Child Development, November/December 2005, Volume 76, Number 6, Pages 1144 1159

The Role of Parenting Styles in Childrens Problem Behavior

Kaisa Aunola and Jari-Erik Nurmi
University of Jyvaskyla

This study investigated the combination of mothers and fathers parenting styles (affection, behavioral control, and psychological control) that would be most influential in predicting their childrens internal and external problem behaviors. A total of 196 children (aged 5 6 years) were followed up six times from kindergarten to the second grade to measure their problem behaviors. Mothers and fathers filled in a questionnaire measuring their parenting styles once every year. The results showed that a high level of psychological control exercised by mothers combined with high affection predicted increases in the levels of both internal and external problem behaviors among children. Behavioral control exercised by mothers decreased childrens external problem behavior but only when combined with a low level of psychological control.

Several studies have been conducted to examine the role of parenting styles in childrens and adolescents internalizing and externalizing problem behavior (Hart, Newell, & Olsen, 2003). Although the three parenting style dimensions, that is, parental warmth/ affection, behavioral control, and psychological control, have each been shown to be associated with child adjustment (Barber, Olsen, & Shagle, 1994; Siequeland, Kendall, & Steinberg, 1996), it has been suggested that it is the combination of parenting characteristics in particular that is influential in child development (Baumrind, 1991; Darling & Steinberg, 1993; Steinberg, 2001). Despite this notion, only a few studies (Gray & Steinberg, 1999; Galambos, Barker, & Almeida, 2003; Pettit & Laird, 2002) have made a systematic effort to examine whether the impact of a particular parenting style dimension is moderated by the other two dimensions. Moreover, all the studies conducted thus far have focused on adolescents. This longitudinal study investigated what combination of mothers affection, behavioral control, and psychological control, on the one hand, and fathers affection, behavioral control, and psychological control, on the other, would show the most influence on their childrens internal and external problem behaviors during their early school years.

This study is part of the ongoing Jyvaskyla Entrance into Pri mary School (JEPS) Study, and was funded by grants from the Finnish Academy (63099, 778230). A version of this article was presented in August 2003 at the XIth European Conference on Developmental Psychology, Milano, Italy. We would like to express our gratitude to all the children and their parents who participated in this study. Correspondence concerning this article should be addressed to Kaisa Aunola, Department of Psychology, University of Jyvaskyla, PO Box 35, 40014 Jyvaskyla, Finland. Electronic mail may be sent to aunola@psyka.jyu.fi.

Problem behavior in children can be manifested in either externalizing or internalizing behavior. Externalizing behavior consists of disinhibited behaviors and other expressions of undersocialization (Kovacs & Devlin, 1998). In this type of problem behavior, negative emotions are directed against others, manifested as anger, aggression, and frustration (Roeser, Eccles, & Strobel, 1998). Children with externalizing problem behaviors have underdeveloped self-regulation skills, leading to undercontrolled behaviors (Cole, Zahn-Waxler, Fox, Usher, & Welsh, 1996). Internalizing behavior, by contrast, includes withdrawal, fearfulness, inhibition, and anxiety (Eisenberg et al., 2001; Roeser et al., 1998). In this type of problem behavior negative emotions are directed at oneself rather than others (Roeser et al., 1998). Internalizing problem behavior originates from overly strong self-regulation (Block & Block, 1980; Cole et al., 1996). Both internal and external problem behaviors have been shown to be rather stable from the early school years onwards (see, e.g., Denham et al., 2000; Kovacs & Devlin, 1998). Furthermore, both these behavior types lead to problems in various life domains, including school, peer relationships, and mental health (Hinshaw, 1992; Roeser et al., 1998). Much research has been published on the role of the family in childrens and adolescents problem behaviors. Of the many parenting variables, parenting styles have been among the most frequently investigated (Steinberg, 2001; Wood, McLeod, Sigman, Hwang, & Chu, 2003). In the dimensional approach to parenting styles the focus has been on the impacts of three dimensions (Hart et al., 2003; Schaefer, 1965): affection (e.g., responsiveness, involvement, supportiveness) refers to parents connectedness to the
r 2005 by the Society for Research in Child Development, Inc. All rights reserved. 0009-3920/2005/7606-0002

Parenting Styles in Childrens Behavior

1145

child and their interactional warmth (Galambos et al., 2003; Wood et al., 2003); behavioral control (e.g., maturity demands, monitoring, limit setting) consists of the regulation of the childs behavior through firm and consistent discipline (Barber, 1996; Galambos et al., 2003); and psychological control (e.g., love withdrawal, guilt induction) refers to parents control of the childs emotions and behavior through psychological means (Barber, 1996). The three parenting style dimensions have each been shown to be associated with child and adolescent problem behaviors. For example, a high level of behavioral control is related to low levels of externalizing problems, such as antisocial behavior and conduct disorders, both among adolescents (Barber & Olsen, 1997; Eccles, Early, Frasier, Belansky, & McCarthy, 1997; Pettit, Laird, Dodge, Bates, & Criss, 2001; Stice & Barrera, 1995) and among elementary school children (Barber, 1996; Lewis, 1981). These results have been ascribed to the fact that behavioral control fosters self-regulation and compliance (Hart et al., 2003; Lewis, 1981). Similarly, it has been suggested that parental affection facilitates childrens adjustment (Gray & Steinberg, 1999; Siequeland et al., 1996). However, the findings on parental affection are contradictory. For example, Miller, Cowan, Cowan, Hetherington, and Clingempeel (1993) and Dodge, Pettit, and Bates (1994) found that maternal warmth was negatively related to externalizing problems among preschoolers, whereas Galambos et al. (2003) showed that parental support was not related to adolescents problem behaviors (see also Stice & Barrera, 1995). Psychological control, in turn, is associated with internalizing problems, such as depressed mood and anxiety, among both adolescents (Barber et al., 1994; Conger, Conger, & Scaramella, 1997; Pettit et al., 2001; Wolfradt, Hempel, & Miles, 2003) and children (Siequeland et al., 1996; Barber & Harmon, 2002; Olsen et al., 2002). Some studies have also found an association between a high level of psychological control and externalizing problem behaviors (for a review, see Barber & Harmon, 2002; Yang et al., 2004). However, many researchers have suggested that it is certain combinations of parenting style variables rather than their unique impacts that contribute to childrens and adolescents adjustment (Baumrind, 1989, 1991; Darling & Steinberg, 1993; Steinberg, 2001). Part of this argumentation has its origins in the traditional parenting style paradigm that described parenting as a combination of various levels of behavioral control and affection (Maccoby & Martin, 1983). For example, authoritative parenting, characterized by a high level of both parental affection and

behavioral control, was shown to be positively associated with adjustment in children of various ages (Baumrind, 1966, 1989; Hart et al., 2003; Maccoby & Martin, 1983). By contrast, authoritarian parenting, characterized as high behavioral control (or, in some conceptualizations, harsh and punitive control; for a review, see Hart et al., 2003) but low affection, and permissive parenting, typified by low behavioral control, are related to various kinds of maladjustment, such as withdrawn behavior, low peer affiliation, and conduct disorders (Baumrind, 1989; Heller, Baker, Henker, & Hinshaw, 1996; Jewell & Stark, 2003; Wolfradt et al., 2003). Previous research on the role of parenting styles, and the importance of their combinations in particular, has at least two limitations. First, in the majority of studies, the role of psychological control has not been examined, or it has not been operationalized as a separate dimension from behavioral control and affection (Barber, 1996). Second, only a few systematic efforts have been made to differentiate between the main effects of the three parenting style dimensions and their interactions. As far as we know, only three studies have examined the extent to which the impacts of the three parenting style dimensions are moderated by the two others. In one of these studies, Galambos et al. (2003) found that parents high level of psychological control combined with a high level of behavioral control was related to externalizing problems (substance use, antisocial behavior, and misconduct at school) among adolescents. They suggested that behavioral control may not be uniformly effective when combined with less desirable parenting. In another study, Pettit and Laird (2002) found that a high level of psychological control combined with a low level of parental involvement was associated with delinquent behavior among adolescents, whereas a high level of psychological control combined with a high level of parental involvement was not. Similarly, Gray and Steinberg (1999) also found that, among adolescents, parental affection prevented internal distress more when combined with a high level of psychological control than when it was combined with a high level of autonomy granting. They suggested that a high level of affection compensated for the negative effects of psychological control. Gray and Steinberg (1999) found further that affection promoted adolescents psychosocial development, particularly when combined with a high level of behavioral control (see also Forehand & Nousiainen, 1993). Earlier studies on the extent to which combinations of the three parenting style variables predict problem behaviors are limited in two ways. First,

1146

Aunola and Nurmi

most of the research in the field has examined adolescents, and hence little is known about how the three parenting style dimensions interact in the prediction of childrens problem behavior. Although in one recent study Gadeyne, Ghesquiere, and Onghena (2004) examined the prospective relationships between parenting styles and childrens problem behaviors from kindergarten to the second grade, their study was limited to two parenting style dimensions (i.e., affection and restrictive control). Moreover, the study did not make a systematic effort to differentiate between the main effects of parenting dimensions and their interactions. The second limitation is that, in most of the previous studies on the interaction between parenting dimensions, maternal and paternal parenting has been aggregated to measure overall parenting in the family (Galambos et al., 2003; Gray & Steinberg, 1999). Thus it has not been possible to examine whether some of the parenting patterns found to influence problem behaviors are different for mothers and fathers (for an exception, see Forehand & Nousiainen, 1993). This study investigated the impacts of both mothers and fathers affection, behavioral control, and psychological control on childrens internal and external problem behaviors, and whether these impacts are moderated by the two other parenting style dimensions. In most research on parenting, the key assumption has been that parenting styles affect childrens adjustment (for a review, see Maccoby, 1992). However, it has been suggested that children and their adjustment may also influence their parents childrearing patterns (Bell, 1968; Harris, 1995; Hart et al., 2003). In the present study, we used cross-lagged longitudinal data to examine the possibility that childrens adjustment also affects their mothers and fathers parenting. The present study sought to answer the following research questions. First, do mothers and fathers parenting styles, that is, affection, behavioral control, and psychological control, predict their childrens subsequent internal and external problem behaviors during the transition from kindergarten to primary school? The following hypotheses were posited: (1a) lack of parental behavioral control would increase childrens externalizing problem behaviors (Barber, 1996; Barber & Olsen, 1997; Galambos et al., 2003); (1b) a high level of psychological control exercised by parents would increase childrens internalizing (Barber & Harmon, 2002) and externalizing problem behaviors (Olsen et al., 2002; Yang et al., 2004); (1c) a high level of parental affection would decrease internal and external child problem behaviors (Dodge et al., 1994; Miller et al., 1993). The second question

was to what extent is the impact of a particular parenting style variable on childrens internal and external problem behaviors moderated by the other two parenting style variables? In other words, is it the combination of parenting style variables rather than their unique impacts that is influential? The following hypotheses were posited: (2a) a high level of affection combined with a high level of behavioral control would decrease internal and external problem behaviors to a greater extent than either a high level of affection with low behavioral control or a low level of affection with high behavioral control (Baumrind, 1989; Forehand & Nousiainen, 1993; Gray & Steinberg, 1999); (2b) a high level of behavioral control combined with a low level of psychological control would decrease a childs externalizing problem behaviors; however, if combined with a high level of psychological control, the impact would be reversed (Galambos et al., 2003); (2c) a high level of psychological control combined with high affection would have a less negative impact on a childs internal and external problem behavior than a high level of psychological control combined with a low level of affection (Gray & Steinberg, 1999; Pettit & Laird, 2002). The third question addressed was to what extent do childrens internal and external problem behaviors predict the kind of parenting their mothers and fathers display?

Method Participants and Procedure Children. The study reported here is part of the Jyvaskyla Entrance into Primary School (JEPS) study (Nurmi & Aunola, 1999). JEPS is an ongoing research project that tracks childrens academic and motivational development during their transition from kindergarten to primary school, and the role that the family and classroom contexts play in this development. The sample selected for the study (N 5 210) consisted of all the 5- to 6-year-old children (age at baseline M 5 75 months, SD 5 3.30 months) born in 1993 in two medium-sized districts in central Finland. Consistent with Finnish school population in general, the sample was homogeneous in terms of race and cultural background (all children were Caucasian and spoke Finnish as their native language). Parental permission to gather data from their children was obtained in August 1999. Permission was given by the parents of 207 children. A representation of the measurement points and variables of the study is shown in Table 1.

Parenting Styles in Childrens Behavior

1147

The children participating in the study were examined six times (Time points 1, 3, 4, 6, 7, and 9; see Table 1): twice during their kindergarten year, that is, in October 1999 (N 5 207) and in April 2000 (N 5 199); twice during their first primary school year, that is, in October 2000 (N 5 196) and in April 2001 (N 5 196); and twice during their second primary school year, that is, in October 2001 (N 5 197) and in March 2002 (N 5 196). Data from all six measurement points were gathered from 196 (104 boys, 92 girls) children. Childrens internalizing and externalizing problem behaviors were assessed by means of structured interviews performed by trained investigators. The interviews took place in a suitable room on the kindergarten or primary school premises. Children who were absent from school on the day of testing (e.g., due to illness) were examined as soon as they were back at school again. The attrition of 11 children was due to the fact that the families of these children moved to other districts and were thus lost to the study. Background information was gathered from the parents of 191 children. A total of 83.2% of the children were from families with two parents, 9.9% of the families consisted of the mother or the father living with her/or his new spouse and their children, and 6.8% of the children were living with their single mother. The number of children per family ranged from 1 to 11 (M 5 2.80, SD 5 1.50). Preliminary analyses showed that children living with two biological parents showed a lower level of externalizing problem behavior at Time 4 and Time 9 than those living with their remarried parents, and a lower level of external problem behavior at Time 9 than those living with their single mother.

Parents. The childrens parents were investigated three times (Time points 2, 5, and 8; see Table 1): during the December of the childrens kindergarten year (Time 2); during the December of the childrens first primary school year (Time 5); and during the December of the childrens second primary school year (Time 8). On these occasions, both of the childrens parents were mailed a questionnaire. They were asked to fill in the questionnaires without discussing the topics with each other. During the December of the kindergarten year the questions concerning parenting styles were filled in by 191 mothers out of 207 (92.3%); during the December of the first grade by 171 out of 196 (87.2%); and during the December of the second grade by 169 out of 196 (86.2%). During the December of the kindergarten year the questions concerning parenting styles were filled in by 167 (80.7%) of the fathers, during the December of the first grade by 157 (74.1%), and during the December of the second grade by 152 (73.4%). A total of 18.0% of the mothers and 13.7% of the fathers had a degree from an institution of university standing, 68.3% of the mothers and 75.2% of the fathers had a qualification from an institution of professional or vocational education, and 13.7% of the mothers and 11.2% of the fathers had no occupational education. To investigate possible selection effects, children whose mothers participated in the study (n 5 197) were compared with children whose mothers did not participate (n 5 10), according to the problem behavior variables. These analyses showed that children whose mothers returned at least one of the three questionnaires showed a higher level of internal problem behavior at Time 1 and Time 3 than those whose mothers did not return any of the questionnaires. Moreover, children

Table 1 Means (M) and Standard Deviations (SD) for the Observed Internal and External Problem Behavior Variables and Mothers and Fathers Parenting Styles at Times 2, 5, and 8 Time 1 M Children Internal problem behavior External problem behavior Mothers Affection Behavioral control Psychological control Fathers Affection Behavioral control Psychological control SD Time 2 M SD Time 3 M SD Time 4 M SD Time 5 M SD Time 6 M SD Time 7 M SD Time 8 M SD Time 9 M SD

1.83 1.41

1.58 1.35 4.16 3.82 2.69 3.92 3.79 2.85 .44 .46 .72 .47 .47 .70

1.52 1.06

1.49 1.21

1.12 0.81

1.29 1.12 4.19 3.77 2.68 3.97 3.76 2.72 .42 .44 .68 .44 .44 .68

1.19 0.89

1.30 1.17

.79 .57

1.01 0.96 4.20 3.70 2.66 3.93 3.69 2.72 .41 .49 .69 .42 .46 .64

.87 .62

1.16 1.12

1148

Aunola and Nurmi

whose fathers did not return any of the three questionnaires (n 5 25) showed a higher level of internal problem behavior at Times 4, 6, and 9, and a higher level of external problem behavior at Times 6 and 9, than those whose fathers returned at least one of the questionnaires (n 5 182). Measurements Problem behavior. Problem behavior was measured by means of a questionnaire (Aunola & Nurmi, 1999a) based on the Johns Hopkins Depression Scale (Joshi, Cappozoli, & Coyle, 1989) and the Strengths and Difficulties Questionnaire (Goodman, Meltzer, & Bailey, 1998). The questionnaire contained 14 statements (9 for internal problem behavior and 5 for external problem behavior), which were read to the children in the structured individual interview situation. Children were asked to respond to each statement in a dichotomized manner, either not true or true. On the basis of confirmatory factor analysis for categorical items, two final summary scores were created. The summary score for internalizing problem behavior consisted of 5 statements focusing on assessment of childrens depressive symptoms (I am worried about many things, I often feel like crying, I often have stomach ache, I often feel tired to do anything, I often get annoyed). The summary score for externalizing problem behavior consisted of four of the statements mainly assessing antisocial behavior and problematic relations with peers (I often end up having trouble with the other kids, Other children annoy me, I often get mad and lose my temper, I often get into fights). The test retest reliability was .82 for internalizing problem behavior and .78 for externalizing problem behavior. Internal consistency assessed with the Kuder Richardson 20 (K R 20) coefficient at successive measurement points was .69, .68, .64, .62, .57, .64 for internalizing problem behavior and .68, .64, .66, .66, .64, .78 for externalizing problem behavior. Split-half reliability (Unequal-length Spearman Brown) at the same points was .72, .79, .68, .64, .63, .69 for internalizing problem behavior and .66, .70, .68, .71, .72, .80 for externalizing problem behavior. The low internal consistency of the scales at some of the measurement points was partly due to the skewed distributions of the individual items. To test the construct validity of the problem behavior measurements (i.e., whether the factor structure would be invariant in different measurement points), confirmatory factor analyses were conducted across different time points using the Mplus sta tistical package (version 3; L. K. Muthen & Muthen,

1998 2004). Separate models were performed for the internalizing (five items) and externalizing (four items) problem behavior constructs. Because the individual items were not normally distributed, the parameters of the models were estimated using a nonnormality robust estimator (MLR; L. K. Muthen n, 1998 2004). Invariance of factor structure & Muthe (i.e., factor loadings) was tested by comparing an unconstrained model with a constrained model (i.e., factor loadings were constrained to be equivalent across time). A test of chi-square difference between these two models using the Satorra Bentler scaled chi-square showed that factor loadings were invariant across six measurement points for both internalizing (w2diff(20, N 5 207) 5 13.99, p ns) and externalizing (w2diff(15, N 5 207) 5 10.08, p ns) problem behaviors. In an earlier data set, the correlation between the scale for internalizing problem behavior and teacher-rated task avoidance among kindergartenaged children was .68 (po.01; Aunola & Nurmi, 1999b). In the present sample, there was a statistically significant correlation between the scale for external problem behavior and received negative peer nominations (from r 5.22, po.01 to r 5.25, po.01). Both scales in the present sample also negatively predicted childrens reading skill development (Halonen, Aunola, Ahonen, & Nurmi, in press). In the present sample, the scale for external problem behavior showed statistically significant correlations with mothers rating of external problem behaviors ranging from .18 (po.05) to .26 (po.01). No correlations were found between child- and mother-reported internalizing problem behaviors. Similarly, in previous research, child and parent evaluations of childrens depression have also shown either no or only low correlations (Engel, Rodrigue, & Geffken, 1994; Ialongo, Edelsohn, & Kellam, 2001; Kazdin, 1990; Kemper, Gerhardstein, Repper, & Kistner, 2003; Leftkowitz & Tesiny, 1984; Weissman, Orvaschel, & Padian, 1980). It has also been suggested that, because the majority of symptoms of depression are private and internal (Kazdin, 1990; Sacco & Graves, 1985) and are not apparent through interactions, self-reports are particularly important in assessing internal problem behaviors (Angold, 1988; Ialongo et al., 2001). Parenting styles. Mothers and fathers parenting styles were measured using a Finnish version of the Blocks Child Rearing Practices Report (CRPR; Roberts, Block, & Block, 1984), which includes items tapping child-rearing attitudes, values, and behaviors. On the basis of the factor analyses (for a

Parenting Styles in Childrens Behavior

1149

more detailed description see Aunola & Nurmi, 2004), summary scores for the three parenting style dimensions were calculated separately for each measurement point. Affection, including items reflecting a positive relationship with the child (10 items; e.g., I often tell my child that I appreciate what he/she tries out or achieves, I often show my child that I love him/her). Behavioral control, including items showing that misbehavior would have clear consequences and parental willingness to confront a child who disobeys (Baumrind, 1991; Maccoby & Martin, 1983) (6 items; e.g., My child should learn that we have rules in our family, When I am angry with my child, I let him/her know about it, If my child misbehaves I usually rebuke him/her). The operationalization of behavioral control has varied from one study to another. For example, in some studies, behavioral control has been operationalized as monitoring (parental knowledge; Kerr & Stattin, 2000) or supervision (Gray & Steinberg, 1999; Pettit et al., 2001), whereas in some other studies it has been operationalized as demandingness, limit setting, or maturity demands (either with or without reasoning about rules and the consequences of misbehavior; Baumrind, 1989, 1991; Maccoby & Martin, 1983; Hart et al., 2003). Moreover, in some studies, the term control has been used to refer to restrictive, harsh, or punitive control (e.g., Gadeyne et al., 2004; Pettit, Bates, & Dodge, 1997). In the present study, the behavioral control scale was characterized by parents limit setting and maturity demands on the child (Baumrind, 1991) rather than by harsh control or restrictiveness. The scale used in the present study assessed directive rather than reasoning-oriented behavioral control. The original questionnaire included items which also measured monitoring and parental knowledge of their childrens whereabouts. However, because in the present sample these items showed only slight variance, they were excluded from the analyses. Psychological control (Control by guilt), including items that reflected parental attitudes appealing to guilt and expressing disappointment (Barber, 1996; Schaefer, 1965) (4 items; e.g., I believe a child should be aware of how much I have done for him/her, I let my child see how disappointed and ashamed I am if he/she misbehaves). The respective Cronbachs alpha reliabilities for the three parenting style dimensions for mothers were .82, .66, and .79 at Time 2; .81, .66, and .77 at Time 5; and .82, .70, and .76 at Time 8. The respective Cronbachs alpha reliabilities for fathers were .82, .70, and

.74 at Time 2; .84, .69, and .72 at Time 5; and .80, .70, and .75 at Time 8. Test retest reliabilities for the three parenting style dimensions for mothers were .89, .82, and .88. The test retest reliabilities for fathers were .89, .85, and .85. Similar reliabilities have been reported in earlier studies on parenting styles (see, e.g., Chen, Liu, & Li, 2000; Dornbusch, Ritter, Leiderman, Roberts, & Fraleigh, 1987; Gadeyne et al., 2004; Glasgow, Dornbusch, Troyer, Steinberg, & Ritter, 1997; Pettit et al., 1997; Gray & Steinberg, 1999). To further validate the factor structure of parenting styles, factor analyses for the scale were performed in another sample (consisting of mothers and fathers of 280 adolescents). The results showed an identical three-factor structure among both mothers and fathers (Aunola, Nurmi, Salmela-Aro, Niemivirta, & Vuori, 2005). The Cronbachs alpha reliabilities for the three parenting style dimensions in this other data set were .88 (mothers) and .86 (fathers) for affection, .72 and .72 for behavioral control, and .79 and .80 for psychological control. In the present sample, the correlation between affection and psychological control for mothers ranged between .03 and .11 (p ns) across the three measurements. Behavioral control correlated positively and statistically significantly with both affection (range from .17 to .18, po.05) and psychological control (range from .28 to .39, po.01). For fathers, behavioral control correlated positively and statistically significantly with psychological control (range from .18, po.05 .29, po.01). Affection did not significantly correlate with either behavioral (r 5.01 .06, p ns) or psychological (r 5 .13 to .04, p ns) control at any of the measurement points. Results The research questions were examined using latent growth curve modeling (LGM) (Duncan, Duncan, Strycker, Li, & Alpert, 1999). This methodology makes it possible to investigate intraindividual changes in a particular variable of interest over time (i.e., growth trajectories) as well as interindividual differences in such changes (B. O. Muthen & Khoo, 1998) by differentiating the initial level component and different growth components (e.g., linear growth and quadratic growth). Moreover, the LGM methodology enables examination of the associations between these growth components across several variables, as well as their interactions. The analyses were performed using the Mplus statistical package (version 3; L. K. Muthen & Muthen, 1998 2004). The parameters of the models were estimated using the missing data method with an MLR estimator (L. K. Muthen & Muthen, 1998

1150

Aunola and Nurmi

2004). This method allowed all the available data to be used in the estimation of the parameters of the models. Goodness of fit was evaluated using four indicators: w2/df; Bentlers (1990) comparative fit index (CFI); the Tucker Lewis index (TLI); and the standardized root mean square residual (SRMR). The means and standard deviations for the observed variables are given in Table 1. LGMs for Individual Variables As a first step, unconditional growth curve models (i.e., models without any predictors) were constructed separately for each variable in order to investigate the extent of individual variation in the initial level and the linear and quadratic growth components of each variable. In the case of childrens internal and external problem behaviors, models were constructed across six measurement points. The model construction was started in both cases by testing a model which included three growth components: initial status, linear trend, and quadratic trend. However, because the quadratic term showed no statistically significant variance in these models, its variance was fixed at zero in the final models. The results for the final models are shown in Table 2. The results for childrens internal problem behavior (w2(15, N 5 207) 5 32.29; CFI 5 0.94; TLI 5 0.94; SRMR 5 0.06) indicated that, at the mean level, childrens internalizing problem behaviors showed a decreasing trend, leveling out over time (see Figure 1). Moreover, the results showed that the variance both of the level and of the linear trend were statistically significant, indicating that there were significant individual differences both in the initial status and in the developmental trend of internal problem behaviors. The correlation between the level and linear trend of internal problem behavior was negative and statistically significant (standardized estimate 5 .67, po.01), suggesting that the higher the initial level of internal problem behavior, the more the level of it decreased during the study period. The results for external problem behavior (w2(15, N 5 207) 5 29.81; CFI 5 0.93; TLI 5 0.93; SRMR 5 0.07) showed a similar pattern (Table 2). At the mean level, externalizing problem behavior showed a decreasing trend, leveling out over time (see Figure 1). Moreover, the results showed a statistically significant variance in the level and linear trend of external problem behaviors, indicating the presence of significant individual differences in both the initial status and developmental trend of external problem behaviors. The correlation between the Level and Linear trend of external problem behavior was negative and statisti-

cally significant (standardized estimate 5 .54, po.05), suggesting that the higher the initial level of external problem behavior, the more the level of it decreased during the study period. Models for mothers and fathers parenting styles, that is, affection, behavioral control, and psychological control, were constructed across three measurement points. Consequently, these models contained only two growth components: initial status and linear trend. These models for parenting styles were tested separately for each dimension and also separately for mothers and fathers. The results of the models are presented in Table 2. The results for mothers parenting styles showed, first, that at the mean level behavioral control slightly decreased over time (Figure 2, Table 2). No meanlevel changes were evident in mothers affection or psychological control. Second, the variances of the level components of all three parenting style dimensions were statistically significant. However, the variances for the linear trends of the three parenting style variables were not significant. This suggests that, although mothers showed different levels of affection, behavioral control, and psychological control, there were no individual differences in the changes in these parenting styles. The results for fathers parenting styles showed a somewhat similar pattern. At the mean level, fathers behavioral and psychological control decreased across time, whereas no mean-level changes were evident in fathers affection (Figure 2, Table 2). As with mothers, the fathers results also showed statistically significant interindividual variance in the Levels of affection, behavioral control, and psychological control. However, no statistically significant variance emerged in the linear trends of any of these variables. The absence of interindividual variation (i.e., individual diversity) in the linear trend of mothers and fathers parenting styles means also that childrens problem behaviors, or any other variable, do not predict changes in mothers and fathers parenting styles in this sample (because there is no individual variation in these changes to be predicted). Because individual variation existed only in the levels of parenting styles, the linear trend components of these variables were not included in the subsequent conditional growth models. Mothers Parenting Styles and Childrens Problem Behavior Internal problem behavior. To examine whether the levels of mothers affection, behavioral control, and psychological control would be associated with the initial level of childrens internal problem behavior

Parenting Styles in Childrens Behavior

1151

Table 2 Parameter Estimates (Unstandardized Forms) of Latent Growth Models for Childrens Internal and External Problem Behaviors and Mothers and Fathers Parenting Styles (Each in Separate Analyses) Means Level Linear trend Quadratic trend Level Variances Linear trend Quadratic trend

Estimate t-value Estimate t-value Estimate t-value Estimate t-value Estimate t-value Estimate t-value Internal problem behavior External problem behavior Mothers affection Mothers behavioral control Mothers psychological control Fathers affection Fathers behavioral control Fathers psychological control 1.84 1.38 4.16 3.82 2.70 3.92 3.80 2.83 17.53 15.48 136.33 122.03 54.40 111.03 111.84 54.64 .38 .31 .01 .05 .01 .00 .05 .06 5.81 4.81 1.04 3.03 0.62 0.16 3.11 2.52 .04 .03 3.13 2.68 1.24 0.72 0.16 0.13 0.35 0.18 0.14 0.31 6.47 5.50 7.32 6.77 8.43 6.97 7.29 5.08 .03 .02 .02 0a 0a .01 0a .02 2.89 2.56 1.90 0a 0a

1.34 0.58

Note. t-Values greater than 1.96 in magnitude indicate a parameter estimate that is significantly different from zero. a Fixed.

and changes in this behavior across the six measurements, an LGM was constructed which included the associations between the level of the parenting style variables (Time 2) and the initial level of childrens internal problem behavior, and the paths from the level of parenting styles to the linear trend of internal problem behavior. The fit of the model was good: w2(27, N 5 207) 5 47.97; CFI 5 0.93; TLI 5 0.92; SRMR 5 0.05. The results showed, however, that the parenting style variables were not associated with the level of childrens internal problem behavior nor did they have any effects on its linear trend. As a second step, the interaction terms of the three parenting style variables (both two- and three-way interactions) were added to the model in separate
Internal Problem Behavior External Problem Behavior

analyses for each. The results showed that the interaction term Affection Psychological Control significantly predicted the linear trend of internal problem behavior (standardized estimate 5 .20, po.05). To investigate this effect in detail, a median split was used to divide the data into two groups according to the level of maternal psychological control at Time 2. Because the parenting style variables were not perfectly normal, medians were thought to better represent the central tendencies of the distributions than means (Owens & Shaw, 2003).
Affection Behavioral Control Psychological Control Affection Behavioral Control Psychological Control

4.3 4.1 3.9 Parenting Style Score 3.7 3.5 3.3 3.1 2.9 2.7 2.5

2.0 1.8 1.6 Problem Behavior Score 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 1 3 4 6 7 9 Measurement Point

5 Measurement Point

Figure 1. Unconditional growth curves for internal and external problem behaviors.

Figure 2. Unconditional growth curves for mothers (straight lines) and fathers (dotted lines) parenting styles.

1152

Aunola and Nurmi

Psychological Control: Low level 0.6 Change in the Linear Trend of Internal Problem Behavior 0.5 0.4 0.3 0.2 0.1 0.0 0.1 0.2 0.3 1 3 Affection Level 5 High level

Growth curve analysis for childrens internal problem behavior and mothers affection was then conducted for these two groups using a multisample approach for the LGM (Curran & Hussong, 2003; Joreskog & Sorbom, 1993; Rigdon, Schumacker, & Wothke, 1998). In this analysis, the path from affection to the linear trend of internal problem behavior was allowed to be estimated separately for each group, w2(47, N 5 191) 5 56.64; CFI 5 0.97; TLI 5 0.97; SRMR 5 0.08. The results showed that, among those children whose mothers reported a high level of psychological control (n 5 102), the higher the level of maternal affection, the greater the increase the children showed in internal problem behavior (standardized estimate 5 .25, po.05). Among those whose mothers showed a low level of psychological control (n 5 89), maternal affection had no effects on the trend of childrens internal problem behavior (standardized estimate 5 0.13, p ns). The results are shown in Figure 3. In order to investigate the Affection Psychological Control interaction further, multisample analyses were performed by dividing the sample into two groups according to maternal affection at Time 2 and then investigating the effects of psychological control on the linear trend of internal problem behavior in these two groups. The results showed that among those children whose mothers reported a low level of affection, maternal psychological control had no effects on the trend of childrens internal problem behavior (standardized estimate 5 .07, p ns). Among those children whose mothers reported a high level of affection the impact of psychological control did not reach statistical significance either (standardized estimate 5 .27, t 5 1.71, p ns). Because of the somewhat low reliabilities of the internalizing problem behavior scale at measurement points 4 (Time 6) and 5 (Time 7), additional analyses were conducted in which the data gathered at these two time points were excluded from all statistical analyses reported previously. In these analyses, the results and major conclusions concerning the role of mothers parenting styles in childrens internalizing problem behaviors remained unchanged. External problem behavior. Next, an analogous LGM was constructed for childrens external problem behavior and mothers parenting styles, w2(27, N 5 207) 5 39.89; CFI 5 0.95; TLI 5 0.94; SRMR 5 0.06. The results showed that the level of maternal affection predicted the Linear trend of external problem behavior (standardized estimate 5 .26, po.05) but that the other parenting dimensions did not. However, examination of the interaction terms showed that the interaction terms between Affection

Figure 3. Change in the linear trend of internal problem behavior at different levels of maternal affection (range 1 5) and psychological control (low/high).

Psychological Control (standardized estimate 5 .26, po.05), and Behavioral control Psychological Control (standardized estimate 5 .26, po.05) significantly predicted the linear trend of external problem behavior. To investigate these effects in detail, the data were divided into two groups according to the level of maternal psychological control at Time 2 (median split). Growth curve analysis for childrens external problem behaviors and mothers affection and behavioral control was then conducted for these two groups, allowing the paths from the level of affection and behavioral control to the linear trend of external problem behavior to be estimated separately for each group, w2(58, N 5 191) 5 65.96; CFI 5 0.97; TLI 5 0.97; SRMR 5 0.09. The results showed, first, that, among children whose mothers reported a high level of psychological control (n 5 102), a high level of maternal affection predicted an increase in childrens external problem behavior (standardized estimate 5 .42, po.001). When combined with a low level of psychological control, no affection effects were found (standardized estimate 5 .09, p ns). Second, among children whose mothers reported a low level of psychological control, behavioral control predicted a decrease in childrens external problem behavior (standardized estimate 5 .27, po.05). When combined with a high level of psychological control, no effects of behavioral control were found (standardized estimate 5 .01, p ns). The results are shown in Figures 4 and 5. In order to investigate the Affection Psychological Control interaction further, and that of Behavioral Control Psychological Control, two subsequent multisample analyses were conducted. First, the sample was divided into two groups according to maternal affec-

Parenting Styles in Childrens Behavior

1153

tion at Time 2, and the effects of psychological control on the linear trend of external problem behavior in these two groups were investigated. The results showed that among those children whose mothers reported a low level of affection, maternal psychological control decreased childrens external problem behavior (standardized estimate 5 .35, po.01). By contrast, among those children whose mothers reported a high level of affection maternal psychological control increased their childrens external problem behavior (standardized estimate 5 .31, po.05). Second, the sample was divided into two groups according to maternal behavioral control at Time 2, and the effects of psychological control on the linear trend of external problem behavior were investigated. The results showed that among those children whose mothers reported a low level of behavioral control, no effects of psychological control on the trend of external problem behavior were found (standardized estimate 5 .04, p ns). Among those children whose mothers reported a high level of behavioral control, the impact of psychological control did not reach statistical significance either (standardized estimate 5 .19, p ns).

Fathers Parenting Styles and Childrens Problem Behavior Next, analogous analyses were performed for fathers parenting styles and childrens internal and external problem behaviors separately. However, no associations between the levels of fathers parenting styles, and childrens level and linear trends of internal or external problem behaviors were found.
Psychological Control:
low level high level

Because of the somewhat low reliabilities of the internalizing problem behavior scale at measurement points 4 (Time 6) and 5 (Time 7), additional analyses were conducted in which the data gathered at these two time points were excluded. The results and major conclusions concerning the role of fathers parenting styles in childrens internalizing problem behaviors remained unchanged, with one exception: when using only four as against all six measurement points for internalizing problem behaviors, fathers psychological control was positively associated with the rate of change of internalizing problem behavior. Because in the present study fewer fathers than mothers participated, it is possible that the absence of statistically significant predictions in the fathers data is due to low sample size. To investigate this possibility, we performed various simulation analyses to test whether any effects of fathers parenting styles would become statistically significant if there had been as many fathers as mothers (N 5 191) in the original sample. These simulation analyses showed that, when the sample of fathers was made equal to that of mothers, fathers high psychological control was found to predict positively the linear trend of childrens externalizing problem behaviors (standardized estimate 5 .26, po.05; R2 5 .07). No other effects turned out to be significant in these simulation analyses.

Discussion This study examined the extent to which the impact of a particular parenting dimension on childrens internal and external problem behaviors would be
Psychological Control:
low level high level

0.1 Change in the Linear Trend of External Problem Behavior 0.0 0.1 0.2 0.3 0.4 0.5

1.0 Change in the Linear Trend of External Problem Behavior 0.8 0.6 0.4 0.2 0.0 0.2 1 3 Affection Level 5

3 Behavioral Control Level

Figure 4. Change in the linear trend of external problem behavior at different levels of maternal affection (range 1 5) and psychological control (low/high).

Figure 5. Change in the linear trend of external problem behavior at different levels of maternal behavioral control (range 1 5) and psychological control (low/high).

1154

Aunola and Nurmi

found to be dependent on the other two dimensions. The results showed that a high level of maternal psychological control combined with high affection predicted increases in the levels of childrens internal and external problem behaviors during their transition from kindergarten to primary school. By contrast, a high level of maternal behavioral control combined with a low level of psychological control predicted decreases in the level of childrens external problem behavior. The finding that mothers psychological control in combination with high affection was detrimental to child adjustment was unexpected. There are at least two possible explanations for this result. The first is that high affection when combined with psychological control, that is, guilt-inducing and manipulative child-rearing, leads to enmeshment in parent child communication (Aunola & Nurmi, 2004). On the one hand, mothers are supportive and close to the child but, on the other hand, they communicate guiltinducing attitudes which manipulate the childs psychological world and increase the childs dependence. Such enmeshment, being characterized as family communication patterns that lead to psychological and emotional fusion among family members (Barber & Buehler, 1996), restricts childrens expression of their own thoughts and emotions. Consequently, it may also lead to various forms of problem behaviors among children (Schaefer, 1965; Humphrey, 1989). The second explanation is that a child-rearing pattern characterized by both a high level of affection and a high level of psychological control may communicate an inconsistent message of maternal approval and love to a child (Barber, 1996). In parenting, such a discrepancy (Punamaki, Qouta, & El-Sarraj, 2001), or double message (Humphrey, 1989), may impact negatively on child adjustment, for example, by provoking anxiety and by diminishing the childs sense of control (Chorpita & Barlow, 1998). The fact that in the present study high psychological control combined with high affection increased not only internal but also external problem behavior among children suggests that internal distress and negative emotions aroused by guilt-inducing mothering also has an influence on a childs undercontrolling behavior (see also Olsen et al., 2002; Yang et al., 2004). The results of this study showed further that, among those children whose mothers reported low affection, maternal psychological control predicted a decrease in childrens external problem behavior. One possible explanation for this result is that low affection combined with low psychological control is a sign of disengaged and neglectful parenting, which

has been shown to be the most detrimental combination in child development (Baumrind, 1989, 1991). By contrast, low affection combined with high psychological control may be a more functional parenting pattern, which also leads to a decrease in external problem behavior. Overall, the results of this study concerning the interaction between maternal affection and psychological control were against our hypothesis (2c) and also different from the findings reported previously for adolescents. For example, Gray and Steinberg (1999) and Pettit and Laird (2002) found that parental affection and involvement decreased problem behaviors, particularly when combined with a high level of psychological control. By contrast, the results of the present study among children showed that maternal affection when combined with high psychological control had a negative impact on child adjustment. We have previously shown a similar pattern of impact of maternal parenting on childrens mathematics skill development using the same sample (Aunola & Nurmi, 2004). One explanation for this discrepancy between the findings of the present study and those of previous research is that the impact of parenting varies according to the childs age. For example, it may be that manipulation of the affective relationship with the child by use of psychological control (Baumrind, 1966; Schaefer, 1965), that is, contingent affection, is harmful, especially during middle childhood, because at that age children have not yet begun to separate emotionally and psychologically from their parents (Barber et al., 1994). By contrast, during adolescence, when individuals are gaining increasing autonomy from their parents, a warm parental relationship may function as a protective factor against the negative impacts of psychological control. However, because there are thus far only few studies conducted among kindergarten and school-aged children, further studies should be performed to support this explanation. The other possible explanation is the fact that in previous studies parenting styles have been measured using adolescents reports (Gray & Steinberg, 1999; Pettit & Laird, 2002), whereas in the present study we used parent-reported parenting styles. It is possible, for example, that children and adolescents are unable to differentiate between psychological control and affection in their mothers and fathers parenting to the same extent as are parents themselves, and therefore child-reported parenting measures are unable to tap the enmeshment consequences of parenting characteristics. The results of the present study showed further that, among mothers, psychological control moder-

Parenting Styles in Childrens Behavior

1155

ated the impact of behavioral control on childrens external problem behaviors: a high level of maternal behavioral control predicted a decrease in the childs external problems, but only if combined with a low level of psychological control. When combined with a high level of psychological control, behavioral control was found to have no impact. This result is consistent with that of Galambos et al. (2003), who found that parental high behavioral control was negatively associated with adolescents external problem behaviors but only if combined with low levels of psychological control. In previous research, parental behavioral control has been shown to be associated with low levels of various kinds of externalizing problems among children and adolescents, such as antisocial behavior and conduct disorders (e.g., Barber, 1996; Hart et al., 2003). It has been suggested that behavioral control exercised by parents prevents externalizing problem behaviors because it fosters childrens own self-regulation in social situations (Lewis, 1981; Hart et al., 2003). The results of the present study, and those found by Galambos et al. (2003), suggest, however, that the positive impacts of behavioral control vanish if mothers show simultaneously a high level of psychological control. One possible explanation for why behavioral control has beneficial effects particularly when combined with a low psychological control is that such a pattern of parenting provides clear rules and expectations for the child but simultaneously also allows him or her to experience and express own thoughts and emotions freely, described also in terms of psychological autonomy (Hart et al., 2003). This kind of autonomy, or feeling of being respected as an individual, has been suggested to be important in the processes of internalizing rules and becoming self-governing (Grolnick, 2003; Hart et al., 2003). The fact that the positive effect of mothers behavioral control on childrens external problem behaviors diminished when combined with a high psychological control may be because high behavioral control together with high psychological control reflects parental overmanagement (Pettit et al., 2001), and in this way intrudes on rather than benefits the childs autonomy and self-regulation (Galambos et al., 2003). This kind of overmanagement resembles in fact the authoritarian parenting style (Barber, 1996; Baumrind, 1966, 1989, 1991; Barber, Bean, & Erickson, 2002). Overall, the results of the present study add to the previous literature by showing that behavioral control is a good strategy for preventing external problem behavior among children (Coopersmith, 1967; Pettit et al., 2001), provided that the childs sense of competence and autonomy are not simultaneously

violated by intrusive and manipulative parenting. The fact that mothers behavioral control combined with low psychological control decreased external problem behaviors rather than internal problem behavior is consistent with previous findings suggesting that behavioral control is an important factor in fostering pro-social behavior and preventing external problem behaviors in particular (Lewis, 1981; Hart et al., 2003). In this study, no interactions were found between parental affection and behavioral control in the prediction of childrens problem behaviors. This result resembles that of Galambos et al. (2003), but conflicts with the results reported by Gray and Steinberg (1999) and Forehand and Nousiainen (1993). The present result is also in contradiction to the traditional typological theory of parenting styles, according to which it is a combination of high affection and high behavioral control that leads to the most positive child outcomes (Maccoby & Martin, 1983; Steinberg, 2001). One possible explanation for the inconsistencies between the results found in the present study and those of some earlier ones is that psychological control in previous studies has typically been included in either the affection (as in measuring negative affect or lack of autonomy granting) or behavioral control (as in measuring control in terms of restrictiveness) dimensions of parenting, or incorporated into the parenting style typology (e.g., as a characteristic of authoritarian parenting) rather than measured as a separate variable (for a review, see Pettit & Laird, 2002). Thus, researchers have not distinguished the impacts of psychological control from the impacts of affection and behavioral control. The results of this study suggest that such differentiation is important, because psychological control in combination with affection, or in combination with behavioral control, is a more powerful predictor of the childs adjustment than is the combination of affection and behavioral control (see also Aunola & Nurmi, 2004). The results of the present study showed further that mothers parenting played an important role in young childrens problem behaviors, whereas fathers parenting had only a marginal role. One explanation for the importance of mothers parenting style is that, because mother child interactions are characterized more than father child interactions by warmth, responsiveness, and intimate exchanges (Collins & Russel, 1991; Forehand & Nousiainen, 1993), children are more open to maternal than paternal influences (Darling & Steinberg, 1993). Mothers, compared to fathers, have also been shown to make a greater effort to maintain dependency in

1156

Aunola and Nurmi

their children (for a review, see Collins & Russel, 1991). It may be that the impact of psychological control in this kind of dependent relationship is greater than in the more autonomous father child relationship. Another possible explanation for fathers marginal role is that, in the present study, the measures of parenting did not cover the kind of interaction that is important in a father child relationship. For example, it has been suggested previously that fathers playful, companionable, and patient interaction styles are particularly important for childrens adjustment (Hart et al., 2003). The third possible explanation for the minor effect of fathers parenting is that the sample of fathers was too small for the effects of fathers parenting to be detected statistically (Chou, Bentler, & Pentz, 2000). The results of this study also showed that it was parenting styles that contributed to childrens problem behaviors rather than vice versa: the fact that there were no interindividual variances in the trends of any of the parenting variables suggest that child outcomes did not influence their parents parenting styles during this early period (see also Aunola & Nurmi, 2004). This result is in accordance with the traditional notion of parents role in childrens socialization (for a review, see Maccoby, 1992), but runs contrary to the notion that child characteristics influence parenting (Bell, 1968; Harris, 1995). Although in the present study childrens problem behaviors were not found to predict changes in mothers and fathers parenting styles, it is still possible that the impact of parenting is dependent on child characteristics. The present study does not, for example, answer the question of what kind of role parenting styles play in modifying the expression of genetically based temperamental differences (Collins, Maccoby, Steinberg, Hetherington, & Bornstein, 2000; Repetti, Taylor, & Seeman, 2002) or in maintaining, or exacerbating, preexisting vulnerabilities in risky families (Dodge & Pettit, 2003; Hart et al., 2003; Repetti et al., 2002). There are six limitations that should be considered in any attempt to generalize the findings of this study. First, problem behaviors were assessed on the basis of childrens self reports. The rationale for using this methodology was that there are some findings suggesting that children are already able to assess their behavior reliably at quite a young age (Edelsohn, Ialongo, Werthamer-Larsson, & Crockett, 1992; Ialongo et al., 2001). However, because it is possible that the results obtained using childrens assessments will differ somewhat from those obtained using other data sources, there is a need in future research to gather information from several

sources, for example, child and parent reports as well as observational data, to form a more complete understanding of the role of parenting. Second, the problem behavior measurement used in this study focused mainly on one aspect of internalizing problem behavior, that is, depressive symptoms, and on one aspect of externalizing problem behavior, that is, antisocial behavior/problematic peer relations. This in turn points to the need in future studies to investigate the effects of parenting styles on other domains of internalizing (e.g., anxiety, fearfulness) and externalizing (e.g., impulsiveness, oppositional and nonconforming behaviors) problem behaviors as well. Third, parent-reported questionnaires were used to measure parenting styles. Although measures of this kind provide information about parental attitudes, there is an obvious advantage to be gained from using observational methods, which study how parents actually behave in interaction situations with their offspring. The fact that findings from studies that have used observational methods and those that have used self-report measures of parenting on childrens problem behaviors are relatively consistent (for a review, see Hart et al., 2003) suggests, however, that different information sources may, at least in part, yield similar information about parental behaviors. Fourth, the instrument used in the present study to measure psychological control emphasized the guilt-inducing aspect rather than, for example, excessive criticism and restrictive or possessive communication. It is possible, therefore, that the present results concerning the impacts of psychological control hold particularly for guilt induction rather than other aspects of psychological control. Fifth, the internal consistencies for the parent-reported behavioral control construct were somewhat low. Clearly, there is a need to develop measures targeted at the parents of kindergarten and school-aged children. Finally, in the present study, fathers data were subject to a selection effect. Children whose fathers did not participate reported a higher level of both internalizing and externalizing problem behaviors than those whose fathers participated. It is therefore possible that, although the effects of fathers parenting were not strong in a sample biased in favor of children who showed a somewhat low level of problem behaviors, these effects might have turn out to be significant among those with higher levels of problem behaviors. Moreover, before making any generalizations concerning the role of fathers parenting styles, there is a need to investigate the role of fathers using larger and more representative samples. Overall, the results of this study add to the previous literature on parenting styles and childrens

Parenting Styles in Childrens Behavior

1157

adjustment by showing, first, that in mothers high psychological control together with a high level of affection appears to be the most detrimental combination for the development of problem behaviors among children. Second, mothers high behavioral control was shown to decrease childrens external problem behaviors but only in combination with a low level of psychological control. These results support the notion that it is not single parenting style variables as such, but rather their combination that is influential in child development (Baumrind, 1991; Darling & Steinberg, 1993). References
Angold, A. (1988). Childhood and adolescent depression: I. Epidemiological and aetiological aspects. British Journal of Psychiatry, 152, 601 617. Aunola, K., & Nurmi, J.-E. (1999a). Problem behavior scale for children (self-rating form). Unpublished test material, University of Jyvaskyla, Finland. Aunola, K., & Nurmi, J.-E. (1999b). Unpublished data, University of Jyvaskyla, Finland. Aunola, K., & Nurmi, J.-E. (2004). Maternal psychological control moderates the impact of affection on childrens math performance. Developmental Psychology, 40, 965 978. Aunola, K., Nurmi, J.-E., Salmela-Aro, K., Niemivirta, M., & Vuori, J. (2005). The role of mothers and fathers parenting styles in adolescents educational plans and aspirations. Manuscript under preparation. Barber, B. K. (1996). Parental psychological control: Revisiting a neglected construct. Child Development, 67, 3296 3319. Barber, B. K., Bean, R. L., & Erickson, L. D. (2002). Expanding the study and understanding of psychological control. In B. K. Barber (Ed.), Intrusive parenting. How psychological control affects children and adolescents (pp. 263 289). Washington, DC: American Psychological Association. Barber, B. K., & Buehler, C. (1996). Family cohesion and enmeshment: Different constructs, different effects. Journal of Marriage and the Family, 58, 433 441. Barber, B. K., & Harmon, E. L. (2002). Violating the self: Parental psychological control of children and adolescents. In B. K. Barber (Ed.), Intrusive parenting. How psychological control affects children and adolescents (pp. 15 52). Washington, DC: American Psychological Association. Barber, B. K., & Olsen, J. A. (1997). Socialization in context: Connection, regulation, and autonomy in the family, school, and neighborhood, and with peers. Journal of Adolescent Research, 12, 287 315. Barber, B. K., Olsen, J. A., & Shagle, S. (1994). Associations between parental psychological control and behavioral control and youth internalized and externalized behaviors. Child Development, 65, 1120 1136.

Baumrind, D. (1966). Effects of authoritative parental control on child behavior. Child Development, 37, 887 907. Baumrind, D. (1989). Rearing competent children. In W. Damon (Ed.), Child development today and tomorrow (pp. 349 378). San Francisco: Jossey-Bass. Baumrind, D. (1991). The influence of parenting styles on adolescent competence and substance use. Journal of Early Adolescence, 11, 56 95. Bell, R. (1968). A reinterpretation of the direction of effects in studies of socialization. Psychological Review, 75, 81 95. Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 99, 229 246. Block, J. H., & Block, J. (1980). The role of ego-control and ego-resiliency in the organization of behavior. In W. A. Collins (Ed.), Minnesota symposia on child psychology: Vol. 13. Development of cognition, affect, and social relations (pp. 39 101). Hillsdale, NJ: Erlbaum. Chen, X., Liu, M., & Li, D. (2000). Parental warmth, control, and indulgence and their relations to adjustment in Chinese children: A longitudinal study. Journal of Family Psychology, 14, 401 419. Chorpita, B. F., & Barlow, D. H. (1998). The development of anxiety: The role of control in the early environment. Psychological Bulletin, 124, 3 21. Chou, C.-P., Bentler, P. M., & Pentz, M. A. (2000). A twostage approach to multilevel structural equation models: Application to longitudinal data. In T. D. Little, K. U. Schnabel, & J. Baumert (Eds.), Modeling longitudinal and multilevel data. Practical issues, applied approaches and specific examples (pp. 33 49). Mahwah, NJ: Lawrence Erlbaum Associates. Cole, P. M., Zahn-Waxler, C., Fox, N. A., Usher, B. A., & Welsh, J. D. (1996). Individual differences in emotion regulation and behavior problems in preschool children. Journal of Abnormal Psychology, 105, 518 529. 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, 218 232. Collins, W. A., & Russell, G. (1991). Mother child and father child relationships in middle childhood and adolescence: A developmental analysis. Developmental Review, 11, 99 136. Conger, K. J., Conger, R. D., & Scaramella, L. V. (1997). Parents, siblings, psychological control, and adolescent adjustment. Journal of Adolescent Research, 12, 113 138. Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco: Freeman. Curran, P. J., & Hussong, A. M. (2003). The use of latent trajectory models in psychopathology research. Journal of Abnormal Psychology, 112, 526 544. Darling, N., & Steinberg, L. (1993). Parenting style as context: An integrative model. Psychological Bulletin, 113, 487 496. Denham, S. A., Workman, E., Cole, P. M., Weissbrod, C., Kendziora, K. T., & Zahn-Waxler, C. (2000). Prediction of externalizing behavior problems from early to middle childhood: The role of parental socialization and emo-

1158

Aunola and Nurmi Halonen, A., Aunola, K., Ahonen, T., & Nurmi, J.-E. (in press). The role of learning to read in the development of problem behaviour: A cross-lagged longitudinal study. British Journal of Educational Psychology. Harris, J. R. (1995). Where is the childs environment?. A group socialization theory of development. Psychological Review, 102, 458 489. Hart, C. H., Newell, L. D., & Olsen, S. F. (2003). Parenting skills and social-communicative competence in childhood. In J. O. Greene & B. R. Burleson (Eds.), Handbook of communication and social interaction skills (pp. 753 797). Mahwah, NJ: Lawrence Erlbaum Associates. Heller, T. L., Baker, B. L., Henker, B., & Hinshaw, S. P. (1996). Externalizing behavior and cognitive functioning from preschool to first grade: Stability and predictors. Journal of Clinical Child Psychology, 25, 376 387. Hinshaw, S. (1992). Externalizing behavior problems and academic underachievement in childhood and adolescence: Causal relationships and underlying mechanisms. Psychological Bulletin, 111, 127 155. Humphrey, L. L. (1989). Observed family interactions among subtypes of eating disorders using structural analysis of social behavior. Journal of Consulting and Clinical Psychology, 57, 206 214. Ialongo, N. S., Edelsohn, G., & Kellam, S. G. (2001). A further look at the prognostic power of young childrens reports of depressed mood and feelings. Child Development, 72, 736 747. Jewell, J. D., & Stark, K. D. (2003). Comparing the family environments of adolescents with conduct disorder and depression. Journal of Child and Family Studies, 12, 77 89. Joshi, P., Cappozoli, J., & Coyle, J. (1989). The Johns Hopkins Depression Scale: Normative data and validation in child psychiatry patients. Journal of American Academy of Child and Adolescent Psychiatry, 29, 283 288. Joreskog, K., & Sorbom, D. (1993). LISREL8: Structural equation modeling with the SIMPLIS command language. Hillsdale, NJ: Lawrence Erlbaum Associates. Kazdin, A. E. (1990). Childhood depression. Journal of Child Psychology and Psychiatry, 31, 121 160. Kemper, T. S., Gerhardstein, R., Repper, K. K., & Kistner, J. A. (2003). Mother child agreement on reports of internalizing symptoms among children referred for evaluation of ADHD. Journal of Psychopathology and Behavioral Assessment, 25, 239 250. Kerr, M., & Stattin, H. (2000). What parents know, how they know it, and several forms of adolescent adjustment: Further support for a reinterpretation of monitoring. Developmental Psychology, 36, 366 380. Kovacs, M., & Devlin, B. (1998). Internalizing disorders in childhood. Journal of Child Psychology and Psychiatry, 39, 47 63. Leftkowitz, M. M., & Tesiny, E. P. (1984). Rejection and depression: Prospective and contemporary analysis. Developmental Psychology, 20, 776 785.

tion expression. Development and Psychopathology, 12, 23 45. Dodge, K. A., & Pettit, G. S. (2003). A biopsychosocial model of the development of chronic conduct problems in adolescence. Developmental Psychology, 39, 349 371. Dodge, K. A., Pettit, G. S., & Bates, J. E. (1994). Socialization mediators of the relation between socioeconomic status and child conduct problems. Child Development, 65, 649 665. Dornbusch, S. M., Ritter, P. L., Leiderman, P. H., Roberts, D. F., & Fraleigh, M. J. (1987). The relation of parenting style to adolescent school performance. Child Development, 58, 1244 1257. Duncan, T. E., Duncan, S. C., Strycker, L. A., Li, F., & Alpert, A. (1999). An introduction to latent variable growth curve modeling. Concepts, issues, and applications. Mahwah, NJ: Lawrence Erlbaum Associates. Eccles, J. S., Early, D., Frasier, K., Belansky, E., & McCarthy, K. (1997). The relation of connection, regulation, and support for autonomy to adolescents functioning. Journal of Adolescent Research, 12, 263 286. Edelsohn, G., Ialongo, N., Werthamer-Larsson, L., & Crockett, L. (1992). Self-reported depressive symptoms in first-grade children: Developmentally transient phenomena? Journal of the American Academy of Child & Adolescent Psychiatry, 31, 282 290. Eisenberg, N., Cumberland, A., Spinrad, T. L., Fabes, R. A., Shepard, S. A., Reiser, M., et al. (2001). The relations of regulation and emotionality to childrens externalizing and internalizing problem behavior. Child Development, 71, 1112 1134. Engel, N. A., Rodrigue, J. R., & Geffken, G. R. (1994). Parent child agreement on ratings of anxiety in children. Psychological Reports, 75, 1251 1260. Forehand, R., & Nousiainen, S. (1993). Maternal and paternal parenting: Critical dimensions in adolescent functioning. Journal of Family Psychology, 7, 213 221. Gadeyne, E., Ghesquiere, P., & Onghena, P. (2004). Longitudinal relations between parenting and child adjustment in young children. Journal of Clinical Child and Adolescent Psychology, 33, 347 358. Galambos, N. L., Barker, E. T., & Almeida, D. M. (2003). Parents do matter: Trajectories of change in externalizing and internalizing problems in early adolescence. Child Development, 74, 578 594. Glasgow, K. L., Dornbusch, S. M., Troyer, L., Steinberg, L., & Ritter, P. L. (1997). Parenting styles, adolescents attributions, and educational outcomes in nine heterogeneous high schools. Child Development, 68, 507 529. Goodman, R., Meltzer, H., & Bailey, V. (1998). The Strengths and Difficulties Questionnaire: A pilot study on the validity of the self-report version. European Child & Adolescent Psychiatry, 7, 125 130. Gray, M. R., & Steinberg, L. (1999). Unpacking authoritative parenting: Reassessing a multidimensional construct. Journal of Marriage and the Family, 61, 574 587. Grolnick, W. S. (2003). The psychology of parental control. Mahwah, NJ: Lawrence Erlbaum Associates.

Parenting Styles in Childrens Behavior Lewis, C. C. (1981). The effects of parental firm control: A reinterpretation of findings. Psychological Bulletin, 90, 547 563. Maccoby, E. E. (1992). The role of parents in the socialization of children: An historical overview. Developmental Psychology, 28, 1006 1017. Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family: Parent child interaction. In P. H. Mussen (Series Ed.) & Hetherington, E. M. (Vol. Ed.), Handbook of child psychology: Vol. 4. Socialization, personality, and social development (4th ed., pp. 1 101). New York: Wiley. Miller, N. B., Cowan, P. A., Cowan, C. P., Hetherington, E. M., & Clingempeel, W. G. (1993). Externalizing in preschoolers and early adolescents: A cross-study replication of a family model. Developmental Psychology, 29, 3 18. Muthen, B. O., & Khoo, S.-T. (1998). Longitudinal studies of achievement growth using latent variable modeling. Learning & Individual Differences, 10, 73 102. Muthen, L. K., & Muthen, B. O. (1998 2004). Mplus. Statistical analyses with latent variables. Users Guide. Los Angeles, CA: Muthen & Muthen. Nurmi, J.-E., & Aunola, K. (1999). Jyvaskyla Entrance into Primary School Study (JEPS) (ongoing). University of Jyvaskyla, Finland. Olsen, S. F., Yang, C., Hart, C. H., Robinson, C. C., Wu, P., Nelson, D. A., et al. (2002). Maternal psychological control and preschool childrens behavioral outcomes in China, Russia, and the United States. In B. K. Barber (Ed.), Intrusive parenting: How psychological control affects children and adolescents (pp. 235 262). Washington, DC: American Psychological Association. Owens, E. B., & Shaw, D. S. (2003). Predicting growth curves of externalizing behavior across the preschool years. Journal of Abnormal Child Psychology, 31, 575 590. Pettit, G. S., Bates, J. E., & Dodge, K. A. (1997). Supportive parenting, ecological context, and childrens adjustment: A seven-year longitudinal study. Child Development, 68, 908 923. Pettit, G. S., & Laird, R. D. (2002). Psychological control and monitoring in early adolescence: The role of parental involvement and earlier child adjustment. In B. K. Barber (Ed.), Intrusive parenting. How psychological control affects children and adolescents (pp. 97 123). Washington, DC: American Psychological Association. Pettit, G. S., Laird, R. D., Dodge, K. A., Bates, J. E., & Criss, M. M. (2001). Antecedents and behavior-problem outcomes of parental monitoring and psychological control in early adolescence. Child Development, 72, 583 598. Punamaki, R.-L., Qouta, S., & El-Sarraj, E. (2001). Resil iency factors predicting psychological adjustment after political violence among Palestinian children. International Journal of Behavioral Development, 25, 256 267.

1159

Repetti, R. L., Taylor, S. E., & Seeman, T. E. (2002). Risky families: Family social environments and the mental and physical health of offspring. Psychological Bulletin, 128, 330 366. Rigdon, E. E., Schumacker, R. E., & Wothke, W. (1998). A comparative review of interaction and nonlinear modeling. In R. E. Schumacker & G. A. Marcoulides (Eds.), Interaction and nonlinear effects in structural equation modeling (pp. 1 16). Mahwah, NJ: Lawrence Erlbaum Associates. Roberts, G. C., Block, H., & Block, J. (1984). Continuity and change in parents child-rearing practices. Child Development, 55, 586 597. Roeser, R., Eccles, J., & Strobel, K. (1998). Linking the study of schooling and mental health: Selected issues and empirical illustrations at the level of the individual. Educational Psychologist, 33, 153 176. Sacco, W. P., & Graves, D. J. (1985). Correspondence between teacher ratings of childhood depression and child selfratings. Journal of Clinical Child Psychology, 14, 353 355. Schaefer, E. S. (1965). A configurational analysis of childrens reports of parent behavior. Journal of Consulting Psychology, 29, 552 557. Siequeland, L., Kendall, P. C., & Steinberg, L. (1996). Anxiety in children: Perceived family environments and observed family interaction. Journal of Clinical Child Psychology, 25, 225 237. Steinberg, L. (2001). We know some things: Parent adolescent relationships in retrospect and prospect. Journal of Research on Adolescence, 11, 1 19. Stice, E., & Barrera, M. (1995). A longitudinal examination of the reciprocal relations between perceived parenting and adolescents substance use and externalizing behaviors. Developmental Psychology, 31, 322 334. Weissman, M. M., Orvaschel, H., & Padian, N. (1980). Childrens symptoms and social functioning self-report scales: Comparison of mothers and childrens reports. Journal of Nervous and Mental Diseases, 168, 736 740. Wolfradt, U., Hempel, S., & Miles, J. N. V. (2003). Perceived parenting styles, depersonalisation, anxiety and coping behaviour in adolescents. Personality and Individual Differences, 34, 521 532. Wood, J. J., McLeod, B. D., Sigman, M., Hwang, W-C., & Chu, B. C. (2003). Parenting and childhood anxiety: Theory, empirical findings, and future directions. Journal of Child Psychology & Psychiatry & Allied Disciplines, 44, 134 151. Yang, C., Hart, C. H., Nelson, D. A., Porter, C. L., Olsen, S. F., & Robinson, C. C. (2004). Fathering in a Beijing, Chinese sample: Associations with boys and girls negative emotionality and aggression. In R. D. Day & M. E. Lamb (Eds.), Conceptualizing and measuring father involvement (pp. 185 215). Mahwah, NJ: Laurence Erlbaum Associates.