Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Jul 22.
Published in final edited form as: Early Child Res Q. 2006 Oct 1;21(4):491–506. doi: 10.1016/j.ecresq.2006.09.002

Parenting quality: Confirmation of a higher-order latent construct with mothers of Head Start children

Laura Hubbs-Tait 1, Melanie C Page 2, Erron L Huey 3, Huei-Juang Starost 4, Anne McDonald Culp 5, Rex E Culp 6, M Elizabeth Harper 7
PMCID: PMC2714192  NIHMSID: NIHMS15157  PMID: 19629192

Abstract

We proposed a higher order latent construct of parenting young children, parenting quality. This higher-order latent construct comprises five component constructs: demographic protection, psychological distress, psychosocial maturity, moral and cognitive reflectivity, and parenting attitudes and beliefs. We evaluated this model with data provided by 199 mothers of 4-year-old children enrolled in Head Start. The model was confirmed with only one adjustment suggested by modification indices. Final RMSEA was .05, CFI .96, and NNFI .94, indicating good model fit. Results were interpreted as emphasizing the interdependence of psychological and environmental demands on parenting. Implications of the model for teachers, early interventionists, and public policy are discussed.

We propose a higher-order latent construct of parenting young children that extends and moves beyond ecological and process models. Ecological and process models (e.g., Belsky, 1984; Bronfenbrenner & Morris, 1998; Kotchik & Forehand, 2002) differentiate family environmental or contextual variables from parental personality and psychological functioning. We, too, subscribed for many years to such a model, proposing that parenting beliefs and attitudes as well as behaviors would be predicted by (1) age of mother; (2) demographic characteristics; (3) maternal affect; and (4) maternal developmental level (Hubbs-Tait, Culp, & Culp, 1997). Our viewpoint changed when we tried to integrate the child's perspective on his/her demands with the parent's perspective on meeting them. From the child's perspective, the identity of predictors of parenting quality is irrelevant. What matters to the child is whether the parent can be responsive to the child's needs and supportive of the child's development. Put most simply, anything that interferes with the parent's ability to address the child's needs is bad; anything that enhances the parent's ability is good. What matters to the parent is having the physical and emotional energy first, to be patient with the child's demands and, second, to respond appropriately to them. Integrating across parent and child perspectives, anything that depletes parenting quality is bad; anything that promotes it is good.

This integrated child and parent perspective suggested to us that traditional separations of psychological distress, demographic risk and protective factors, and parenting attitudes and beliefs were artificial. All contribute to but are also an integral part of parenting quality. Furthermore, like others, we thought that moral and cognitive reflectivity (Powers, 1988) and psychosocial maturity (Erikson, 1959, 1968; Peterson, Smirles, & Wentworth, 1997; Pratt, Danso, Arnold, Norris, & Filyer, 2001) are an integral part of parenting quality. An example will illustrate the latent construct of parenting quality (see Figure 1a). Consider a mother with high moral reflection and cognitive skills; well-developed identity and capacity for intimate relationships; married for 10 years, with two children, and a dual income; minimal depression, hostility, and anxiety; and highly positive parenting attitudes. This mother's parenting quality is excellent; she is emotionally supportive, responsive, cognitively stimulating, and involved in the range of her children's activities. However, consider what happens when this mother's parents are both killed in a car accident and she becomes too depressed to continue working. As a result, her joy in her interactions with her husband and children is diminished; however, due to her high moral and cognitive reflectivity, she has the ethical strength and cognitive awareness to continue to provide the basics of care for her children. The change in family income decreases the number of recreational activities for her children and the mother's depression limits her involvement in those that remain. Although she remains cognitively stimulating and her capacity for intimate relationships has not changed, her parenting quality as a whole has altered.

Figure 1a.

Figure 1a

Open in a new tab

Five Factor Hypothetical Higher Order Latent-Construct Model

Metaphorically, the resulting change in parenting quality is similar to what happens in an aquifer when a major spring dries up, reducing the number of sources for recharging it. Not only is the amount of water reduced, but its artesian quality has also been altered. Even when rainfall increases to make up the amount of water, the change in water quality remains. In the paragraphs below, we discuss the individual components of our higher-order parenting quality construct.

Moral and Cognitive Reflectivity

We begin with parental ethical strength and thoughtfulness, which we call moral and cognitive reflectivity. Parents with greater moral and cognitive reflectivity are better able to see their child's perspective, to consider the consequences of their behavior for the child, and to put their child's needs before their own. We operationalized this construct as maternal sociomoral reflection and education.

Sociomoral reflection on moral values (e.g., importance of telling the truth, keeping a promise, saving a life, and not taking things that belong to others) constitutes a developmental continuum that ranges from superficial to profound (Gibbs, Basinger, & Fuller, 1992). As individuals develop, their sociomoral reflective capacity increases from reflections that are unilateral and focused on power, status, and authority (stage 1); to reflections that are instrumental, pragmatic, and quid pro quo (stage 2); to reflections that recognize relationships, depend on empathic role taking, and include prosocial intentions and caring (stage 3); to reflections that affirm basic rights and values, accept societal responsibility, and reveal individual character and standards of conscience (stage 4, Gibbs et al., 1992).

There are precedents for including sociomoral reflection in parenting quality. First, in developing questions to stimulate sociomoral reflection, Gibbs et al. (1992) incorporated one question that specifically addressed a parenting issue: “How about keeping a promise to a child? How important is it for parents to keep promises, if they can, to their children?” (p. 150). Thus, sociomoral reflection includes reflections on the importance of a contract between the parent and his/her child. Second, Powers (1988) found that in family moral discussions, mothers with higher levels of moral reasoning were more likely to be encouraging of other family members' contributions and were more accepting of family members' voicing different perspectives. According to Powers, “the moral stage of a parent influences the parent's behavior with the child” (1988, p. 216). Third, in a group of foster parents, Richardson, Foster, and McAdams (1998) found that moral reasoning as measured by the Defining Issues Test and positive parenting attitudes were significantly positively correlated. These studies are consistent with our hypothesis that sociomoral reflection contributes to moral and cognitive reflectivity, which, in turn, enhances parenting quality.

Education is the other component of moral and cognitive reflectivity in our model. Although there are many measures of specific parental cognitive functions, we selected educational attainment precisely because it is associated with a wide range of higher cognitive and social cognitive functioning and behavior. For example, educational attainment was positively and significantly related to prosocial helpfulness in a random national sample of approximately 5,000 adults (Sares, 1998). Among urban biological and foster parents of Head Start children, Coolahan, McWayne, Fantuzzo, and Grim (2002) found that parental education was related to active-responsive parenting behaviors, which included such prosocial behaviors as demonstration of affection (hugging and kissing), praising the child, and responsivity to the child's own emotions. In a study of 386 families participating in the national Early Head Start Evaluation study, Tamis-LeMonda, Shannon, Cabrera, and Lamb (2004) found that greater maternal educational attainment was linked with higher cognitive stimulation, positive regard, and sensitivity as well as lower intrusiveness. Thus, mothers with more education have more parenting skills and parenting skills of higher quality. Thus, parental educational attainment is part of parenting quality.

Psychosocial Maturity

Psychosocial development from late adolescence through adulthood, the periods of development relevant to mothers of young children, reflects both general growth and individual differences in identity, intimacy, and generativity (Erikson, 1963). General growth in Erikson's theory refers to the maturation of particular capacities for experiencing reality. As each capacity emerges and attempts to express itself, it has to be integrated into the social and cultural contexts of the individual's life. This attempt occasions a crisis, an acute period that is similar to a medical crisis, in which psychosocial functioning may take a turn for the better or worse, with the healthiest resolution being a balance between the extremes of the continuum of psychosocial functioning appropriate to the particular stage (Erikson, 1959, 1963; Evans & Hubbs-Tait, 1991, p. 325). In adolescence, the crisis is between identity and identity confusion. Turning for the better involves a healthy balance between self-certainty and self-consciousness, apprenticeship and work paralysis, and ideological commitment versus confusion of values, among other things (Erikson, 1968, p. 94). In young adulthood, the crisis is between intimacy and isolation. Individual differences in Erikson's theory refer to each individual's history of resolutions of crises in previous stages that increase the likelihood of turning for the better (or worse) in the current stage (Erikson, 1959, 1968). In adolescence and young adulthood, the positive resolutions of previous crises affect the degree of identity and intimacy arising from those two stages and continuing on into adulthood. In turn, the degree of identity and intimacy arising in adolescence and young adulthood affect the resolution of the crisis of middle age between generativity and stagnation (Erikson, 1959, 1963, 1968).

There are precedents for including psychosocial maturity in parenting quality. In a sample of 76 mothers, McKeering and Pakenham (2000) found that both identity and intimacy were closely related to generativity. In turn, Pratt, et al. (2001) and Peterson et al. (1997) found that mothers' generativity was linked to their authoritative parenting. The latter two samples consisted of mothers of adolescents whereas the former also included mothers of preadolescent children, suggesting that identity and intimacy are relevant indicators of the psychosocial development of mothers of young children and should be related to their parenting. Consistent with this assumption, O'Brien and Peyton (2002) found that, among mothers of young children, marital intimacy was related to mothers' parenting attitudes. Thus, in samples of mothers of young children, identity and intimacy are part of parenting quality.

Psychological Distress

Precedents for including freedom from maternal psychological distress in parenting quality abound. With respect to depression, in their meta-analysis of 46 studies Lovejoy, Graczyk, O'Hare, and Neuman (2000) found moderate effect sizes for the relation between maternal depression and maternal negative or disengaged behaviors and a small effect size for the relation between maternal depression and positive behaviors. More recent studies continue to confirm these findings (e.g., Campbell, Brownell, & Hungerford, 2004; Cummings, Keller, & Davies, 2005) With respect to anxiety, Whaley, Pinto, and Sigman (1999) found that anxious mothers compared to non-anxious controls scored higher on measures of criticism and catastrophizing and lower on measures of enthusiasm, warmth, and granting of autonomy. Similarly, Woodruff-Borden, Morrow, Bourland and Cambron (2002) found that compared to control parents, more anxious parents were more withdrawn and disengaged and exhibited fewer productive behaviors like agreeing or praising their children. With respect to hostility, maternal reports of hostile parenting are highly correlated with children's reports on a puppet task of their mothers' hostile practices such as yelling and hitting (Silk, Sessa, Morris, Steinberg, & Avenevoli, 2004).

Although depression, hostility, and anxiety each have unique features, there are precedents for including them in a single construct, psychological distress (Coyne & Gotlib, 1983). Furthermore psychological distress as composed of symptoms of depression, hostility, and anxiety has been argued to be positively related to problematic parenting and negatively related to parenting quality (Lovejoy et al., 2000; Mammen, Kolko, & Pilkonis, 2002). Thus, we include psychological distress in parenting quality.

Demographic Protection

We have included four indicators of demographic protection as part of parenting quality: income, number of children in the family, age of the mother when her first child was born, and marital status. Each of these variables has been recognized as contributing to family risk or protection depending on the positive or negative impact of the variable. Low family income has long been recognized as a risk factor in terms of its negative impact on parenting with higher family income regarded as a protective factor (e.g., Bradley, Corwyn, McAdoo, & Garcia-Coll, 2001; Conger, Ge, Elder, Lorenz, & Simons, 1994; Conger et al., 2002; Klebanov, Brooks-Gunn, & Duncan, 1994; McLoyd, 1990). Low family income is regarded by many as impacting parenting not only in terms of lacking both the necessities and the stimulating toys that money can buy, but also as a correlate of multiple other stressors on parents (Leinonen, Solantus, Punamaeki, 2002; Mayer, 1997; Shonkoff & Phillips, 2000).

Parenting quality differs as a function of marital status in that single parents may experience stress related to role-strains from being the primary parent and provider for their children (Amato & Fowler, 2002; Hilton & Desrochers, 2000). Because single parent status and income are often linked, it is important to note that research also demonstrates that significant effects exist for residing in a single parent household independent of income (Dodge, Petit, & Bates, 1994; Lipman, Boyle, Dooley, Offord, 2002). Like marital status, the number of children in the family is a drain on parenting quality. Even increasing from one to two children in the family affects the quality of parenting offered to preschool children (Benigno & Ellis, 2004). More children are linked to less supportive home environments (Luster & Dubow, 1990) and greater abuse and neglect (Zuravin, 1988).

Parenting quality also differs as a function of the mothers' birth age, particularly whether she was an adult or an adolescent mother when her first child was born. For example, adolescent mothers' perceptions of infant emotions are less accurate and positive than the perceptions of adult mothers and non-pregnant adolescents (Osofsky & Culp, 1993). When compared to adult mothers, adolescent mothers are less verbal with their infants (Culp, Osofsky, & O'Brien, 1996).

Parenting Attitudes and Beliefs

Parenting attitudes are “childrearing opinions” (Holden & Buck, 2002, p. 552) whereas parenting beliefs are more cognitive in nature (Sigel & McGillicuddy-De Lisi, 2002). Unfortunately, and as noted by Holden and Buck, measures of parenting attitudes are rarely pure indicators of attitudes but include items evaluating beliefs, values, attributions, and other constructs. We have named our construct “attitudes and beliefs” not only to reflect the reality of measurement but also because we think that the cognitive aspect of beliefs is important to parenting quality. Parents with more positive attitudes and beliefs not only have positive opinions about their children but also have more accurate cognitions about them. They are less likely to endorse such items as “children will quit crying faster if they are ignored” and “children under three years of age should be expected to feed, bathe, and clothe themselves” (Bavolek, 1984, 1989a, 1989b). In fact, research on a nationally representative sample shows that parents with attitudes that devalue children are much more likely to use physical discipline with their children (Thompson et al., 1999). Mothers who endorsed items reflecting the negative attitude of parent-child role reversal when their children were age 2 were more likely to show rejecting parenting behaviors in videotaped interactions at the same time (Shaw, Criss, Schonberg, & Beck, 2004). Further, parent-child conflict at age 5 was related to parental endorsement of items reflecting low empathy when the child was age 2 (Criss, Shaw, & Skuban, 2003). Thus, positive attitudes are linked to improved parenting quality whereas negative attitudes are associated with deteriorating quality.

We do recognize that it is possible that positive attitudes are an expression of parenting quality rather than a part of the construct of parenting quality itself. This modified construct of parenting quality is depicted in Figure 1b and is consistent with research showing significant differences in parenting attitudes and beliefs associated with differences in a measure of cumulative risk that included psychological distress, measures of demographic risk, and other measures (Nair, Schuler, Black, Kettinger, & Harrington, 2003). Figure 1b is also consistent with research predicting parenting attitudes and beliefs from maternal depression (Pascoe & Solomon, 1994), socioeconomic status (Pinderhughes, Dodge, Bates, Pettit, & Zelli, 2000), maternal education, depression, and other variables (McKenry, Kotch, & Browne, 1991), and number of children in the household (Strom, Hathaway, & Slaughter, 1981). Note that the models in Figure1a and Figure 1b are conceptually equivalent in proposing that parenting quality is a higher order latent construct. They differ only in that 1a includes parenting attitudes and beliefs as part of parenting quality whereas 1b does not.

Figure 1b.

Figure 1b

Open in a new tab

Four Factor Hypothetical Higher Order Latent-Construct Model

Previous Research

To our knowledge no previous research group has proposed or tested a higher-order latent construct of parenting young children. A number of researchers have proposed empirical links between two of the component constructs. For example, the inverse relation between parental psychological distress and demographic protection, particularly as indicated by income, is well recognized (e.g., Brody, Murry, Kim, & Brown, 2002; Klebanov, Brooks-Gunn, McCarton, & McCormick, 1998). However, empirical links between components do not necessarily indicate that all of the components constitute a unified whole, which is what a single latent factor is (e.g., Halliday-Boykins & Graham, 2001).

The construct that comes closest to parenting quality is cumulative risk. Sameroff, Seifer, Barocas, Zax, and Greenspan (1987) conceptualized environmental risk as the total number of risk factors present from a set consisting of maternal mental health, maternal anxiety, maternal rigidity/flexibility in parenting attitudes, maternal spontaneity in interactive behavior, maternal education, occupational status of head of household, minority group status, presence of father, number of children, and stressful life events. Thus, this measure of risk included elements of demographic protection/risk, psychological distress, moral and cognitive reflectivity, and parenting attitudes and behavior. However, the identity of the component constructs is lost in the computation of cumulative risk. Most research groups sum the indicators of risk (0=absent; 1=present) so that all measures of cumulative risk are the sum of the number of risk factors (e.g., Burchinal, Roberts, Hooper, & Zeisel, 2000; Dumka, Roosa, & Jackson, 1997; Nair et al., 2003). Thus, the cumulative risk model follows two rules: all or nothing (for each contributing risk factor) and whole equals the sum of the parts (for the whole model). If the measured risk factor is sufficiently high, it is present; otherwise it is not. However, its identity is not important nor is any variation in the factor above or below the threshold that indicates the risk is present. In contrast, our latent construct conserves the identity of component constructs and component indicators, includes all variation in each indicator, and allows testing of whether all of the components can constitute a unified whole of parenting quality.

In sum, we proposed the higher order latent construct of parenting quality, because we think that parenting quality is a single factor. We proposed the two variants (Figures 1a and 1b), because sufficient research exists to suggest that parenting attitudes and beliefs may be predicted by the construct (referred to hereafter as “alternative theoretical model) rather than being a part of the construct (referred to hereafter as “primary theoretical model”).

Method

Participants

Participants were 199 mothers of 4-year-old children (102 boys, 97 girls) attending their pre-kindergarten year of Head Start in 1995-1996 or 1996-1997. Data from four grandmothers, four stepmothers, and one great-grandmother were excluded from this report. Mothers' ages on the date on which they completed all the information on which this report is based ranged from 19.75 to 46.23 (M = 29.05, SD = 5.62). Mothers' ethnicity was 78% European American, 17% American Indian, 2% African American, 2% Hispanic, and 1% multiethnic. Mothers were asked to report on the ethnicity of the father of their child who was attending Head Start. Fathers' ethnicity was 67% European American, 22% American Indian, 6% African American, 3% Hispanic, and 2% multiethnic.

Forty-six percent of the mothers were married; 20% were divorced; 17% were remarried; 10% had never been married; 4% were separated; and 3% were widowed. Twenty-three percent of the mothers had dropped out of school in high school or earlier (8% before 10th grade); 73% completed high school or obtained an equivalency diploma; 4% had attended or completed college (see Table 1). All families lived in rural, north-central Oklahoma. Monthly household income ranged from $0 (one family) to $4000 (one family); median monthly household income was $1250 (see Table 1 for mean). Household size ranged from 2 to 8 persons (Mdn = 4 persons), with the number of adults ranging from 1 to 4 (Mdn = 2) and the number of children ranging from 1 to 6 (Mdn = 2).

Table 1.

Descriptive Statistics for all Measures

Construct and Variable Mean SD Possible range Actual range
Demographic Protection
Monthly household income 1344.47 740.39 $0 - $4,000
Number of children 2.5 1.01 1 - 6
Mother's age at birth of first 21.39 4.40 14.48 - 39.55
Mother's age at birth of target 24.29 5.60 14.48 - 41.34
Psychological Distress
Depression 15.39 11.71 0 - 60 0 - 55
Anxiety 37.13 10.10 20 - 80 20 - 77
Hostility 59.46 16.11 29 - 145 30 - 133
Psychosocial Maturity
Intimacy 46.34 6.79 12 - 60 25 - 60
Identity 49.06 8.60 12 - 60 21 - 60
Moral and Cognitive Reflectivity
Education 11.96 1.47 5th grade - college
Sociomoral reflection 300.13 32.81 100 - 400 207 - 373
Parenting Attitudes and Beliefs
Endorsement of physical punishment 20.79 6.14 10 - 50 10 - 38
Inappropriate expectations 10.20 3.11 6 - 30 6 - 20
Lack of empathy 13.72 4.73 8 - 40 8 - 30
Role reversal 17.86 6.39 8 - 40 8 - 38
Open in a new tab

Procedure

Mothers were recruited in the Fall of their child's pre-kindergarten year in Head Start. At the time of recruitment, mothers completed a demographic information form, the Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977, 1991), the Aggression Questionnaire (AQ; Buss & Perry, 1992), the State-Trait Anxiety Inventory-Trait form (STAI-T, Spielberger, 1983), the Erikson Psychosocial Inventory (EPSI, Rosenthal, Gurney, & Moore, 1981), the Adult-Adolescent Parenting Inventory (AAPI; Bavolek, 1984, 1989), and the Sociomoral Reflection Measure-Short Form (SRM-SF, Gibbs et al., 1992).

Measures

Operationalization of constructs

Family demographic protection was assessed by answers to questions on the demographic information form. Maternal psychological distress was assessed by scores on the CES-D, the AQ, and the STAI-T. Maternal psychosocial maturity was assessed by scores on two subscales of the EPSI: identity and intimacy. Maternal moral and cognitive reflectivity were assessed by scores on the SRM-SF and responses to a question about number of years of education on the demographic information form. Maternal parenting attitudes and beliefs were assessed by scores on the four subscales of the AAPI: endorsement of physical punishment, lack of empathy, inappropriate expectations, and role reversal.

Demographic information form

On the demographic information form, mothers indicated the range of monthly household income that characterized their family. We followed the procedures used in the Infant Health and Development Program studies (e.g., Duncan, Brooks-Gunn, & Klebanov, 1994; Klebanov et al., 1998) to convert the categorical responses to a continuous measure by assigning the midpoint of each interval. Income midpoints were $0, $50, $250, $750, $1250, $1750, $2250, $2750, $3750, and $4000 per month. On the same form, mothers indicated their marital status. Mothers who were married, remarried, or widowed were classified as “married”; mothers who were separated, divorced, or never married were classified as “not married”. The criterion for classifying widows as “married” was that the marital bond was never dissolved. Switching the six widows to the “single” group did not affect the final model. Mothers listed their birth date and the birth date of their pre-kindergarten child as well as the ages of all of their children. Dates and ages were used to compute each mother's age at the birth of her first child and at the date of her target (i.e., Head Start) child's birth.

Center for Epidemiologic Studies Depression Scale

The CES-D is a 20-item self-report instrument scored on a 4-point (0 to 3) Likert scale that measures the degree to which participants have been bothered by depressive symptoms in the previous week (Radloff, 1977). Items include “My sleep was restless;” “I felt lonely;” “I had crying spells;” “I did not feel like eating; my appetite was poor;” “I was bothered by things that usually don't bother me.” Internal consistency and test-retest reliability of the CES-D are within the acceptable range for samples that include females of the same age as the current sample (Radloff, 1977, 1991). Internal consistency of the 20 items in the current sample was .92.

Aggression Questionnaire

The AQ (Buss & Perry, 1992) is a 29-item questionnaire consisting of four subscales: hostility, anger, verbal aggression, and physical aggression. Respondents rate each item on a 1 (“extremely uncharacteristic of me”) to 5 (“extremely characteristic of me”) Likert-type scale. Test-retest reliability for a sample of women in the same age range as the current sample is moderate to high for all four subscales over 7 months (.67 to .82, Harris, 1997). Convergent validity is indicated by moderate to high correlations (.32 to .76) of all four subscales with measures of affect instability, aggression, negative interpersonal relationships, and low frustration tolerance (Harris, 1997). Internal consistency of the 29 items in the current sample was .90.

State-Trait Anxiety Scale-Trait Form

The STAI-T consists of 20 items rated by participants on a 4-point scale to reflect how they “generally feel” (Spielberger, 1983). Items include 11 statements directly reflecting anxiety (e.g., I feel nervous and restless; “I wish I could be as happy as others seem;” “I lack self-confidence”), and 9 reverse-coded statements (e.g., “I am happy;” “I feel secure”). In addition to the evidence for construct validity offered by Spielberger, independent research shows that mothers classified as high versus low or moderate on anxiety by the state and trait versions of the STAI differed significantly 4.5 years later on autonomic anxiety symptoms as well as on social phobia and lack of confidence in mothering capabilities (Barnett, Schaafsma, Guzman, & Parker, 1991). We analyzed total scores computed from all 20 items; internal consistency of the 20 items was .92.

Erikson Psychosocial Inventory

The EPSI is a 72-item scale, consisting of six 12-item subscales, each assessing one of Erikson's first six stages or psychosocial functioning (Rosenthal et al. 1981). We administered only two of the subscales, identity and intimacy, using the modifications recommended by Arehart and Smith (1990). Sample items from the identity subscale include “I can't decide what to do with my life (reverse);” I like myself and am proud of what I stand for.“ Sample items from the intimacy subscale include “I have a close physical and emotional relationship with another person;” “I care deeply for others;” “I think it's crazy to get too involved with people (reverse).” Test-retest and internal consistency reliability and construct validity are acceptable (e.g., Rosenthal et al., 1991; Dyk & Adams, 1990). Internal consistency for identity and intimacy in the current sample was .89 and .73, respectively.

Sociomoral Reflection Measure - Short Form

The SRM-SF is an 11-item production task that assesses judgments and justifications of the importance of such values as contract and truth, affiliation, life, property and law, and legal justice. Participants are first given real-life scenarios (e.g., “Think about when you've made a promise to a friend of yours”) and are then asked to provide judgments of importance (“How important is it for people to keep promises, if they can to friends?”) using the scale, “very important,” “important,” or “not important.” Finally, participants are asked to give a moral justification for why that value is (or is not) important. Justifications are scored for stage of sociomoral reasoning according to their degree of match with stage-diagnostic justifications (see Gibbs et al., 1992). The resulting Sociomoral Reflection Maturity Score (SRMS) ranges from 100 (Stage 1) to 400 (Stage 4). Test-retest reliability and internal consistency are acceptable (Basinger, Gibbs & Fuller, 1995; Gibbs et al., 1992). Construct validity has been demonstrated by a high correlation with Kohlberg's Moral Judgment Inventory (e.g., Basinger et al., 1995) and criterion validity by its discrimination of delinquent from non-delinquent samples (Gregg, Gibbs, & Basinger, 1994). In samples of university students and community-dwelling adults, there are no gender differences in SRMS (Garmon, Basinger, Gregg, & Gibbs, 1996), supporting the appropriateness of the instrument for the current sample of adult females. Every SRM-SF protocol in the current sample was scored by two of the developers of the measure, Drs. John Gibbs and Karen Basinger. One served as the primary and the other as the reliability rater (Gibbs et al., 1992). Both raters independently scored written responses by 27 mothers (14%) randomly selected by the fourth author. Inter-rater reliability for the present sample was acceptable, r(27) =. 96, p <. 01.

Education

Mothers filled in their highest grade completed on the demographic information form. Because many students enroll in vocational-technical institutions and high school classes concurrently, we did not give greater value to completion of vocational training than completion of high school.

Adult-Adolescent Parenting Inventory

The AAPI is a 32-item inventory of negative parenting attitudes and beliefs related to child abuse (Bavolek, 1984, 1989b), consisting of four subscales: endorsement of physical punishment (“children should always be spanked when they misbehave”), lack of empathy (“parents who are sensitive to their children's feelings and moods often spoil their children”), inappropriate expectations (“children under three years should be expected to feed, bathe, and clothe themselves), and role reversal (“young children should be expected to comfort their mother when she is feeling blue”). We scored the Likert-scale (1 to 5) responses so that higher subscale scores reflect more negative attitudes. Test-retest reliability and construct validity are reported to be acceptable (Bavolek, 1989b). According to Holden and Buck (2002), the power of attitudes as a construct is their relation to behaviors. Thus, support for construct validity of the AAPI as a measure of parenting beliefs and attitudes is provided by the link between rejecting maternal behavior and AAPI role reversal (Shaw et al., 2004). Further, Kiang, Moreno, and Robinson (2004) found that negative parenting attitudes and beliefs measured prenatally by total AAPI scores predicted both later maternal insensitive behavior and mothers' perceptions of their children as having more difficult temperament. Internal consistency for the subscales in the current sample was as follows: role reversal (α = .87); physical punishment (α = .80); inappropriate expectations (α = .73); and lack of empathy (α = .85).

Results

Preliminary Analyses

Means, standard deviations, and ranges for all of the variables are in Table 1. The correlation matrix of all variables is provided in the Appendix. In terms of univariate normality, anxiety, marriage, and mother's education were nonnormally distributed. In terms of multivariate normality, the assumption of multivariate normality was violated (Mardia's criterion (a measure of multivariate kurtosis) = 23.77). As this value was above the recommended cutoff of 21 (West, Finch, & Curran, 1995), we first examined the influence of outliers. After deleting the two most highly influential outlying cases, Mardia's criterion was 19.37 indicating that the multivariate normality assumption for SEM was met. However, because all of the analyses reported below yielded the same results with the outliers included or excluded, we retained the two outlying cases.

There are several different estimation methods that can be used in SEM. We used maximum likelihood estimators because the alternatives were less satisfactory. For example, there are several alternate estimation techniques, but a minimum sample size of 500 is required in order to use them (West et al., 1995). Another solution to nonnormality is to transform the variables. However all conclusions are then based on the transformed variables, rendering inferences about theoretical and measurement models based on the original variables difficult.

Primary Analyses

Measurement model

Following the approach recommended in Anderson and Gerbing (1988) and demonstrated in Halliday-Boykins and Graham (2001) and Newcomb and Loeb (1999), we tested a measurement model before evaluating our theoretical model of interest. Using confirmatory factor analysis (CFA) we hypothesized that five constructs - psychological distress, moral and cognitive reflectivity, parenting attitudes and beliefs, psychological maturity, and demographic protection - generated the observed variables. We used several indices to determine model fit. We initially looked at the model χ2, which, if the data fit the model well, is nonsignificant. However, as early as 1969, it was recognized that this χ2 oftentimes would remain significant even if the data fit the model (Jöreskog, 1969; see also Mueller, 1997). Thus three additional fit indices were examined: the Comparative Fit Index (CFI), the Nonnormed Fit Index (NNFI; recommendation for use in Hu & Bentler, 1995) and the Root Mean Square Error of Approximation (RMSEA; developed by Browne & Cudek, 1993). Ideally based on a sample size of at least 250, an NNFI greater than .90 and RMSEA less than .05 are indicative of good model fit. The measurement fit statistics are presented in the top row of Table 2. In a measurement model, the main concern is how the observed variables load on their respective factors rather than on model fit. Although it is possible to make modifications to the model, post hoc adjustments that are not theoretically justifiable are likely to be based on idiosyncrasies in the data and are not recommended (Hoyle & Smith, 1994). Standardized factor loadings of this measurement model are presented in Table 3. With the exception of mother's age at first birth, all of the factor loadings were high, in the expected direction (except for total number of children), and significant (p < .05). Birth age (regardless of whether it was measured at age of birth of first child or at birth of the target child) did not load on the demographic protection factor and was thus dropped from the model.

Table 2.

Measurement Fit Statistics

Model χ2 df p value CFI NNFI RMSEA
Measurement 111.86 67 .00 .95 .93 .06
Initial theoretical 120.86 71 .00 .94 .92 .06
Primary five-factor theoretical 108.93 70 .00 .96 .94 .05
Alternative four-factor theoretical 49.21 39 .13 .98 .98 .03
Open in a new tab
Table 3.

Loadings for Measurement Model

Indicator Component Construct Standardized Factor Loadings
Income Demographic Protection .716
Number of children Demographic Protection .216
Marital status Demographic Protection .584
Depression Psychological Distress .741
Anxiety Psychological Distress .874
Hostility Psychological Distress .628
Intimacy Psychosocial Maturity .663
Identity Psychosocial Maturity .934
Sociomoral reflection Moral and Cognitive Reflectivity .602
Mother's education Moral and Cognitive Reflectivity .586
Endorsement of physical punishment Parenting Attitudes and Beliefs .566
Inappropriate expectations Parenting Attitudes and Beliefs .760
Lack of empathy Parenting Attitudes and Beliefs .789
Role reversal Parenting Attitudes and Beliefs .734
Open in a new tab

Intercorrelations among the latent factors are presented in Table 4. Most correlations of latent factors were in the expected direction and all were significant with the exception of the relation between demographic protection and moral and cognitive reflectivity.

Table 4.

Intercorrelations of Latent Factors

Factor/variable (1) (2) (3) (4) (5)
(1) Demographic Protection 1
(2) Psychological Distress -.33* 1
(3) Psychosocial Maturity .38* -.90* 1
(4) Moral and Cognitive Reflectivity .21* -.42* .29* 1
(5) Parenting Attitudes and Beliefs -.26* .40* -.33* -.68* 1
Open in a new tab
*

Note: p < .05

Primary Theoretical Model

The hypothesized model depicted in Figure 1a was tested using CFA. Note that this model represents a theoretical model and is not a direct illustration of the statistical model. For example, none of the error terms are included in this depiction. It should be noted though that in order for the model to be estimable, a path between the error terms for psychological maturity and psychological distress was included. The fit statistics for the initial model examining whether or not there is a higher order construct, parenting quality, that generates the latent constructs from the measurement model are presented in the second row of Table 2. As seen, the fit although adequate, is not ideal. To improve the model fit, we examined the modification indices making suggested changes that were theoretically plausible, one change at a time. We made one adjustment to the model based upon these indices. Specifically, we correlated the error terms of sociomoral reflection and endorsement of physical punishment. The standardized loadings for this final model are presented in Figure 2 and the fit statistics are in the third row of Table 2. Although the chi-square is still significant, the RMSEA suggests good fit (i.e., = .05), the NNFI is .94, and the CFI is .96.

Figure 2.

Figure 2

Open in a new tab

Final Model and Standardized Loading

Alternative Theoretical Model

The hypothesized model in Figure1b was also tested using CFA. Note that this model represents a theoretical model and is not a direct illustration of the statistical model. For example, none of the error terms are included in this depiction. It should be noted though that in order for the model to be estimable, a path between the error terms for psychological maturity and psychological distress was included. We did not include a figure of this model with its coefficients because the only difference was that parenting attitudes and beliefs switched to an observed variable, which is depicted in Figure 1b. However, the other four component constructs of parenting quality remained the same. The fit statistics for the initial model examining parenting attitudes and beliefs as an observed variable are presented in the fourth row of Table 2. As seen the fit is very good. The chi-square is nonsignificant, the RMSEA = .03, and the NNFI is .98.

Discussion

The results support our higher-order latent construct, parenting quality. The paths from all proposed component constructs to parenting quality are significant. With the exception of mothers' age at the time of her child's birth, all indicators contributed to their respective component constructs.

Nonetheless, the model of parenting quality that provided the best fit was the model in which parenting attitudes and beliefs were not included in the construct, but instead were predicted by parenting quality. Thus, it could be that parenting attitudes and beliefs are a reflection or expression of parenting quality and may best be conceptualized as lying outside the construct of parenting quality. Alternatively, it could be that this model fits better as it has fewer constraints than the original model (i.e., parenting attitudes and beliefs as a latent construct has more parameters to estimate than parenting attitudes and beliefs as an observed variable). In both models, there is a path leading from parenting quality to parenting attitudes and beliefs, so they are very similar mathematically. However, they differ in whether parenting attitudes and beliefs are an observed outcome or a component construct of parenting quality. Both models support the construct of parenting quality, but the support for the alternate model suggests that parenting attitudes and beliefs may mediate between parenting quality and parenting behaviors.

Confirmation of this model extends the literature on parenting in two ways. First, both models emphasize that resources for parenting are interdependent. Isolating the impact of psychological distress from that of demographic protection, psychosocial maturity, parenting attitudes and beliefs, or moral and cognitive reflectivity is an artificial division, as is any other separation of component constructs. We emphasize this interdependence, as it suggests that assigning environmental factors such as family economic status to a different level or sphere of influence than parental psychological characteristics is a conceptual error. Instead, demographic protective factors and a variety of psychological constructs - distress, maturity, and attitudes - all contribute significantly to the same higher-order construct. Second, the model includes components that have heretofore received insufficient attention in the early childhood and parenting literatures, specifically parental psychosocial maturity and sociomoral reflection. Sociomoral reflection contributes not only to its component construct and the higher-order latent construct, but also, in the current sample, is specifically associated with endorsement of physical punishment. That is, mothers with higher SRM scores have lower scores on endorsement of physical punishment.

As we have noted, the specific characteristics of the current sample dictated several modifications to the proposed theoretical model: associations between psychosocial maturity and psychological distress and between sociomoral reflection and endorsement of physical punishment. Thus although only a few modifications are data set specific, the model will need to be replicated across more samples. Finally, in the current sample, number of children is positively associated with demographic protection. This is likely accounted for by the fact that mothers who were married had larger families and lived in households with higher incomes. Testing of our higher-order model with other less economically disadvantaged families is necessary to see if similar relations hold.

The higher order latent construct of parenting quality shares a few features of general systems models. Both models of parenting quality are interdependent wholes in that all of the parts are linked together (von Bertalanffy, 1968), both in the conceptual framework we proposed and in the results of the confirmatory factor analyses. However, these models are static, not changing and not dynamic. Thus, although they are consistent with two of the features of organized systems, wholeness and hierarchical structure, they are not consistent with the other two features of organized systems: adaptive self-stabilization and adaptive self-organization (Cox & Paley, 1997). Parenting quality subsumes its component constructs (hierarchical order) and is a coherent factor as demonstrated by the analyses. But, evaluating the other two features of systems requires information on change over time, information that we do not have on this sample of 199 mothers.

Implications for Policy and Practice

The current research supporting parenting quality as a single coherent factor lends support to innovative programs offering parent education or family-centered prevention programs in schools (Kumpfer & Collings, 2004). Our model of parenting quality includes parental moral reasoning and psychosocial maturity with more traditional constructs - psychological distress, demographic protection, and perhaps parenting attitudes and beliefs. Teachers are neither trained nor do they have the time to deliver mental health services much less moral education to parents. However, when schools, parents, and communities join together to intervene (Kumpfer & Collings, 2004), interventions can change families. Thus, it is important to consider parenting quality in the risk factors families bring to preschool programs, in order to develop effective interventions for young children (Hubbs-Tait et al., 2002).

Recent research on the effectiveness of attachment interventions provides some important lessons for the new trend toward family-centered prevention in schools. Effecting systemic change is difficult, and usually requires targeting more than one component of the system (e.g., Lewis, Lamey, & Douglas, 1999), although it is possible to identify single control parameters that can produce change throughout an entire system (see Thelen & Smith, 1998). Representative of interventions that target multiple components of the attachment system is the Circle of Security (COS) intervention implemented with Head Start and Early Head Start children and their families. The COS targets a panoply of maternal characteristics simultaneously, including empathy and reflective functioning (Cooper, Hoffman, Powell, & Marvin, 2005) In contrast, meta-analyses by Bakermans-Kranenburg, van IJzendoorn, and Juffer (2003) have demonstrated that intervention programs targeting only maternal sensitivity are effective in promoting infant attachment security, a different component in the attachment system. As far as the construct of parenting quality is concerned, we do not yet know whether interventions targeting multiple components or focusing on single components, such as maternal sociomoral reflection alone, will be more effective. Future intervention research will be needed to answer this question.

Acknowledgments

We gratefully acknowledge the support of the participating mothers as well as the teachers and administrators of United Community Action Program, Inc. Head Start. Funding for this study was provided by grants from the Department of Health and Human Services, Administration on Children, Youth and Families (90-YD-0036) and the National Institute of Mental Health (R03MH52115) to L. Hubbs-Tait, A. M. Culp, and R. E. Culp. The John and Sue Taylor Professorship in Human Environmental Sciences also supported the contribution of L. Hubbs-Tait. Elizabeth Harper's work was supported by a National Science Foundation Research Experiences for Undergraduates grant (SES-0097643) to M. Page. Portions of these data were presented at the 2003 meeting of the Society for Research in Child Development in Tampa. We are grateful for assistance in data collection to Layle Reese Krogel, Jane Flett, Kerri Horton, Kim Austin, Pamela Carr, and Tara Saathoff-Wells.

Appendix

Correlation Matrix of all Variables

Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Income
Age at target child's birth .07
Marital status .42** .09
Number of children .12 .26** .21**
Hostility -.10 -.07 -.16* -.07
Depression .29** .00 -.18* -.02 .46**
Anxiety -.18* .12 -.15* .07 .54** .65**
Intimacy .17* -.08 .12 .04 -.38** -.45** -.52**
Identity .27* .00 .20** .03 -.54** -.62** -.73** .62**
Physical punishment -.10 .08 -.08 -.02 .32** .14* -.26** -.22** -.24**
Inappropriate behavior -.18* .00 -.12 -.09 .19** .27** .34** -.24** -.26** .44**
Lack of empathy -.15* .04 -.04 -.07 .24** .27** .35** -.33** -.25** .44** .61**
Role reversal -.19** -.04 -.06 -.06 .19** .23** .19** -.08 -.11 .47** .55** .56**
Sociomoral reflection .07 .11 .10 -.01 -.13 -.16* -.22** .12 .15* -.06 -.28** -.39** -.36**
Education .17* .13 .02 -.13 -.12 -.16* -.26* .19** .16* -.14** -.30** -.31** -.33** .35**
Open in a new tab

Note:

*

p ≤ .05;

**

p < .01

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Amato PR, Fowler F. Parenting practices, child adjustment, and family diversity. Journal of Marriage and Family. 2002;64:703–716. [Google Scholar]
  2. Anderson JC, Gerbing DW. Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin. 1988;103:411–423. [Google Scholar]
  3. Arehart DM, Smith PH. Identity in adolescence: Influences of dysfunction and psychosocial task issues. Journal of Youth and Adolescence. 1990;19:63–72. doi: 10.1007/BF01539446. [DOI] [PubMed] [Google Scholar]
  4. Bakermans-Kranenburg MJ, van IJzendoorn MH, Juffer F. Less is more: Meta-analyses of sensitivity and attachment interventions in early childhood. Psychological Bulletin. 2003;129:195–215. doi: 10.1037/0033-2909.129.2.195. [DOI] [PubMed] [Google Scholar]
  5. Barnett B, Schaafsma MF, Guzman A-M, Parker GB. Maternal anxiety: A 5-year review of an intervention study. Journal of Child Psychology & Psychiatry. 1991;32:423–438. doi: 10.1111/j.1469-7610.1991.tb00321.x. [DOI] [PubMed] [Google Scholar]
  6. Basinger KS, Gibbs JC, Fuller D. Context and measurement of moral judgment. International Journal of Behavioral Development. 1995;18:537–556. [Google Scholar]
  7. Bavolek SJ. Handbook for the Adult-Adolescent Parenting Inventory (AAPI) Family Development Resources; Park City, UT: 1984. [Google Scholar]
  8. Bavolek SJ. Assessing and treating high-risk parenting attitudes. Early Child Development and Care. 1989a;42:99–112. [Google Scholar]
  9. Bavolek SJ. Research and validation report of the Adult-Adolescent Parenting Inventory (AAPI) Family Development Resources; Eau Claire, WI: 1989b. [Google Scholar]
  10. Belsky J. The determinants of parenting: A process model. Child Development. 1984;55:83–96. doi: 10.1111/j.1467-8624.1984.tb00275.x. [DOI] [PubMed] [Google Scholar]
  11. Benigno JP, Ellis S. Two is greater than three: Effects of older siblings on parental support of preschoolers' counting in middle-income families. Early Childhood Research Quarterly. 2004;19:4–20. [Google Scholar]
  12. Bertalanffy L. von. General system theory: Foundations, development, applications. George Braziller; New York: 1968. [Google Scholar]
  13. Bradley RH, Corwyn RF, McAdoo HP, Garcia-Coll C. The home environments of children in the United States. Part I: Variations by age, ethnicity, and poverty status. Child Development. 2001;72:1844–1867. doi: 10.1111/1467-8624.t01-1-00382. [DOI] [PubMed] [Google Scholar]
  14. Brody GH, Murry VM, Kim S, Brown AC. Longitudinal pathways to competence and psychological adjustment among African American children living in rural single-parent households. Child Development. 2002;73:1505–1516. doi: 10.1111/1467-8624.00486. [DOI] [PubMed] [Google Scholar]
  15. Bronfenbrenner U, Morris PA. The ecology of developmental processes. In: Damon W, Lerner RM, editors. Handbook of child psychology: Vol. 1. Theoretical models of human development. 5th ed. John Wiley & Sons; New York: 1998. pp. 993–1028. Series Ed. Vol. Ed. [Google Scholar]
  16. Browne MW, Cudek R. Alternative ways of assessing model fit. In: Bollen KA, Long JS, editors. Testing structural equation models. Sage; Newbury Park, CA: 1993. pp. 136–162. [Google Scholar]
  17. Burchinal MR, Roberts JE, Hooper S, Zeisel SA. Cumulative risk and early cognitive development: A comparison of statistical risk models. Developmental Psychology. 2000;36:793–807. doi: 10.1037//0012-1649.36.6.793. [DOI] [PubMed] [Google Scholar]
  18. Buss AH, Perry M. The Aggression Questionnaire. Journal of Personality and Social Psychology. 1992;63:452–459. doi: 10.1037//0022-3514.63.3.452. [DOI] [PubMed] [Google Scholar]
  19. Campbell SB, Brownell CA, Hungerford A. The course of maternal depressive symptoms and maternal sensitivity as predictors of attachment security at 36 months. Development and Psychopathology. 2004;16:231–252. doi: 10.1017/s0954579404044499. [DOI] [PubMed] [Google Scholar]
  20. Conger RD, Ge X, Elder G, Lorenz F, Simons R. Economic stress, coercive family process, and developmental problems of adolescents. Child Development. 1994;65:541–561. [PubMed] [Google Scholar]
  21. Conger RD, Wallace LE, Sun R, Simons R, McLoyd VC, Brody GH. Economic pressure in African American families: A replication and extension of the family stress model. Developmental Psychology. 2002;38:179–193. [PubMed] [Google Scholar]
  22. Coolahan K, McWayne C, Fantuzzo J, Grim S. Validation of a multidimensional assessment of parenting styles for low-income African-American families with preschool children. Early Childhood Research Quarterly. 2002;17:356–373. [Google Scholar]
  23. Cooper G, Hoffman K, Powell B, Marvin R. The Circle of Security Intervention: Differential diagnosis and differential treatment. In: Berlin LJ, Ziv Y, Amaya-Jackson L, L., Greenberg MT, editors. Enhancing early attachments: Theory, research, intervention, and policy. Guilford; New York: 2005. [Google Scholar]
  24. Cox MJ, Paley B. Families as systems. Annual Review of Psychology. 1997;48:243–267. doi: 10.1146/annurev.psych.48.1.243. [DOI] [PubMed] [Google Scholar]
  25. Coyne JC, Gotlib IH. The role of cognition in depression: A critical appraisal. Psychological Bulletin. 1983;94:472–505. [PubMed] [Google Scholar]
  26. Criss MM, Shaw DS, Skuban EM. Open communication and conflict in the parent-child dyad: Anexamination of antecedents and developmental trajectories. Poster presented at the biennial meeting of the Society for Research in Child Development; Tampa, FL. Apr, 2003. [Google Scholar]
  27. Culp AM, Osofsky JD, O'Brien M. Language patterns of adolescent and older mothers and their one-year-old children: A comparison study. First Language. 1996;16:61–75. [Google Scholar]
  28. Cummings EM, Keller PS, Davies PT. Towards a family process model of maternal and paternal depressive symptoms: Exploring multiple relations with child and family functioning. Journal of Child Psychology and Psychiatry. 2005;46:479–489. doi: 10.1111/j.1469-7610.2004.00368.x. [DOI] [PubMed] [Google Scholar]
  29. Dodge KA, Pettit GS, Bates JE. Socialization mediators of the relation between socioeconomic status and child conduct problems. Child Development. 1994;65:649–665. [PubMed] [Google Scholar]
  30. Dumka LE, Roosa MW, Jackson KM. Risk, conflict, mothers' parenting, and children's adjustment in low-income, Mexican immigrant, and Mexican American families. Journal of Marriage and the Family. 1997;59:309–323. [Google Scholar]
  31. Duncan GJ, Brooks-Gunn J, Klebanov PK. Economic deprivation and early childhood development. Child Development. 1994;65:296–318. [PubMed] [Google Scholar]
  32. Dyk PH, Adams GR. Identity and intimacy: An initial investigation of three theoretical models using cross-lag panel correlations. Journal of Youth and Adolescence. 1990;19:91–110. doi: 10.1007/BF01538715. [DOI] [PubMed] [Google Scholar]
  33. Erikson EH. Identity and the life cycle. Psychological Issues. 1959;1:18–171. [Google Scholar]
  34. Erikson EH. Childhood and society. Norton; New York: 1963. [Google Scholar]
  35. Erikson EH. Identity, youth and crisis. Norton; New York: 1968. [Google Scholar]
  36. Evans RG, Hubbs-Tait L. Applying psychology in today's world. Prentice-Hall; Englewood Cliffs, NJ: 1991. [Google Scholar]
  37. Garmon LC, Basinger KS, Gregg VR, Gibbs JC. Gender differences in stage and expression of moral judgment. Merrill-Palmer Quarterly. 1996;42:418–437. [Google Scholar]
  38. Gibbs JC, Basinger KS, Fuller D. Moral maturity: Measuring the development of sociomoral reflection. Erlbaum; Hillsdale, NJ: 1992. [Google Scholar]
  39. Gregg VR, Gibbs JC, Basinger KS. Patterns of developmental delay in moral judgment by male and female delinquents. Merrill-Palmer Quarterly. 1994;40:538–553. [Google Scholar]
  40. Halliday-Boykins CA, Graham S. At both ends of the gun: Testing the relationship between community violence exposure and youth violent behavior. Journal of Abnormal Child Psychology. 2001;29:383–402. doi: 10.1023/a:1010443302344. [DOI] [PubMed] [Google Scholar]
  41. Harris JA. A further evaluation of the Aggression Questionnaire: Issues of validity and reliability. Behaviour Research and Therapy. 1997;35:1047–1053. doi: 10.1016/s0005-7967(97)00064-8. [DOI] [PubMed] [Google Scholar]
  42. Hilton JM, Desrochers S. The influence of economic strain, coping with roles, and parental control on the parenting of custodial single mothers and custodial single fathers. Journal of Divorce & Remarriage. 2000;33:55–76. [Google Scholar]
  43. Holden, Buck . Parental attitudes toward childrearing. In: Bornstein MH, editor. Handbook of parenting. 2nd ed. Erlbaum; Mahwah, NJ: 2002. pp. 537–562. [Google Scholar]
  44. Hoyle RH, Smith GT. Formulating clinical research hypotheses as structural equation models: A conceptual overview. Journal of Consulting and Clinical Psychology. 1994;62:429–440. doi: 10.1037//0022-006x.62.3.429. [DOI] [PubMed] [Google Scholar]
  45. Hu L, Bentler PM. Evaluating model fit. In: Hoyle R, editor. Structural equation modeling: Issues and applications. Sage; Thousand Oaks, CA: 1995. pp. 76–99. [Google Scholar]
  46. Hubbs-Tait L, Culp AM, Culp RE. Maternal affect and developmental level predict parenting practices. Poster presented at the biennial meeting of the Society for Research in Child Development; Washington, DC. Apr, 1997. [Google Scholar]
  47. Hubbs-Tait L, Culp AM, Huey E, Culp RE, Starost H-J, Hare C. Relation of Head Start attendance to children's cognitive and social outcomes: Moderation by family risk. Early Childhood Research Quarterly. 2002;17:539–558. [Google Scholar]
  48. Jöreskog KG. A general approach to confirmatory maximum likelihood factor analysis. Psychometrika. 1969;34:183–202. [Google Scholar]
  49. Kiang L, Moreno AJ, Robinson JL. Maternal preconceptions about parenting predict child temperament, maternal sensitivity, and children's empathy. Developmental Psychology. 2004;40:1081–1092. doi: 10.1037/0012-1649.40.6.1081. [DOI] [PubMed] [Google Scholar]
  50. Klebanov PK, Brooks-Gunn J, Duncan GJ. Does neighborhood and family poverty affect mothers' parenting, mental health, and social support? Journal of Marriage and the Family. 1994;56:441–455. [Google Scholar]
  51. Klebanov PK, Brooks-Gunn J, McCarton C, McCormick MC. The contribution of neighborhood and family income to developmental test-scores over the first 3 years of life. Child Development. 1998;69:1420–1436. [PubMed] [Google Scholar]
  52. Kotchik BA, Forehand R. Putting parenting in perspective: A discussion of the contextual factors that shape parenting practices. Journal of Child and Family Studies. 2002;11:255–269. [Google Scholar]
  53. Kumpfer KL, Collings SJ. Effectiveness of family focused interventions for school-based prevention. In: Robinson KE, editor. Advances in school-based mental health interventions: Best practices and program models. Civic Research Institute; Kingston, NJ: 2004. pp. 1–14. [Google Scholar]
  54. Leinonen JA, Solantus TS, Punamaeki R-L. The specific mediating paths between economic hardship and the quality of parenting. International Journal of Behavioral Development. 2002;26:423–435. [Google Scholar]
  55. Lewis MD, Lamey AV, Douglas L. A new dynamic systems method for the analysis of early socioemotional development. Developmental Science. 1999;2:457–475. [Google Scholar]
  56. Lipman EL, Boyle MH, Dooley MD. Child well-being in single-mother families. Journal of the American Academy of Child & Adolescent Psychiatry. 2002;41:75–82. doi: 10.1097/00004583-200201000-00014. [DOI] [PubMed] [Google Scholar]
  57. Lovejoy MC, Graczyk PA, O'Hare E, Neuman G. Maternal depression and parenting behavior: A meta-analytic review. Clinical Psychology Review. 2000;20:561–592. doi: 10.1016/s0272-7358(98)00100-7. [DOI] [PubMed] [Google Scholar]
  58. Luster T, Dubow E. Predictors of the quality of the home environment that adolescent mothers provide for their school-aged children. Journal of Youth & Adolescence. 1990;19:475–494. doi: 10.1007/BF01537476. [DOI] [PubMed] [Google Scholar]
  59. Mammen OK, Kolko DJ, Pilkonis PA. Negative affect and parental aggression in child physical abuse. Child Abuse & Neglect. 2002;26:407–424. doi: 10.1016/s0145-2134(02)00316-2. [DOI] [PubMed] [Google Scholar]
  60. Mayer SE. What money can't buy. Harvard University Press; Cambridge, MA: 1997. [Google Scholar]
  61. McKeering H, Pakenham KI. Gender and generativity issues in parenting: Do fathers benefit more than mothers from involvement in child care activities? Sex Roles. 2000;43:459–480. [Google Scholar]
  62. McKenry PC, Kotch JB, Browne DH. Correlates of dysfunctional parenting attitudes among low-income adolescent mothers. Journal of Adolescent Research. 1991;6:212–234. [Google Scholar]
  63. McLoyd VC. The impact of economic hardship on black families and children: Psychological distress, parenting, and socioemotional development. Child Development. 1990;61:311–346. doi: 10.1111/j.1467-8624.1990.tb02781.x. [DOI] [PubMed] [Google Scholar]
  64. Mueller RO. Structural equation modeling: Back to basics. Structural Equation Modeling. 1997;4:353–369. [Google Scholar]
  65. Nair P, Schuler ME, Black MM, Kettinger L, Harrington D. Cumulative environmental risk in substance abusing women: Early intervention, parenting stress, child abuse potential and child development. Child Abuse & Neglect. 2003;27:997–1017. doi: 10.1016/s0145-2134(03)00169-8. [DOI] [PubMed] [Google Scholar]
  66. Newcomb MD, Loeb TB. Poor parenting as an adult problem behavior: General deviance, deviant attitudes, inadequate family support and bonding, or just bad parents? Journal of Family Psychology. 1999;13:175–193. [Google Scholar]
  67. O'Brien M, Peyton V. Parenting attitudes and marital intimacy: A longitudinal analysis. Journal of Family Psychology. 2002;16:118–127. [PubMed] [Google Scholar]
  68. Osofsky JD, Culp AM. Perceptions of infant emotions in adolescent mothers. In: Emde RN, Osofsky JD, Butterfield PM, editors. The I FEEL pictures: A new instrument for interpreting emotions. International Universities Press; Madison, CT: 1993. pp. 149–160. [Google Scholar]
  69. Pascoe JM, Solomon R. Prenatal correlates of indigent mothers' attitudes about spoiling their young infants: A longitudinal study. Journal of Developmental & Behavioral Pediatrics. 1994;15:367–369. [PubMed] [Google Scholar]
  70. Peterson BE, Smirles KA, Wentworth PA. Generativity and authoritarianism: Implications for personality, political involvement, and parenting. Journal of Personality and Social Psychology. 1997;72:1202–1216. [Google Scholar]
  71. Pinderhughes EE, Dodge KA, Bates JE, Pettit GS, Zelli A. Discipline responses: Influences of parents' socioeconomic status, ethnicity, beliefs about parenting, stress, and cognitive-emotional processes. Journal of Family Psychology. 2000;14:380–400. doi: 10.1037//0893-3200.14.3.380. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Powers SI. Moral judgment development within the family. Journal of Moral Education. 1988;17:209–219. [Google Scholar]
  73. Pratt MW, Danso HA, Arnold ML, Norris JE, Filyer R. Adult generativity and the socialization of adolescents: Relations to mothers' and fathers' parenting beliefs, styles, and practices. Journal of Personality. 2001;69:89–120. doi: 10.1111/1467-6494.00137. [DOI] [PubMed] [Google Scholar]
  74. Radloff LS. The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401. [Google Scholar]
  75. Radloff LS. The use of the Center for Epidemiologic Studies Depression Scale in adolescents and young adults. Journal of Youth and Adolescence. 1991;20:149–166. doi: 10.1007/BF01537606. [DOI] [PubMed] [Google Scholar]
  76. Richardson B, Foster VA, McAdams CR. Parenting attitudes and moral development of treatment foster parents: Implications for training and supervision. Child & Youth Care Forum. 1998;27:409–431. [Google Scholar]
  77. Rosenthal DA, Gurney RM, Moore SM. From trust to intimacy: A new inventory for examining Erikson's stages of psychosocial development. Journal of Youth and Adolescence. 1981;10:525–537. doi: 10.1007/BF02087944. [DOI] [PubMed] [Google Scholar]
  78. Sameroff AJ, Seifer R, Barocas R, Zax M, Greenspan S. Intelligence quotient scores of 4-year-old children: Social-environmental risk factors. Pediatrics. 1987;79:343–350. [PubMed] [Google Scholar]
  79. Sares TA. Sociopolitical viewpoints as narrated by family and educational background. Journal of Social Psychology. 1998;138:637–644. [Google Scholar]
  80. Shaw DS, Criss MM, Schonberg MA, Beck JE. The development of family hierarchies and their relation to children's conduct problems. Development and Psychopathology. 2004;16:483–500. doi: 10.1017/s0954579404004638. [DOI] [PubMed] [Google Scholar]
  81. Shonkoff JP, Phillips DA. From neurons to neighborhoods: The science of early childhood development. National Academy Press; Washington, DC: 2000. [PubMed] [Google Scholar]
  82. Sigel IE, McGillicuddy-DeLisi AV. Parent beliefs are cognitions: The dynamic belief systems model. In: Bornstein MH, editor. Handbook of parenting. 2nd ed. Erlbaum; Mahwah, NJ: 2002. pp. 485–508. [Google Scholar]
  83. Silk JS, Sessa FM, Morris AS, Steinberg L, Avenevoli S. Neighborhood cohesion as a buffer against hostile maternal parenting. Journal of Family Psychology. 2004;18:135–146. doi: 10.1037/0893-3200.18.1.135. [DOI] [PubMed] [Google Scholar]
  84. Spielberger CD. Manual for the State-Trait Anxiety Inventory (Form Y) Consulting Psychologists Press; Palo Alto, CA: 1983. [Google Scholar]
  85. Strom R, Hathaway C, Slaughter H. The correlation of maternal attitudes and preschool children's performance on the McCarthy Scales of Children's Abilities. Journal of Instructional Psychology. 1981;8:139–145. [Google Scholar]
  86. Tamis-LaMonda CS, Shannon JD, Cabrera NJ, Lamb ME. Fathers and mothers at play with their 2- and 3-year-olds: contributions to language and cognitive development. Child Development. 2004;75:1806–1820. doi: 10.1111/j.1467-8624.2004.00818.x. [DOI] [PubMed] [Google Scholar]
  87. Thelen E, Smith LB. Dynamic systems theories. In: Damon W, Lerner RM, editors. Handbook of child psychology: Vol. 1. Theoretical models of human development. 5th ed. John Wiley & Sons; New York: 1998. pp. 563–634. Series Ed. Vol. Ed. [Google Scholar]
  88. Thompson RA, Christiansen EH, Jackson S, Wyatt JM, Colman RA, Peterson RL, et al. Parent attitudes and discipline practices: Profiles and correlates in a nationally representative sample. Child Maltreatment. 1999;4:316–330. doi: 10.1016/s0145-2134(98)00108-2. [DOI] [PubMed] [Google Scholar]
  89. West SG, Finch JF, Curran PJ. Structural equation models with nonnormal variables: Problems and remedies. In: Hoyle R, editor. Structural equation modeling: Issues and applications. Sage; Thousand Oaks, CA: 1995. pp. 56–75. [Google Scholar]
  90. Whaley SE, Pinto A, Sigman M. Characterizing interactions between anxious mothers and their children. Journal of Consulting and Clinical Psychology. 1999;67:I 826–836. doi: 10.1037//0022-006x.67.6.826. [DOI] [PubMed] [Google Scholar]
  91. Woodruff-Borden J, Morrow C, Bourland S, Cambron S. The behavior of anxious parents: Examining mechanisms of transmission of anxiety from parent to child. Journal of Clinical Child and Adolescent Psychology. 2002;31:364–374. doi: 10.1207/S15374424JCCP3103_08. [DOI] [PubMed] [Google Scholar]
  92. Zuravin SJ. Child maltreatment and teenage first births: A relationship mediated by chronic sociodemographic stress? American Journal of Orthopsychiatry. 1988;58:91–103. doi: 10.1111/j.1939-0025.1988.tb01569.x. [DOI] [PubMed] [Google Scholar]

RESOURCES