This article was downloaded by: [Weis, Robert] On: 8 May 2009 Access details: Access Details: [subscription number 905003565] Publisher Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Early Child Development and Care Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t713640830
Parenting across cultural contexts in the USA: assessing parenting behaviour in an ethnically and socioeconomically diverse sample Robert Weis a; Erin E. Toolis a a Department of Psychology, Denison University, Granville, Ohio, USA First Published on: 31 October 2008
To cite this Article Weis, Robert and Toolis, Erin E.(2008)'Parenting across cultural contexts in the USA: assessing parenting
behaviour in an ethnically and socioeconomically diverse sample',Early Child Development and Care, To link to this Article: DOI: 10.1080/03004430802472083 URL: http://dx.doi.org/10.1080/03004430802472083
PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.
Early Child Development and Care 2008, 1–18, iFirst Article
Parenting across cultural contexts in the USA: assessing parenting behaviour in an ethnically and socioeconomically diverse sample Robert Weis* and Erin E. Toolis Department of Psychology, Denison University, Granville, Ohio, USA (Received 12 August 2008; final version received 10 September 2008) Taylor and Francis GECD_A_347376.sgm
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Early 10.1080/03004430802472083 0300-4430 Original Taylor 02008 00
[email protected] RobertWeis 000002008 Childhood &Article Francis (print)/1476-8275 Development(online) and Care
This study examined the factorial validity of the Parent Behavior Inventory (PBI), a brief measure of parenting behaviour, in a diverse sample of mothers and compared differences in parenting as a function of ethnicity and socioeconomic status (SES). Exploratory and confirmatory factor analyses were performed on data generated by mothers of young children (N = 1705). Analyses indicated a tripartite structure with dimensions reflecting Warmth, Control and Hostility, respectively. Scores differed as a function of ethnicity and SES: European American and high-SES mothers reported greatest Warmth whereas African American and high-SES mothers reported greatest Control. A significant ethnicity X SES interaction was found for Hostility: Low-SES African American and high-SES Latina mothers reported the most hostile–coercive behaviours. Hostility among Latina mothers was associated with employment outside the home. Results support the factorial validity of the PBI with diverse populations and offer normative data for ideographic assessment of parenting behaviour across sociocultural contexts. Keywords: parenting behaviour; mothers; measurement; ethnicity; race; socioeconomic status (SES)
The last four decades have seen considerable developments in parenting research. Baumrind’s (1967) description of authoritative, authoritarian and permissive parenting provided an initial framework upon which to organise our understanding of parenting behaviour and its effects on child development. More recently, attention has been directed at identifying the dimensions of parenting behaviour that might be useful to researchers interested in studying family functioning or clinicians charged with the task of improving problematic parent–child interactions (Maccoby & Martin, 1983). Most studies suggest that a combination of parental responsiveness and demandingness leads to optimal academic and social–emotional development in children (Leyendecker, Harwood, Comparini, & Yalcinkaya, 2005). However, generalisations about the dimensions of parenting that lead to the best child outcomes downplay the great variability in parenting behaviour seen in families from diverse cultural backgrounds. Unfortunately, researchers have only recently turned their attention to the measurement of parenting behaviour and the quality of parent–child interactions in ethnically and socioeconomically diverse families. Most existing measures of parenting behaviour have not been validated for use with parents who are not members of the dominant culture. This is unfortunate, because European American, middle-class families constitute a shrinking percentage of the US population. Furthermore, parenting behaviour is best understood in the family’s *Corresponding author. Email:
[email protected] ISSN 0300-4430 print/ISSN 1476-8275 online © 2008 Taylor & Francis DOI: 10.1080/03004430802472083 http://www.informaworld.com
2
R. Weis and E. E. Toolis
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
sociocultural context (Hill, 2006). The strategies and tactics that parents use to socialise their children are a product of culturally specific values and customs (e.g. ethnicity, language) and the family’s ecological niche in society (e.g. socioeconomic status (SES); García Coll et al., 1996). The assessment of parenting in ethnically and socioeconomically diverse families depends on measures of parenting behaviour that have been developed from samples of these families. The primary purpose of this study was to examine the factorial validity of a measure of parenting behaviour designed for parents from diverse ethnic and socioeconomic backgrounds. Dimensions of parenting behaviour Previous research has identified two dimensions of parenting behaviour important to children’s functioning. The first dimension has been labelled parental ‘responsiveness’ or ‘warmth’ and reflects the degree to which parents show acceptance of their children through affection, shared activities and emotional or tangible support (Maccoby & Martin, 1983). Responsive parents are sensitive to their children’s needs and signals, provide developmentally appropriate assistance and communicate to their children that they are worthy of the attention and care of others. Parental responsiveness is associated with children’s social–emotional well-being across development. The second dimension of parenting, ‘demandingness’, is a more complex construct that likely reflects at least two facets of parental behaviour. The first facet, behavioural control, refers to parents’ attempts to regulate children’s behaviour in a manner that promotes individuation and self-determination (Barber, 2002; Barber, Stolz, & Olsen, 2005). Behavioural control involves clearly communicated rules, high expectations for child behaviour, parental supervision, consistent discipline and rewards to encourage age-appropriate actions. Elements of behavioural control have been described by several researchers using slightly different terms including firm control (Schaefer, 1965), authoritative control (Baumrind, 1966) and parental demandingness (Steinberg, 1990). Regardless of the label, the goal of behavioural control is to promote the regulation of children’s behaviour, to foster the internalisation of rules, to increase behavioural inhibition and emotion regulation, and to permit self-determination of child behaviour within the constraints of society. The second facet, hostility-coercion, refers to socialisation tactics that are not sensitive and responsive to children’s needs and that inhibit the development of children’s autonomy and capacity for self-direction (Barber, 2002). Threats, power assertion, love withdrawal, guilt, criticism, ridicule, nagging or harsh physical discipline can reflect hostility-coercion. Aspects of parental hostility-coercion can be seen in previous authors’ descriptions of psychological discipline (Becker, 1964), the authoritarian parenting type (Baumrind, 1967), power assertion (Hoffman, 1970), coercive family process (Patterson, Reid, & Dishion, 1992), threat-oriented parenting (Bugental, 1992) and intrusive parenting (Barber, 2002). Rather than encouraging children’s behaviour regulation, hostile-coercive actions are largely designed to increase parents’ power over their children. Parents’ hostile-coercive displays are associated with the development of mood and externalising behaviour problems in children. Parenting behaviour across cultural contexts Although considerable research has been directed at understanding and measuring the dimensions of parenting behaviour in European American, middle-class families, far less attention has focused on examining these dimensions in other ethnic and socioeconomic
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Early Child Development and Care
3
groups. Indeed, previous research has shown that parenting behaviour varies as a function of ethnicity. African American parents are more likely than European American parents to adopt ‘no-nonsense’, authoritarian parenting styles (Brody & Flor, 1998; Hill, McBride Murry, & Anderson, 2005). Specifically, African American parents tend to show lower levels of overt affection and emotional support towards their children than European American parents and they more often rely on physical discipline to gain children’s compliance (Dodge, McLoyd, & Lansford, 2005; Hill et al., 2005; McLoyd & Smith, 2002). These differences in parenting behaviour likely reflect the fact that African American parents tend to value respect for authority and quick compliance with adults’ commands (Hill, 2006). Studies examining parenting behaviour in Latino families have yielded mixed results (Hill, 2006). Some studies indicate that Latino parents are more likely than European American parents to rely on harsh, physical discipline whereas other studies suggest that Latino parents are more permissive than their European American counterparts (Cauce & DomenechRodríguez, 2002; Eisenberg, 1999; Hill, Bush, & Roosa, 2003). These inconsistent findings may reflect the heterogeneity of the Latino samples that have been studied. For example, low levels of responsiveness and high levels of physical discipline are more common among less acculturated Latinos than more acculturated Latinos, perhaps reflecting the importance many recent Latino immigrants place on respect for parents and conformity to traditional roles in the family (Roosa, Morgan-Lopez, Cree, & Specter, 2002). Emerging data also indicate that the effects of parenting behaviour on children’s development may be partially dependent on ethnicity. For example, harsh physical discipline is consistently associated with externalising behaviour problems in European American children, whereas physical discipline may protect certain African American children from conduct problems in adolescence (Gershoff, 2002; Lansford, Deater-Deckard, Dodge, Bates, & Pettit, 2004). Similarly, highly acculturated Latino parents who rely on physical discipline may place their children at risk for behaviour and emotional problems; in contrast, less acculturated Latinos who use physical discipline have children with fewer conduct problems (Hill, 2006). Physical discipline administered by ethnic minority parents might communicate concern for children’s welfare and, therefore, carry a different meaning than most instances of physical discipline administered by European American parents (McLoyd & Smith, 2002). Parenting behaviour also differs as a function of SES. SES is a multifaceted variable that reflects the family’s financial capital (e.g. income, material resources), human capital (e.g. parental education, knowledge of child development) and social capital (e.g. social networks and support; Entwisle & Astone, 1994). For the parents of young children, maternal education and, to a somewhat lesser extent, family income are two of the best predictors of the quality of parental care (Bornstein, Hahn, Suwalsky, & Haynes, 2003; Leyendecker et al., 2005). Low-SES parents are more likely than high-SES parents to value obedience, conformity and self-reliance in their children. On average, low-SES parents adopt authoritarian parenting styles, rely on physical discipline as a primary means to gain child compliance, display less warmth and greater disapproval towards their children, and discipline more inconsistently. In contrast, high-SES parents are more likely than low-SES parents to value independence, self-direction and open communication with their children. On average, high-SES parents adopt authoritative parenting styles, rely on discussion to resolve parent–child disputes, give greater attention to children’s emotions and structure activities around the needs of their children (Duncan & Magnuson, 2003; Hoff, Laursen, & Tardif, 2002). Unfortunately, many studies have confounded ethnicity and SES, by comparing middleclass European American parents with low-income ethnic minority parents (Hill, 2006).
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
4
R. Weis and E. E. Toolis
Recently, however, several well-designed studies have investigated the unique effects of ethnicity and SES on parenting, by holding either ethnicity or SES constant. Some studies found few differences in parenting after controlling for SES. For example, Hill and Bush (2001) found no differences in the affection or disciplinary strategies shown by African American and European American parents who had comparable incomes and educational attainment. Similarly, Solis-Camara and Fox (1995) found no differences in parental nurturance, discipline or expectations placed on children in a socioeconomically matched sample of Latino and European American parents. On the other hand, other studies have found significant differences in parenting as a function of ethnicity, even after controlling for SES. For example, African American parents may be more likely to endorse the use of mild physical punishment than European American parents regardless of socioeconomic background (Lansford et al., 2004; McLoyd & Smith, 2002; Pinderhughes, Dodge, Bates, Pettit, & Zelli, 2000). Furthermore, Mexican American parents may be more likely to use spanking, scolding and criticism than European American parents even after controlling for SES (MacPhee, Fritz, & Miller, 1996). These unique effects of ethnicity on parenting behaviour may reflect culturally specific values and beliefs about socialisation (Deater-Deckard, 2004). Minority parents of all socioeconomic backgrounds may be more likely to engage in physical discipline with their children to prepare them for life in a world that will judge them more harshly because of the colour of their skin, their native language or their customs (Dodge et al., 2005). The current study In this study, we administered a measure of parenting behaviour, the Parent Behavior Inventory (PBI; Lovejoy, Weis, O’Hare, & Rubin, 1999), to a large diverse sample of mothers of young children. We hypothesised that the PBI would show factorial validity with this sample by adequately measuring three orthogonal dimensions of parenting behaviour: warmth, control and hostility-coercion. Evidence of the PBI’s factorial validity, based on exploratory and confirmatory factor analysis, would provide initial justification for its use as a broad-band measure of parenting behaviour in families from diverse backgrounds. We also sought to compare mothers’ reports of warmth, control and hostility-coercion across ethnic and socioeconomic contexts. PBI ratings across ethnic and socioeconomic groups would be important in two ways. First, we hoped that the normative data generated by this study might be useful for professionals interested in the ideographic assessment of parenting behaviour in clinical settings. Our findings might allow clinicians to make inferences about ‘warm’ or ‘hostile-coercive’ parenting based on actuarial data rather than subjective impressions. Second, our data might be important to researchers interested in the nomothetic assessment of parenting behaviour in mothers from diverse cultural backgrounds. Indeed, we hypothesised that ethnicity and SES would be independently associated with maternal reports of caregiving. Based on previous research, we expected European American mothers to show greater warmth but lower control and hostility-coercion than African American and Latina mothers. Furthermore, we expected high-SES mothers to show greater warmth and control, but lower hostility-coercion, than low-SES mothers. Finally, we anticipated significant ethnicity X SES interactions; specifically, we expected ethnic differences in parenting behaviour to be greatest among low-SES families compared to high-SES families.
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Early Child Development and Care
5
Method Participants A total of 1705 mothers of preschool age children provided complete data for this study. Mothers ranged in age from 19 to 52 years (M = 31.46, SD = 6.54 years). Approximately 78% were married. Ethnicities included European American (39.8%), Latina (31.9%), African American (15.2%), Far Eastern (4.6%), Indian/Pakistani (3.8%), African (2.7%) and Middle Eastern (2%). Approximately 10% of mothers completed elementary school, 50% completed high school, 30% completed college and 10% completed graduate or professional school. Annual family incomes were as follows: 14.8% earned less than $20,000; 26.8% earned $20,000–$39,000; 25.2% earned $40,000–59,000; 21.5% earned $60,000–69,000; and 11.8% earned more than $70,000. Approximately 41% of mothers were employed outside the home. Among Latina mothers, 51.6% were born outside the USA. Latina immigrants were predominantly from Mexico (84%), El Salvador (3.9%), Columbia (3.6%) and Guatemala (3.1%). Mothers were asked to complete the PBI regarding their preschool age child. If mothers had more than one preschool age child, research assistants selected a target child to diversify the age and gender of the sample. Children ranged in age from two to six years (M = 4.11, SD = 1.46 years). Approximately 51% were boys. Approximately 32% of mothers had one child, 40% had two children, 18% had three children and 10% had four or more children.
Measure A modified version of the PBI (Lovejoy et al., 1999) was used to assess parenting behaviour. The original version of the PBI measured two broad-band, independent dimensions of parenting: support/engagement and hostility/coercion. Support/engagement reflects ‘behavior which demonstrates the parent’s acceptance of the child through affection, shared activities, and emotional and instrumental support’ (Lovejoy et al., 1999, p. 535). Hostility/coercion reflects behaviours ‘which express negative affect or indifference toward the child and may involve the use of coercion, threat, or physical punishment to influence the child’s behavior’ (Lovejoy et al., 1999, p. 535). Each dimension consisted of ten items, each rated on a six-point Likert scale. The modified version of the PBI, used in this study, included 10 additional items believed to measure ‘behavioural control’. These items reflect parents’ attempts to regulate children’s overt actions in a manner that promotes individuation and self-direction. To generate items for this dimension, we examined previously published measures of parenting behaviour designed for mothers of preschool and young school age children. Items that reflected parental limit-setting, monitoring and consistent use of discipline, but did not describe the use of hostile-coercive responses to child misbehaviour (e.g. spanking, yelling), were deemed appropriate for the current ‘behavioural control’ dimension. Specifically, we identified and modified items from the Parental Monitoring and Supervision subscale of the Alabama Parenting Questionnaire (Shelton, Frick, & Wootton, 1996), the Expectations subscale of the Parent Behavior Checklist (Brenner & Fox, 1999), the Laxness subscale of the Parenting Scale (Arnold, O’Leary, Wolff, & Acker, 1993), the Rigidity subscale of the Child Abuse Potential Inventory (Milner, 1986) and the Parent Practices Scale (Strayhorn & Weidman, 1988). Evidence for the reliability and factorial validity of the original version of the PBI comes from several studies involving predominantly European American, middle-class mothers of young children (Lovejoy et al., 1999). Construct validity is supported by meaningful
6
R. Weis and E. E. Toolis
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
correlations with other self-report measures of parenting behaviour and parenting stress, spousal reports of parenting behaviour and observations of parenting conducted in a laboratory setting.
Procedure Mothers were recruited from urban, suburban and rural areas in central and southeastern Ohio, central and southeastern Texas, and central and southern Wisconsin. Effort was made to gather data from an ethnically and socioeconomically diverse sample by recruiting from a wide range of neighbourhoods. Primary means of recruitment included solicitation outside grocery stores, shopping centres, restaurants, parks, zoos, public family events and paediatricians’/dentists’ offices. Research assistants described the study to mothers and requested oral consent to participate. Mothers were also given a written informed consent form to keep. Participation was anonymous. Mothers completed a one-page demographic questionnaire and the PBI and returned the questionnaires to research assistants immediately upon completion. Mothers were given the option to complete the PBI in English or Spanish; however, only data from the English version of the PBI are presented in this study. Complete data were provided by 1705 mothers. Mothers were randomly divided into two groups to investigate the factorial validity of the PBI. Data from approximately one-half of the sample were used to perform exploratory factor analysis (EFA; n = 853) whereas the remaining data were used to perform confirmatory factor analysis (CFA; n = 852).
Results Exploratory factor analysis We conducted a principal components analysis using data from the first subsample to estimate the number of latent variables assessed by the modified PBI. The number of components was determined using three criteria: previous research, the scree test (i.e. eigenvalues greater than two; Tabachnick & Fidell, 2007) and the interpretability of the component solution. Varimax rotation was used to produce more easily interpretable, orthogonal dimensions. The rotated solution (Table 1) yielded three interpretable factors: Warmth, Control, and Hostility. These components accounted for 16.16%, 14.96% and 14.33% of the variance, respectively. Cronbach’s alpha was .83 for the Warmth scale, .88 for the Control scale and .85 for the Hostility scale. Standardised parameter estimates and corrected item–total correlations for each scale are also presented in Table 1. The average item–total correlation for the Warmth scale was .53 with a range of .38 to .65. The average item–total correlation for the Control scale was .61 with a range of .47 to .70. The average item-total correlation for the Hostility scale was .56 with a range of .38 to .63. The bivariate correlation between the Warmth and Control scales was .37 (p < .001); the correlation between the Warmth and Hostility scales was −.33 (p < .001); the correlation between the Control and Hostility scale was .11 (p = .002).
Confirmatory factor analysis We performed CFA on responses from the second subsample of parents. The data were screened for violations of the assumptions of structural equation modelling. The data contained no multivariate outliers and conformed to the assumption of normality. Assumptions of homogeneity of variance–covariance matrices and linearity were met
Early Child Development and Care Table 1.
7
Standardised parameter estimates for the three-factor model. Factor loading
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Item number and content
Warmth Control Hostility
M
SD
Corrected item–total correlation
10. I laugh with my child about things we find funny. 16. I thank or praise my child. 7. My child and I hug and kiss each other. 4. I try to teach my child new things. 22. I comfort my child when he/she seems scared or upset. 13. My child and I play games, do crafts or read together. 25. I hold or touch my child in a loving way. 28. I listen to my child’s feelings and try to understand them. 19. I help my child with things he/she is doing. 1. I have pleasant talks with my child.
.700
—
—
4.60
.691
.594
.692 .663 .647 .641
— — — —
— — — —
4.51 4.49 4.40 4.55
.756 .880 .822 .813
.651 .527 .596 .507
.602
—
—
4.32
.873
.549
.557 .539
— —
— —
4.45 4.34
.929 .912
.380 .550
.503
—
—
4.29
.908
.471
.487
—
—
4.41
.824
.471
14. When I ask my child to do something, I expect him/her to obey. 20. I have very strict rules for my child and he/she knows them. 5. I expect my child to listen to me when I give a command. 2. When my child makes a mess, I make him/ her clean it up. 8. When my child disobeys me, I am quick to discipline. 17. I do not allow my child to talk back. 23. I expect my child to use good manners while eating. 11. When my child’s clothes or face is dirty, I make him/her wash up. 26. I expect my child to be polite to adults. 29. I do not give in to my child’s temper tantrums.
—
.759
—
3.71 1.17
.543
—
.728
—
2.89 1.37
.395
—
.714
—
3.71 1.16
.496
—
.713
—
3.37 1.21
.414
—
.702
—
3.18 1.28
.446
— —
.669 .660
— —
3.44 1.27 3.69 1.20
.354 .462
—
.648
—
2.89 1.39
.320
.380 —
.615 .534
— —
3.85 1.32 3.23 1.33
.471 .349
—
—
.770
1.72 1.26
.627
—
—
.756
1.88 1.30
.590
— — — —
— — — —
.678 .664 .660 .659
1.47 0.81 1.08 1.21
.580 .612 .597 .582
9. I lose patience with my child and yell at him/her. 3. I lose my temper when my child disobeys me. 12. I complain about my child’s behaviour. 6. I grab or handle my child roughly. 18. I use physical punishment with my child. 15. I spank my child when he/she frustrates me.
1.30 1.06 1.25 1.33
8
R. Weis and E. E. Toolis
Table 1.
(Continued). Factor loading
Item number and content
Warmth Control Hostility
21. I threaten my child when he/she disobeys me. 24. I say mean things to my child that can make him/her feel bad. 27. When my child acts up, I remind my child how much I’ve sacrificed for him/ her. 30. When my child asks for help or attention, I ignore him/her or make him/her wait until later.
M
SD
Corrected item–total correlation
—
—
.637
1.50 1.42
.552
—
—
.588
0.69 1.07
.572
—
—
.577
0.92 1.28
.505
—
—
.403
1.10 1.40
.376
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Note: N = 853. Only loadings ≥ .30 are shown. Items are ordered by magnitude of factor loading for each scale.
(Tabachnick & Fidell, 2007). Confirmatory factor analysis using LISREL 8.5 (Jöreskog & Sörbom, 2003) was conducted to examine the fit of mothers’ responses to the three-factor model developed through principal components analysis. For comparison, nested alternative null, one- and two-factor models were also tested. The one-factor model posited a single latent variable, Parenting Behaviour, accounting for all maternal responses. The two-factor model posited two latent variables, Warmth-Control and Hostility, as separate but correlated factors. The interitem covariance matrix was analysed using the maximum likelihood method of estimation (Hu & Bentler, 1995). The following fit criteria were selected to evaluate the model: chi-square, goodness of fit index (GFI), comparative fit index (CFI), incremental fit index (IFI), normed fit index (NFI), root-mean-square residual (RMR) and root-mean-square error of approximation (RMSEA). Because of the broad-band nature of the PBI dimensions, the following criteria were used to indicate acceptable fit: GFI ≥ .80, CFI ≥ .80, IFI ≥ .80, NFI ≥ .80, RMR ≤ .08 and RMSEA ≤ .08 (Hu & Bentler, 1995). Results (Table 2) showed acceptable fit for the proposed three-factor model. The threefactor model provided significantly better fit than the null, ∆χ2(33) = 8305.86, p < .001, one-factor, ∆χ2(3) = 4537.72, p < .001 and two-factor, ∆χ2(2) = 2034.42, p < .001, models. Bivariate correlations for the three latent variables were as follows: .37 between Warmth and Control, −.33 between Warmth and Hostility, and .17 between Control and Hostility. Table 2.
Fit indices for confirmatory factor analyses.
Model Null One-factor Two-factor Three-factor
χ2
df
GFI
CFI
IFI
NFI
RMR
RMSEA
10391.70 6623.56 4120.26 2085.84
435 405 404 402
.33 .45 .63 .85
.00 .38 .63 .84
.00 .38 .63 .84
.00 .33 .60 .84
.25 .18 .13 .06
.26 .21 .14 .06
Note: N = 852; df = degrees of freedom; GFI = goodness of fit index; CFI = comparative fit index; IFI = incremental fit index; NFI = normed fit index; RMR = root-mean-square residual; RMSEA = root-mean-square error of approximation.
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Early Child Development and Care
9
Parenting differences as a function of ethnicity and SES We performed a multivariate analysis of covariance (MANCOVA) to examine differences in parenting behaviour as a function of ethnicity and SES. Ethnicity was determined by mothers’ self-identification of being European American, African American or Latina on the demographics form. Mothers who identified as some other ethnicity were not included in these analyses. SES was determined by a combination of mothers’ education and family income. Specifically, mothers were assigned one point for every level of educational attainment (i.e. elementary school, high school, college, professional) and one point for every level of family income (i.e. < $20,000; $20,000–$39,000; $40,000–59,000; $60,000–69,000; ≥ $70,000). Each mother’s educational and family income points were added together to calculate an estimate of SES ranging from two to nine. Then, mothers were divided into three SES groups based on their total scores: low-SES (scores 2–4), middle-SES (scores 5–7) and high-SES (scores 8–9). For the MANCOVA, the independent variables were maternal ethnicity (i.e. European American, African American, Latina) and SES (i.e. low, middle, high). The dependent variables were maternal ratings of parenting behaviour (i.e. Warmth, Control, Hostility). Marital status and child gender served as covariates. Assumptions of normality and linearity were met, although inspection showed possible heterogeneity of variance–covariance matrices (Tabachnick & Fidell, 2007). Consequently, we decided to use Pillai’s Trace to evaluate the multivariate test, given its robustness against heterogeneity and unequal sample sizes across cells (Olson, 1976). Descriptive statistics for the sample are presented in Table 3. Results showed a significant multivariate effect for ethnicity, Pillai’s Trace = .048, F(6, 2940) = 12.13, p < .001, η2 = .024, and a significant multivariate effect for SES, Pillai’s Trace = .021, F(6, 2940) = 5.22, p < .001, η2 = .011. We also found a significant ethnicity X SES interaction, Pillai’s Trace = .029, F(12, 4413) = 5.22, p < .001, η2 = .011. Table 3.
Parenting behaviour as a function of maternal ethnicity and socioeconomic status. Maternal ethnicity European American (n = 679)
Parenting dimension/SES
African American (n = 259)
Latina (n = 544)
M
SD
n
M
SD
n
Warmth Low Middle High
44.97 45.27 46.19
5.56 5.14 3.34
145 258 276
42.77 44.55 45.89
5.74 4.55 4.36
71 116 72
Control Low Middle High
31.17 33.06 36.59
8.86 9.15 6.90
35.24 36.44 36.40
7.79 8.96 7.84
31.47 9.10 32.58 8.65 35.12 5.13
Hostility Low Middle High
10.27 11.03 11.25
7.04 6.42 6.16
16.93 15.40 11.10
9.75 9.54 7.13
12.08 9.30 13.26 8.42 14.80 8.13
Note: N = 1482.
M
SD
42.57 6.13 43.78 5.73 45.50 3.97
n 331 163 50
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
10
R. Weis and E. E. Toolis
Maternal warmth We conducted follow-up analyses of covariance (ANCOVAs), evaluated at p < .017 to control for familywise error, to examine differences in maternal Warmth as a function of ethnicity and SES. Results showed a significant main effect for ethnicity F(2, 1471) = 10.73, p < .001, η2 = .014. Planned contrasts, evaluated at p < .017, showed that European American mothers reported greater Warmth than African American and Latina mothers, with no differences in Warmth between the latter two groups. Results also showed a significant main effect for SES, F(2, 1471) = 11.86, p < .001, η2 = .016. Planned contrasts, evaluated at p < .017, showed that high-SES mothers reported more Warmth than middle-SES mothers, whereas middle-SES mothers reported more Warmth than low-SES mothers. We did not find a significant ethnicity X SES interaction for maternal Warmth, F(4, 1471) = 1.67, p = .155. However, we conducted planned contrasts, evaluated at p < .005, despite the non-significant interaction, given that these comparisons were a focus of interest a priori (Keppel, Saufley, & Tokunaga, 1992). Planned contrasts indicated that low-SES European American mothers reported greater Warmth than low-SES African American or Hispanic mothers. However, high-SES mothers reported no differences in Warmth as a function of ethnicity. Maternal control We also conducted follow-up ANCOVAs, evaluated at p < .017, to examine differences in maternal Control as a function of ethnicity and SES. As before, results showed a significant main effect for ethnicity, F(2, 1471) = 10.83, p < .001, η2 = .015. Planned contrasts, evaluated at p < .017, revealed that African American mothers reported greater Control than European American and Latina mothers, with no differences in Control between the latter two groups. We also found a significant main effect for SES, F(2, 1471) = 6.43, p = .002, η2 = .01. Planned contrasts, evaluated at p < .017, indicated that high-SES mothers reported greater Control than middle- and low-SES mothers, with no differences between the latter two groups. We did not find a significant ethnicity X SES interaction for maternal Control, F(4, 1471) = 1.79, p = .128. However, planned contrasts, evaluated at p < .005, indicated that low-SES African American mothers reported significantly greater Control than lowSES European American and Latina mothers. However, high-SES parents did not differ in their reports of Control as a function of ethnicity. Maternal hostility We conducted a follow-up ANCOVA, evaluated at p < .017, to examine differences in maternal Hostility as a function of ethnicity and SES. We found a significant main effect for ethnicity, F(2, 1471) = 22.65, p < .001, η2 = .031. African American and Latina mothers reported comparable levels of Hostility towards their children and significantly higher levels than those reported by European American mothers. We did not find a significant main effect for SES, F(2, 1471) = .871, p = .419. We found a significant ethnicity X SES interaction for maternal Hostility, F(4, 1471) = 6.95, p < .001, η2 = .02. Planned contrasts, evaluated at p < .005, showed that low-SES African American mothers reported significantly greater Hostility than low-SES Latina and European American mothers. Furthermore, high-SES Latina mothers reported greater Hostility than high-SES European American and African American mothers (Figure 1).
Early Child Development and Care
11
16.00
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Mean Hostility Score
Maternal Ethnicity European American African American Latina
14.00
12.00
10.00
Low
Middle
High
SES Figure 1.
Hostility scores as a function of SES and ethnicity.
Figure 1. Hostility scores as a function of SES and ethnicity.
Hostility reported by Latina mothers We conducted two post-hoc analyses to investigate the significant difference between highSES Latina and non-Latina mothers. Specifically, we wanted to identify a mechanism that might explain why high-SES Latina mothers would report significantly greater Hostility than their European American and African American counterparts. In our first analysis, we conducted an ANCOVA (controlling for marital status and child gender) examining differences in Hostility as a function of SES and maternal country of birth among Latina mothers. We reasoned that high-SES mothers who were not born in the USA might display greater Hostility towards their children due to the double pressures of acculturation and upward social mobility (García Coll & Pachter, 2002). However, results did not indicate a significant effect for SES, F(2, 536) = 2.84, p = .059, country of birth, F(1, 536) = .293, p = .589, or their interaction, F(2, 536) = .646, p = .525. In our second analysis, we conducted an ANCOVA (controlling for marital status and child gender), examining differences in maternal Hostility as a function of SES and maternal employment status among Latina mothers. We reasoned that Latina mothers who were employed outside the home might display greater Hostility due to combined pressures associated with work- and child-related responsibilities. Indeed, previous research has indicated that Latina mothers who are employed outside the home might be at particular risk for
12
R. Weis and E. E. Toolis 20.00
Mother Employed Outside the Home?
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Mean Hostility Score
18.00
Yes No
16.00
14.00
12.00
10.00
Low
Middle
High
SES Figure 2.
Hostility scores as a function of SES and employment status for Latina mothers.
parenting stress because such employment may be at odds with traditional gender roles endorsed by some Latino families (Roehling, Hernandez Jarvis, & Swope, 2005). Marital and parenting stress might be especially pronounced for upwardly mobile Latina mothers who are attempting to manage both caregiving and work-related tasks. Results partially supported our hypothesis. Although we failed to find a significant main effect for SES, F(2, 536) = 2.11, p = .122, or a significant SES X employment status interaction, F(2, 536) = 1.96, p = .141, we did discover a significant main effect for maternal employment, F(1, 536) = 25.77, p < .001, η2 = .046. At all three levels of the SES spectrum (i.e. low, middle, high), Latina mothers who worked outside the home reported greater Hostility than their counterparts who did not work outside the home (Figure 2). In contrast, Hostility scores did not differ significantly between working (M = 11.35, SD = 6.56) and non-working (M = 10.71, SD = 6.38) European American mothers, t(677) = 1.26, p = .207, or between working (M = 15.01, SD = 9.20) and non-working (M = 14.29, SD = 9.30) African American mothers, t(257) = .623, p = .534. Figure 2. Hostility scores as a function of SES and employment status for Latina mothers.
Discussion Our findings provide initial evidence for the factorial validity of the modified PBI with a large sample of mothers from diverse ethnic and socioeconomic backgrounds. Results of
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Early Child Development and Care
13
EFA suggest that the PBI can be meaningfully understood in terms of three, relatively independent parenting dimensions reflecting Warmth, Control and Hostility, respectively. The addition of the Control scale adds to the PBI’s utility by allowing professionals to assess traditional dimensions of parenting (i.e. responsiveness, behavioural control) as well as aspects of parenting associated with children’s behaviour and emotional problems (i.e. hostility-coercion). Results of CFA indicate that the proposed tripartite division of the modified PBI provides an acceptable fit for the data generated by mothers from a wide range of sociocultural backgrounds. This finding extends the initial validation studies of the PBI factor structure, which were based on data provided by predominantly European American, middle-class parents (Lovejoy et al., 1999). Although future research might be directed at examining the concurrent and discriminant validity of the modified PBI with non-European American and low-SES parents, our results provide some of the first evidence that the measure reflects theoretically meaningful dimensions of parenting in a diverse sample. Professionals interested in the ideographic assessment of parenting might use the normative data presented in this study to interpret the ratings of individual mothers of specific ethnicities or socioeconomic groups. We found significant differences in maternal Warmth as a function of ethnicity and SES. Consistent with predictions, European American mothers reported significantly greater Warmth than did African American and Latina mothers. Similarly, high-SES mothers reported significantly greater Warmth than middle-SES mothers whereas middle-SES mothers reported greater Warmth than low-SES mothers. These findings are consistent with several previous studies indicating greater affection, emotional support and responsiveness shown by European American or high-SES mothers compared to other groups (Harwood, Leyendecker, Carlson, Asencio, & Miller, 2002; Hill et al., 2003; Hoff et al., 2002; McLoyd & Smith, 2002). We also found that differences between European American and non-European American mothers were most pronounced at the low end of the SES spectrum. As SES increased, differences in maternal Warmth as a function of ethnicity were reduced. Similar results have been observed in studies of African American and Latino parents, suggesting that many apparent ethnic differences in parenting behaviour may be partially attributable to differences in SES (Hill et al., 2003; Leyendecker et al., 2005). Also consistent with our hypotheses, we found that ethnicity and SES were related to the degree of Control reported by mothers. African American mothers reported greater Control than European American and Latina mothers; furthermore, high-SES mothers reported greater Control than low- and middle-SES mothers. These findings are consistent with previous research showing that African American mothers are more likely than European American parents to adopt ‘no-nonsense’ parenting styles that emphasise obedience (Dodge et al., 2005). Our results also agree with research suggesting that higher-SES parents are more likely to place high expectations on their children’s behaviour and discipline in a consistent manner (Hoff et al., 2002). As with the case of Warmth, we found that differences in maternal Control as a function of ethnicity were most pronounced among low-SES mothers. As SES increased, differences between African American and non-African American mothers were attenuated. These findings are consistent with previous research showing few differences between European American and African American mothers with similar incomes and educational attainment (Hill & Bush, 2001). Our results are also consistent with research suggesting that low-SES African American mothers, in particular, may rely on control to protect children from environmental risks associated with low-income, high-crime neighbourhoods (McLoyd, 1998). As expected, European American mothers reported significantly less Hostility than African American mothers overall. Furthermore, this significant main effect for ethnicity
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
14
R. Weis and E. E. Toolis
was partially attributable to the large difference between European American and African American mothers at the low end of the SES spectrum. As SES increased, differences in European American and African American mothers’ hostility towards their children were reduced. This finding is consistent with previous research suggesting that differences between European American and African American mothers’ harsh, coercive or inconsistent parenting practices may be largely due to differences in SES (Hill, 2006; Hill & Bush, 2001). However, our results are somewhat inconsistent with other studies showing that African American mothers are more likely to endorse higher levels of physical discipline than European American mothers, even after controlling for SES (Pinderhughes et al., 2000). It is likely that the meaning of physical discipline is different for European American and African American children, with physical discipline serving a protective function for African American youths (McLoyd & Smith, 2002). Future research might be directed at examining whether the meaning of hostile-coercive parenting behaviour that does not involve physical discipline (e.g. threats, love withdrawal) differs as a function of ethnicity. The role of ethnicity and SES in predicting maternal Hostility among Latina mothers was more complicated. Contrary to our expectations, high-SES Latina mothers reported significantly greater Hostility than their European American and African American counterparts. We initially suspected that mothers’ country of birth might explain this finding; perhaps high-SES Latina mothers experienced greater parenting stress due to the dual pressures of upward social mobility and difficulties adjusting to the dominant culture (García Coll & Pachter, 2002; García Coll et al., 1996; Roosa et al., 2002). This explanation was not supported. Our second hypothesis, that maternal employment was associated with the higher levels of Hostility reported by high-SES Latina mothers was partially supported. We found a significant main effect for maternal employment; low-, middle- and high-SES Latina mothers who worked outside the home reported greater Hostility than Latina mothers who stayed at home with their children. The percentage of Latina mothers employed outside the home increased markedly across the low- (29%), middle- (40%) and high-SES (58%) groups. Therefore, the tendency for high-SES Latina mothers to report greater levels of Hostility than low- and middle-SES Latina mothers was partially driven by high-SES Latina mothers’ greater likelihood of employment. Why might employment place Latina mothers at risk for hostile-coercive displays towards their children? In a nationally representative and ethnically diverse sample of working adults, Roehling et al. (2005) found an association between maternal employment and stress in women’s marital and caregiving relationships. The authors used the term ‘work-tofamily spillover’ to describe the phenomena in which strains and conflicts associated with employment adversely affect mothers’ mood and interactions with her family. In their analyses, Latina working mothers showed greater work-to-family spillover than European American and African American mothers. The authors found that Latina mothers who are employed outside the home were at particular risk for work-to-family spillover because Latino families, on average, endorse more traditional gender-roles than European American and African American families (Kane, 2000). For example, many Latina mothers endorse the traditional gender role of the marianista woman, who sacrifices her needs for the welfare of her family, who is devoted to her husband, and who views her primary responsibility as the care of her children. Working Latina mothers who endorse the marianismo gender role may be overburdened with the dual responsibilities of homemaking and contributing to the family’s financial welfare. Consequently, role overload may make these mothers more likely to engage in harsh, coercive or inconsistent parenting behaviours towards their children (Roehling et al., 2005). African American and European American mothers,
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Early Child Development and Care
15
in contrast, show more egalitarian gender roles. Indeed, Hostility scores did not differ between working and non-working African American and European American mothers in our study. One of the primary limitations of this study was its reliance on maternal self-reports to estimate parenting behaviour. Although self-reports are a time- and cost-effective method to gather data from large numbers of parents, they are susceptible to response biases (e.g. social desirability, demand characteristics) that can compromise the internal validity of the study. Future research might be directed at examining whether the differences in parenting behaviour reported in this study might also be independently observed by others. The demographic makeup of participants in our study may also limit its generalisability. First, 78% of the mothers in our sample were married. In comparison, 74% of European American, 66% of Latina and 37% of African American mothers in the USA population are married (U.S. Census Bureau, 2008). Although martial status is not consistently associated with the quality of maternal parenting behaviour (Gibson-Davis, 2008), unmarried mothers are at risk for financial and psychosocial stressors that can compromise the care they afford their children (DeKlyen, McLanahan, Brooks-Gunn, & Knab, 2006). Furthermore, the deleterious effects of stress on parenting behaviour may be especially pronounced for ethnic minority mothers without partners (Ceballo & McLoyd, 2002). The results of our study, based largely on reports from married mothers, may not generalise to non-married parents. Second, we examined only mothers’ reports of parenting behaviour; the factorial validity of the revised PBI with fathers is unknown. Previous research suggests that mothers and fathers often engage in the same types of parenting behaviours, albeit to different degrees (Adamsons & Buehler, 2007; Verhoeven, Junger, Van Aken, Dekovic, & Van Aken, 2007). For example, mothers of young children often display greater warmth and responsiveness than fathers (Calzada, Eyberg, Rich, & Querido, 2004). However, mother–father differences in parenting also depend heavily on the cultural context of the family (Gamble, Ramakumar, & Diaz, 2007). Future research may be directed at investigating the factorial invariance and mean-level similarity of the revised PBI with both mothers and fathers from diverse ethnic and socioeconomic backgrounds. Despite these limitations, our study provides initial evidence for the factorial validity of the modified PBI as a measure of maternal Warmth, Control and Hostility in an ethnically and socioeconomically diverse sample. We hope that the normative data generated by this study might be used as a starting point for professionals interested in assessing individual parents from diverse backgrounds. Our research also highlights the importance of considering parents’ cultural contexts while assessing their interactions with their children, given the unique effects of ethnicity and SES on maternal behaviour. Finally, we hope that our findings can be added to the growing body of quantitative research on parenting in Latino families. Our data indicate that Latino ethnicity may present unique benefits and challenges to the process of socialisation not experienced by many European American and African American mothers. Perhaps our findings may be used to direct future research on Latino families of all socioeconomic strata who have been neglected in the research literature for too long.
Acknowledgements This research was supported by a grant from the Carl Arvid Anderson Family Fund for Science Research. Portions of the data were presented in a symposium on parenting and ethnicity at the 2008 meeting of the Midwestern Psychological Association, Chicago, Illinois.
16
R. Weis and E. E. Toolis
Notes on contributors Robert Weis is a Licensed Psychologist and an Assistant Professor of Clinical Psychology at Denison University, a selective liberal arts college in central Ohio. His research interests are in psychometrics, program evaluation, and parent-child interactions. Erin E. Toolis is currently a Programme Manager at Sanctuary Art Center, a non-profit outreach programme for disadvantaged youths in Seattle, Washington. Her research and clinical interests are in the assessment at treatment of at-risk adolescents.
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
References Adamsons, K., & Buehler, C. (2007). Mothering and fathering versus parenting: Measurement equivalence in parenting measures. Parenting: Science and Practice, 7, 271–303. Arnold, D.S., O’Leary, S., Wolff, L.S., & Acker, M.M. (1993). The parenting scale: A measure of dysfunctional parenting in discipline situations. Psychological Assessment, 5, 137–144. Barber, B.K. (2002). Intrusive parenting: How psychological control affects children and adolescents. Washington, DC: American Psychological Association. Barber, B.K., Stolz, H.E., & Olsen, J.A. (2005). Parental support, psychological control, and behavioral control: Assessing relevance across time, culture, and method. Monographs of the Society for Research in Child Development, 70, 1–137. Baumrind, D. (1966). Effects of authoritative control on child behavior. Child Development, 37, 887–907. Baumrind, D. (1967). Childcare practices anteceding three patterns of preschool behavior. Genetic Psychology Monographs, 75, 43–88. Becker, W.C. (1964). Consequences of different kinds of parental discipline. In M.L. Hoffman & L.W. Hoffman (Eds.), Review of child development research (pp. 509–535). New York: Russell Sage Foundation. Bornstein, M.H., Hahn, C., Suwalsky, J.T.D., & Haynes, O.M. (2003). Socioeconomic status, parenting, and child development: The Hollingshead four-factor index of social status and the socioeconomic index of occupations. In M.H. Bornstein & R.H. Bradley (Eds.), Socioeconomic status, parenting, and child development (pp. 29–82). Mahwah, NJ: Erlbaum. Brenner, V., & Fox, R. (1999). An empirically derived classification of parenting practices. Journal of Genetic Psychology, 160, 343–347. Brody, G., & Flor, D. (1998). Maternal resources, parenting practices, and child competence in rural, single-parent African American families. Child Development, 69, 803–816. Bugental, D.B. (1992). Affective and cognitive processes within threat-oriented family systems. In I.E. Sigel, A.V. McGillicuddy-DeLisi, & J.J. Goodnow (Eds.), Parental belief systems: The psychological consequences for children (pp. 219–248). Hillsdale, NJ: Erlbaum. Calzada, E.J., Eyberg, S.M., Rich, B., & Querido, J.G. (2004). Parenting disruptive preschoolers: Experiences of mothers and fathers. Journal of Abnormal Child Psychology, 32, 203–213. Cauce, A.M., & Domenech-Rodríguez, M. (2002). Latino families: Myths and realities. In J.M. Contreras, K.A. Kearns, & A.M. Neal-Barnett (Eds.), Latino children and families in the United States (pp. 3–25). Westport, CT: Praeger. Ceballo, R., & McLoyd, V.C. (2002). Social support and parenting in poor, dangerous neighborhoods. Child Development, 73, 1310–1321. Deater-Deckard, K. (2004). Parenting stress. New Haven, CT: Yale University Press. DeKlyen, M., McLanahan, S., Brooks-Gunn, J., & Knab, J. (2006). The mental health of married, cohabitating, and non-coresident parents with infants. American Journal of Public Health, 96, 1836–1841. Dodge, K.A., McLoyd, V.C., & Lansford, J.E. (2005). The cultural context of physically disciplining children. In V.C. McLoyd, N.E. Hill, & K.A. Dodge (Eds.), African American family life: Ecological and cultural diversity (pp. 245–263). New York: Guilford. Duncan, G.J., & Magnuson, K.A. (2003). Off with Hollingshead: Socioeconomic resources, parenting, and child development. In M.H. Bornstein & R.H. Bradley (Eds.), Socioeconomic status, parenting, and child development (pp. 83–106). Mahwah, NJ: Erlbaum.
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
Early Child Development and Care
17
Eisenberg, A.R. (1999). Emotional talk among Mexican American and Anglo American mothers and children from two social classes. Merrill-Palmer Quarterly, 45, 267–284. Entwisle, D.R., & Astone, N.M. (1994). Some practical guidelines for measuring youths’ race/ ethnicity and socioeconomic status. Child Development, 65, 1521–1540. Gamble, W.C., Ramakumar, S., & Diaz, A. (2007). Maternal and paternal similarities and differences in parenting: An examination of Mexican-American parents of young children. Early Childhood Research Quarterly, 22, 72–88. García Coll, C.G., Lamberty, G., Jenkins, R., McAdoo, H.P., Crnic, K., Wasik, B.H., et al. (1996). An integrative model for the study of developmental competencies in minority children. Child Development, 67, 1891–1914. García Coll, C.G., & Pachter, L.M. (2002). Ethnic and minority parenting. In M.H. Bornstein (Ed.), Handbook of parenting: Vol. 4. Social conditions and applied parenting (pp. 1–20). Mahwah, NJ: Erlbaum. Gershoff, E.T. (2002). Corporal punishment by parents and associated child behaviors and experiences: A meta-analytic and theoretical review. Psychological Bulletin, 128, 539–579. Gibson-Davis, C.M. (2008). Family structure effects on maternal and paternal parenting in lowincome families. Journal of Marriage and the Family, 70, 452–465. Harwood, R., Leyendecker, B., Carlson, V., Asencio, M., & Miller, A. (2002). Parenting among Latino families in the U.S. In M.H. Bornstein (Ed.), Handbook of parenting: Vol. 4. Social conditions and applied parenting (pp. 21–46). Mahwah, NJ: Erlbaum. Hill, N.E. (2006). Disentangling ethnicity, socioeconomic status and parenting: Interactions, influences and meaning. Vulnerable Children and Youth Studies, 1, 114–124. Hill, N.E., & Bush, K. (2001). Relations between parenting environment and children’s mental health among African-American and Euro-American mothers and children. Journal of Marriage and the Family, 63, 954–966. Hill, N.E., Bush, K.R., & Roosa, M.W. (2003). Parenting and family socialization strategies and children’s mental health: Low-income Mexican-American and Euro-American mothers and children. Child Development, 74, 189–204. Hill, N.E., McBride Murry, V., & Anderson, V.D. (2005). Sociocultural contexts of African American families. In V.C. McLoyd, N.E. Hill, & K.A. Dodge (Eds.), African American family life: Ecological and cultural diversity (pp. 21–44). New York: Guilford. Hoff, E., Laursen, B., & Tardif, T. (2002). Socioeconomic status and parenting. In M.H. Bornstein (Ed.), Handbook of parenting: Vol. 2. Biology and ecology of parenting (pp. 231–252). Mahwah, NJ: Erlbaum. Hoffman, M.L. (1970). Conscience, personality, and socialization techniques. Human Development, 13, 90–126. Hu, L., & Bentler, P.M. (1995). Evaluating model fit. In R.H. Hoyle (Ed.), Structural equation modeling: Concepts, issues, and applications (pp. 76–99). Thousand Oaks, CA: Sage. Jöreskog, K.G., & Sörbom, D. (2003). LISREL 8.5. Chicago, IL: Scientific Software International. Kane, E. (2000). Racial and ethnic variations in gender-related attitudes. Annual Review of Sociology, 26, 419–439. Keppel, G., Saufley, W.H., & Tokunaga, H. (1992). Introduction to design and analysis. New York: W.H. Freeman. Lansford, J.E., Deater-Deckard, K., Dodge, K.A., Bates, J.E., & Pettit, G.S. (2004). Ethnic differences in the link between physical discipline and later adolescent externalizing. Journal of Child Psychology and Psychiatry, 45, 801–812. Leyendecker, B., Harwood, R.L., Comparini, L., & Yalcinkaya, A. (2005). Socioeconomic status, ethnicity, and parenting. In T. Luster & L. Okagaki (Eds.), Parenting: An ecological perspective (pp. 319–341). Mahwah, NJ: Erlbaum. Lovejoy, M.C., Weis, R., O’Hare, E., & Rubin, E.C. (1999). Development and initial validation of the Parent Behavior Inventory. Psychological Assessment, 11, 534–545. Maccoby, E.E., & Martin, J.A. (1983). Socialization in the context of the family: Parent–child interaction. In E.M. Hetherington (Ed.), Handbook of child psychology: Vol. 4. Socialization, personality, and social development (pp. 1–101). New York: Wiley. MacPhee, D., Fritz, J., & Miller, J. (1996). Ethnic variations in personal social networks and parenting. Child Development, 67, 3278–3295. McLoyd, V.C. (1998). Socioeconomic disadvantage and child development. American Psychologist, 53, 185–204.
Downloaded By: [Weis, Robert] At: 16:28 8 May 2009
18
R. Weis and E. E. Toolis
McLoyd, V.C., & Smith, J. (2002). Physical discipline and behavior problems in African American, European American, and Hispanic children: Emotional support as a moderator. Journal of Marriage and the Family, 64, 40–53. Milner, J.J. (1986). The Child Abuse Potential Inventory: Manual (2nd ed.). DeKalb, IL: Psytec. Olson, C.L. (1976). On choosing a test statistic in multivariate analysis of variance. Psychological Bulletin, 83, 579–586. Patterson, G.R., Reid, J.B., & Dishion, T.J. (1992). Antisocial boys. Eugene, OR: Castalia. Pinderhughes, E.E., Dodge, K.A., Bates, J.E., Pettit, G.S., & Zelli, A. (2000). Discipline responses: Influences of parents’ socioeconomic status, ethnicity, beliefs about parenting, stress, and cognitive-emotional processes. Journal of Family Psychology, 14, 380–400. Roehling, P.V., Hernandez Jarvis, L.H., & Swope, H.E. (2005). Variations in negative work-family spillover among European American, African American, and Hispanic American men and women: Does ethnicity matter? Journal of Family Issues, 26, 840–865. Roosa, M.W., Morgan-Lopez, A.A., Cree, W.K., & Specter, M.M. (2002). Ethnic culture, poverty, and context: Sources of influence on Latino families and children. In J.M. Contreras, K.A. Kearns, & A.M. Neal-Barnett (Eds.), Latino children and families in the United States (pp. 27–44). Westport, CT: Praeger. Schaefer, E.S. (1965). Children’s reports of parental behavior: An inventory. Child Development, 36, 413–424. Shelton, K.K., Frick, P.J., & Wootton, J. (1996). Assessment of parenting practices in families of elementary school-age children. Journal of Clinical Child Psychology, 25, 317–329. Solis-Camara, P., & Fox, A. (1995). Parenting among mothers with young children in Mexico and the United States. Journal of Social Psychology, 135, 591–599. Steinberg, L. (1990). Autonomy, conflict, and harmony in the family relationship. In S.S. Feldman & G.R. Elliott (Eds.), At the threshold: The developing adolescent (pp. 255–276). Cambridge, MA: Harvard University Press. Strayhorn, J.M., & Weidman, C.S. (1988). A Parent Practices Scale and its relation to parent and child mental health. Journal of the American Academy of Child & Adolescent Psychiatry, 27, 613–618. Tabachnick, B.G., & Fidell, L.S. (2007). Using multivariate statistics. Boston, MA: Pearson. U.S. Census Bureau. (2008). Current population survey: Annual social and economic supplement. Washington, DC: Author. Verhoeven, M., Junger, M., Van Aken, C., Dekovic, M., & Van Aken, M.A.G. (2007). A short-term longitudinal study of the development of self-reported parenting during toddlerhood. Parenting: Science and Practice, 7, 367–394.
