency distribution table of emotion responses by picture, using a reference sample of mothers provided
in the manual. This measure has shown acceptable reliability and validity with non-maltreating (Applebaum, Butterfield, &
Culp, 1993) and with neglectful mothers (Hildyard & Wolfe, 2007).

Procedure

The study procedure was approved by the Institutional Review Board for Human Subjects at the University of Western
Ontario. Potential participants were screened by the primary investigator and/or social worker to ensure that they met all
criteria. Men were either invited to participate by the researcher at a community agency for abusive men, or by their CPS
social worker. All men recruited from the community agency for violent men were also involved with CPS. With permission
of the community agency, an investigator presented the research project at a regularly offered initial orientation session and
interested men were recruited at that point. If recruited by a worker, CPS social workers and community agency personnel
made the initial contact with potential participants who met the inclusion criteria. The comparison group was recruited
by community agency personnel or through newspapers. Men were asked if they would be interested in participating in
a study looking at “fathers’ relationships with their children, and the experiences of fathers in the parenting of young
children.” Names of interested fathers were passed on to the primary investigator, who subsequently contacted the parent
to explain the study and schedule an initial home or office visit. Written informed consent was obtained at the outset
of the initial visit. Measures were randomly presented to control for order effects and administered during one session.
Screening and background measures were also administered: the CAPI, the Childhood Trauma Questionnaire-Short Form
(Bernstein, Ahluvalia, Pogge, & Handelsman, 1997), and the CAGE alcohol dependence screening questionnaire (Ewing, 1984).
All participants were debriefed, provided with contact information for various community agencies, and compensated $50.

Recruitment of maltreating fathers

The maltreating fathers were referred and recruited through the Children’s Aid Societies (CAS), Probation and Parole
offices of the Ministry of Community Safety and Correctional Services, and an agency providing intervention for men who
abuse their partners and/or their children. A small number of referrals came through a children’s tertiary mental health
facility (n = 3). As noted above, all fathers in the maltreating group were identified and followed by child protective service
due to physical maltreatment concerns. The vast majority of abusive fathers were involved with both CPS and the community
agency for men who abuse their family members.
It was very difficult to recruit and foster participation by the maltreating group. Thirteen maltreating fathers rescheduled at
least one of their appointments. Of these fathers, the average rate of rescheduling was 4.23 times. Only two of the maltreating
fathers called in advance to cancel their sessions; the others simply did not show for their session. Three maltreating fathers
were rescheduled more than 10 times, and one father rescheduled sessions 18 times (and failed to show for any of them
despite insisting that he was invested in participating). With two abusive fathers, a period of 8 months elapsed before their
initial expressed interest and their ability to attend for the study.
Reasons for choosing not to begin or complete the study were numerous. Some fathers could not be contacted due to
telephone disconnection or change of contact information (n = 4), loss of housing (n = 1), moving out of the city (n = 1), or
incarceration (n = 2). Others (n = 5) seemed unable to attend due to the chaotic nature of their lives: they forgot appointments,
other commitments arose, they were unable to arrange for travel (although taxi fare and/or bus fare was provided), and so
on. Still others simply said they no longer desired to participate. Three fathers also cited their conflict with the referring
agency as associated with decisions not to pursue research participation.

Recruitment of non-maltreating fathers

The non-maltreating comparison fathers were primarily recruited from neighborhood daycares, agencies, after-school
programs, and family service programs within the same community as the abusive fathers. Advertisements inviting partic-
ipation were posed throughout the community and in three local newspapers. In addition, 1350 advertisements were sent
home with Junior Kindergarten through Grade 6 students of neighborhood schools; the latter approach received a response
of only three fathers. In total, 30 fathers responded to these recruitment efforts. In comparison to the maltreating fathers,
it was significantly less complex to arrange for participation of the non-maltreating fathers (only one father requested that
his session be rescheduled, and he called in advance).
 K.J. Francis, D.A. Wolfe / Child Abuse & Neglect 32 (2008) 1127–1137 1131

Results

Approach to data analysis

To examine differences between abusive and non-abusive fathers on continuous measures of individual functioning (i.e.,
mental health concerns, parenting stress, and anger) data were analyzed using Mancova to reduce Type I error, covarying
on education due to differences noted below. All multivariate effects were investigated using Pillai’s Trace statistic, followed
by univariate tests. We then used Chi-square analyses to examine categorical differences (clinical versus non-clinical range)
between groups. Chi-square was also used to test the hypothesis that abusive fathers would show less empathy toward their
child and to perceive children’s emotional cues more negatively. Because the distributions of three categories (e.g., Passive,
Shame, and Disgust) on the IFEEL test were significantly non-normal, the Mann–Whitney (non-parametric) rank sums test
was used for these comparisons.

Demographic and background characteristics

Table 1 provides demographic and background characteristics of the abusive and non-abusive fathers. Overall the two
groups were reasonably similar on demographic factors, but differed on background characteristics as predicted. The two
groups did not differ with respect to age, race/ethnicity, or employment status; however, abusive fathers’ highest attained
educational grade level and income were significantly lower than that of non-abusive fathers (because education and income
were highly correlated we co-varied on education only). As expected, abusive fathers were significantly more likely to have
a history of charges related to both property offenses and violent offenses, and were more likely to report owning a weapon.
More than half (56%) of the abusive fathers had charges for violent offenses, such as assault (including woman abuse), assault
with a weapon, armed robbery, extortion, and sexual assault.
Abusive fathers also reported experiencing significantly more maltreatment in childhood than non-abusive fathers
(Table 1), comprised mostly of multiple forms of severe maltreatment. The nature of the abuse they reported from their
own childhood was qualitatively different in the extent of cruelty, compared to that experienced by non-abusive fathers. For
example, several abusive fathers described being locked in the bedroom or closets for extended periods of time when they
were “bad,” and being forced to urinate and defecate in the closet; several described being beaten with fishing rods, extension

Table 1
Demographic and background characteristics of abusive and non-abusive fathers.

Variable Abusive (n = 23) Non-abusive (n = 25) F(1, 46) Chi-square (1, n = 48)

Age (M, S.D., range) 38.44 (7.54; 25–61) 39.20 (5.63; 29–50) .27
Caucasian (%) 87.0 80.0 5.93
Number of Children (range) 3.72 (1–7) 2.32 (1–5) 8.89**

Marital Status (%) 15.47**

Single 4.3 0
Married 47.8 84.0
Separated 39.1 4.0
Divorced 0 12.0
Living with partner 8.7 0

Years of school completed (M, S.D., range; category below) 11.87 (1.01; 10–13) 13.68 (1.35; 12–16) 27.35*** 15.17***
Did not complete high school (%) 39.1 0
Completed high school (%) 26.1 16.0
Degree or years beyond high school (%) 34.8 84.0

Employed (%) 65.2 64.0 .01

Annual household income (%)

Annual income over $39,999 9.5 60.0 12.48***
Annual income under $20,000 52.4 16.0 6.87**
Annual income under $10,000 9.5 4.0 .57

Arrested for non-violent offense (%) 47.8 16.0 5.65*

Arrested for violent Offense (%) 56.5 4.0 16.00***

History of maltreatment as childa (M, S.D.) 86.24 (26.41) 55.81 (14.63) 20.28b , **
History of alcohol abusec 47.6% 8.3% 8.84**

Note: one of the abusive fathers declined to provide this information.

a
Based on the Childhood Trauma Questionnaire (Bernstein et al., 1997).
b
d.f. = 1, 36.
c
Based on the CAGE (Ewing, 1984).
*
p ≤ .05.
**
p ≤ .01.
***
p ≤ .001.
1132 K.J. Francis, D.A. Wolfe / Child Abuse & Neglect 32 (2008) 1127–1137

Table 2
Means and standard deviations of BSI Clinical Scales, and percentage falling in clinical range, for abusive and non-abusive fathers.

BSI clinical scales Abusive fathers (n = 20) Non-abusive fathers (n = 19) Statistics
M (S.D.) % clinical range M (S.D.) % clinical range F(1, 43) Chi-square (1)

Somatization 59.84 (12.40) 33.3 48.21 (7.62) 5.0 8.21** 5.24*

Interpersonal sensitivity 57.64 (10.47) 40.0 53.13 (7.64) 15.0 3.16
Depression 61.75 (8.84) 50.0 50.20 (8.43) 10.0 10.86** 7.62**
Anxiety 86.48 (12.81) 47.6 55.50 (10.61) 20.0 3.48
Hostility 59.39 (7.45) 52.8 56.03 (8.46) 15.0 6.34**
Paranoid ideation 60.37 (9.65) 61.9 54.42 (8.37) 10.0 11.90***
Psychoticism 59.25 (11.59) 42.9 54.65 (8.57) 10.0 5.63*
GSI 62.22 (8.88) 55.0 56.10 (8.79) 15.8 6.51**

Note: GSI is the Global Severity Index. Profiles with incomplete data were removed from analyses due to the small number of items per scale.
*
p ≤ .05.
**
p ≤ .01.
***
p ≤ .001.

cords, and hangars. Eight of the abusive fathers had themselves been taken in by child protective services when they were
children. In contrast, only one of the non-abusive fathers described abuse characterized by this level of degradation. Finally,
almost half (47.6%) of abusive fathers reported a history of problematic alcohol use, and approximately 60% had stepchildren.

Mental health and parenting stress

A significant multivariate effect was observed for abusive and non-abusive fathers’ self-reported mental health functioning
on the BSI scales [Pillai’s = .52, F(8, 36) = 3.72, p ≤ .01]. Examination of the univariate effects revealed that abusive fathers
reported significantly more Somatic Distress and Depression than non-abusive fathers (see Table 2). Fathers’ scores were
also compared with a series of chi-square analyses corrected with a Bonferroni adjustment for multiple comparisons, based
on their categorization in the elevated (clinically significant) range on the BSI (Table 2). Significantly more abusive fathers’
scores fell in the clinically significant range for the Paranoia, Hostility, Depression, Psychoticism, Somatization and Global
Severity Scales. Compared to non-abusive fathers, more than six times as many abusive fathers’ scores fell in the clinically
significant range for the Paranoia and Somatization scales, and five times as many scored in the elevated range on the
Depression scale. Over 60% of abusive fathers reported elevated scores on at least one of the clinical scales assessing mental
health functioning, in comparison with only 20% of non-abusive fathers. Also, abusive fathers were significantly more likely
to demonstrate elevated scores on almost all clinical scales, including depression, hostility, paranoid ideation, psychoticism
and somatization.
A significant multivariate effect was also obtained for the PSI/SF scale analyses [Pillai’s = .38, F(5, 33) = 4.05, p ≤ .01].
Univariate analyses revealed that abusive fathers showed significantly greater stress on each subscale: Parental Distress,
Parent–Child Dysfunctional Interaction, and Difficult Child, as well as the Total Parenting Stress Scale (Table 3). Table 3 also
shows fathers’ responses on the PSI/SF categorically, based on the percent who scored in the clinically elevated range (above
the 85th percentile) on each scale. Chi-square analyses revealed that significantly more abusive fathers reported scores in
the clinically elevated range for the Parental Distress, Difficult Child, and Total Parenting Stress scales.

Empathy and anger

A significant multivariate effect was found [Pillai’s = .184, F(2, 43) = 4.85, p ≤ .05] on fathers’ scores on the Empathic Concern
and Perspective Taking scales of the IRI. Consistent with our hypothesis, univariate tests indicated that compared with non-
abusive fathers (M = 28.00, S.D. = .83), abusive fathers (M = 25.06, S.D. = .80) rated themselves as having significantly less
empathic concern for their child [F(1, 45) = 5.27, p ≤ .05]. Similarly, abusive fathers (M = 22.69, S.D. = .91), as compared to non-

Table 3
Means and standard deviations of Parenting Stress Index (PSI) Scale Scores, and percentage falling in clinical range, for abusive and non-abusive fathers.

PSI clinical scales Abusive fathers (n = 21) Non-abusive fathers (n = 19) Statistics

M (S.D.) % clinical range M (S.D.) % clinical range F(1, 37) Chi-square (1)

Parental distress 32.10 (6.18) 11.0 27.10 (4.61) 4.0 4.80* 3.68*
Parent–child dysfunctional interaction 32.07 (9.48) 14.0 19.03 (5.84) 7.0 17.34***
Difficult child 37.62 (8.32) 16.0 26.53 (7.13) 5.0 12.45*** 8.79**
Total parenting stress 102.21 (21.95) 17.0 74.19 (14.67) 5.0 14.07*** 10.65***

Note: seven PSI profiles were removed from analyses due to defensive responding (two abusive and five comparison fathers), with no significant change in
the pattern of results.
*
p ≤ .05.
**
p ≤ .01.
***
p ≤ .001.
 K.J. Francis, D.A. Wolfe / Child Abuse & Neglect 32 (2008) 1127–1137 1133

Table 4
Means and standard deviations of STAXI-2 Scale, and percentage falling in clinical range, for abusive and non-abusive fathers.

STAXI-2 clinical scales Abusive fathers (n = 23) Non-abusive fathers (n = 25) Statistics

M (S.D.) % clinical range M (S.D.) % clinical range F(1, 45) Chi-square (1)
*
State anger 18.06 (5.22) 13.0 14.91 (.80) 0.0 5.59 3.48*
Trait anger 18.39 (4.85) 43.5 15.76 (3.25) 12.0 3.08 6.01**
Anger expression out 17.05 (2.70) 52.2 13.92 (3.18) 20.0 8.34** 5.42*
Anger expression index 49.84 (20.82) 52.2 27.86 (13.94) 24.0 12.16*** 4.06*

Note: higher scores represent higher levels on the specified scale. One profile from the abusive group was removed from analyses due to incomplete data.
*
p ≤ .05.
**
p ≤ .01.
***
p ≤ .001.

abusive fathers (M = 26.19, S.D. = .94), rated themselves as being significantly less likely to consider their child’s perspective
(F(1, 45) = 5.84, p ≤ .05).
The omnibus test for the STAXI-2 scales (see Table 4) revealed a significant multivariate effect [Pillai’s = .34, F(4, 42) = 5.29,
p ≤ .01]. Examination of the univariate effects indicated that abusive fathers reported significantly higher levels of State Anger.
Consistent with predictions, abusive fathers also were significantly more likely to express their anger through verbally or
physically aggressive behavior (Anger Expression Out), and reported a significantly higher level of total anger experience and
expression (Anger Expression Index). Chi-square analyses were conducted to determine whether more abusive than non-
abusive fathers reported scores that fell in the clinically significant range of the STAXI-2 anger scales. Again, significantly
more abusive fathers reported clinically elevated scores on all four scales (State Anger, Trait Anger, Anger Expression Out,
and Anger Expression Index; see Table 4).

Perception of child emotions

Descriptive information and test statistics for each of the emotion categories are presented in Table 5. Consistent with
previous findings of non-normal distributions of IFEEL responses, a number of the emotion categories were not normally
distributed (Shame, Disgust, and Passive emotion categories had skewness or kurtosis above an absolute value of 5);
non-parametric tests were used to examine group differences for these three categories. Multivariate analysis of variance
comparing the frequency of responses of abusive and non-abusive fathers on the Surprise, Interest, Joy, Content, Sad, Cautious,
Disgust, Anger, Distress, and Fear categories was significant [Pillai’s = .40, F(10, 35) = 2.44, p ≤ .05]. Examination of the means
revealed that abusive fathers gave more responses falling in the Anger, Fear, and Other categories, and fewer falling in the
Interest category. “Other” responses reported by abusive fathers included “ugly,” “snaky,” “devious,” “deprived,” “spooky,”
and “attitude” (one abusive father said “put horns on that kid” in response to one of the photographs). In contrast, “other”
responses provided by non-abusive fathers included “wow,” “matter of fact,” “don’t know,” “I’ve done something and I’m not
sure whether it’s good or bad,” and “is there any expression?” The Mann–Whitney (non-parametric) rank sums test was used
to examine group differences between abusive fathers and non-abusive fathers on the Shame, Disgust, and Passive categories.
The z-statistic was significant for the Disgust (p ≤ .05) category (five abusive fathers provided “Disgust” responses, compared
to no non-abusive fathers). Abusive fathers also provided significantly fewer “Passive” responses (p ≤ .05); no difference
between the groups was found for the Shame category.

Table 5
Means and standard deviations of abusive and non-abusive fathers’ responses to IFEEL pictures by emotion category.

Emotion category Abusive fathers (n = 22) Non-abusive fathers (n = 25) F(2, 33) Mann–Whitney U
M (S.D.) M (S.D.)

Surprise 2.31 (1.53) 2.31 (2.23) .00

Interest 5.75 (2.72) 8.17 (2.99) 6.03*
Joy 3.21 (1.67) 3.29 (1.39) .03
Content 2.07 (1.75) 3.29 (2.08) 3.32
Passive 1.14 (.38) 2.04 (.35) 193.5; Z = −1.98*
Sad 3.32 (2.38) 3.81 (2.50) .34
Cautious 2.56 (2.06) 1.89 (1.65) 1.27
Shame .18 (.09) .19 (.08) 283.0; Z = −.09
Disgust .19 (.06) 0.00 (0.0) 234.0; Z = −2.25*
**
Anger 2.02 (1.96) .69 (.72) 6.71
Distress 3.01 (2.68) 2.57 (1.47) .38
Fear 2.38 (1.63) .92 (1.37) 8.08**
Other 1.86 (2.29) .73 (.87) 4.04*

Note: the profiles from two abusive fathers were removed due to incomplete data.
*
p ≤ .05.
**
p ≤ .01.
1134 K.J. Francis, D.A. Wolfe / Child Abuse & Neglect 32 (2008) 1127–1137

Discussion

As expected, abusive fathers had less educational achievement, lower financial status, higher levels of childhood trauma,
and significantly more current mental health concerns than non-abusive fathers. They also reported a greater history of
alcohol abuse, more arrests for violent offenses, more weapon ownership, and described themselves as more hostile and
suspicious. Abusive fathers reported more parenting stress and anger, and a greater likelihood of aggression when angry.
They also reported being less likely to take their child’s perspective and feel empathic concern for their child. Abusive fathers’
responses reflected higher levels of parent–child relationship distress. They perceived their children as more difficult and also
perceived children’s emotional displays as more negative and as reflecting more threat-related emotions. Further, abusive
fathers had more children, specifically more stepchildren, than non-abusive fathers, which is of particular concern in light
of research indicating that children are at significantly greater risk of homicide from stepparents (Daly & Wilson, 2000).

Mental health and parenting stress

Abusive fathers in this study were higher on scales assessing depression, paranoia, and hostility (among other scales),
which suggest that when faced with child-related stimuli they may form quick, negative attributions and engage in behavior
reflecting this. It is likely that the experience of mental health problems interferes with men’s ability to cope effectively
with parenting-related difficulties, and perhaps renders them more vulnerable to such difficulties. Depression has been
linked to biased perceptions and negative beliefs about others, and research with depressed mothers demonstrates an
increased risk of problematic parenting (Azar & Wolfe, 2006). As well, previous research illustrates that severely mentally ill
individuals who are high in paranoia are more likely to make quick judgments, fail to seek disconfirming evidence, and be
more certain of their judgments (Peer, Rothmann, Penrod, Penn, & Spaulding, 2004). These patterns have been documented
as particularly pronounced when tasks involve emotional material (Mujica-Parodi, Malaspina, & Sackeim, 2000). Further,
individuals who are high in hostility are more likely to experience anger, disgust, and contempt, and exaggerate and distort
hostile or threatening aspects of others (Fenigstein, 1997). The elevations of mental health symptoms are concerning, and
further research is necessary to determine whether these findings are reflective of the larger group of abusive fathers, or
if this represents a particularly impaired sample. Not all abusive fathers reported impaired mental health functioning. It
may be that a sub-group of abusive fathers struggles with mental health issues, while other sub-groups do not. It is also
notable that the abusive fathers in this sample were all involved with child protective services; such involvement may lead to
heightened levels of hostility and paranoia. Future studies using structured clinical interviews may yield more comprehensive
information about abusive fathers’ mental health functioning.
Abusive fathers also reported significantly more stress directly related to parenting, with approximately 80% scoring in the
elevated level on a measure assessing overall parenting-related distress. The experience of stress related to parenting has a
strong influence on abusive behavior (Wolfe, 1999). The abusive fathers scored higher on a subscale assessing parent distress,
suggesting a poorer sense of parenting competence, role restriction, conflict with the other parent, depression, and lack of
social support. Their scores on a scale assessing dysfunctional parent–child interactions were also significantly higher than
those of non-abusive fathers, suggesting that they do not perceive the parent–child relationship to be reinforcing. Rather,
an abusive father may view his or her child negatively, see himself as mistreated or rejected by the child, and perceive the
child as not meeting his (the parent’s) expectations (Abidin, 1995). This perception may increase the likelihood of abusive
behavior. This finding seems particularly noteworthy since abusive fathers were less likely to have custody of their child.
Abusive fathers also scored higher on an index assessing the experience of intense angry feelings. Although the experience
of anger in itself does not explain aggression (i.e., an individual may be angry about something without acting on the anger
aggressively), the experience of high levels of negative affect increases risk for engaging in abusive behavior as it interferes
with effective, tolerant, and flexible information-processing (Holtzworth-Munroe & Hutchinson, 1993). In the parent–child
relationship, anger interferes with parents’ ability to respond to their child in a proactive, responsive and balanced manner
(Denham et al., 2000; Hakim-Larson et al., 1999). Decreased flexibility in information-processing impedes parents’ ability
to generate alternative explanations for and responses to anger, which may result in increased aggression. Indeed, abusive
fathers reported higher scores on a scale assessing outward expression of anger. High scores on this scale reflect the frequent
aggressive expression of anger, such as through verbal aggression or physical aggression toward other people and/or objects
(Spielberger, 1998). More than half of the abusive fathers reported levels of the outward expression of anger that were above
that of 75% of their same-sex age-related peers. Thus, the general experience of anger by abusive fathers is more likely to be
associated with aggressive behavior, similar to findings with child abusive mothers (Wolfe, 1999) and batterers (Norlander
& Eckhardt, 2005).

Empathy and perceptions of the child

On a measure assessing fathers’ empathic abilities, abusive fathers rated themselves as having less cognitive and affec-
tive empathy for their children. Specifically, abusive fathers indicated that they were less likely to try to consider their
child’s perspective and were less likely to feel empathic concern for their child. Lower levels of perspective-taking have
been associated with high levels of impulsivity (Beven et al., 2004), which serves to place the child at greater risk, par-
ticularly in the face of abusive fathers’ heightened anger and general aggression. Perez-Albeniz and de Paul (2004) also
 K.J. Francis, D.A. Wolfe / Child Abuse & Neglect 32 (2008) 1127–1137 1135

found that abusive fathers reported lower levels of perspective taking but they did not find differences in empathic concern,
which may be due to their small sample (n = 9). Empathy for others’ emotional experience is an important component in
managing any social context, including the parenting context (Saarni, 1999). An appreciation of how his child may feel or
view events allows the father to see the child as an individual with a unique viewpoint and feelings. In addition, empathy
for the child’s experience may result in greater flexibility and less negative affect in the face of challenging circumstances,
and in turn enhance the father’s support of his child’s endeavors. However, if the father is unable or unwilling to recog-
nize and appreciate a child’s differing viewpoint, feelings, and needs, his ability to respond to the child in a sensitive and
responsive manner is compromised and action becomes focused on meeting the father’s needs rather than the needs of
the child.
Abusive fathers also perceived their child’s behavior as significantly more difficult than non-abusive fathers on the PSI,
with almost three-quarters of their scores falling in the clinically elevated range. High scores suggest that the parent per-
ceives their child to be demanding, not adaptable, and difficult to manage. Although children of abusive parents may exhibit
more difficult behavior as a result of dysregulation stemming from maltreatment and poor parenting, research with abu-
sive mothers suggests that the adult’s perception of the child is distorted. Abusive mothers perceive the behavior of their
child (as well as other children) as more difficult and negative during behavioral/dyadic interaction tasks than indepen-
dent raters (Wolfe, 1999), and over-report their child’s negative behavior (Haskett, Scott, & Fann, 1995). Similar findings
have been reported regarding male batterers’ distorted negative perceptions of their partners (Murphy & Eckhardt, 2005).
Abusive fathers also interpreted children’s socio-emotional signals more negatively on the IFEEL pictures task, and per-
ceived more negative or hostile emotions on children’s faces. For example abusive fathers were more likely to label the
children’s facial expressions as portraying feelings of anger and disgust, which suggests they may have a hostile attri-
bution bias and are vigilant to signs of threat. A cognitive style marked by bias and vigilance is consistent with their
elevations on paranoia and hostility. That is, emotions associated with hostility include anger, disgust, and contempt, and
the tendency to perceive others’ emotions as those of anger, disgust, and contempt as well (Brummett et al., 1998; Izard,
1991).
Abusive fathers’ heightened perceptions of anger and fear expressions in children have significant implications for their
parenting. Facial expressions do not simply serve the purpose of signaling an emotion state; they also serve to communicate
behavioral response requirements (Walden & Smith, 1997). For example, perceiving another’s emotional signal as anger
prompts a response to handle conflict (Niedenthal, Brauer, Robin, & Innes-Ker, 2002). Because abusive fathers are more likely
to perceive anger and hostile intent they may feel that they are often in the defensive position of having to manage struggles.
Over time such vigilance can lead to chronic feelings of frustration and emotion dysregulation, and increase the likelihood
of an aggressive or punishing response. At a minimum, the perception of anger in a child’s face signals that the child is less
available to meet the needs of the father, which is likely to be threatening to their sense of entitlement (Scott & Crooks, 2004).
Abusive fathers were also less likely to perceive interest in children’s facial expressions. Children’s displays of interest and
curiosity can be very rewarding for parents, and usually evoke an “approach” response to the child (Niedenthal et al., 2002).
If fathers do not perceive their child as interested they are less likely to foster children’s interest and respond to the child’s
needs and preferences for stimulation. This failure to respond with “approach” and stimulating behavior can also affect the
attachment relationship (Egeland & Erickson, 1987). Further, if an abusive father does not perceive a child as interested in
something he thinks the child should be interested in, the father may engage in more forceful attempts to draw the child’s
interest (e.g., forcing the child to look in a certain direction, overstimulating the child, etc.).
These results of the IFEEL pictures support the premise that abusive fathers have a negative attribution bias toward
children, even children with whom they are unfamiliar (Bugental, 2004). Although they had no information about the
behavior of the children depicted in the IFEEL pictures, abusive fathers often perceived the children’s expressions more
negatively than non-abusive fathers. Similar findings of attribution bias have been reported with neglectful (Hildyard & Wolfe,
2007) and abusive mothers (Dadds, Mullins, McAllister, & Atkinson, 2003). This suggests that the abusive fathers’ negative
perceptions of children do not simply reflect experiences with their own children’s difficult emotions and behavior. Rather,
these perceptions may also be a function of the fathers’ underlying biases and deficits in perceptions and/or interpretations
of children’s socio-emotional signals.

Implications for intervention

Despite evidence that abusive fathers struggle with a myriad of difficulties stemming from their own maltreatment histo-
ries and current mental health concerns, none was receiving treatment beyond child protective services. This discrepancy, as
well as the significance of their problems and their impact on others, underscores the importance of increasing therapeutic
assistance for fathers who abuse their children. Treatment approaches must be responsive to the multiple mental health
concerns and trauma histories among these men, and avoid a more narrow problem-solving focus that may ignore the signif-
icant influence of cognitive and emotional factors affecting their parenting role (Kelly & Wolfe, 2004). Cognitive-behavioral
strategies have been effective in altering hostile perceptions and negative emotional arousal with abusive mothers (Azar &
Wolfe, 2006) and thus may be beneficial in reducing these critical factors among abusive fathers. For example, Scott, Francis,
Crooks, & Kelly (2006) and Scott, Francis, Crooks, et al. (2006) are evaluating a program for abusive fathers that incorporates
cognitive behavioral and motivational interviewing strategies designed to encourage them to examine their views, attitudes,
and expectations of children, and to improve their responses to typical childrearing situations.
1136 K.J. Francis, D.A. Wolfe / Child Abuse & Neglect 32 (2008) 1127–1137

Limitations and future directions

Some design and methodological limitations should be noted. The recruitment strategies used in this study pose certain
limitations. Attempts were made to broadly solicit participation through a variety of means (e.g., schools, advertisements,
community postings, visits to agencies), yet the response rate was low (similar to other studies of abusive fathers; Bugental,
2004; Duggan et al., 2004). Abusive men who did elect to participate may differ on important variables from those who did
not. In addition, larger sample sizes would enable researchers to match the groups on additional variables such as education
and socioeconomic status, rather than statistically control for them. Because abusive fathers were initially required to have
contact with their children, those whose access had been terminated (e.g., because of abuse or imprisonment) were not
able to participate. Thus, the sample may under-represent more severely abusive fathers. A further limitation is the lack of
measures validated specifically for use with this population. Although the comparison group excluded fathers who scored
high on abuse potential, this cut-off score was not derived from father-specific norms. Similarly, the reference sample for the
IFEEL pictures is comprised solely of mothers. In addition, although the modifications to the IRI were minor, there has been
no psychometric testing of the validity of this measure in a parent population. Future research needs to establish valid and
reliable measures normed on male samples. As well, most of the measures were self-report, which are subject to retrospective
bias and demand characteristics. Finally, this study focused on individual factors related to fathers’ abuse of their children.
There are likely to be sub-groups of fathers for whom anger, cognitive deficits and distortions, abuse in the family of origin,
and mental health difficulties manifest in different ways. In addition, actual and perceived power and control have been
identified as significant factors in the perpetration and impact of men’s abusive behavior (e.g., Peled, 2000; Scott, Francis,
Crooks, & Kelly, 2006; Scott, Francis, Crooks, et al., 2006); this dynamic warrants investigation in fathers who abuse their
children. Further empirical studies to examine the mechanisms by which cognitive and affective factors interact with family,
community, and social factors will provide a firmer theoretical understanding of the different pathways to maltreatment in
different fathers.

Acknowledgements

The authors thank Tim Kelly and Michele Paddon of Changing Ways London, and Maureen Reid of London Middlesex
Children’s Aid Society, for their invaluable assistance in recruiting fathers for this project. The authors also wish to thank
Marcus Juodis, Adam Jaffe, and Hammad Khan for their assistance in data collection, and Katreena Scott and Claire Crooks
for their encouragement and conceptual assistance in undertaking the study.

References

Abidin, R. R. (1995). Parenting stress index (3rd ed.). Lutz, FL: Psychological Assessment Resources, Inc.
Ammerman, R. T., Kolko, D. J., Kirisci, L., Blackson, T. C., & Dawes, M. A. (1999). Child abuse potential in parents with histories of substance use disorder. Child
Abuse & Neglect, 23(12), 1225–1238.
Applebaum, M. I., Butterfield, P. M., & Culp, R. E. (1993). Operating characteristics and psychometric properties of the IFEEL Pictures. In R. N. Emde, J. D.
Osofsky, & P. M. Butterfield (Eds.), The IFEEL pictures: A new instrument for interpreting emotions (pp. 97–130). Madison, CT: International Universities
Press.
Atkins, M., & Steitz, J. (2002). The assessment of empathy: An evaluation of the Interpersonal Reactivity Index.
http://www.uu.edu/union/academ/tep/research/atkins.htm Accessed 15.11.04.
Azar, S. (1997). A cognitive behavioral approach to understanding and treating parents who physically abuse their children. In D. A. Wolfe, R. J. McMahon,
& R. D. Peters (Eds.), Child abuse: New directions in prevention and treatment across the lifespan (pp. 79–101). Thousand Oaks, CA: Sage.
Azar, S., & Wolfe, D. (2006). Child physical abuse and neglect. In E. J. Mash & R. A. Barkley (Eds.), Treatment of childhood disorders (3rd ed., pp. 595–646). New
York: Guilford.
Bancroft, L., & Silverman, J. G. (2002). The batterer as parent: Addressing the impact of domestic violence on family dynamics. Thousand Oaks, CA: Sage.
Bernstein, D. P., Ahluvalia, T., Pogge, D., & Handelsman, L. (1997). Validity of the Childhood Trauma Questionnaire in an adolescent psychiatric population.
Journal of the American Academy of Child and Adolescent Psychiatry, 36, 340–348.
Beven, J. P., O’Brien-Malone, A., & Hall, G. (2004). Using the Interpersonal Reactivity Index to assess empathy in violent offenders. International Journal of
Forensic Psychology, 1, 33–41.
Brummett, R., Maynard, K., Babyak, M., Haney, T., Siegler, I., Helms, M., & Barefoot, J. (1998). Measures of hostility as predictors of facial affect during social
interaction: Evidence for construct validity. Annals of Behavior Medicine, 20, 168–173.
Bugental, D. B. (1993). Communication in abusive relationships: Cognitive constructions of interpersonal power. American Behavioral Scientist, 36, 288–308.
Bugental, D. B. (2004). Finding ways to reduce the prevalence of child maltreatment among fathers: A comment on the alternative approaches. Clinical
Psychology: Science and Practice, 11, 112–115.
Butterfield, P. M., & Ridgeway, D. (1993). The IFEEL pictures: Description, administration, and lexicon. In R. N. Emde, J. D. Osofsky, & P. M. Butterfield (Eds.),
The IFEEL pictures: A new instrument for interpreting emotions (pp. 73–95). Madison, CT: International Universities Press.
Coohey, C. (2006). Physically abusive fathers and risk assessment. Child Abuse & Neglect, 30(5), 467–480.
Dadds, M. R., Mullins, M. J., McAllister, R. A., & Atkinson, E. (2003). Attributions, affect, and behavior in abuse-risk mothers: A laboratory study. Child Abuse
& Neglect, 27, 21–45.
Daly, M., & Wilson, M. (2000). Not quite right. American Psychologist, 55(6), 679–680.
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology,
44(1), 113–126.
Davis, M. H. (1994). Empathy: A social psychological approach. Madison, CT: Brown & Benchmark.
De Bellis, M. D., Broussard, E. R., Herring, D. J., Wexler, S., Moritz, G., & Benitez, J. G. (2001). Psychiatric co-morbidity in caregivers and children involved in
maltreatment: A pilot research study with policy implications. Child Abuse & Neglect, 25, 923–944.
Delsol, C., & Margolin, G. (2004). The role of family-of-origin violence in men’s marital violence perpetration. Clinical Psychology Review, 24, 99–123.
Denham, S. A., Workman, E., Cole, P. M., Weissbrod, C., Kendziora, K. T., & Zahn-Waxler, C. (2000). Prediction of externalizing behavior problems from early
to middle childhood: The role of parental socialization and emotion expression. Development and Psychopathology, 12(1), 23–45.
 K.J. Francis, D.A. Wolfe / Child Abuse & Neglect 32 (2008) 1127–1137 1137

Derogatis, L. (1993). Brief Symptom Inventory: Administration, scoring and procedures manual. Minneapolis, MN: National Computer Systems Inc.
Dubowitz, H. (2006). Where’s Dad? A need to understand fathers’ role in child maltreatment. Child Abuse & Neglect, 30(5), 461–465.
Duggan, A. K., Fuddy, L., McFarlane, E., Burrell, L., Windham, A., Higman, S., & Sia, C. (2004). Evaluating a statewide home visiting program to prevent child
abuse in at-risk families of newborns: Fathers’ participation and outcomes. Child Maltreatment, 9, 3–17.
Eckhardt, C. I., Barbour, K. A., & Stuart, G. L. (1997). Anger and hostility in maritally violent men: Conceptual distinctions, measurement issues, and literature
review. Clinical Psychology Review, 17, 333–358.
Eckhardt, C., Norlander, B., & Deffenbacher, J. (2004). The assessment of anger and hostility: A critical review. Aggression and Violent Behavior, 9, 15–43.
Egeland, B., & Erickson, M. F. (1987). Psychologically unavailable caregiving. In M. Brassard, B. Germain, & S. Hart (Eds.), Psychological maltreatment of children
and youth (pp. 110–120). Elmsford, NY: Pergamon.
Ewing, J. A. (1984). Detecting alcoholism: The CAGE questionnaire. Journal of the American Medical Association, 252(14), 1905–1907.
Fenigstein, A. (1997). Paranoid thought and self-schematic processing. Journal of Social and Clinical Psychology, 16, 77–94.
Guterman, N. B., & Lee, Y. (2005). The role of fathers in risk for physical child abuse and neglect: Possible pathways and unanswered questions. Child
Maltreatment, 10(2), 136–149.
Hakim-Larson, J., Dunham, K., Vellet, S., Murdaca, L., & Levenbach, J. (1999). Parental affect and coping. Canadian Journal of Behavioral Science, 31, 5–18.
Haskett, M. E., Scott, S. S., & Fann, K. D. (1995). Child Abuse Potential Inventory and parenting behavior: Relationships with high-risk correlates. Child Abuse
& Neglect, 19, 1483–1495.
Hildyard, K., & Wolfe, D. A. (2007). Understanding child neglect: Cognitive processes underlying neglectful parenting. Child Abuse & Neglect, 31, 895–907.
Holtzworth-Munroe, A., & Hutchinson, G. (1993). Attributing negative intent to wife behavior: The attributions of maritally violent versus nonviolent men.
Journal of Abnormal Psychology, 102, 206–211.
Izard, C. (1991). The psychology of emotions. New York, NY: Plenum.
Jacobson, N. S., Gottman, J. M., Waltz, J., Rushe, R., & Babcock, J. (2000). Affect, verbal content, and psychophysiology in the arguments of couples with a
violent husband. Prevention and Treatment, 3, 1–12.
Kelly, T., & Wolfe, D. A. (2004). Advancing change with maltreating fathers. Clinical Psychology Science and Practice, 11, 116–119.
Mammen, O. K., Kolko, D. J., & Pilkonis, P. A. (2002). Negative affect and parental aggression in child physical abuse. Child Abuse & Neglect, 26, 407–424.
Milner, J. S. (1986). The Child Abuse Potential Inventory manual (2nd ed.). Webster, NC: PSytec Corporation.
Milner, J. S. (2003). Social information processing in high-risk and physically abusive parents. Child Abuse & Neglect, 27, 7–20.
Moss, H. B., Mezzich, A., Yao, J. K., Gavaler, J., & Martin, C. S. (1995). Aggressivity among sons of substance-abusing fathers: Association with psychiatric
disorder in the father and son, paternal personality, pubertal development, and socioeconomic status. American Journal of Drug and Alcohol Abuse, 21(2),
195–208.
Mujica-Parodi, L., Malaspina, D., & Sackeim, H. (2000). Logical processing, affect, and delusional thought in schizophrenia. Harvard Review of Psychiatry, 8,
73–83.
Murphy, C. M., & Eckhardt, C. I. (2005). Treating the abusive partner: An individualized cognitive behavioral approach. New York: Guilford.
National Research Council. (1993). Understanding child abuse and neglect. Washington, DC: National Academy Press.
Niedenthal, P. M., Brauer, M., Robin, L., & Innes-Ker, A. H. (2002). Adult attachment and the perception of facial expression of emotion. Journal of Personality
and Social Psychology, 82, 419–433.
Norlander, B., & Eckhardt, C. (2005). Anger, hostility, and male perpetrators of intimate partner violence: A meta-analytic review. Clinical Psychology Review,
25, 119–152.
Peer, J. E., Rothmann, T. L., Penrod, R. D., Penn, D. L., & Spaulding, W. D. (2004). Social cognitive bias and neurocognitive deficit in paranoid symptoms:
Evidence for an interaction effect and changes during treatment. Schizophrenia Research, 71, 463–471.
Peled, E. (2000). Parenting by men who abuse women: Issues and dilemmas. British Journal of Social Work, 30, 25–36.
Perez-Albeniz, A., & de Paul, J. (2004). Gender differences in empathy in parents at high- and low-risk of child physical abuse. Child Abuse & Neglect, 28,
289–300.
Saarni, C. (1999). The development of emotional competence. New York: Guildford Press.
Schaeffer, C. M., Alexander, P. C., Bethke, K., & Kretz, L. S. (2005). Predictors of child abuse potential among military parents: Comparing mothers and fathers.
Journal of Family Violence, 20, 123–129.
Schumacher, J. A., Feldbau-Kohn, S., Slep, A. M., & Heyman, R. E. (2001). Risk factors for male-to-female partner physical abuse. Aggression and Violent
Behavior, 6, 281–352.
Scott, K. L., & Crooks, C. (2004). Effecting change in maltreating fathers: Critical principles for intervention planning. Clinical Psychology: Science & Practice,
11, 95–111.
Scott, K. L., Francis, K. J., Crooks, C., & Kelly, T. (2006). Caring Dads: Helping fathers value their children. Victoria, BC: Trafford Publishing.
Scott, K. L., Francis, K. J., Crooks, C., Paddon, M., & Wolfe, D. A. (2006). Guidelines for intervention with abusive fathers. In J. Edleson & O. J. Williams (Eds.),
Parenting by men who batter: New directions for assessment and intervention (pp. 102–117). New York, NY: Oxford University Press.
Slep, A. M. S., & O’Leary, S. G. (2001). Examining partner and child abuse: Are we ready for a more integrated approach to family violence? Clinical Child and
Family Psychology Review, 4, 87–107.
Spielberger, C. D. (1998). State Trait Anger Expression Inventory. 2: Professional manual. Lutz, FL: Psychological Assessment Resources.
Straus, M. A., & Smith, C. (1990). Family patterns and child abuse. In M. A. Straus & R. J. Gelles (Eds.), Physical violence in American families (pp. 245–261).
New Brunswick, NJ: Transaction.
Trocme, N., Fallon, B., MacLaurin, B., Daciuk, J., Felstiner, C., Black, T., Tonmyr, L., Blackstock, K. B., Turcotte, D., & Cloutier, R. (2005). Canadian incidence study
of reported child abuse and neglect – 2003 Major findings. Ottawa, ON: Minister of Public Works and Government Services Canada.
US Department of Health and Human Services. (2006). Child maltreatment 2004. Washington, DC: Government Printing Office.
Walden, T. A., & Smith, M. C. (1997). Emotion regulation. Motivation & Emotion, 21, 7–25.
Wolfe, D. A. (1999). Child abuse: Implications for child development and psychopathology (2nd ed.). Thousand Oaks, CA: Sage.
Zahn-Waxler, C., & Wagner, E. (1993). Caregivers’ interpretations of infant emotions: A comparison of depressed and well mothers. In R. N. Emde, J. D.
Osofsky, & P. M. Butterfield (Eds.), The IFEEL pictures: A new instrument for interpreting emotions (pp. 175–184). Madison, CT: International Universities
Press.