To examine the association between father involvement and child neglect.
Participants were recruited from an inner-city pediatric primary care clinic and a clinic for children at risk for human immunodeficiency virus infection in a teaching hospital.
Mothers and fathers or father figures, and 244 five-year olds participating in a longitudinal study.
Main Outcome Measures
Child neglect measured via home observation, a videotaped mother-child interaction, and child protective services reports.
A father or father figure was identified for 72% of the children. Rates of neglect ranged between 11% and 30%. Father absence alone was not associated with neglect. However, in families with an identified and interviewed father, a longer duration of involvement (P<.01), a greater sense of parenting efficacy (P<.01), more involvement with household tasks (P<.05), and less involvement with child care (P<.05) were associated with less neglect. The overall model explained 26.5% of the variance in neglect.
There is substantial involvement of fathers in a subset of this high-risk sample, although more than a quarter of the children lacked a father or father figure. The mere presence of a father did not significantly influence the neglect of the children; rather, the nature of his involvement did. Fathers who felt more effective as parents were less likely to have neglected their children. A greater sense of efficacy may reflect parenting skills and be important in enhancing the contribution of fathers to their children's well-being. Pediatric health care providers can play a valuable role in enhancing the involvement and skills of fathers.
THE NEW morbidity" in the 1970s1 and "Bright Futures" in the 1990s2 have drawn health care providers' attention to psychosocial problems confronting many families. However, there has been little attention to the specific roles of family members, in particular fathers, in the pediatric literature. To our knowledge, the issue of child neglect, a major problem for many families, has also been ignored.3
Concern has been expressed about the roles fathers play, or do not play, in their children's lives.4 There are no clear expectations of what fathers' roles should be, and many children live in fatherless homes. In 1987, 16% of all white children and 50% of African American children in the United States lived with single mothers.5 At the same time, many fathers have taken increasingly active roles in their children's lives,6 including many who do not live with their children.7 We must consider the roles of the nonbiological fathers who may act like fathers through their involvement with a child.
The involvement of African American fathers in their families has received increasing attention. Many African American teenaged fathers are involved in their children's lives, even though the fathers are not married to the mothers and are not living in the home.8 Nevertheless, the breadwinner role is an important function of all fathers, and high rates of joblessness among low-income fathers may strain their role as providers.9 Additionally, McAdoo10 takes issue with stereotypes of African American fathers based on studies of those at the lowest economic level. He suggests there is a need for more observational studies of fathers and children. In spite of increasing study, the role of African American men in families ". . . is one of the most conspicuously neglected areas of family research."11
Studies on the effects of a father's presence or absence have yielded mixed results; some have not found the father's presence to affect children.7,12 In general, however, the father's involvement does influence the lives and development of their children,13 although much of the research in this area has been done on middle-class white families.14 Paternal involvement enhances the development of preschool-aged 15- 17 and school-aged children,18 and children in single-parent families are more likely to grow up in poverty, less likely to finish high school, and more likely to have children as teenagers.19 Black et al20 found that while presence of the father was not associated with a child's behavior and development, the nature of the father's involvement was. Fathers who were satisfied with parenting, who contributed financially to the family, and who were nurturing during play had children with better cognitive and language development.
Lamb6 has described ways in which fathers may influence their children's lives. First, they are usually the primary breadwinner. Second, fathers interact directly by caring for, teaching, and playing with their children. Third, fathers provide emotional and material support to mothers, which affects children indirectly.21 These relationships are complex; a reciprocal relationship may also exist whereby paternal involvement is facilitated by a positive marital relationship. Yet, increased paternal involvement may not always be beneficial.22 For example, a child's emotional adjustment may improve when contact with an abusive or neglectful father is reduced or eliminated.23
Lamb6 has also called attention to the context of paternal involvement. A father's motivation to be involved, his evaluation of his parenting, and the mother's perception of his involvement may all affect a child's well-being. If father involvement can enhance children's health and well-being, it may also protect them from an important form of child maltreatment—neglect.
Neglect is usually defined as a significant omission in care by a parent or caregiver. It is the most prevalent form of child maltreatment, accounting for approximately two thirds of identified maltreatment cases. For example, almost 2 million children, or 3% of children in the United States, were estimated to have been neglected in 1993.24 Neglect has serious consequences. 25 It is associated with dysfunctional interpersonal relationships,25 speech and language disorders,26,27 academic problems,28 and mental health problems.29 In addition, approximately half of the estimated 2000 annual fatalities from child maltreatment are due to neglect.30
In the field of child abuse and neglect, very little research has been conducted on fathers, and the report on the status of child maltreatment research by the National Research Council recommended that this gap be addressed.31 This lack of research is particularly striking, given the acceptance of the ecological model in the conceptualization of abuse and neglect.32 Fathers should not be ignored in analyses of the multiple, interacting factors contributing to child maltreatment.33 Involved fathers may protect their children against adverse circumstances.
To our knowledge, this is the first study examining the association of father involvement and child neglect. We have focused on low-income, urban, African American fathers and their 5-year-old children. We have used a broad definition of father to include "someone who is like a father to this child," regardless of their biological relationship, in recognition of the important roles father figures play.34 This study expands on the issue of child neglect, and on our earlier work on fathers' influence on children's development, behavior, and home environment.20 In this study, we tested the hypothesis that greater involvement of fathers (living in the home, contributing financially, nurturing, and participating in household tasks and child care) protects children from neglect, with less involvement being associated with neglect. We also examined if the fathers' contributions were significant after taking the mothers' parenting into account.
The sample of 244 five-year-old African American children and their families were participants in a longitudinal study of child health and development, among families at risk for maltreatment. The families were recruited from pediatric clinics when the children were younger than 2 years on the basis of nonorganic failure to thrive, risk for human immunodeficiency virus infection, or low income without other specific risks.
Mothers consented to participate following institutional review board–approved procedures. When the children were 5 years old, mothers were asked about their child's contact with their biological father. If contact occurred at least monthly, permission was requested to interview the father. If he was not involved, mothers were asked if the child had "someone like a father" and the same procedure was followed. Unless otherwise stated, "father" refers to both biological fathers and father figures.
After informed consent was obtained, fathers were interviewed in the laboratory. The interview assessed health, satisfaction with parenting, involvement with their child, relationship with the child's mother, perception of the child and the neighborhood, and supports and stresses. Fathers were also videotaped playing with their child. The men were paid $25 for participation.
A similar maternal interview was usually conducted on the same day. There was also a 40-minute home visit, usually involving only mother and child. In 1997, computerized child protective services (CPS) records were searched for any past relevant record data concerning maltreatment and were coded by a trained abstractor unaware of other aspects of the study.
Four measures were used to assess paternal roles: (1) demography (whether the father or father figure lived with the child), (2) relationship to the child, (3) employment, and (4) contribution to child support.
A modified version of the "Who Does What"35,36 scale was used; the 24 items yielded 2 scales measuring involvement in child care and household tasks, with Cronbach αs of .92 and .77, respectively. Respondents rated each item on a 9-point scale ranging from "I do it all" (1), to "shared" (5), to "partner does it all" (9). Scores were averaged to obtain an overall score.
Paternal nurturing was coded from videotapes of fathers playing with their children. The videotapes were coded by research assistants who were unaware of the children's risk status. A schema based on parenting style37 was used to rate the interaction of parent and child.38 Individual items were coded using behaviorally anchored, ordinal scores. High scores represent the most positive interactions. The raters were trained until percent agreement exceeded 90%, and reliability was maintained through weekly reviews. Four factors were derived: warmth, engagement, structure, and negative emotion, each with a Cronbach α greater than .80. The first 3 were highly correlated and were averaged to form a construct of parent nurturing.
Fathers' views of their parenting were rated on the Parenting Sense of Competence Scale.39,40 This 16-item scale has 2 subscales—parenting satisfaction and parenting efficacy. Scores were summed and high scores represent greater satisfaction and sense of efficacy.
As recommended by the National Research Council,31 we used 6 measures of neglect, with data from 5 different sources. We developed a composite Neglect Index:
These scales measure child neglect.41,42 Research assistants used 14 scales that were completed after a home visit. Three dimensions were assessed: the home (furnishings, overcrowding, sanitation, utilities, and safety), observations of the child (clothing and hygiene), and maternal behavior toward the child (supervision, child care, acceptance, approval, expectations, discipline, and stimulation). We considered a seriousness score of less than 60 (on a 0-100 scale) on any scale as inadequate care or neglect, except for safety, where the cutoff was 50 to adjust for the children's age. These scales (eg, utilities) include conditions experienced by the children without inferring parental responsibility, in keeping with a child-focused definition of neglect.43
This was also completed after the home visit.44 This 45-item scale covers 6 aspects of the home environment relevant to the well-being of young children (eg, emotional and verbal responsivity of mother). Interrater reliability, assessed by having 2 research assistants jointly code 25% of the visits, was greater than .90. The lowest quartile (25%) on the overall score in our sample reflected the least adequate home environments and was construed as a form of neglect.
The Maryland central computer files for CPS reports on the study children were searched in 1997. Neglect was determined from abstracted records for each child between the ages of 4 and 6 years who was the subject of a CPS report, whether substantiated or not. Eight families with a report for abuse only were excluded; those who had been abused and neglected were included.
The videotapes of the mothers playing with their 5-year-old children were coded in the same way as for the fathers. The average score of the same 3 scales was used, with the lowest quartile representing neglect.
This factor was measured by a maternal questionnaire regarding aspects of the neighborhood, particularly safety, drug use, and violence. A factor analysis using maximum likelihood extraction with oblimin rotation yielded an 11-item scale with adequate internal consistency (Cronbach α = .92). This measure was used as a continuous variable with low scores reflecting a less adequate environment.
Finally, a 3-point composite Neglect Index was developed based on the the Child Well-Being Scales, Home Observation for Measurement of the Environment scale, CPS, and videotaped mother-child interaction. Dimensions were "probable neglect" (CPS and 2 other measures or no CPS but 3 other measures indicating neglect), "possible neglect" (CPS and either no or one other measure or no CPS but 2 other measures indicating neglect), and "unlikely neglect" (no CPS and either no or one other measure indicating neglect.) Environmental neglect was not included.
There were 3 categories of father involvement: father identified (by mother) and interviewed, no father identified, and father identified but not interviewed. Demographic characteristics of families by father availability were examined by χ2 and 1-way analyses of variance. The relationship between father availability and neglect was examined with factorial analyses of covariance for each measure of neglect, controlling for the number of adults in the home, the only variable that was significantly different by father availability.
Next, we focused on only those fathers who were interviewed (n = 117), examining the hypothesis that greater paternal involvement protected children from neglect. We used hierarchical multiple regression analyses and, after controlling for group status (failure to thrive, human immunodeficiency virus risk, and comparison), maternal age, maternal education, child's sex, the number of adults in the home, and the number of children in the home, the following variables were examined for their association with neglect: whether the child lived with the father, the father's involvement in years, father's marital status, father's education, whether the father was the biological father, father's employment status and financial contribution, paternal parenting effectiveness, paternal involvement in child care and household tasks, and paternal nurturing. Nonsignificant variables were removed and the model was rerun. We also examined the role of maternal parenting and its association with neglect to assess what the father added after controlling for maternal parenting, using the 2 maternal subscales of the Parenting Sense of Competence Scale. Of the 2 subscales, only parenting satisfaction was associated with neglect, so the above regression analyses were repeated controlling for it.
Of the 244 families, 176 fathers were identified by the mothers (72%) and 117 (66%) were interviewed. Demographic characteristics of the families are summarized in Table 1. Approximately half of the children were male and there were an average of almost 4 children in each household. The only significant demographic difference according to the availability of a father figure was that the families in which the father was interviewed had more adults in the household compared with the other 2 groups (no father identified or father identified but not interviewed). There were no significant differences in any of the neglect measures according to the presence of fathers (Table 2). In addition, 72% of the men had daily contact with the children. As expected, rates varied across the 6 measures. Associations of child, maternal, household, and demographic characteristics with child neglect were examined for families of interviewed fathers. Table 3 provides the findings using the Neglect Index. Child, maternal, and household characteristics did not differ.
Hierarchical multiple regression analyses predicting the Neglect Index revealed that after controlling for maternal education, age, and parenting satisfaction, child sex and group, and the number of adults in the home, several father involvement variables were associated with neglect. Neglect was less likely when the father was involved for a longer period (P<.01), expressed more parenting effectiveness (P<.05), and was more involved in household tasks (P<.05), but was less involved in child care (P<.05). The overall model explained 26.5% of the variance (Table 4). Paternal cohabitation with the child, marital status, education, financial contribution, and biological status were not associated with the Neglect Index. The other measures of neglect were not significantly associated with father involvement.
Fathers or father figures are actively involved in the lives of many of the high-risk 5-year-old children in this study. Among families in which a father was identified and interviewed, approximately two thirds lived with the child. Indeed, Furstenburg and Harris45 reported that while only 11% of their African American sample of children in Baltimore, Md, resided with a male adult throughout childhood, only 8% never lived with a father or father figure. However, in one fourth of the families, mothers did not identify either a father or father figure involved in the child's life.
The study also found relatively high rates of neglect occurring between 4 and 6 years of age—16% based on CPS records. This is substantially higher than the 2.9% rate based on community professionals' assessment.24 Given the limitations of any single neglect measure, the 4 measures combined in the Neglect Index may be a useful indicator. Using this index, approximately one third of families had "probable" or "possible" neglect. Although caution is warranted in generalizing these findings, rates may be similar in comparable communities.
Is paternal involvement in children's lives associated with reduced neglect? In the broadest sense, the findings suggest that the father's role might be filled by another adult in the household. Neglect rates did not differ by father or father figure presence or absence. There were no group differences in number of adults in the household; mothers not living with a man were more likely to be living with another major support provider.46 Other research concludes that having a father at home enhances child development, but absence is not harmful if a grandmother is present.47
There were, however, significant findings within the group of families in which a father was identified and interviewed, presumably a select subgroup of the most involved fathers. Father involvement over an extended period was associated with less neglect, suggesting that a stronger relationship may have developed over time. The longer duration may also be a marker of family stability and consistency in caregivers. In addition, women may be less likely to remain involved with men who do not treat their children well. One might expect that the longer period reflects the involvement of biological fathers instead of father figures, but this relationship did not differ significantly, illustrating that nonbiological father figures often play stable roles in this population. Fathers' sense of efficacy regarding their parenting was associated with less neglect, suggesting that their rating reflected a more successful role in nurturing their child. The importance of fathers' positive view of their parenting roles in protecting against neglect supports the emphasis by Lamb6 on the context of paternal involvement. Aside from motivation, having relevant skills and confidence may promote involvement, enjoyment, and sensitivity. However, causality cannot be inferred; fathers may have felt more effective because neglect had not been a concern. In the current study, fathers' contribution was significant after controlling for mothers' satisfaction with parenting.
Greater father involvement in household tasks was also associated with lower neglect risk. Helping with tasks could ease mothers' workloads,48 and could also be a proxy for overall paternal family involvement, supporting other research indicating fathers contribute to family and child well-being.6 The association of greater father involvement in child care with more neglect was unanticipated. Perhaps fathers played a greater role when mothers were relatively unavailable, a conclusion supported by other research.49
This study had a power of 80% to detect significance in variables that accounted for 5% of the variance.50 It is interesting to consider the variables that were not significantly associated with neglect. Employed fathers' financial contributions were not a factor, perhaps because most of the families were living in poverty and the men made meager financial contributions. Even though more than half of the men worked and provided support, most families still received public assistance. Support by noncustodial, low-income fathers typically is irregular, informal,14 and less than $2000 a year.51 It seemed likely that having a biological father involved would be more protective than a father figure. However, neglect rates did not differ between these 2 groups. Furstenburg and Harris45 also found that the biological relationship of the father was unrelated to child outcomes. Similarly, having the father living in the home was not significant. This finding was also reported by Furstenburg and Harris,45 although children benefited if the father or father figure lived in the home and had a close relationship with the child. Similarly, when the children in this sample were 3 years of age, child behavior and development improved when fathers were more nurturing of their children.20 The present findings are consistent with those results. In low-income communities, many men may play important roles in their children's lives even if they do not live in the home.52 Both the quality of the relationship and fathers' involvement seem to be more important than the biological relationship of the father or where he resides.
The study has several limitations. First, the men interviewed may not be representative of all African American fathers or father figures. At the same time, there were few differences between families in which a father was interviewed and those in which he was identified but not interviewed. In addition, because the participants were from low-income, inner-city families, generalizability to other groups is limited. Second, the data are cross-sectional, precluding inferences of causality. We do not know whether greater involvement of fathers protected children from neglect or whether men were more likely to participate in better-functioning families. Third, the children were 5 years old at the time of this study, restricting the "opportunity" for maltreatment. Maltreatment could emerge as an even larger problem in future years and the influence of fathers could become more evident.
Finally, the measurement of neglect remains problematic. While direct observation seems promising, particularly for assessing physical neglect, it is unlikely that psychological neglect will occur during a brief observation. Child protective services data are limited by the biases of who gets identified, reported, and substantiated. The composite Neglect Index attempts to avoid these limitations.
The high rate of neglect in this population is alarming and warrants serious attention to address children's needs and prevent the sequelae related to neglect. Most of the children did have fathers or father figures in their lives and this study suggests an association between greater father involvement and a lower risk for neglect. The main question may be how to encourage fathers to be more involved with their children in ways that are optimally nurturing.
Major efforts have been made to ensure that fathers fulfill child support obligations.14 Because economic security supports marriage and father involvement, tax code reforms improving the financial benefits of marriage could aid families. Similar improvements could come from allowing fathers to remain in the homes of women receiving welfare. An array of programs to improve the educational, parenting, and job skills of young fathers have been developed, but few have been rigorously evaluated.14 Theoretically sound, promising interventions need to be evaluated. More modest interventions may also be effective. For example, it is striking how professionals and agencies serving children often focus on mothers and ignore fathers. Health care providers should actively invite fathers to participate in visits and nurture their involvement in their children's lives. In addition, men, especially young men, may not know how to nurture their children, especially if they were raised in fatherless households. We need more programs for fathers that build specific skills to meet children's child care and developmental needs. Pediatric health care providers should seek opportunities to impart parenting knowledge and skills to fathers. In addition, pediatric health care providers can facilitate fathers' emotional and material support of their children's mothers. They can also encourage mothers to permit fathers to be more involved in their children's lives by explaining the valuable roles that fathers can play.
Finally, tax code changes, child support enforcement, and efforts to help men become good fathers are insufficient. A cultural change acknowledging the roles fathers can play is needed. The media increasingly show men as nurturing parents and men indicate that they want more time with their children.14 A concerted public education campaign would include skill-building activities that teach parents how to incorporate men into fathering roles.
Editor's Note: It's important to have studies, like this, that emphasize the role of fathers in the lives of children. It's a little disconcerting, however, to define a father or father figure as someone who has contact at least monthly. I guess we take what we can get.—Catherine D. DeAngelis, MD
Accepted for publication June 22, 1999.
This research was supported by grant 90CA1401 from the National Center on Child Abuse and Neglect, US Department of Health and Human Services, Washington, DC (Dr Dubowitz).
Corresponding author: Howard Dubowitz, MD, MS, Department of Pediatrics, University of Maryland School of Medicine, 700 W Lombard St, Baltimore, MD 21201 (e-mail: firstname.lastname@example.org).
Dubowitz H, Black MM, Kerr MA, Starr RH, Harrington D. Fathers and Child Neglect. Arch Pediatr Adolesc Med. 2000;154(2):135-141. doi:10.1001/archpedi.154.2.135