The Journal of Genetic Psychology, 2008, 169(3), 209226 Copyright 2008 Heldref Publications

Parenting Style as a Mediator Between Childrens Negative Emotionality and Problematic Behavior in Early Childhood
MARJA C. PAULUSSEN-HOOGEBOOM GEERT JAN J. M. STAMS JO M. A. HERMANNS THEA T. D. PEETSMA GODFRIED L. H. VAN DEN WITTENBOER University of Amsterdam

ABSTRACT. Negative emotionality is considered to be the core of the difficult temperament concept (J. E. Bates, 1989; R. L. Shiner, 1998). In this correlational study, the authors examined whether the relations between childrens negative emotionality and problematic behavior (internalizing and externalizing) were partially mediated by parenting style (authoritative and authoritarian) in a community sample of 196 3-year-old children and their mothers. The authors assessed maternal perception of child negative emotionality using the Childrens Behavior Questionnaire (M. K. Rothbart, S. A. Ahadi, K. L. Hershey, & P. Fisher, 2001) and assessed problematic child behavior by means of maternal report using the Child Behavior Checklist (T. M. Achenbach, 1992). The results showed that the relations between child negative emotionality and internalizing and externalizing behaviors were partially mediated by mothers authoritative parenting style. Moreover, when the authors used confirmatory factor analysis to decontaminate possible overlap in item content between measures assessing temperament and problematic behavior, the association between negative emotionality and internalizing behavior was fully mediated by authoritative parenting. Keywords: externalizing, internalizing, parenting, temperament

CHILDREN DIFFER FROM EACH OTHER beginning early in life, and these differences may have implications for parentchild interactions. Some important differences pertain to childrens temperament (Putnam, Sanson, & Rothbart, 2002). Although ideas about temperament go back to ancient Greco-Roman times, Thomas, Chess, Birch, Hertzig, and Korn (1963) conducted the first major study of temperament in children. Their New York Longitudinal Study (NYLS) identified
Address correspondence to Marja C. Paulussen-Hoogeboom, Department of Education, Faculty of Social and Behavioral Sciences, University of Amsterdam, PO Box 94208, 1090 GE Amsterdam, The Netherlands; [email protected] (e-mail).
209

210

The Journal of Genetic Psychology

nine dimensions of temperament: activity level, approachwithdrawal, adaptability, mood, threshold, intensity, distractibility, rhythmicity, and attention-span persistence (Thomas et al., 1963). They also developed a difficult temperament concept that included the negative poles of the dimensions approachwithdrawal, adaptability, mood, intensity, and rhythmicity. They concluded that difficult preschoolers in the NYLS were at increased risk for later behavioral and emotional problems (Thomas, Chess, & Birch, 1968). This finding has spurred much research examining the association between temperament characteristics and developmental outcomes. Although researchers have debated the definition of temperament over the past several decades, a consensus has emerged that the term refers to constitutionally based differences in behavioral style that are visible from the childs earliest years (Sanson, Hemphill, & Smart, 2004). Three broad aspects of temperament are gaining wide acceptance: negative emotionality, self-regulation, and a dimension variously labeled as approachwithdrawal, inhibition, or sociability (Sanson et al., 2004). Negative emotionality can be considered the core of the difficult temperament concept (Bates, 1989; Lee & Bates, 1985; Prior, 1992; Shiner, 1998). A widely accepted definition of negative emotionality is the childs tendency to react to stressors with high degrees of emotionality, including anger, irritability, fear, or sadness (Rothbart, Ahadi, & Hershey, 1994). Temperament is considered more a function of the biological makeup of the child and less a product of the childs interaction with caregivers (Zeanah & Fox, 2004). Researchers support the view of temperament as innate or biological and have found the genetic component of temperament to be fairly large. For example, Bokhorst et al. (2003) compared temperament in monozygotic and dizygotic twins and calculated that 77% of the variance in temperamental reactivity could be explained by genetic factors. Accordingly, temperament also has been shown to be moderately stable over time, with correlations ranging from .2 to .4, although stability may be higher (.7 < r < .8) if measurement error is taken into account (Sanson et al., 2004). Temperament is generally measured using parent questionnaires because researchers tend to be primarily interested in parental perceptions of temperament. Although some researchers have questioned the validity of parent report as an objective measure of child temperament (Mangelsdorf, Schoppe, & Buur, 2000), other researchers have argued that there is a strong objective component in parent ratings of child temperament (Rothbart & Bates, 1998). This last claim is supported by empirical studies showing convergence between parent ratings and observational assessments of temperament (e.g., Kochanska, 1995; Pauli-Pott, Mertesacker, Bade, Bauer, & Beckmann, 2000; Schuler, Black, & Starr, 1995). Temperament and Problematic Behavior In the decades following the NYLS, empirical evidence has accumulated showing that difficult temperament in early childhood is both concurrently and

Paulussen-Hoogeboom, Stams, Hermanns, Peetsma, & van den Wittenboer

211

prospectively related to internalizing behavior, such as anxiety, sadness, social withdrawal, and fearfulness, and externalizing behavior, such as overactivity, poor impulse control, noncompliance, aggression toward peers, and tantrums (Rothbart & Bates, 1998; Sanson et al., 2004). Most researchers focus on externalizing behavior, presumably because pure internalizing symptoms are difficult to identify and thus harder to study in young children (Campbell, 1995). When measured at the preschool age, however, externalizing and internalizing behaviors are substantially correlated in clinical and nonclinical samples (Campbell; Gilliom & Shaw, 2004; Gjone & Stevenson, 1997). This co-occurrence is not well understood. Although shared risk factors, genetic influences, and risk factors that stem from the development of an initial disorder constitute some of the proposed explanatory factors, few researchers to date have explored this phenomenon (Oland & Shaw, 2005). Regarding gender, Campbell concluded that the bulk of evidence suggests that gender differences in internalizing and externalizing behaviors are not marked in preschool-aged children. Although school-aged boys have a higher incidence of externalizing behavior, by early adolescence, girls shift toward more internalizing behavior (Achenbach, Howell, Quay, & Conners, 1991). Like temperament assessment, assessment of problematic behavior in preschool-aged children often involves adult-rated questionnaires (mostly by parents, sometimes combined with preschool or daycare teachers). The validity of parent report of problematic behavior has been demonstrated in several studies (Campbell, 1995). According to Campbell, preschool-aged children whom parents or teachers rate higher on externalizing behaviors are also more difficult to handle when they are observed interacting with teachers, parents, and peers in structured and unstructured situations. Parenting Behavior Researchers have consistently described parenting behavior along two dimensions: support and control (Maccoby & Martin, 1983; Rollins & Thomas, 1979). Support can be designated as parental behavior that makes the child feel comfortable in the relationship with his or her parent, fostering an internal representation in the child that he or she is basically accepted (Rollins & Thomas). Supportive parenting relates to constructs such as warmth, sensitivity, responsiveness, and acceptance and is considered essential for the formation of secure attachments and other positive developmental outcomes (e.g., Coplan, Hastings, Lagace-Seguin, & Moulton, 2002; van IJzendoorn & Bakermans-Kranenburg, 2004), whereas lack of support may contribute to problematic behavior (Rothbaum & Weisz, 1994). The second dimension, control, is behavior of the parent toward the child with the intent of directing the behavior of the child in a manner desirable to the parents (Rollins & Thomas, p. 321). Control strategies may vary from positive to negative, depending on the parent and the situation. One negative strategy, restrictive

212

The Journal of Genetic Psychology

control, is characterized by high power assertion, negativity, intrusiveness, hostility, overcontrol, or overinvolvement. More restrictive control has been associated with increased externalizing behavior (Calkins, 2002; Campbell, 1995). Maccoby and Martin (1983) proposed a fourfold scheme in which combinations of support (high and low) and control (high and low) describe four parenting styles: authoritative, authoritarian, permissive, and neglecting. The most common parenting styles in this fourfold scheme are authoritative and authoritarian parenting. Authoritative parenting is characterized by a combination of high warmth, firm but fair control, and the use of explanations and reasoning (Campbell, 1995). Authoritarian parenting involves power assertion without warmth, nurturance, or two-way communication. Authoritative parenting is generally advantageous to child development (Coplan et al., 2002; Rothbart & Bates, 1998). In addition, this method also captures the sense of mutuality that may be particularly important for parenting children who have high levels of negative emotionality (Rothbaum & Weisz, 1994). In contrast, authoritarian maternal parenting that is arbitrary, negative, or uninvolved is associated with noncompliance, defiance, and low internalization of control (Campbell). Fathers generally report less authoritative parenting than do mothers (e.g., Winsler, Madigan, & Aquilino, 2005). An explanation for this finding may be that mothers are generally more emotionally invested in parenting, face relatively strong societal expectations about parenting, and, as a rule, have the most responsibility for parenting (Corwyn & Bradley, 1999; Geary, 2000). Parenting style is more authoritative and child-centered in families of higher socioeconomic status (SES), in contrast to the authoritarian parent-centered style that characterizes lower SES families (Hoff, Laursen, & Tardif, 2002). Furthermore, children showing problematic behavior at preschool age are more likely to come from families of lower SES (Campbell, 1995). These findings may be explained by the more stressful family circumstances in lower SES families, such as financial problems, housing problems, and difficulty in accessing child welfare, medical facilities, or other social services that can be used as a source of support. Mediation by Parenting In the majority of studies, researchers report small to moderate associations between childrens negative emotionality and problematic behavior (Sanson, Oberklaid, Pedlow, & Prior, 1991). In addition, most researchers only focus on direct effects. What appear to be direct effects, however, could also result from more complex and indirect relations, such as mediation processes. Mediation by parenting may be a plausible indirect pathway connecting temperament with problematic behavior (Collins, Maccoby, Steinberg, Hetherington, & Bornstein, 2000; Lee & Bates, 1985; Rothbart & Bates, 1998). Mediation by parenting implies that the childs negative emotionality affects parenting and that parenting subsequently affects the childs behavior. Several researchers already have investigated these separate paths. Empirical evidence has shown that higher levels of child negative

Paulussen-Hoogeboom, Stams, Hermanns, Peetsma, & van den Wittenboer

213

emotionality are associated with (a) more authoritarian parenting behaviors, such as power assertion, low emotional support, punitiveness, and general unresponsiveness, and (b) less authoritative parenting (e.g., Sanson et al., 2004). Therefore, these authoritarian parenting behaviors are related to more externalizing behavior (Paterson & Sanson, 1999; Rothbaum & Weisz, 1994; Rubin, Burgess, Dwyer, & Hastings, 2003) and internalizing behavior (Morris et al., 2002). Present Study In the present study, we attempted to provide more insight into the mechanism that relates preschoolers negative emotionality to their problematic internalizing and externalizing behaviors, as measured by maternal perceptions. We examined direct and indirect associations in this community sample. We focused on 3-year-olds to fill a gap in the existing literature. Based on our proposed mediation model (see Figure 1), we expected to find (a) direct associations between

Negative emotionality e1 1 Authoritarian parenting e3 1 Externalizing behavior

e2 1 Authoritative parenting e4 1 Internalizing behavior Socioeconomic status FIGURE 1. Theoretical model relating preschoolers negative emotionality to their internalizing and externalizing behaviors.

214

The Journal of Genetic Psychology

child negative emotionality and both internalizing and externalizing behaviors, (b) direct associations among higher levels of negative emotionality, less authoritative parenting, and more authoritarian parenting, (c) direct associations between less authoritative and more authoritarian parenting and higher levels of internalizing and externalizing behavior, (d) indirect, mediated paths from negative emotionality through parenting styles to more problematic behavior, (e) associations between lower levels of SES and higher and lower levels of authoritative parenting, and (f) associations between lower levels of SES and higher levels of internalizing and externalizing behaviors. Because we considered parenting style to be only one of the processes linking temperament with problematic behavior, we hypothesized partial mediation, which indicates that the effects of negative emotionality on internalizing and externalizing behaviors remain significant in a model that accounts for mediation by parenting. We note that confounding of measures due to item-content overlap in the assessment of temperament and problematic behavior may invalidate conclusions about the relation between negative emotionality and both internalizing and externalizing behaviors. However, Lemery, Essex, and Smider (2002) and Lengua, West, and Sandler (1998) showed that even after the removal of contaminated questionnaire items, there continued to be significant and interpretable relations between temperament and problematic behavior. In this study, we accounted for possible contamination of the measures for negative emotionality and problematic behavior by repeating the analyses of our final model using the decontaminated measures derived from a confirmatory factor analysis. Method Participants In collaboration with Dutch child health centers in the province NorthHolland, we recruited 196 preschool-aged children (98 girls, 98 boys; M age = 3.4 years, SD = 0.4 years) and their mothers (M age = 35.9 years, SD = 3.7 years) to participate in this study. In The Netherlands, child health centers serve the general population. Of the participating families, 99% were two-parent families; mothers were either married to or living with the father of the child. Ninety-two percent of the mothers were born in The Netherlands, and 8% of the mothers were born elsewhere. Forty-seven percent of the children were firstborn and 53% were later born. Procedure We contacted 750 randomly selected families through a letter from the child health centers and asked if they would participate in a study of temperament and development. A registration form and a postage-paid envelope (to be sent to the

Paulussen-Hoogeboom, Stams, Hermanns, Peetsma, & van den Wittenboer

215

university) were enclosed. Participants were assured confidentiality and that they would receive a brief report about the results of the study. The 267 families (36%) who agreed to participate were sent a set of questionnaires to be completed at home by the parent who was most involved in raising the child. The parents returned the questionnaires by mail. We received 201 (75%) sets of completed questionnaires. Because all but 5 of the sets of questionnaires were completed by mothers, we report only data with regard to mothers (N = 196). Due to Dutch privacy legislation, it was not possible to investigate whether a nonresponse bias existed. Measures Negative emotionality. For the composite measure of negative emotionality, mothers completed five scales of the Childrens Behavior Questionnaire (CBQ; Majdandzic & van den Boom, 2001; Rothbart, Ahadi, Hershey, & Fisher, 2001). 1. Angerfrustration scale (13 items) referred to interruption of ongoing tasks or goal blocking. Cronbachs alpha estimating internal consistency for this scale was .83. 2. Discomfort scale (12 items, = .69) concerned reactions to sensory qualities of stimulation, including intensity, rate, or complexities of light, movement, sound, and texture. 3. Fear scale (12 items, = .71) included unease, worry, or nervousness, which are related to anticipated pain or distress or potentially threatening situations. 4. Sadness scale (12 items, = .68) concerned exposure to suffering, disappointment, and object loss and hence lowered mood and energy. 5. Soothability scale (13 items, = .81) concerned the rate of recovery from peak distress, excitement, or general arousal. Mothers rated their child on a 7-point Likert-type scale ranging from 1 (extremely untrue of your child) to 7 (extremely true of your child). Mothers also had a nonapplicable response option to be used when the child had not (yet) been observed in the situation described. We assessed construct validity using principal component analysis, which revealed a one-dimensional solution with soothability loading negatively, factor loadings ranging from .69 to .82, and an explained variance of 58%. Internal consistency for the composite negative emotionality measure was = .82. The mean score for the composite negative emotionality measure was 3.32 (SD = 0.70, N = 196). Parenting styles. We derived the composite measures for authoritative and authoritarian parenting styles from six scales that the mother completed. 1. Responsiveness scale (8 items, = .92) referred to the extent to which the mother considered herself responsive to the needs and signals of her child.

216

The Journal of Genetic Psychology

It was derived from items from the Nijmegen Parenting Questionnaire (Gerris et al., 1993; Gerrits, Dekovic, Groenendaal, & Noom 1996). 2. Consistency scale (8 items, = .70) referred to the extent to which the mothers behavior was predictable for the child and was measured by items from the Parenting Dimensions Inventory (Gerrits, Groenendaal, Dekovic, & Noom, 1996; Slater & Power, 1987). 3. Acceptance scale (12 items, = .62) referred to the extent to which the child lived up to the mothers physical, intellectual, and emotional expectations. The items came from the Parenting Stress Index (translated and revised by Brock, Vermulst, Gerris, & Abidin, 1992; Groenendaal, Gerrits, & Rispens, 1996). Mothers rated their parenting behavior on a 6-point Likert-type scale ranging from 1 (totally disagree) to 6 (totally agree). 4. Induction scale (5 items, = .83) indicated disciplining techniques, such as pointing out consequences of misbehaving to the child and reasoning with the child. 5. Power assertion scale (5 items, = .79) indicated disciplining techniques such as raising ones voice, the use of physical punishment, taking away privileges, or giving the child an extra chore. 6. Love withdrawal scale (5 items, = .81) referred to disciplining techniques such as punishing the child by ignoring him or her or sending the child away. Induction, power assertion, and love withdrawal were measured by items from the Parenting Dimensions Inventory (Gerrits, Groenendaal et al., 1996; Slater & Power, 1987). Mothers read five vignettes describing parenting situations and 10 possible reactions. Subsequently, they used a 4-point Likert-type scale ranging from 1 (very improbable) to 4 (very probable) to rate how likely they considered themselves to react in these ways. Principal component analysis revealed a two-dimensional solution with 29% of the variance explained by a factor dealing with authoritative parenting and 28% of the variance explained by a factor that could be labeled authoritarian parenting. Scales loading on authoritative parenting were Responsiveness (factor loading = .59), Acceptance (.71), Induction (.52), and Consistency (.74). Scales loading on authoritarian parenting were Power assertion (.88) and Love withdrawal (.87). We standardized the six separate scale scores before computing composite scores. Internal consistency for the composite authoritative parenting measure was = .83. Internal consistency for the composite authoritarian parenting measure was = .76. Problematic behavior. We assessed problematic behavior using the Child Behavior Checklist (CBCL; Achenbach, 1992; Koot, van den Oord, Verhulst, & Boomsma, 1997). The CBCL contains 100 problem items that are scored on 3-point Likert-type scales. Mothers indicated whether behavioral descriptions

Paulussen-Hoogeboom, Stams, Hermanns, Peetsma, & van den Wittenboer

217

were 0 (not at all true), 1 (somewhat true), or 2 (very true) of their child. Overall scores were obtained for the broadband syndromes of internalizing behavior (anxiousdepressed and withdrawn; 26 items, = .88) and externalizing behavior (aggressive and destructive behavior; 25 items, = .74). The mean score for internalizing behavior was 6.00 (SD = 4.25, N = 196). The mean score for externalizing behavior was 11.44 (SD = 7.54, N = 196). SES. SES of the family was a combination of the educational and vocational backgrounds of both parents as reported by the mother and was computed on the basis of sample-specific factor loadings and standard deviations. Mean scores corresponded to socioeconomic strata: 39 was lower class, 912 was middle class, and 1216 was upper class (Bernstein & Brandis, 1970). Principal component analysis revealed a one-dimensional solution with factor loadings ranging from .71 (vocational background of mother) to .82 (educational background of mother) and an explained variance of 63%. Internal consistency of the scale was = .80. The mean score for SES was 12.13 (SD = 2.57, N = 196). The sample consisted primarily (90%) of middle- and upper-class families. Results Correlations and covariances for the study variables are presented in Table 1. In preliminary analyses, we examined gender and birth-order differences with respect to the composites of negative emotionality, authoritative and authoritarian parenting styles, and internalizing and externalizing behaviors. There were no significant gender or birth-order differences in mean scores on any of the variables. We analyzed data on the sample covariance matrix estimates using maximum-likelihood estimation. We used AMOS, version 4.0 (Arbuckle & Wothke, 1999). We performed tests of indirect, mediated effects using bootstrap procedures. Bootstrapping results in
TABLE 1. Correlations and Covariances Between the Study Variables Variable 1. Negative emotionality 2. Authoritative parenting 3. Authoritarian parenting 4. Internalizing behavior 5. Externalizing behavior 6. Socioeconomic status 1 .40** .07 .47** .52** .08 2 .11 .01 .34** .36** .10 3 .02 .00 .09 .18** .18** 4 .06 .02 .01 .62** .24** 5 .11 .04 .02 .03 .23** 6 .15 .10 .20 .10 .17

Note. Correlations are presented below the diagonal. Covariances are presented above the diagonal. ** p < .01, one-tailed.

218

The Journal of Genetic Psychology

enhanced statistical power for testing mediational models when sample sizes are moderate or small (Shrout & Bolger, 2002). Overall model fit was assessed using chi-square, the adjusted goodness-of-fit index (AGFI), and Steigers root mean square error of approximation (RMSEA). The AGFI is a measure of the relative amount of variance and covariance that are jointly accounted for by the model. It ranges from 0 to 1, and values close to 1 indicate a good fit (Jreskorg & Srbom, 1989). The RMSEA is a population-based index and is relatively insensitive to sample size (Loehlin, 1998). According to Loehlin, an RMSEA of .05 or less would indicate a close fit of the model. We started with the full theoretical model (see Figure 1), which we found fit the data well, 2(2, N = 196) = 1.91, p = .38; AGFI = .97; RMSEA = .00. However, the direct paths from SES to authoritative parenting, negative emotionality to authoritarian parenting, and authoritarian parenting to both internalizing and externalizing behavior were not significant. Omitting these paths from the model for parsimony left authoritarian parenting disconnected from both predictor (negative emotionality) and outcome variables (problematic behavior). Therefore, we removed authoritarian parenting from the model. The adjusted model (see Figure 2) still fit the data well, 2(2, N = 196) = 2.49, p = .29; AGFI = .96; RMSEA = .04. All direct paths in this model were significant, so no further adjustments were made. Childrens negative emotionality was positively associated with childrens externalizing behavior ( = .44, p = .002) and internalizing behavior ( = .39, p = .002). We found a negative association between childrens negative emotionality and maternal authoritative parenting
e3 .44 Negative emotionality .17 .40 e1 .16 Authoritative parenting .39 .17 Socioeconomic status .19 .26 Internalizing behavior e2 .18 .32 Externalizing behavior .46

FIGURE 2. Adjusted model with regression weights and explained variance.

Paulussen-Hoogeboom, Stams, Hermanns, Peetsma, & van den Wittenboer

219

style ( = .40, p = .002). Maternal authoritative parenting style was also negatively associated with childrens externalizing behavior ( = .17, p = .015), as was the association between maternal authoritative parenting style and childrens internalizing behavior ( = .17, p = .008). Family SES was negatively associated with both childrens internalizing ( = .19, p = .007) and externalizing behaviors ( = .18, p = .002). However, SES was not associated with maternal authoritative parenting style. Childrens externalizing and internalizing behaviors were moderately correlated (r = .46, p = .002). With regard to the hypothesized indirect effects, the association between childrens negative emotionality and externalizing behavior was partially mediated by maternal authoritative parenting style ( = .07, p < .05), and the association between childrens negative emotionality and internalizing behavior was also partially mediated by maternal authoritative parenting style ( = .07, p < .01). The model accounted for 26% of the variance in internalizing behavior and 32% of the variance in externalizing behavior. In additional analyses, we tested the final mediation model after decontamination of the CBQ and CBCL items through confirmatory factor analysis (CFA; see Lengua et al., 1998). CFA of the CBQ and CBCL yielded three meaningful constructs (negative emotionality, internalizing behavior, and externalizing behavior) with satisfactory internal consistencies (.68 < < .85). Item-content overlap proved to be most prominent in the scale for internalizing behavior. The CFA model showed an acceptable fit to the data, 2(1376, N = 196) = 2611.31, p < .00; AGFI = .66; RMSEA = .07: The ratio between the chi-square statistic and the degrees of freedom was lower than 2.5 (Hu & Bentler, 1999). Subsequently, we tested the final mediation model using the decontaminated CBQ and CBCL. The decontaminated mediation model still showed an adequate fit to the data, 2(3, N = 196) = 4.14, p = .25; AGFI = .96; RMSEA = .04. The direct association between negative emotionality and externalizing behavior was less strong in the decontaminated model but still significant. The direct association between negative emotionality and internalizing behavior, however, was reduced to nonsignificance. The mediated effects that we found in the adjusted model were also present in the decontaminated model and were of similar size. However, the finding that the direct association between negative emotionality and internalizing behavior was reduced to nonsignificance indicates that this association was then fully mediated by authoritative parenting instead of partially mediated. Discussion Our aim in the present study was to investigate parenting as a possible mechanism linking childrens negative emotional reaction style to their problematic behavior in families drawn from the general population. In line with earlier findings, the results of this study show a relation between child negative emotionality and problematic behavior. More important, though, is the finding that this relation was mediated by mothers authoritative parenting style. As relations between temperament

220

The Journal of Genetic Psychology

and problematic behavior may be inflated because of item-content overlap, we used CFA as a means of decontamination. The decontaminated model differed from the previous best-fitting model in that the direct relation between negative emotionality and internalizing behavior was reduced to nonsignificance, indicating that the association between negative emotionality and internalizing behavior was fully mediated by authoritative parenting. It seems that decontamination of measures assessing negative emotionality and problematic behavior yields a clearer picture of the mediation process. Moreover, our findings match those of Sanson, Prior, and Kyrios (1990), who found a significant amount of overlap between temperament and internalizing behavior, indicating that parents may infer their childs internalizing behavior partly from perceptions of their childs temperament. Theoretical Relevance The coefficients of the mediated paths were small but of theoretical interest because this study demonstrates that parenting can mediate the relation between childrens difficult temperament and problematic behavior. Furthermore, with authoritarian parenting styles not being a mediator, our results suggest that authoritative parenting may be more important for the behavioral adjustment of preschoolers than is authoritarian parenting. This is in line with findings by Pettit and Bates (1989) and Rothbaum and Weisz (1994), who have suggested that, despite both authoritative and authoritarian parenting styles being high on the control dimension, parental control should be exercised in a sensitive way to be effective. In this way, it appears that love withdrawal and power assertion are aspects of control that do not contribute to the prevention of problematic behavior. Coplan et al. (2002) suggested that an authoritarian parenting style functions as a lens through which all childrens behaviors are perceived and evaluated. The researchers found the default emotional response of authoritarian mothers to be negative, regardless of child behavior. It is possible that a parent-centered parenting style also generates a default behavioral response and consequently does not show any variation in negative parenting behavior when compared with varying degrees of negative emotionality. Our findings agree with those of studies showing associations between internalizing and externalizing behaviors at the preschool age (Campbell, 1995; Gilliom & Shaw, 2004; Gjone & Stevenson, 1997). The moderate relation between internalizing and externalizing behaviors may partly be accounted for by a process of multifinality, namely, a similar initial condition (difficult temperament) that leads to different effects (internalizing and externalizing behavior; Cicchetti & Rogosch, 1996). We found empirical evidence for a link between negative emotionality and both internalizing and externalizing behaviors. It is interesting that the relation between negative emotionality and externalizing behavior proved to be partially mediated by parenting, whereas the relation between negative emotionality and internalizing behavior was fully mediated by parenting in the

Paulussen-Hoogeboom, Stams, Hermanns, Peetsma, & van den Wittenboer

221

model that was not affected by item-content overlap in measures of temperament and problematic behavior. Last, as hypothesized, family SES proved to be modestly related to problematic behavior. The relations between SES and authoritative and authoritarian parenting styles were in the expected direction, though not significant, indicating that parenting did not act as a mediator here. Stronger relations may be found in more heterogeneous samples. Practical Relevance The finding that an authoritative parenting style mediates the relations between negative emotionality and problematic behaviors underscores the importance of providing effective parenting support to parents who have difficulties in dealing with their young childs negative emotionality on a daily basis. When parents can be trained and encouraged to react to their childrens negative emotionality in an adaptive way, parentchild interactions may become more enjoyable, thereby reducing the occurrence of problematic behaviors and preventing more serious behavioral problems later in life (Campbell, 1995; Patterson, 1982). We note that even in general population samples, a substantial percentage of children (up to 10%) may develop internalizing- and externalizing-behavior problems in the clinical range (Achenbach et al., 1991; Verhulst, van der Ende, & Koot, 1996). The present study adds to the body of literature showing how normal development may go awry. Limitations Some limitations of this study should be mentioned. The first is the correlational nature of the design, which sets limits on the causal interpretation of our results. Second, it is possible that the mediations we found may be accounted for by genetic similarities between parent and child. However, researchers focusing on individual differences in internalizing and externalizing behaviors have shown that both genes and environment play a role and that shared environmental factors, especially parenting, can be considered important for the stability of internalizing and externalizing behavior from preschool to middle childhood and early adolescence (Bartels, 2004; Stams, Juffer, & van IJzendoorn, 2002; van der Valk, van den Oord, Verhulst, & Boomsma, 2003). A third limitation of the study is its limited generalizability because the participants predominantly came from middle to high socioeconomic backgrounds. Notably, a recent meta-analysis of negative emotionality and parenting showed the association between more child negative emotionality and less supportive parenting to be stronger in lower SES families than in families from middle or higher socioeconomic backgrounds (Paulussen-Hoogeboom, Stams, Hermanns, & Peetsma, 2007). Last, the present findings may be generalized only to mothers and not to fathers. For example,

222

The Journal of Genetic Psychology

researchers investigating the ecology of fathering have shown that fathers and mothers differ in both caregiving sensitivity and play interaction, which may have different developmental consequences for the child (Grossmann et al., 2002; Lamb, Frodi, Frodi, & Hwang, 1982; Parke, 1995). Suggestions for Additional Research Despite the increasing involvement of fathers in child rearing in many Western countries, researchers in this field still tend to focus almost exclusively on mothers. However, a more complete examination of the associations among negative emotionality, parenting style, and problematic behavior would require a greater number of studies that include data from fathers. Thus, although fathers are generally less willing to participate in research than are mothers, it is essential that researchers make every effort to get and keep fathers involved. In addition to including fathers, we also recommend that future researchers use longitudinal designs to test the proposed mediational model in the most robust way. In sum, this study revealed that difficult temperament in preschoolers, characterized by expressions of negative emotionality, was related to both externalizing and internalizing behaviors. Moreover, we found initial evidence that an authoritative parenting style mediates the relation between childrens difficult temperament and problematic behavior. Although replication of our results is needed, this study adds to a greater understanding of the processes that are responsible for the progression of difficult temperament into problematic behavior.
AUTHOR NOTES Marja C. Paulussen-Hoogeboom is a developmental psychologist. Her research interests are in temperament, specifically negative emotionality, parenting, and behavior problems in young children. Geert Jan J. M. Stams has conducted longitudinal research, intervention studies, and several meta-analyses in the areas of socioemotional and moral development, focusing on familial and extrafamilial determinants of behavior problems in early childhood, middle childhood, and adolescence. Jo M. A. Hermanns is a developmental psychologist. His research interests are family support, parenting, and developmental risks. Thea T. D. Peetsma is senior researcher at the SCO-Kohnstamm Institute at the University of Amsterdam, The Netherlands, and coordinator of the division for basic research of this institute. She specializes in psychosocial development, learning motivation of children, and children with special educational needs. Godfried L. H. van den Wittenboer has researched the methodological issues of educational and psychological measurement, varying from problems in facet theory, such as reducing complex research designs, response scalability, the structure of measurement in facet designs and interaction problems in simple research designs, to issues in longitudinal data analysis. REFERENCES Achenbach, T. M. (1992). Manual for the Child Behavior Checklist/2-3 and 1992 profile. Burlington: University of Vermont, Department of Psychiatry.

Paulussen-Hoogeboom, Stams, Hermanns, Peetsma, & van den Wittenboer

223

Achenbach, T. M., Howell, C. T., Quay, C. T., & Conners, C. K. (1991). National survey of problems and competencies among four- to sixteen-year-olds: Parents reports from normative and clinical samples. Monographs of the Society for Research in Child Development, 56(3, Serial No. 225). Arbuckle, J. L., & Wothke, W. (1999). Amos 4.0 users guide. Chicago: SmallWaters. Bartels, M. (2004). De oorzaken van individuele verschillen in stabiliteit en verandering van gedragsproblemen [Causes of individual differences in stability and change of behavior problems]. Neuropraxis, 8, 107115. Bates, J. E. (1989). Concepts and measures of temperament. In G. A. Kohnstamm & J. E. Bates (Eds.), Temperament in childhood (pp. 326). New York: Wiley. Bernstein, B., & Brandis, W. (1970). Social class differences in communication and control. In W. Brandis & D. Henderson (Eds.), Social class, language and communication (pp. 93153). London: Routledge. Bokhorst, C. L., Bakermans-Kranenburg, M. J., Pasco Fearon, R. M., van IJzendoorn, M. H., Fonagy, P., & Schuengel, C. (2003). The importance of shared environment in motherinfant attachment security: A behavioral genetic study. Child Development, 74, 17691782. Brock, A. J. L. L., Vermulst, A. A., Gerris, J. R. M., & Abidin, R. R. (1992). Nijmeegse ouderlijke stress index: Handleiding experimentele versie [Nijmegen Parenting Stress Index: Manual experimental version]. Lisse, The Netherlands: Swets & Zeitlinger. Calkins, S. D. (2002). Does aversive behavior during toddlerhood matter?: The effects of difficult temperament on maternal perceptions and behavior. Infant Mental Health Journal, 23, 381402. Campbell, S. B. (1995). Behavior problems in preschool children: A review of recent research. Journal of Child Psychology and Psychiatry and Allied Disciplines, 36, 113149. Cicchetti, D., & Rogosch, F. A. (1996). Equifinality and multifinality in developmental psychopathology. Development and Psychopathology, 8, 597600. Collins, W. A., Maccoby, E. E., Steinberg, L., Hetherington, E. M., & Bornstein, M. H. (2000). Contemporary research on parenting: The case for nature and nurture. American Psychologist, 55, 218232. Coplan, R. J., Hastings, P. D., Lagace-Seguin, D. G., & Moulton, C. E. (2002). Authoritative and authoritarian mothers parenting goals, attributions and emotions across different childrearing contexts. Parenting: Science and Practice, 2, 126. Corwyn, R. F., & Bradley, R. H. (1999). Determinants of paternal and maternal investment in children. Infant Mental Health Journal, 20, 238256. Geary, D. C. (2000). Evolution and proximate expression of human paternal investment. Psychological Bulletin, 126, 5577. Gerris, J. R. M., Vermulst, A. A., Boxtel, D. A. A. M., Janssens, J. M. A. M., Zutphen, R. A. H., & Felling, A. J. A. (1993). Parenting in Dutch families: A representative description of Dutch family life in terms of validated concepts, representing characteristics of parents, children, the family as a system and parental sociocultural value orientations. Nijmegen, The Netherlands: Radboud University Nijmegen, Institute of Family Studies. Gerrits, L. A. W., Dekovic, M., Groenendaal, J. H. A., & Noom, M. J. (1996). Opvoedingsgedrag [Parenting behavior]. In J. Rispens, J. M. A. Hermanns, & W. H. J. Meeus (Eds.), Opvoeden in Nederland [Parenting in the Netherlands] (pp. 4169). Assen, The Netherlands: Van Gorcum. Gerrits, L. A. W., Groenendaal, J. H. A., Dekovic, M., & Noom, M. J. (1996). Interne rapportage over een nederlandse bewerking van de Parenting Dimensions Inventory (PDI) [Internal report of a Dutch translation of the Parenting Dimensions Inventory (PDI)]. Utrecht, The Netherlands: Universiteit Utrecht.

224

The Journal of Genetic Psychology

Gilliom, M., & Shaw, D. S. (2004). Codevelopment of externalizing and internalizing problems in early childhood. Development and Psychopathology, 16, 313333. Gjone, H., & Stevenson, J. (1997). The association between internalizing and externalizing behavior in childhood and early adolescence: Genetic or environmental common influences? Journal of Abnormal Child Psychology, 25, 277286. Groenendaal, J. H. A., Gerrits, L. A. W., & Rispens, J. (1996). Opvoeding en ontwikkeling en de kinderperiode [Parenting and development in childhood]. In J. Rispens, J. M. A. Hermanns, & W. H. J. Meeus (Eds.), Opvoeden in Nederland [Parenting in the Netherlands] (pp. 181206). Assen, The Netherlands: Van Gorcum. Grossmann, K., Grossmann, K. E., Fremmer-Bombik, E., Kindler, H., Scheuerer-Englisch, H., & Zimmerman, P. (2002). The uniqueness of the child-father attachment relationship: Fathers sensitive and challenging play as a pivotal variable in a 16-year longitudinal study. Social Development, 11, 307331. 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 (2nd ed., pp. 231252). Mahwah, NJ: Erlbaum. Hu, L-T., & Bentler, T. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 155. Jreskorg, K. G., & Srbom, D. (1989). Lisrel 7: A guide to the program and applications. Mooresville, IN: Scientific Software. Kochanska, G. (1995). Childrens temperament, mothers discipline, and security of attachment: Multiple pathways to emerging internalization. Child Development, 66, 597615. Koot, H. M., van den Oord, E. J. C. G., Verhulst, F. C., & Boomsma, D. I. (1997). Behavioral and emotional problems in young preschoolers: Cross-cultural testing of the validity of the Child Behavior Checklist/2-3. Journal of Abnormal Child Psychology, 25, 183196. Lamb, M. E., Frodi, A. M., Frodi, M., & Hwang, C. P. (1982). Characteristics of maternal and paternal behavior in traditional and nontraditional Swedish families. International Journal of Behavioral Development, 5, 131141. Lee, C. L., & Bates, J. E. (1985). Mother-child interaction at age two years and perceived difficult temperament. Child Development, 56, 13141325. Lemery, K. S., Essex, M. J., & Smider, N. A. (2002). Revealing the relation between temperament and behavior problem symptoms by eliminating measurement confounding: Expert ratings and factor analyses. Child Development, 73, 867882. Lengua, L. J., West, S. G., & Sandler, I. N. (1998). Temperament as a predictor of symptomatology in children: Addressing contamination of measures. Child Development, 69, 164181. Loehlin, J. C. (1998). Latent variable models: An introduction to factor, path, and structural analysis (3rd ed.). Mahwah, NJ: Erlbaum. Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family: Parent-child interaction. In P. H. Mussen (Series Ed.) & E. M. Hetherington (Vol. Ed.), Handbook of child psychology: Vol. 4. Socialization, personality, and social development (pp. 1101). New York: Wiley. Majdandzic, M., & van den Boom, D. C. (2001). Dutch version of the Childrens Behavior Questionnaire. Unpublished manuscript, University of Amsterdam, The Netherlands. Mangelsdorf, S. C., Schoppe, S. J., & Buur, H. (2000). The meaning of parental reports: A contextual approach to the study of temperament and behavior problems in childhood. In V. J. Molfese & D. L. Molfese (Eds.), Temperament and personality development across the life span (pp. 121140). Mahwah, NJ: Erlbaum.

Paulussen-Hoogeboom, Stams, Hermanns, Peetsma, & van den Wittenboer

225

Morris, A. S., Silk, J. S., Steinberg, L., Sessa, F. M., Avenevoli, S., & Essex, M. J. (2002). Temperamental vulnerability and negative parenting as interacting predictors of child adjustment. Journal of Marriage and Family, 64, 461471. Oland, A. A., & Shaw, D. S. (2005). Pure versus co-occurring externalizing and internalizing symptoms in children: The potential role of socio-developmental milestones. Clinical Child and Family Psychology Review, 8, 247270. Parke, R. D. (1995). Fathers and families. In M. H. Bornstein (Ed.), Handbook of parenting: Vol. 3. Status and social conditions of parenting (pp. 2763). Hillsdale, NJ: Erlbaum. Paterson, G., & Sanson, A. (1999). The association of behavioural adjustment to temperament, parenting and family characteristics among 5-year-old children. Social Development, 8, 293309. Patterson, G. R. (1982). Coercive family processes. Eugene, OR: Castalia. Pauli-Pott, U., Mertesacker, B., Bade, U., Bauer, C., & Beckmann, D. (2000). Contexts of relations of infant negative emotionality to caregivers reactivity/sensitivity. Infant Behavior and Development, 23, 2339. Paulussen-Hoogeboom, M. C., Stams, G. J. J. M., Hermanns, J. M. A., & Peetsma, T. T. D. (2007). Child negative emotionality and parenting from infancy to preschool: A meta-analytic review. Developmental Psychology, 43, 438453. Pettit, G. S., & Bates, J. E. (1989). Family interaction patterns and childrens behavior problems from infancy to 4 years. Developmental Psychology, 25, 413420. Prior, M. (1992). Childhood temperament. Journal of Child Psychology and Psychiatry and Allied Disciplines, 33, 249279. Putnam, S. P., Sanson, A. V., & Rothbart, M. K. (2002). Child temperament and parenting. In M. H. Bornstein (Ed.), Handbook of parenting: Vol. 1. Children and parenting (pp. 255277). London: Erlbaum. Rollins, B. C., & Thomas, D. L. (1979). Parental support, power, and control techniques in the socialization of children. In W. R. Burr, R. Hill, F. I. Nye, & I. L. Reiss (Eds.), Contemporary theories about the family (Vol. 1, pp. 317364). London: Free Press. Rothbart, M. K., Ahadi, S. A., & Hershey, K. L. (1994). Temperament and social behavior in childhood. Merrill Palmer Quarterly, 40, 2139. Rothbart, M. K., Ahadi, S. A., Hershey, K. L., & Fisher, P. (2001). Investigations of temperament at three to seven years: The Childrens Behavior Questionnaire. Child Development, 72, 13941408. Rothbart, M. K., & Bates, J. E. (1998). Temperament. In W. Damon & N. Eisenberg (Eds.), Handbook of child psychology: Vol 3. Social, emotional, and personality development (5th ed., pp. 105176). New York: Wiley. Rothbaum, F., & Weisz, J. R. (1994). Parental caregiving and child externalizing behavior in nonclinical samples: A meta-analysis. Psychological Bulletin, 116, 5574. Rubin, K. H., Burgess, K. B., Dwyer, K. M., & Hastings, P. D. (2003). Predicting preschoolers externalizing behaviors from toddler temperament, conflict, and maternal negativity. Developmental Psychology, 39, 164176. Sanson, A., Hemphill, S. A., & Smart, D. (2004). Connections between temperament and social development: A review. Social Development, 13, 142170. Sanson, A., Oberklaid, F., Pedlow, R., & Prior, M. (1991). Risk indicators: Assessment of infancy predictors of pre-school behavioural maladjustment. Journal of Child Psychology, Psychiatry and Allied Disciplines, 32, 609626. Sanson, A., Prior, M., & Kyrios, M. (1990). Contamination of measures in temperament research. Merrill Palmer Quarterly, 36, 179192. Schuler, M. E., Black, M. M., & Starr, R. H., Jr. (1995). Determinants of mother-infant interaction: Effects of prenatal drug exposure, social support, and infant temperament. Journal of Clinical Child Psychology, 24, 397405.

226

The Journal of Genetic Psychology

Shiner, R. L. (1998). How shall we speak of childrens personalities in middle childhood? A preliminary taxonomy. Psychological Bulletin, 124, 308332. Shrout, E. P., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7, 422445. Slater, M. A., & Power, T. G. (1987). Multidimensional assessment of parenting in singleparent families. In J. P. Vincent (Ed.), Advances in family intervention, assessment and theory (pp. 197228). Greenwich, CT: JAI Press. Stams, G. J. J. M., Juffer, F., & van IJzendoorn, M. H. (2002). Maternal sensitivity, infant attachment, and temperament in early childhood predict adjustment in middle childhood: The case of adopted children and their biologically unrelated parents. Developmental Psychology, 38, 806821. Thomas, A., Chess, S., & Birch, H. G. (1968). Temperament and behavior disorders in children. New York: New York University Press. Thomas, A., Chess, S., Birch, H. G., Hertzig, M. E., & Korn, S. (1963). Behavioral individuality in early childhood. New York: New York University Press. van der Valk, J. C., van den Oord, E. J. C. G., Verhulst, F. C., & Boomsma, D. I. (2003). Genetic and environmental contributions to stability and change in childrens internalizing and externalizing problems. Journal of the American Academy of Child & Adolescent Psychiatry, 42, 12121220. van IJzendoorn, M. H., & Bakermans-Kranenburg, M. J. (2004). Maternal sensitivity and infant temperament in the formation of attachment. In G. Bremner & A. Slater (Eds.), Theories of infant development (pp. 233257). Oxford, England: Blackwell. Verhulst, F. C., van der Ende, J., & Koot, H. M. (1996). Handleiding voor de CBCL/4-18 [Manual for the CBCL/4-18]. Rotterdam, The Netherlands: Erasmus Universiteit Rotterdam. Winsler, A., Madigan, A. L., & Aquilino, S. A. (2005). Correspondence between maternal and paternal parenting styles in early childhood. Early Childhood Research Quarterly, 20, 112. Zeanah, C. H., & Fox, N. A. (2004). Temperament and attachment disorders. Journal of Clinical Child and Adolescent Psychology, 33, 3241.

Submitted January 12, 2006 Accepted February 4, 2008