The Influence of Maternal History of Abuse on Parenting Knowledge and Behavior

Abstract

This study examined the intergenerational transmission of abuse among a sample of 681 teen, adult low, and adult high resource first-time mothers. Participants ranged in age from 14 to 36 years, with a mean of 20 years. Exposure to childhood emotional and physical abuse was associated with 6-month parenting behavior; but not parenting knowledge. Teen mothers, as opposed to adult mothers, had higher mean scores for exposure to childhood emotional and physical abuse. Adult high resource mothers reported lower mean scores on each abuse outcome than both teen and adult low resource mothers. For the total sample of mothers, as past exposure to emotional and physical abuse increased, maternal responsivity decreased, and opinions towards, and propensities for, abusive behavior increased.

A history of physical abuse is associated with punitive parenting style and discipline practices (Dixon, Brown, & Hamilton-Giachritsis, 2005; Dixon, Hamilton-Giachritsis, & Brown, 2005; Haapasalo & Aaltonen, 1999; Pears & Capaldi, 2001). In attempting to explain the prevalence of harsh, insensitive parenting behavior documented among high risk populations, such as teen and low-resource mothers (e.g., Borkowski et al., 2007), it is particularly important to examine past exposure to abuse and how it relates to present parenting knowledge and behavior. Examinations of past childhood experiences combined with examinations of reported opinions and behaviors aid in clarifying mechanisms involved in strategies parents use when raising their children. The current study was designed to assess maternal history of abuse and how it subsequently relates to parenting knowledge and behavior among a diverse sample (i.e., teen, adult low resource - older than 21 with less than 2 years of college, and adult high resource - older than 21 with more than 2 years of college) of first time mothers.

Intergenerational Transmission of Child Abuse

One factor implicated in the etiology of child abuse is parents' own childhood experiences of abuse. It is estimated that up to one third of the variance predicting child maltreatment is accounted for by maternal history of abuse (Haapasalo & Aaltonen, 1999). It has been shown that mothers who had experienced some form of abuse during their lifetime were more likely to engage in negative responses and abusive behavior toward their children than mothers who did not experience abuse (Dixon, Brown, et al., 2005; Dixon, Hamilton-Giachritsis, et al., 2005; Scaramella & Conger, 2003). Consequently, insight into an individual's upbringing is an important factor in understanding the development of beliefs about parenting and parenting practices (Miller, 1988). Investigations of abusive parenting have led researchers to examine parents' concurrent attitudes and behaviors as well as childhood experiences in an attempt to better understand the occurrence of abuse. It has been well documented that individuals with a history of severe physical abuse are more likely to endorse the use of harsh punishment than those with a mild history of abuse (Pears & Capaldi, 2001). In short, positive relationships have been documented between childhood exposure to abuse/neglect and later propensities towards less optimal parenting practices.

Unfortunately, evidence supporting the intergenerational transmission of abuse is limited (Ertem, Leventhal, & Dobbs, 2000). For example, it has been found that 90% of abused mothers do not go on to abuse their own children within the first year of life (Dixon, Hamilton-Giachritsis et al., 2005). However, in this study only 135 of the 4351 participants reported experiencing an abusive childhood. The large difference in group size, and the relatively small number of individuals in the abused group, may have led to confounds in the analyses. Additionally, the possibility still remains that these mothers may abuse their children at a later date (Belsky, 1993). Inconsistent findings regarding the intergenerational cycle of child abuse may also be due to differences in study design and methodology (Pears & Capaldi, 2001). In particular, Langeland and Dijkstra's (1995) review of the literature found that rates of transmission of intergenerational child abuse varied by the researchers' definition of abuse, whether the individuals were abused by one or both parents, and whether or not parents had access to supportive relationships. These authors reported wide variations in rates of intergenerational transmission of abuse (7-56% in retrospective studies, and 3-70% in prospective studies). Overall, the literature supports the assertion that parents who were maltreated in childhood are at increased risk for abusing their own children. This underlies the importance of examining historical factors that influence current parenting behavior.

Theoretical frameworks

Although there is no clear evidence regarding the mechanisms involved in the intergenerational transmission of child abuse (Ertem et al., 2000), it may be that mothers who have been abused have a lower threshold for reacting to their children's misbehavior, which in turn, leads to the use of harsh discipline. This reactive propensity creates a situation in which even minor infractions are likely to set off a series of negative mother – child interactions. In reference to their conceptual model of at-risk parenting (see Whitman, Brokowski, Keogh, & Weed, 2001 see Whitman, Brokowski, Keogh, & Weed, 2001), Borkowski and colleagues (2007) mentioned that in order to parent effectively, at-risk (e.g., teen) mothers must be cognitively prepared to raise and properly discipline their children. Of most importance, cognitive readiness has been shown to be related to a variety of maternal factors; specifically a history of exposure to abuse and neglect (Whitman et al.). Therefore, it may also be the case that mothers who engage in abuse are likely to have access to fewer positive disciplinary strategies, thereby increasing their propensity to use harsh punishment (Zaidi, Knutson, & Mehm, 1989). Pears and Capaldi (2001) suggested that parents with poor discipline skills experience a great deal of stress and frustration in dealing with their children. When this stress is coupled with a history of physical abuse, the outcome is more likely to result in the transmission of abuse from one generation to the next (Pears & Capaldi).

Furthermore, it is less likely that children raised in abusive families are provided sufficient opportunities to observe models of consistent and fair parenting (Pears & Capaldi, 2001). Therefore, Bandura's (1973) modeling paradigm of social learning may also help clarify this “cycle of abuse” by suggesting that abusive and neglectful parenting are learned behaviors passed on from parent to child. From this perspective, abused children learn that harsh parenting is a successful method of getting one's needs met (e.g., obedience and release of frustration), which later translates into how they react in current and future negative interactions, particularly negative interactions between themselves and their own children.

Group Differences in Abusive Parenting

Age, education level, and financial variations may further help to clarify differences documented between abusive and non-abusive mothers. Extant research has highlighted the increased propensity of teen mothers to engage in abusive acts towards their children (e.g., Shaw, Gilliom, Ingoldsby, & Nagin, 2003; Whitman et al., 2001; Borkowski et al., 2007). More specifically, younger mothers and those with lower education levels have been shown to be more likely to reject their children (Shaw et al.) and were also more likely to engage in negative parenting practices than older mothers and/or those with higher education (Bernardi et al., 1992). Whitman and colleagues (2001) also reported that when interacting with their children, adolescent mothers as opposed to adult mothers, were more likely to be intolerant, impatient, insensitive, and engage in more punitive parenting practices and behavior. The authors suggest that a lack of sufficient parenting knowledge and cognitive readiness to parent may help to explain the parenting differences between teen and adult mothers. In line with this contention, it is important to acknowledge that teen mothers are more likely to live in poverty and have lower educational attainment than adult mothers (Whitman et al.). Poverty, financial status, and/or educational level may also serve as an important function in explaining parenting differences.

Although the mechanism by which poverty impacts parenting is still unclear, studies have shown a correlation between income, years of education, and parenting behavior (e.g., Linver, Brooks-Gunn, & Kohen, 2002; Borkowski et al., 2007). Child Trends (2008) reported that abuse and neglect are more common in poor families than in families with higher incomes. In their study of 755 preschool children (mean age = 3.5 years), Lohman, Pittman, Coley, and Chase-Lansdale (2004) found that impoverished parents (i.e., those with low income and receiving Medicaid or food stamps), were less responsive, not as firm, and inconsistent in their interactions with their children. Additional studies have found that impoverished mothers, and/or mothers with lower educational attainment, demonstrate reduced levels of involvement, higher levels of punitive parenting (Bradley, Corwyn, McAdoo, & Garcia Coll, 2001), and less warmth toward their children than those from less impoverished backgrounds (Lohman et al.).

Although there are higher incidents of child abuse among high-risk groups (i.e, teen mothers, those living in poverty, or those with lower educational attainment) mentioned above, this behavior is not relegated exclusively to impoverished or young parents. Nevertheless, it is likely that there are differences in maternal history of abuse and abusive parenting practices between impoverished or low resource adult, teen, and high resource adult parents. Specifically, studies have found that families with greater risk factors and less access to external resources and/or sources of support were more likely to engage in child abuse as compared with families with fewer risk factors (e.g., Borkowski et al., 2007). Since the risk factors associated with poverty and teen parenting generally outweigh difficulties faced by middle and upper income families (Kohl, Edelson, English, & Barth, 2005), a cumulative stress model would suggest that high resource adult mothers who engage in abusive parenting behaviors may have been exposed to comparable stressors with similar impacts on parenting as those experienced by low resource and teen mothers throughout their development. Abusive high resource mothers may have been exposed to abusive parenting themselves, and/or other traumatic events that many low resource individuals experience throughout their lifetimes such as domestic and community violence.

Aims and Objectives

The current study was retrospective and cross-sectional and examined the intergenerational transmission of child abuse utilizing data collected in the Parenting for the First Time Project (Centers for Prevention of Child Neglect, 2002). Based on Bandura's (1973) modeling paradigm, the aim of this study was to examine how maternal histories of physical, emotional, and sexual abuse are associated with parenting knowledge and behavior. Data were obtained during laboratory interviews six months after the birth of the mother's first child. The following research questions were addressed: 1) What is the incidence of exposure to childhood abuse among a sample of first-time teen, adult low resource, and adult high resource mothers? 2) Does having a history of emotional, physical, or sexual abuse affect maternal parenting knowledge and behavior differently for the different types of first-time mothers. Analyses were also conducted by combining (i.e., pooling over) subjects, controlling for group membership. These analyses were conducted to determine whether a history of abuse was related to current parenting, specifically among a representative sample of first-time mothers.

Methods

Participants

Participants were drawn from the Parenting for the First Time Project (Centers for the Prevention of Child Neglect, 2002), a larger longitudinal prospective study designed to understand sub-threshold levels of neglectful parenting among a population-based, representative sample of first-time teen mothers and an ethnically matched sample of first-time adult mothers and their children. Mothers were asked to participate in the project if the pregnancy would be their first live birth and if they met age and education requirements (see below). Based on these criteria, 681 mothers were recruited during their pregnancies through primary care facilities in four communities that varied in size and ethnic composition including: South Bend, Indiana; Kansas City, Kansas and Missouri; Washington, D.C.; and Birmingham, Alabama.

Studies by the U.S. Census Bureau (2002) have consistently shown that individuals with higher levels of education have higher income levels and more access to external sources of support (i.e., financial and social) than individuals with less education. With the understanding that education level relates to access to external sources of support, mothers were recruited into three groups: teens (less than 19 years of age at the time of the child's birth, n = 396), adult low resource (older than 21 with less than 2 years of college, n = 169), and adult high resource (older than 21 and more than 2 years of college, n = 117).

The overall sample of mothers ranged in age from 14 to 36 years of age at the time of the child's birth, with a mean age of 19.8 years. The average age of the teen mothers was 17.5 years, with adult low-resource mother's average age at 25.5 years, and adult high resource mothers averaging 36.3 years. The sample was 68.5% African American, 14.4% European American, 14.3% Hispanic/Latina, and 2.8% of other ethnicities. Nearly half of the children were male (46.5%). Sixty-three percent of the sample completed the 12^th grade, 11 percent completed community college or a vocational program, and 25 percent completed a 4 year degree or graduate school. For the total sample of first-time mothers, 17.6% were married, 58.7% were single-parents, and 22.6% were in a romantic relationship with someone other than their child's father. Each group had mothers with monthly family incomes in the lowest range (i.e., less than $415 per month), however the upper limit of family incomes were different between groups with the highest monthly family income reported for adult high resource mothers ($16,666+), followed by teen ($8,336 - $12,500) and adult low resource ($4,166 - $5,000) mothers.

Measures

History of abuse

At 6-months postpartum, mothers were administered the Childhood Trauma Questionnaire (CTQ; Bernstein et al. 1994), a retrospective measure of childhood abuse and neglect. It consists of 70 questions that begin with the phrase, “When I was growing up…” and then end with a query about different types of abuse and neglect. Each item is rated for frequency on a 5-point Likert scale with responses ranging from, ‘never true’ to ‘very often true’, with higher scores indicating an abusive history. The instrument yields five different factors: physical and emotional abuse, emotional neglect, sexual abuse, and physical neglect. For the current investigation, the physical, emotional, and sexual abuse subscales were used. The instrument has been found by Bernstein and colleagues (1994) to have a high test-retest reliability over a period of months (intraclass correlation = .88). In addition, good convergent validity was shown through a high correlation with an alternate measure of childhood trauma (Childhood Trauma Interview; Draijer & Langeland, 1999) and non-significant correlations with measures of vocabulary and social desirability. For the current study, the CTQ was shown to have good internal consistency with a total test coefficient alpha of .88. This measure is a valuable research tool as it is quick and easy to administer, highly reliable and valid, and allows for distinctions between different types of abuse and neglect (Bernstein et al.).

Parenting knowledge

Maternal knowledge of infant development was assessed using the Knowledge of Infant Development Inventory – Short Form (KIDI; MacPhee, 1981). Using a modified version of the original 75-item measure, mothers indicated their degree of agreement on a 5-point scale for 14 items, with low scores indicating poor knowledge of developmental milestones. Example items include: “All infants need the same amount of sleep” and, “A good way to train children not to hit is to hit them.” This measure has shown high test-retest reliability (α = .92) and good internal consistencies (α = .82) (MacPhee). The internal consistency of the KIDI was adequate in the current sample (α = .60)

Parenting behavior

Maternal child abuse potential was measured with an abbreviated version of the Child Abuse Potential Inventory (CAPI; Milner, 1986). This 25-item measure assesses parental rigidity, such as attitudes about children's behavior, and unhappiness. Mothers responded either “agree” or “disagree” to statements such as “People expect too much of me” and “Children should never cause trouble.” Higher scores (agreeing with the statements) indicate a higher likelihood of abuse. The CAPI has a test-retest reliability index of .90 (Milner). For the current study, the CAPI was shown to have good internal consistency with a total test coefficient alpha of .77 at 6 months.

Mothers' style and philosophy of parenting were assessed using the Parenting Style Expectations Questionnaire (PS; Bavolek, 1984). The 27-item questionnaire required mothers to indicate their degree of agreement with statement that reflected their empathetic awareness, attitudes toward the use of physical punishment, abuse/neglect, and authoritarianism. The total measure was found to have internal consistency of .89 with a test-retest reliability index of .87 (Bavolek). For the current investigation, four subscales were used: responsivity/empathy (reliability α = .83), punishment (reliability α = .86), abuse/neglect (reliability α = .64), and authoritarian parenting (reliability α = .52). For the responsivity/empathy subscale, higher scores are associated with positive parenting opinions; however, for the other three subscales higher scores represent less positive parenting opinions. For congruency in interpretations of the results, the punishment, abuse/neglect, and authoritarian subscales were reversed scored to coincide with the responsivity/empathy subscale.

Analysis

Data analyses were conducted as follows: First, correlation analyses were conducted to determine the interrelationships between childhood exposure to emotional, physical, and sexual abuse, 6-month knowledge of infant development, and parenting behavior. Second, descriptive analyses were used to describe mean childhood emotional, physical, and sexual abuse scores separately for each type of first-time mother (i.e., teen, adult low resource, and adult high resource). Next, an analysis of variance (ANOVA) was conducted to determine whether there were mean differences between types of mother on their exposure to abuse. Lastly, multiple regression analyses were used to examine the predictive relationships between maternal history of emotional and physical abuse and parenting knowledge and behavior 6 months after the birth of their first child.

Results

Relationships Among History of Abuse, Parenting Knowledge, and Behavior

Interrcorrelations among all variables used in subsequent analyses are reported in Table 1. The upper portion of Table 1 demonstrates that there were significant relationships between maternal reports of childhood exposure to emotional, physical, and sexual abuse, with r values ranging from .27 to .56. The reporting of exposure to abuse in either one of these domains was highly associated with the reporting of exposure to abuse in other domains as well. The lower left portion of Table 1 reveals that exposure to childhood emotional and physical abuse were significantly related to maternal reports of their 6-month responsiveness to their children; r(681) = -.12, p < .05 and r(681) = -.16, p < .01, respectively. Increases in maternal report of childhood emotional and physical abuse were associated with decreased responsivity and empathy toward their 6-month old children. Childhood emotional and physical abuse were also significantly associated with maternal punishment and abuse/neglect scores with r values ranging from .12 to .17; as reports of abuse increased, maternal opinions towards punishment and abuse/neglect increased. Finally, exposure to emotional and physical abuse were significantly related to maternal reports of child abuse potential; r(681) = .15, p < .01 and r(681) = .18, p < .01, respectively. In summary, as maternal reports of exposure to abuse increased, reports of propensities towards abusing their children increased.

Table 1.

Correlations Matrix of Childhood Emotional, Physical, and Sexual Abuse, on Parenting Knowledge and Behaviors 6 months Postpartum (N = 681).

	1	2	3	4	5	6	7	8	9
1. CTQ Emotional Abusea	1
2. CTQ Physical Abusea	.56*	1
3. CTQ Sexual Abusea	.39**	.27**	1
4. Knowledge of Infant Development	.06	-.02	.11	1
5. PS - Responsivity/Empathy Scoreb	-.12*	-.16**	.00	.40**	1
6. PS - Punishment Scoreb	.15**	.17**	.07	-.26**	-.46**	1
7. PS - Abuse/Neglect Scoreb	.15**	.14*	.01	-.36**	-.52**	.64**	1
8. PS - Authoritarian Scoreb	-.05	-.06	.02	-.47**	-.69**	.50**	.47**	1
9. CAPI Total Scorec	.15**	.18**	-.06	-.40**	-.48**	.47**	.43**	.54**	1

Note.

^**p < .01,

^*p < .05;

^a= Childhood Trauma Questionnaire;

^b= Parenting Style Questionnaire;

^c= Child Abuse Potential Inventory.

The right portion of Table 1 presents associations between parenting knowledge and behaviors 6 months after the birth of the mothers' first child. As knowledge increased, maternal responsiveness increased (r(681) = .40, p < .01), and opinions and behaviors regarding punishment (r(681) = -.26, p < .01), abuse/neglect (r(681) = -.36, p < .01), authoritarian parenting (r(681) = -.47, p < .01), and abuse potential (r(681) = -40, p < .01) decreased. Overall, Table 1 reveals significant associations among the predictor variables themselves and parenting knowledge and behaviors. Childhood exposure to sexual abuse was not significantly related to any outcome variable, and therefore, was not tested as a predictor in subsequent regression analyses.

Childhood Exposure to Emotional, Physical, and Sexual Abuse Among First-time Mothers

Our next research goal was to describe mean childhood abuse scores for teen, adult low resource, and adult high resource mothers. Within the sample of teen mothers, 186 (47.0 %) reported experiencing some level of physical abuse (M = 7.88, SD = 3.82), and 189 (47.7 %) reported experiencing emotional (M = 8.66, SD = 4.14) and sexual abuse (M = 6.37, SD = 3.96). For the low resource sample, 81 (48.0 %) mothers reported experiencing physical (M = 7.00, SD = 2.30) and emotional abuse (M = 7.92, SD = 3.86), and 77 (45.6 %) reported experiencing sexual abuse (M = 6.91, SD = 4.73). Within the high resource sample, 96 (82.1 %) reported experiencing physical abuse (M = 6.63, SD = 2.12), 95 (81.2 %) reported experiencing emotional abuse (M = 6.76, SD = 2.24), and 94 (80.3 %) reported experiencing sexual abuse (M = 5.90, SD = 2.21). Teen mothers, as opposed to adult low and high resource mothers, had higher mean scores for exposure to childhood emotional and physical abuse. Adult low resource mothers had higher means than teen and adult high resource mothers for exposure to childhood sexual abuse. Overall, although adult high resource mothers had a higher frequency of exposure to abuse, they reported lower mean scores on each abuse outcome than both teen and adult low resource mothers.

Analyses of variance (ANOVA) found that teen, adult low resource, and adult high resource mothers differed significantly in their reported exposure to emotional abuse, F(2, 679) = 8.20, p < .001; post hoc contrasts using the Bonferroni procedure found that mean emotional abuse scores for teen mothers was significantly different from those of adult high resource mothers. In addition, a significant difference was documented for mean physical abuse scores as a function of type of mother, F(2, 679) = 5.59, p < .01; post hoc contrast found that mean physcial abuse scores for teen mothers was significantly different from those of adult high resource mothers. Mothers did not differ in their mean reports of exposure to childhood sexual abuse.

Childhood Exposure to Emotional Abuse and its Impact on Parenting Knowledge, and Behaviors 6-months Postpartum

Using simple regression, our main research objective was to examine the effects of childhood exposure to abuse on parenting knowledge and behavior. Table 2 presents the results of regressing 6-month parenting outcomes on childhood exposure to emotional abuse. Emotional abuse did not separately predict 6-month parenting outcomes for the different classifications of first-time mothers. However, analyses conducted for the total sample of first-time mothers (controlling for type of mother) found that childhood exposure to emotional abuse accounted for a significant proportion of the variance found among their responsivity/empathy, punishment, and abuse neglect scores. Emotional abuse also accounted for a significant proportion of the variance found among the total sample's child abuse potential (CAPI) scores. As exposure to emotional abuse increased, maternal responsivity decreased, whereas opinions towards, and propensities for, abusive behavior increased.

Table 2.

Results of Regressing 6-month Parenting Knowledge and Behavior on Childhood Exposure to Emotional Abuse (N = 681).

	Teen (n = 396)		Adult Low Resource (n = 169)		Adult High Resource (n = 117)		Total (n = 681)
	R2	β	R2	β	R2	β	R2	β
Knowledge of Infant Development	.04	.19	.00	.03	.04	.19	.01	.06
PS - Responsivity/Empathy Score	.01	-.11	.01	.11	.00	-.04	.12	-.12*
PS - Punishment Scorea	.01	-.11	.02	-.13	.01	-.11	.05	.15**
PS - Abuse/Neglect Scorea	.02	-.13	.00	-.04	.01	.10	.05	.15**
PS - Authoritarian Scorea	.00	.04	.00	.06	.00	-.02	.13	-.05
CAPI Total Scoreb	.02	.13	.00	.02	.10	.10	.12	.15**

Note.

^**p < .01, p < .05;

^a= Parenting Style Questionnaire;

^b= Child Abuse Potential Inventory.

Childhood Exposure to Physical Abuse and its Impact on Parenting Knowledge, and Behaviors 6-months Postpartum

Table 3 presents the results of regressing 6-month parenting outcomes on childhood exposure to physical abuse. Physical abuse did not predict 6-month parenting outcomes for teen and adult low resource mothers. However, childhood exposure to physical abuse predicted 6-month parenting outcomes for adult high resource mothers. More specifically, physical abuse accounted for a significant proportion of the variance found among adult high resource mother's responsivity/empathy scores, punishment scores, and moderately predicted authoritarian parenting scores. Physical abuse also accounted for a significant proportion of the variance found among adult high resource mother's propensities to engage in abusive behavior. As exposure to physical abuse increased, maternal responsivity decreased; whereas propensities for abusive behavior increased.

Table 3.

Results of Regressing 6-month Parenting Knowledge and Behavior on Childhood Exposure to Physical Abuse on (N = 681).

	Teen (n = 396)		Adult Low Resource (n = 169)		Adult High Resource (n = 117)		Total (n = 681)
	R2	β	R2	β	R2	β	R2	β
Knowledge of Infant Development	.00	.03	.04	.19	.02	.12	.08	.04
PS - Responsivity/Empathy Score	.00	-.07	.00	-.06	.06	-.25*	.14	-.16**
PS - Punishment Score	.01	-.11	.02	-.12	.07	.26*	.06	.14**
PS - Abuse/Neglect Score	.01	-.09	.00	.04	.03	-.16	.12	.14*
PS - Authoritarian Score	.00	.04	.00	.07	.04	.21	.13	.13*
CAPI Total Score	.00	.07	.04	.19	.17	.41**	.12	-.06

Note.

^**p < .01, p < .05;

^a= Parenting Style Questionnaire;

^b= Child Abuse Potential Inventory.

For the total sample of first-time mothers, childhood exposure to physical abuse accounted for a significant proportion of the variance found among their responsivity/empathy, punishment, and abuse neglect scores. Physical abuse also accounted for a significant proportion of the variance found among the total sample's authoritarian parenting scores. As exposure to physical abuse increased, maternal responsivity decreased, whereas propensities for abusive behavior increased.

Discussion

The current study explored the relationship between past exposure to emotional, physical, and sexual abuse and its impact on later parenting knowledge and a propensity for abusive behavior. The findings of the present study are in accord with previous studies documenting the impact of childhood experiences of abuse on current parenting behavior (i.e., Dixon, Brown et al., 2005; Markowitz, 2001; Narang & Contreras, 2004; Pears & Capaldi, 2001), suggesting that individuals with a history of abuse are at increased risk for maltreating their children. Correlation analyses found that exposure to both childhood emotional and physical abuse were significantly associated with 6-month parenting opinions and a propensity for abusive behavior, as opposed to parenting knowledge. Results of the current investigation suggest that the type of childhood environments mothers' experience impacts their future parenting behavior but has less of an impact on their parenting knowledge.

Exposure to sexual abuse was not significantly related to parenting outcomes; therefore, it was not tested as a predictor in analyses. Investigations of abuse have suggested that specific forms of abuse may be related to specific maladaptive, emotional, and behavioral outcomes in adulthood (Lindsay, Law, Quinn, Smart, & Smith, 2001). Physical abuse has been consistently shown to be linked to aggression towards others (e.g., Dixon, Hamilton-Giachritsis et al., 2005), while exposure to sexual abuse has been shown to relate to maladaptive sexual behavior in adulthood (Lindsay et al., 2001). Gore-Felton, Koopman, and Hernandez (2001) found that exposure to emotional abuse was the only form of abuse that was significantly associated with emotional problems. Taken together, extant research supports the present study's finding that sexual abuse was not related to any of the outcomes measures since exposure to sexual abuse has been linked to sexual behavior in adulthood but not parenting knowledge and discipline practices.

Childhood Exposure to Emotional, Physical, and Sexual Abuse among First-time Mothers

The current investigation is the first of its kind to concurrently examine the frequency of exposure to specific types (i.e., physical, emotional, and sexual) of abuse in childhood versus collapsing over categories. Rates of exposure to childhood abuse for adult high resource mothers (∼80%), as opposed to teen and adult low resource mothers (∼47%), were consistent with those of Narang and Contreras (2005) who found an 85% prevalence of physical abuse during childhood. Of particular interest was the finding that teen mothers, as opposed to adult high resource mothers, had significantly higher mean scores for exposure to childhood emotional and physical abuse. Studies have found that exposure to abuse in childhood is associated with teen pregnancy (Hillis et al., 2007). Additionally, individuals living in poverty are exposed to a greater number of risk factors relative to those who live above the poverty line. Exposure to numerous risk factors is in turn associated with more punitive parenting (Olson, Ceballo, & Park, 2002). Teen mothers are in a particularly high risk category relative to older mothers because they are more likely to come from a higher poverty background (Moore & Brooks-Gunn, 2002), and have the added stresses of balancing parenthood with adolescence (Borkowksi et al., 2007). It is therefore not surprising that teen parents reported increased exposure to abuse relative to their adult counterparts.

Exposure to Childhood Abuse and its Expression in Later Parenting

A general goal of the present study was to demonstrate that the parenting behaviors of teens and adult low resource mothers would be more negatively impacted by past exposure to abuse than would adult high resource mothers because of poor cognitive readiness to parent and less access to external sources of support. Unexpectedly, although there were significant differences between the different types of mothers on their mean scores of abuse, emotional abuse did not differentially predict parenting outcomes separately for the different classifications of mothers. Therefore, type of mother was not a significant factor accounting for the intergenerational transmission of emotional abuse.

However, in line with extant research (e.g., Dixon, Brown et al., 2005), both emotional and physical abuse predicted parenting behavior after pooling over the different classifications of first-time mothers (i.e., combining the sample), suggesting a more direct link between past exposure to abuse and current parenting behaviors. Adding further evidence for the direct relationship between parental history of abuse and current child abuse, Pears and Capaldi (2001) failed to demonstrate indirect relationships for the transmission of abuse through parental socioeconomic status, antisocial behavior, psychopathology, consistency of discipline, and child behavior problems; however, they were able to document a direct path from past exposure to abuse and current parenting. In short, support was found for the possibility that childhood emotional and physical abuse may serve as a predictor of parenting outcomes for the total sample of first-time mothers as opposed to teen, adult low resource, and adult high resource mothers separately. Specifically for exposure to emotional abuse, combining the sample resulted in a larger sample size, which may have increased the power to predict parenting outcomes (O'Rourke, 2003).

Another interesting finding was that unlike emotional abuse, childhood physical abuse influenced parenting behaviors for adult high resource mothers. In addition, findings suggest that as exposure to physical abuse increased, maternal responsivity decreased and opinions towards, and propensities for abusive behavior increased. It appears that for adult high resource mothers, childhood exposure to emotional abuse has less of an impact on parenting behaviors than childhood exposure to physical abuse. Bandura's (1973) modeling paradigm may help to explain this occurrence by suggesting that observable behaviors have more of an impact on an individual's perception of themselves through the act of imitation than does verbal learning.

A major strength of this study was our ability to measure the impact of past exposure to specific types of abuse and its impact on later parenting knowledge and practices in a diverse sample of first-time mothers. This study allowed us to discern that exposure to specific types of abuse (i.e., emotional and physical) may play a major role in explaining parenting differences documented in the extant literature for teen and adult mothers (e.g., Whitman et al., 2001), thereby serving as a stepping stone for future research comparing the impact of abuse among these samples. Potential research efforts could focus on determining whether there are cultural (historical) differences between the samples that lead teens and low resource adult mothers to interpret harsh physical punishment as reflective of caregiver attention and concern compared to the same behaviors considered non-normative and abusive by high resource mothers.

Study Limitations

The present interpretations need to be made in light of three limitations. First, maternal report was used as the sole source of gaining insight to exposure to abuse and current parenting beliefs and behavior. The use of multiple informants may have provided more accurate information regarding both actual events that occurred in these mothers' lives, and their current level of functioning. Future studies examining a broad range of factors may serve to further elucidate the underlying causal mechanisms of parenting behavior, such as direct observations of the quality of the home environment and mother-child interactions. Obtaining more objective measures of maternal history of abuse, possibly through police records, would also provide valuable information. Objective measures of negative parenting behavior, such as involvement with Child Protective Services, number of trips to the hospital, or clinical assessments of potential for abuse may also provide useful information with respect to the association between maternal history of abuse and present parenting behavior.

Second, the use of a retrospective measure of exposure to abuse may have been contaminated by individual confounding factors such as age, memory capacity, and/or intelligence. A common criticism of retrospective methodology is that high levels of accuracy may be lacking. In fact, the current investigation found that although adult high resource mothers had lower means scores for their exposure to abuse, their frequencies of reported abuse (∼ 80% versus ∼ 47%) were substantially higher than that of teen and adult low resources mothers. Despite its critics, retrospective reporting has been utilized extensively in studies examining the effects that exposure to traumatic and negative life events has on current adjustment (i.e., Dixon, Brown et al., 2005). These studies have found that reporting of case histories were reliable predictors of current adjustment and behavior, demonstrating that retrospective methods are essential and beneficial when studying certain sensitive topics.

Third, although the current investigation accounted for age and education differences in exposure to abuse between first-time teen and adult mothers, it failed to account for and/or test for the effects of other possible factors (e.g., race, gender, culture, psychopathology, and child's temperament) found in the extant literature to be related to abuse and parenting. In particular, ethnicity and/or race have been shown to be related to self-reported abuse. Meston, Heiman, Trapnell, and Carlin (1999) found that Asians reported higher levels of physical and emotional abuse than did their European counterparts. Child Trends (2008) reported that non-Hispanic Black, American Indian/Alaskan Native, and Pacific Islander children have higher rates of reported child maltreatment than do other children. Consequently, further investigations among first-time mothers that account for the influence of such factors as culture and race on child-rearing and disciplinary practices is warranted.

Conclusions and Implications

Despite the above mentioned limitations, the findings of this study provide implications for clinical practice and intervention and prevention efforts aimed at improving the parenting skills and behaviors of first-time mothers, particularly for teen and adult high resource mothers who have experienced childhood abuse. Programs and policies should focus on parents' mental health and their ability to discipline their children in appropriate manners. Preventing problems before they start is generally thought to be more cost-effective and less time-consuming than waiting to intervene after a problem has already developed (Cummings, Davies, & Campbell, 2000). Therefore, it becomes particularly important for community health nurses and family practitioners, who are in frequent contact with mothers prior to the birth of the children, to offer parent coaching/training, which have the potential to improve parenting behaviors and enhance developmental outcomes in children (Cowan & Cowan, 2002), specifically to those mothers who have been, and/or are, currently exposed to abuse.

The delivery of parent-training has evolved over the years, with the primary aim of increasing the cost-effectiveness of services for the largest number of “at-risk” families (Nixon, 2002). As an alternative to face-to-face interventions, parenting booklets, coupled with counseling services, may be useful and convenient methods to provide parenting education. The National Institute of Child Health and Human Development (NICHD) developed an informational booklet, Adventures in Parenting How Responding, Preventing, Monitoring, Mentoring, and Modeling Can Help You be a Successful Parent (NICHD, 2001), designed to educate parents on the basic principles. The specific parenting skills discussed and taught in the Adventures in Parenting booklet are particularly relevant to the findings of the current investigation which found tentative support of a possible link between past exposure to abuse and poor parenting responsiveness and discipline practices among teen and adult high resource mothers. Findings suggest that these mothers lack adequate parenting skills related to responding to their children's needs in an appropriate manner and ways to address inappropriate behavior. Upon completion of a parent-training program, such as the self-administered program created by NICHD, teen and adult high resource mothers who have been exposed to childhood abuse could become more effective, consistent, active, and attentive parents (NICHD).

Regardless of the specific focus and design of a prevention program, its ultimate goal is to bring about improvements in the developmental trajectories of a group thought to be in danger of poor developmental outcomes. In this regard, parent-training programs can be thought of as prevention programs designed to teach optimal parenting attitudes, beliefs, and behaviors, which in combinations are likely to bring about optimal child development. The preventive and intervention efforts of family counselors and researchers should focus on enhancing positive parenting skills and discipline strategies, improving mother's mental health by offering referrals and counseling services such as anger management, and assist families in forming positive relationships (i.e., support networks) as a way of expanding access to services and information; thereby counteracting the effects of childhood exposure to abuse on current parenting. Furthermore, the early identification of first-time mothers who have been exposed to abuse and the on-going development of individualized intervention and effective prevention strategies and treatments can have highly significant public health implications.