The Relationship between Parenting Stress and Child Psychopathology: Differences between Mothers and Fathers
Fien M. Peters University of Amsterdam
Abstract
Different theories exist that try to explain the associations between parenting stress and child psychopathology. However, the majority of studies, that often use these theories as a framework, primarily included mothers, and not fathers, in their research. The present study investigated whether there are differences between stress levels of mothers and fathers. It was also examined whether the association between
perceptions of parenting stress and child psychopathology differed depending on gender of the parent and the type of problem behaviour. In a group of 96 families, which were registered at a treatment centre, the relationship between parenting stress and child psychopathology appeared to be positive and equal for mothers and fathers. No differences were found for externalizing or internalizing problems. With the use of structural equation modelling no evidence was found for one of the models in which the direction of the effects was modelled differently. The effects were the same for the different types of problems and for mothers and fathers. In contrast, the analysis of variance suggested that paternal stress was affected by internalizing and externalizing problems, whereas maternal stress was only affected by externalizing problems. Exploratory analyses revealed that the correlations between parenting stress and psychopathology were different for boys and girls. Results are discussed by
recommending the inclusion of fathers in treatment as well as in research, and in light of the concept ‘gender stereotyping’. Further longitudinal research is needed to examine the nature and direction of the effects into further detail.
The Relationship between Parenting Stress and Child Psychopathology: Differences between Mothers and Fathers
Parenting stress is a concept that is often investigated in research on the development of psychopathology in children (Baker, 1994; Crnic & Low, 2000). Parenting stress can be defined as “a set of processes that leads to aversive psychological and physiological reactions arising from attempts to adapt to the demands of parenthood” (Deater-Deckard, 2004, p. 6). The association between parenting stress and child psychopathology seems to be bidirectional: child
psychopathology may lead to higher stress levels and parenting stress may increase child problems.
There is evidence for these bidirectional associations. On the one hand, evidence suggests that parents of children with attention-deficit/hyperactivity disorder (ADHD, Johnston & Mash, 2001), autism spectrum disorder (ASD, Lecavalier, Leone, & Wiltz, 2006), and oppositional defiant disorder (ODD, Ross, Blanc, McNeil, Eyberg, & Hembree-Kirgin,1998) report higher levels of parenting stress than parents of children without psychopathology. Likewise, the link between parenting stress and the more general concept of externalizing problems, which includes different kinds of child psychopathology like ADHD, ODD, and Conduct Disorder (CD), has been thoroughly investigated. In their review of the literature, Morgan, Robinson, and Aldridge (2002) state that the stress levels of parents of children with externalizing problems tend to be significantly higher, as reported by parents themselves, than those of parents of non-externalizing children (normal control groups). Deater-Deckard (2004) argues that the presence of externalizing problems in children has the strongest link with parenting stress in comparison to other behavioural and emotional child problems: they give rise to heightened levels of parenting stress.
Although it seems logical to focus on externalizing problems, as they are easy to detect by parents, it is important to pay attention to internalizing problems and their relation with parenting stress, because internalizing problems are salient in the population as well (around 15% under Dutch youth, Dorsselaer, 2010). Internalizing problems include withdrawn, depressive, anxious, isolative, and dependent behaviours. There is evidence to suggest that parenting stress is higher in families where children show internalizing problems than in families with children without problems (e.g. Mesman & Koot, 2000; Ashford, Smit, van Lier, Cuijpers, & Koot, 2008; Anthony et al., 2005).
In addition to the type of problem behaviour that affects the levels of parenting stress, the severity of the symptoms of the disorder influences parenting stress levels as well: the more severe the symptoms, the higher the reported stress levels (e.g. Dunn, Burbine, Bowers, & Tantleff-Dunn, 2001). Similarly, Deater-Deckard (2004) states that, in addition to the type of disorder, it is the overall level of dysfunction or impairment that is related to the level of parenting stress instead of the presence or absence of a behavioural disorder per se.
On the other hand, regarding the previously mentioned bi-directional relationship, there is evidence to suggest that parenting stress may be a causal and/or aggravating factor in the development of child psychopathology. Costa, Weems, Pellerin, and Dalton (2006), investigated this relationship using the short form of the Parenting Stress Index (Abidin, 1995), which consists of 13 scales that can be grouped into two domains. Firstly, stress in the child domain, which measures stress as a result of children’s temperamental and behavioural characteristics that make them difficult to manage. Secondly, the parent domain, that reflects levels of distress following from
the parent’s perception of themselves and their personal life (Abidin & Wilfong, 1989).
The authors found that parental stress experienced on the child domain could indeed predict internalizing and externalizing problems of their child, whereas stress on the parent domain, which was measured at the same time, could not (beyond parental psychopathology). This result is in concordance with the results of Crnic, Gaze, and Hoffman (2005) who showed that there was a direct relationship between parenting stress and negative child functioning, and that this relationship was not mediated by parent-child interactions. However, it is in contrast with the idea that stronger feelings of incompetence, used as a stress measure, might increase the risk for child psychopathology (Teti & Gelfand, 1991), and the finding that the component ‘competence’ of parenting skills appeared to be an important factor in child
development (Noh, Dumas, Wolf, & Fisman, 1989).
Modelling the Interaction between Stress and Child Psychopathology
There are several models that provide a framework for understanding the directionality between parenting stress and its antecedents and consequences. It is important to explain the processes of cause and effect, because it is known that stress can have adverse effects (Crnic & Low, 2002). These frameworks are principally based on the family stress theory of which the foundations lie in the ABC-X model (Hill, 1949; 1958). In this model the family crisis/parental stress (X) is the result of a three-way interaction between A: the stressor event (e.g. child psychopathology), B: the resources a family uses to be able to cope with the stressor (personal
characteristics as well as social network), and C: the meaning the family attributes to the stressor (attitudes, expectations) (Peterson & Hennon, 2005). The family stress
theory can help to understand the influence that parents and children have on each other, more specifically the effects of child psychopathology on parenting stress and vice versa.
Several models include the same basic principles to explain the development of parenting stress and its effects. Firstly, they assume that factors, like parenting
behaviours and family context that are involved in the process are dynamic. Secondly, these factors can have direct and/or indirect effects. Finally, the relationships are assumed to be reciprocal (e.g. the P-C-R theory, Deater-Deckard, 2004; daily hassles theory, Crnic & Low, 2002). An illustration of a combination of these examples is presented in Figure 1.
Figure 1
An integrated model of parenting stress based on the ABC-X model, the P-C-R theory and the daily hassles theory
A: Stressor/event: - child psychopathology - other child characteristics (e.g. gender) B: Resources/ coping: - personal characteristics parent - social network - parental psychopathology - cognitive dysfunctional thoughts
C: Attitudes and expectations
Parent-child interactions/ relationship
X: Stress Societal
Additionally, the magnitude of the influence of minor daily stressors is dependent on how parents cope and adapt to daily hassles. Parental beliefs are fundamental to coping and adaptation strategies (Crnic & Low, 2002), and those beliefs appear to have strong relations with parenting stress. A concrete example is that to one father, refusal of his child to do something is connected to the child’s age and immaturity, while another father sees this as his own incompetence as a father. Scientific grounding for this idea is for example given by McBride (1989), who found that fathers experienced more stress when they reported a low perceived sense of
competence in parenting skills. Likewise, for mothers it was found that relationships between perceptions of infant temperament and maternal behaviour were mediated by maternal self-efficacy (Teti & Gelfand, 1991). Teti and Gelfand also showed that the ‘competence scale’, being part of the ‘parent domain’ of the PSI mentioned earlier, is one of the more important scales of the parenting stress measure, because it had a strong association with other measures of stress (mother self-report and report by a clinical professional). Moreover Noh, Dumas, Wolf, and Fisman, (1989), argue that mothers with strong feelings of incompetence have higher risks of suffering from elevated personal dysfunction. Consequently, heightened levels of personal
dysfunction seemed to be a risk factor for higher scores on the parent domain of the PSI, hence experiencing more stress. Finally, in the light of the abovementioned theories, more stress might lead to a decrease in effective parenting and an increase of child problems.
Although, many theories exist about the nature of parenting stress and child psychopathology, the direction of effects between these concepts remains unclear. There is some strong evidence of a causal child effect on parents (Anastopoulos, Guevremont, Shelton, & DuPaul, 1992), although evidence for parent effects on
children exists as well (Crnic, Gaze, & Hoffman, 2005; Deater-Deckard, 2004). It seems that parents and children influence each other via bidirectional parent and child effects, and it is hard to disentangle what comes first: the stressful behaviour of the parent or the problematic behaviour of the child. Therefore, in this research several models will be tested: parent effects, child effects and a simultaneous effect: parent and child effects in one model.
Lack of Fathers in Research
Although the term parenting stress refers to parents in general, that is, mothers as well as fathers, most studies on parenting stress and child psychopathology are conducted with mothers or female caregivers as respondents (e.g. Lecavalier, Leone, & Wiltz, 2006; Johnston & Mash, 2001). The scarcity of including fathers as
respondents in developmental research seems to be a general trend. A recent meta-analysis showed that there continues to be a shortage of studies on the role of fathers in abnormal child development (Phares, Lopez, Fields, Kamboukos, & Duhig, 2005). Also, Cassano et al. (2006) report that ‘an increase in the proportion of studies that examine differences and commonalities between maternal and paternal influence [on risk and protective factors that affect the developmental pathways of psychopathology] was not identified’ (p. 587). In addition to this, even though the number of studies that included mother reports has increased in comparison to a meta-analysis of Phares and Compas (1992), in the years from 1997 to 2005 the number of studies including only fathers as an informant remained the same (1%).
The lack of the inclusion of fathers in research is in contrast with the seemingly increasing attention for fathers, and the acknowledgement of their role in the
illustrated by Lamb and Tamis-Lemonda (2004) who reported that fathers play an important role in the development of their children and affect the course of their children’s development. Also, Connell and Goodman (2002) argued that fathers are just as important as mothers in the development of child psychopathology and their mental health. In their meta-analysis, the authors investigated the difference between psychopathology of mothers and fathers as a risk factor for a child to develop
internalizing and/or externalizing problems. As the bidirectional association between child psychopathology and parental psychopathology was the same for mothers and fathers, the authors conclude that their results indicate that fathers play a very important role in the development of children. Furthermore, the results of Kahn, Brandt, and Whitaker (2004), which state that the influence of the mental health of the mother on the child’s well-being can vary depending on the mental health of the father, add to the importance of fathers in the development of children. They found that when fathers reported a better mental health, the adverse effects of the poor mental health of the mother on the child’s behavioural and emotional problems were reduced.
To the author’s knowledge, only a small number of studies on parenting stress included fathers, and the conclusions that can be drawn from these studies are limited. For example Baker (1994) found that mothers experienced a little and significantly more stress than fathers of children diagnosed with ADHD, but he used a small and male sample size. Deater-Deckard and Scarr (1996) found no significant differences in stress levels for mothers and fathers, but these results cannot be generalized to a clinical population. According to Deater- Deckard (2004) the results on differences between maternal and paternal stress are mixed. Evidence exists for mothers reporting
slightly more stress, but also no differences between mothers and fathers have been reported in clinical as well as in general populations.
In sum, the evidence for associations between parenting stress and child
psychopathology is mainly based on reports of mothers or research in which very few fathers participated in comparison to mothers (e.g. Lecavalier, Leone, & Wiltz, 2006; Johnston & Mash, 2001). What seems to be unclear is whether the relationship between parenting stress and child psychopathology is as strong for fathers as it is for mothers. Considering the evidence that exists on the important contributions fathers make to the development of a child, it is important to know what these relationships between stress and child psychopathology look like for fathers, and whether the associations between paternal stress and child internalizing and/or externalizing problems are different than those for mothers.
In summary, evidence exists for links between parenting stress of mothers and child psychopathology, but fathers are mostly excluded in these studies. Reports about differences between mothers and fathers regarding their stress levels are scarce, although there seem to be results suggesting either that no difference exists or mothers experiencing a little more stress. Also, evidence for differences between maternal and paternal parenting stress and its association with specific child psychopathology is lacking. The overall lacuna in developmental research is the inclusion of fathers, even though it is demonstrated that they play a very important role in the development of children. The aim of this research is to gain more insight into differences in parenting stress levels of fathers and mothers and its association with the internalizing and externalizing problems of their child. This will be done by investigating the following research questions:
different levels of parenting stress?
2. What is the relationship between parenting stress and child psychopathology? 3a. Is the relationship between parenting stress and child psychopathology different depending on the type of problems?
3b. Do parents report different levels of parenting stress depending on whether they assessed their child’s (problem) behaviour as more internalizing or externalizing? 4. Is the relationship between parenting stress and specific child psychopathology different for fathers and mothers?
5. How can we explore the direction of the associations between parenting stress and child psychopathology with the use of structural equation modelling?
Method Participants and Procedure
Participants were parents of children who were referred to the clinical academic treatment centre UvA-Virenze in Amsterdam, The Netherlands. Upon registration, parents were obliged to fill out behaviour questionnaires that are necessary for the treatment course. They received a letter with information about the use of the data for research purposes and the opportunity to withdraw consent for the use of their data. The research has been approved by the local Ethics Committee of the University of Amsterdam.
From May to December 2010, 156 children were registered by their parents at UvA-Virenze. Inclusion criteria were completed questionnaires by mother and father for parenting stress as well as for child psychopathology. 96 families met the
inclusion criteria. Mothers and fathers were 24-61 years old (M = 42.50, SD = 5.23, and M = 45.45, SD = 5.90, respectively) and their children’s age ranged from 6 to 20
(M = 10.52, SD = 2.55). The sample consisted of 54 boys and 42 girls. Participating families consisted for 16.7 % of 1 child, 57.3 % of 2 children and for 26 % of 3 or more children. 92 mothers and 88 fathers were the biological parent, four fathers were stepfathers, and one father filled out he had a ‘different’ relationship with the
registered child. Four mothers and three fathers had missing values on this variable due to technical problems with the online survey tool. The median education level was higher vocational education for mothers and fathers. 64 fathers worked fulltime and 24 worked part-time (number of working hours M = 38.88, SD = 8.46), while 14 mothers had fulltime jobs versus 69 part-time working mothers (M = 27.38, SD = 7.39).
Measures
Parenting stress.
Parenting stress will be measured with the subscale ‘Competence’ of the Dutch Nijmeegse Parental Stress Index (Nijmeegse Ouderlijke Stress Index (NOSI)). The NOSI is based on the Parenting Stress Index (PSI) (Abidin, 1995), and translated by de Brock, Vermulst, Gerris, and Abidin (1992). The subscale ‘competence’ consists of 13 items, measuring the extent to which the parent feels he or she is (in)competent in raising his/her child. The scale is focussed on parenting skills that are closely related to the behaviour of the child and to what extent parents feel they can manage this behaviour.
Parents rate each item on a six-point Likert scale ranging from 1 (strongly disagree) to 6 (strongly agree). Tenitems are negatively worded (e.g. “Parenthood of this child is much more difficult than I thought it would be”), and threeare positively worded (e.g. “I feel that I am competent in caring for my child”). Therefore, these
three items were recoded, and total scores are calculated by summing up the scores on all thirteen (recoded) items. This means that higher scores reflect higher levels of parenting stress. In this sample the scale had a reliability of α = .87 and α = .78 for fathers and mothers respectively.
Child psychopathology.
To measure child psychopathology the Child Behaviour Checklist for children in the age of 6 and 18 (CBCL; Achenbach & Rescorla, 2001) is used. Although the CBCL is not intended to measure psychopathology at an older age, for children between 18 and 23 years in this sample the questionnaire was used to get insight in psychopathology of those children as well.
The CBCL consists of 113 items describing child behavioural, emotional, and social problems. Fathers and mothers rate each item on a three-point Likert scale where 0 is ‘not true’, 1 is ‘somewhat or sometimes true’, and 2 is ‘very true or often true’, now or within the past six months. The items of the CBCL can be totalled for summary indices of internalizing, externalizing, and total problems, and narrowband scales. In the analyses the sum scores for internalizing and externalizing problems were used. The good reliability and validity of the American version of the CBCL were confirmed for the Dutch translation (De Groot, Koot, & Verhulst, 1994). For this sample, internal consistency for the two scales scored by mothers and fathers was high with Cronbach’s alphas ranging from .80 to .90 (M = .85).
Statistical analyses
Student’s t-test for paired samples will be used to compare stress levels of mothers and fathers.
The association between parenting stress, using the separate measurements for mothers and fathers, and mother and father reports of child psychopathology, will be calculated with the use of Pearson’s correlations. Tests will be two-sided at an alpha level of .05.
Analyses of variance (ANOVA) will be used to compare the mean stress levels of mothers and fathers of children with more internalizing problems with mean stress levels of mothers and fathers of children with more externalizing problems.
Structural equation modelling (Kline, 2005) will be used to test the equality of different correlations through chi-square (χ2) tests: the equality of the correlations between parenting stress and internalizing and externalizing problems will be tested as well as whether correlations are different for mothers and fathers. Finally, the
directionality of the associations will be investigated by fitting several path models to the covariance matrix.
The models will be tested with the maximum likelihood (ML) estimation method using the computer programme Mx (Neale, Boker, Xie, & Maes in Kline, 2005). This method yields estimates, correlation residuals, as well as a chi-square (χ2) measure of overall goodness of (exact) fit and a root mean square error of approximation
(RMSEA). A significant χ2(α = .05) indicates that the model does not fit the data
exactly. The RMSEA is a measure of approximate fit. Browne & Cudeck (in Kline, 2005) suggest rules of thumb with an RMSEA of ≤ .05 indicating close approximate fit, .05 - .08 suggesting reasonable error of approximation, and an RMSEA ≥ .10 suggesting poor fit. Chi-square difference tests will be used to compare model fit and examine whether improvement/deterioration of the fit is significant.
Results Descriptives
Prior to analyses all variables were examined through various SPSS procedures for accuracy of data entry, missing values, and fit between their distributions and the assumptions of Pearson’s correlations, Student’s t-test, and structural equation modelling. The variables were examined separately for mothers and fathers. For 60 children either mother, father or both parents had missing values on parenting stress, hence these cases were deleted. Two children were older than 18 years. Analyses for descriptives, t-tests and correlations were run with these two cases, and were checked with the analyses without these two cases. After deletion of the 60 cases, there were no cases with missing values on the main variables parenting stress and child psychopathology. This led to a total of N = 96.
Nine univariate outliers were replaced by the second highest score on that
variable. To reduce skewness and improve the normality and linearity of the variables parenting stress, internalizing problems rated by father, and externalizing problems rated by mother and father, a logarithmic transformation was executed for these variables. However, results of the t-tests and correlations were not significantly different with these transformed variables. Given that the interpretation of results becomes more difficult after the application of transformations, in this research the skewed variables were used. Beside the difficult interpretation, the skewed variables are a better representation of the clinical population. With the use of a p < .001 criterion for Mahalanobis distance, no multivariate outliers among the cases were identified.
T-test
A t-test, conducted to test whether fathers and mothers differed on their level of parenting stress, demonstrated no significant difference, t (95) = .611, p = .543. Main values are represented in Table 1. For the sake of completeness, results of the
dependent t-tests for mother and father report of internalizing and externalizing problems are presented in Table 1 as well, showing that fathers report less
internalizing problems than mothers (p = .017). Results indicated that in- or exclusion of the two cases with children older than 18 years did not lead to different results, and, therefore, these were included in the analysis.
Table 1
Means, standard deviations, correlations, and t-values (N = 96) of main variables
Mothers Fathers
Measures Mean SD Mean SD R t-value df p-value
Parenting stress 25.23 8.52 24.50 9.68 .179 .611 95 .543 Internalizing problems 11.67 6.75 10.01 7.45 .565*** 2.438 95 .017* Externalizing problems 8.77 6.53 7.85 6.06 .643*** 1.683 95 .096
Note. Values are based on untransformed variables.
***p < .001. * p < .05.
Correlations
Pearson’s correlations were used to explore the bivariate associations between
parents. For exploratory reasons, the variables age and gender were included in the analysis as well. Correlations are presented in Table 2.
Analyses showed that parents reported more parenting stress when they assessed their child with more internalizing and/or externalizing problems. This was true both for mothers and fathers. The associations between mother’s stress and internalizing and externalizing problems (respectively, r = .285, p = .005; r = .425, p = .000) and father’s stress and externalizing problems (r = .283, p = .005) were significant at an alpha level of .01, as opposed to father’s stress and internalizing problems (r = .242, p = .017) being significant at α = .05. The agreement of mothers and fathers when assessing the problem behaviour of their child was confirmed by the correlations between mother and father report of child problem behaviour which were significant at an alpha level of .001 (Table 2).
Notably, there were no significant correlations between stress levels of fathers and mothers (as is found in Hastings, 2003). Also, no significant correlations were found between stress levels and psychopathology assessed by the other parent.
Mothers of older children reported more stress and more internalizing problems as is shown by the significant positive correlations with age (r = .249, p = .015 for stress levels, and r = .205, p = .045 for internalizing problems). No significant correlations were found between child gender and any of the variables.
Table 2
Correlations between parenting stress and child behaviour problems including child gender and age (N = 96)
Measure
Mother Father Gender Age
Measure 1. 2. 3. 4. 5. 6. 7. 8. Mother 1. Stress - .285** (.005) .425*** (.000) .179 (.081) .075 (.468) .177 (.085) .080 (.437) .249* (.015) 2. Internalizing - .303** (.003) .073 (.480) .565*** (.000) .144 (.163) .103 (.317) .205* (.045) 3. Externalizing - .161 (.117) .069 (.501) .643*** (.000) .102 (.322) -.037 (.723) Father 4. Stress - .242* (.017) .283** (.005) .037 (.720) .114 (.269) 5. Internalizing - .373** (.000) .180 (.079) .183 (.074) 6. Externalizing - .171 (.096) -.027 (.793) 7. Child Gender - -.189 (.065) 8. Child Age - SD 8.517 6.752 6.530 9.683 7.452 6.063 .499 2.552
Note. Cells give correlations and p-values within parentheses; SD = standard deviation.
For exploratory purposes, correlations between the variables were also calculated for boys and girls separate. Results showed that neither mother’s nor father’s stress levels correlated significantly with internalizing problems in girls. Stress levels of fathers did not correlate with externalizing problems in boys either. From Table 3 it can be inferred that higher levels of stress in mothers were associated with higher levels of externalizing problems in girls, and internalizing and externalizing problems in boys. For fathers, higher levels of internalizing problems in boys and externalizing problems in girls were associated with higher levels of stress.
Table 3
Correlations between parenting stress and child behaviour problems for boys and girls separate
Boys (N = 54) Girls (N = 42)
Internalizing Externalizing Internalizing Externalizing
Stress mother .309** (.023) .484***(.000) .240 (.126) .342** (.027)
Stress father .393*** (.003) .209 (.129) .035 (.827) .418*** (.006)
Note. Cells give correlations and p-values within parentheses.
*** p < .001. ** p < .01.
Structural Equation Modelling and Analyses of Variance
Firstly, a model in which the observed correlations were modelled was fitted to the covariance matrix of the variables parenting stress of mothers and fathers, and internalizing and externalizing problems as assessed by both parents. This was a saturated model (hence, χ2 (0) = 0). To be able to compare the different correlations, a
saturated model was needed, so the results would not be influenced by any other fixed value in the model.
Secondly, the correlations of internalizing and externalizing problems with stress were constrained to be equal for mothers as well as for fathers. This did not lead to a significant worse fit, with a chi-square difference test of χ2D (2) = 1.189, p = .552.
These results suggest that the correlations between child psychopathology and parenting stress are the same for externalizing and internalizing problems.
To investigate whether mothers and fathers report different levels of parenting stress depending on the type of problem, analyses of variance were used. Mean scores are presented in Table 4. The results suggested that stress levels of mothers were significantly different when they reported more externalizing problems (F (1, 96) = 9.082, p = .003), but that stress levels were not affected by the amount of internalizing problems (F (1, 96) = 2.269, p = .135). For fathers, main effects were found for internalizing and externalizing problems on parenting stress, with F (1, 96) = 4.716, p = .032 and F (1, 96) = 4.064, p = .047, respectively. It seems that stress levels were different depending on the type of problem behaviour reported by parents. However, no interaction effects were found. It is, therefore, indicated that the effect of the level of internalizing or externalizing problems is not dependent on each other.
Thirdly, in turn, the correlations of internalizing and externalizing problems were constrained to be equal for mothers and fathers. Both models did not lead to a
significant worse fit and, hence, it is suggested that the correlations between
internalizing and externalizing problems and parenting stress are the same for mothers and fathers. Results can be found in Table 5.
Table 4
Mean parenting stress levels of mothers and fathers in categorized groups
Mother Father
Type of problem Externalizing
Low High Low High
Internalizing Low 22.114 (35) 26.167 (18) 21.000 (34) 24.357 (14) High 23.588 (17) 29.846 (26) 24.048 (21) 29.333 (27)
Note. Cells give means, and within parentheses number of participants.
Table 5
Fit indices for the comparison of correlations between child psychopathology and parenting stress of mothers and fathers (N = 96)
χ2
(df) p-value RMSEA (90% CI) Internalizing problems1 .013 (1) .909 .000 (.000,.113) Externalizing problems1 .731 (1) .393 .000 (.000, .256) Internalizing and externalizing problems1 1.189 (2) .552 .000 (.000, .175)
Note. CI = confidence interval.
1
Correlations are constrained to be equal for mothers and fathers.
Finally, the direction of the associations was explored by fitting three models to the covariance matrix: the first model included effects from psychopathology to
parenting stress, in the second model effects from parenting stress on child
psychopathology were tested, and the third model had bidirectional effects between parenting stress and child psychopathology. To account for possible effects of gender and age, these were included as explanatory variables. Connel and Goodman (2002) suggested that age may be a factor that explains the different effects of child
psychopathology over the years.
The first model led to a moderate fit with χ2
(10) = 12.531, p = .251; RMSEA = .052 (90% confidence interval [.000, .129]). Correlation residuals higher than .1 were solved by adding one direct effect and one correlation (between disturbances) which led to a significant better fit, χ2D (2) = 7.452, p = .024. Moreover, to make the
model more parsimonious, two direct effects smaller than .05 were deleted. This led to a final fit of χ2
(10) = 5.155, p = .881; RMSEA = .000 (90% confidence interval [.000, .055]). The final model can be found in Figure 2.
The second model, with effects directed from stress to psychopathology, led to a close fit as well, χ2
(14) = 15.632, p = .336; RMSEA = .035 (90% confidence interval [.000, .108]). Step by step direct effects were added between variables with the largest correlation residual (higher than .1). After the addition of three direct effects and one variance, even more direct effects from gender and age to psychopathology and from psychopathology to stress were necessary to reduce correlation residuals. However, this addition would lead to a bi-directional and nonrecursive model. Such a model would be tested during the third attempt, and therefore, the respecification of the second model was ceased. The fit of the adjusted model was not significantly better than the starting model: χ2
(10) = 7.823, p = .646; RMSEA = .000 (90% confidence interval [.000, .092]), and χ2
D (4) = 7.809, p = .099. Figure 3 represents the final
Finally the model with reciprocal effects was fitted to the covariance matrix. An empirical check for identification was executed and revealed that the model was identified. Initially, this model led to a moderate fit χ2
(6) = 10.455, p = .107; RMSEA = .088 (90% confidence interval [.000, .176]). Once again, correlation residuals higher than .1 were considered and three direct effects were added to the model which led to a χ2
(3) = .881, p = .830; RMSEA = .088 (90% confidence interval [.000, .101]), which was a significant improvement of the model, χ2D (3) = 9.574, p = .023. A direct
effect smaller than .01 (standardized) was deleted to make the model more
parsimonious, and led to a final fit of χ2 (4) = .882, p = .927; RMSEA = .000 (90% confidence interval [.000, .048]). Other small standardized effects (< .1) were retained, because these were needed to compare the size of the different effects (the comparison of the effects of internalizing and externalizing problems on parenting stress). The empirical check for identification revealed that this final model was identified as well. In Figure 4 the final model is illustrated.
Figure 2. Final version of the first model. D = disturbance. Internalizing mother Externalizing mother Internalizing father Externalizing father Age child Gender child Stress father Stress mother D D D D D D
Figure 3. Final version of the second model. D = disturbance. Internalizing mother Externalizing mother Internalizing father Externalizing father Age child Gender child Stress father Stress mother D D D D D D
Figure 4. Final version of the third model. D = disturbance. Internalizing mother Externalizing mother Internalizing father Externalizing father Age child
Gender child Stress father
Stress mother D D D D D D
All three attempts led to a good final fit, but the third model was chosen because this was considered to be the best representation of reality. Table 6 provides
unstandardized and standardized parameter estimates of the final model. Confidence intervals revealed that five direct and five total effects were significant. Mother’s stress levels were directly affected by age, gender, and externalizing problems as reported by mother. Age and gender had direct effects on father reports of internalizing problems. The total indirect effect of age on externalizing problems reported by mother was significant as well. Values of the estimated variances and covariances are presented in Table 7.
To make the analyses complete, the comparison of the size of the effects was investigated using the bi-directional model as well. The chi-square values for these constrained models did not become significantly worse, and so the results indicated that the effects of internalizing and externalizing problems on parenting stress are the same, χ2 (6) = .3.819, p = .701; RMSEA = .000 (90% confidence interval [.000, .101]), and χ2
D (2) = 2.937, p = .230 . Moreover, the results suggested that these effects are
the same for mothers and fathers, given that the fit of the model did not deteriorate with the inclusion of equality constraints, χ2 (6) = .1.019, p = .985; RMSEA = .000 (90% confidence interval [.000, .000]), and χ2D (2) = .137, p = .934.
Table 6
Standardized and unstandardized parameter estimates of the bidirectional model
Endogenous variables
Mother Father
Internalizing Externalizing Stress Internalizing Externalizing Stress
Causal variable Unst. Std. Unst. Std. Unst. Std. Unst. Std. Unst. Std. Unst. Std.
Age child Direct effect .421 .160 -.209 -.082 .998* .299* .600* .208* -- -- .468 .123 Total indirect effects .195 .074 .168* .066* -.083 -.025 .058 .020 .021 .009 -.008 -.002 Total effect .616 .234 -.042 -.016 .915* .274* .658* .228* .021 .009 .460 .121 Gender child Direct effects 1.505 .112 .881 .068 1.995 .117 .3171* .216* 2.040* .169* -- -- Total indirect effects .469 .035 .402 .031 .200 .012 .084 .006 .030 .003 .670 .035 Total effect 1.974 .147 1.283 .099 2.195 .129 3.255* .221* 2.071* .171* .670 .035 Mother Internalizing Direct effect -- -- -.113 -.089 Total indirect effects -.021 -.022 -.004 -.003 Total effect -.021 -.022 -.117 -.092 Externalizing Direct effect -- -- .329* .250* Indirect effects .073 .070 .012 .009 Total effect .073 .070 .341 .260 Stress Direct effect .214 .271 .183 .241 Total indirect effects .008 .010 .007 .009
Mother Father
Internalizing Externalizing Stress Internalizing Externalizing Stress
Causal variable Unst. Std. Unst. Std. Unst. Std. Unst. Std. Unst. Std. Unst. Std.
Total effect .222 .281 .190* .250* Father Internalizing Direct effect -- -- -.024 .018 Total indirect effects .001 -.001 .000 -.000 Total effect .001 -.001 -.024 -.018 Externalizing Direct effect .360 .225 Total indirect effects .046 .038 .005 .003 Total effect .046 .038 .365 .228 Stress Direct effect .126 .166 .045 .072 Total indirect effects .002 .002 .001 .001 Total effect .128 .168 .046 .073
Table 7
Unstandardized and standardized estimated variances and covariances
Parameter Unstandardized 99% CI Standardized
Age child 6.503*** 4.572, 9.867 1
Gender child .250*** .176, .372 1
Gender Age -.241* -.481, -.028a -.189
Dinternalizing mother 40.417*** 28.478, 60.234 .894
Dexternalizing mother 35.188*** 24.439, 54.182 .837
Dinternalizing mother Dexternalizing mother 6.182** .900, 13.263
b
.142
Dinternalizing father 47.437*** 33.726, 69.592 .877
Dinternalizing father Dinternalizing mother 23.809*** 13.083, 40.326 .482
Dexternalizing father 34.309*** 23.965, 52.216 .936
Dexternalizing father Dinternalizing father 11.143*** 3.897, 24.147 .250
Dexternalizing father Dexternalizing mother 22.320*** 13.463, 36.670 .569
Dstress mother 58.363*** 38.763, 1000.00 .803
Dstress father 85.644*** 59.311, 1000.00 .914
Dstress mother Dstress father 8.54 -3.436, 22.064
a
.104
Note. D = disturbance, = covariance.
a 90% confidence interval. b 95% confidence interval.
* Significant at a 90% confidence interval. ** Significant at a 95% confidence interval. *** Significant at a 99% confidence interval.
Discussion
The goals of this study were to compare the parenting stress levels of mothers and fathers, and to investigate the associations between parenting stress and child
psychopathology. In particular, it was examined whether these associations were different for or depending on the gender of the parent and/or the type of problems of the child. Moreover, the final goal was to explore the direction of the associations that were found in this study.
Firstly, the results suggested that in this sample of mothers and fathers of children registered at a clinical treatment center, there was no difference between their stress levels. These results are similar to the results from previous studies that found no difference between stress levels of parents from a general population (e.g. Deater-Deckard & Scarr, 1996).
Secondly, the correlations between child psychopathology and parenting stress were positive and significant, meaning that higher levels of parenting stress were associated with higher levels of child problems.
Thirdly, the results revealed that the correlations between child psychopathology and parenting stress were the same for mothers and fathers. Also, there was no difference, neither for mothers nor fathers, between associations between parenting stress and the type of problems, i.e. internalizing or externalizing problems. Taking the variables gender and age into consideration did alter the correlations. Fathers seemed to experience more stress when they assessed sons as more internalizing and daughters as more externalizing. For mothers, stress levels seemed to be higher when boys had internalizing and/ or externalizing and girls had externalizing problems.
Strikingly, with the use of analyses of variance, it was suggested that stress levels of mothers were affected by externalizing problems, and that internalizing as well as externalizing had effects on paternal stress.
Finally, the direction of the associations was explored, and it seems that effects run from child psychopathology to parenting stress, and vice versa. Hence, parenting
stress probably affects child psychopathology, while, at the same time, child
psychopathology has an effect on parenting stress. The inclusion of child gender and age in the model led to the suggestion that mother’s stress levels are directly
influenced by the age of the child, as opposed to fathers whose stress levels did not seem to be affected directly by any of the variables. Gender affected internalizing and externalizing problems reported by fathers. Age had a direct effect on father report of internalizing problems as well.
Implications
The finding that fathers and mothers did not differ in the amount of parenting stress might have implications for the treatment of children with psychopathology, namely that both father and mother should be involved during the treatment process instead of only one of the parents, most of the mothers or female caregivers. This can be explained in light of an important treatment option like parent training, for example the ‘Barkley training’ for parents of children with ADHD (Barkley, 1997). In this kind of treatment, parents are trained in restricting their child’s behaviour or when (not) to punish or to reward their child. For these kinds of training it seems to be important that the parents have a good mental health, as they are supposed to learn new skills and pass the new rules on to their children. When parents suffer from a form of parental psychological maladjustment, such as parenting stress, the effectiveness of parent training might be limited (Barkley, 2006; Anastopoulos, Hennis Rhoads, & Farley, 2006). Parenting stress might impede improvement of the child’s behaviour, because parenting stress can be related to parenting behaviour which the training is aimed to adjust. When the parent is too stressed to adapt the skills, the parent training will probably not be as effective as it is for parents without parenting stress. Moreover,
the effects of the training might be stronger when both parents are participating. With the inclusion of both parents, the risk that one of the stressed parents that stays home and limits the beneficial effects of the training (such as generalization beyond the treatment setting), is diminished.
Although, it is recommended to include both parents, in practice it appears that primarily mothers are involved during treatment (Duhig, Phares, & Birkeland, 2002). It is suggested that clinicians are less likely to involve fathers in therapy, but also fathers are more unwilling than mothers to be engaged in the therapeutic process when they are invited (Phares, Lopez, Fields, Kamboukos, & Duhig, 2005). Several studies and books are available for clinical practitioners to become more aware of the importance of fathers, and in which suggestions are offered to include mothers as well as fathers in the therapeutic process (Duhig, Phares, & Birkeland).
Beside the treatment process, this result has implications for the research on developmental psychopathology concerning its origin and maintenance. Stress can have adverse effects on child behaviour/development (Crnic & Low, 2002). However, most research that found such effects used mother reports primarily. The finding that fathers do not show different stress levels than mothers might suggest they might have the same negative (or positive) influence on a child’s development as mothers. Further research is needed to find out whether such effects indeed exist.
Besides the comparable stress levels of fathers and mothers, this investigation also showed that the associations between parenting stress and child psychopathology were positive and strong. These results reveal that parents, who report higher levels of child psychopathology, seem to experience higher levels of stress as well.
Elaborated tests showed that the correlations between child psychopathology and parenting stress were not different for mothers and fathers. This may mean that
mothers and fathers are equally influenced by internalizing as well as externalizing problems in children.
Similarly, the correlations between psychopathology and parenting stress appeared to be the same for internalizing and externalizing problems. This is a remarkable finding because in previous research it was generally found that
externalizing problems had the strongest link with parenting stress (Deater-Deckard, 2004), and is in contrast with the finding in this research that maternal stress levels were only affected by externalizing problems. Father’s stress levels, on the other hand, were affected by internalizing as well as externalizing problems. These findings may be an indication that parents, or at least fathers, are equally susceptible for a child’s psychopathology, regardless of the type of problems. Or the other way around: parenting stress may influence children’s psychopathology, whether they show internalizing or externalizing problems. This should be a demonstration for the importance of including internalizing as well as externalizing problems in
developmental research. Although, it appears that measures of internalizing problems are often included in research on child psychopathology, to the authors knowledge, up until now no review of the literature exists about internalizing problems and their association with parenting stress. Such a review is needed to examine the importance of including internalizing problems in research.
Strikingly, the exploratory analyses showed that the associations of child psychopathology and parenting stress were different for boys and girls. There seems to be a difference for mothers and fathers and their stress levels depending on the type of problem behaviour and gender of their child. Significant associations were found between parenting stress and internalizing problems in boys and externalizing
stress, and between internalizing girls and paternal and maternal stress were non-significant. These results may be explained with the consideration of the concept ‘gender stereotyping’. It is a widespread idea that boys are more active and show more aggressive behaviour than girls. Girls are expected to play with dolls and prefer manual play, like drawing and painting. It is also suggested that boys are less
restricted during play than girls (Pomerleau, Bolduc, Malcuit, & Cossette, 1990). Research has found that parents tend to judge it more natural for girls to play with dolls and stay quiet, and for boys to explore and display externalizing behaviours (Fagot, 1978; Smith & Daglish, 1977). It could be argued that externalizing behaviour is expected from boys, whereas girls are expected to be more quiet. Consequential, if girls show externalizing behaviour, this might elicit more stressful feelings from parents. However, this cannot explain why mother’s stress level was, but father’s stress level was not associated with externalizing behaviour in boys. It could be helpful to consider the idea that fathers often function as a parent for the so called ‘rough and tumble play’. Substantial evidence exists which shows that fathers have a specific role in the family in preparing their child for the external world (Bögels & Phares, 2008). If it is true, that fathers principally have a physical function in the family, whereas mothers have a caring role, it could be that fathers are better in managing boys that show externalizing behaviour. Fathers have experience with externalizing behaviour and expect their sons to show this type of behaviour. It could, therefore, be that fathers experience the deviant behaviour of their son as less stressful than mothers might do (Paquette, 2004; Bögels & Phares, 2008; Bögels & Perotti, 2010). In addition, it is suggested that fathers tend to use more stereotyping than mothers and, therefore, the results may be different for mothers and fathers (Lytton & Romney, 1991; Plant, Hyde, Keltner, & Devine, 2000).
Hence, the absence of an association between internalizing problems in girls and parenting stress seems logical, because this type of behaviour suits the expectations from the outside world. This could be a positive finding as parents possibly know very well how to manage internalizing behaviour of girls and do not experience this as stressful. However, the negative consequence could be that parents might overlook internalizing problems in girls and, as a result, will not look for help when needed. The risk may be that parents are inclined to underestimate the seriousness of daughters with internalizing problems. This may be another indication that it is important to include internalizing problems in developmental research.
Considering both results regarding type of problem and gender of the parent, it was important to consider the directionality of psychopathology and parenting stress for internalizing as well as for externalizing problems, and subsequently the strength of the effects. Unfortunately, the analyses of the correlational model were done without considering the gender of the child, because this was the initial goal of the research. Moreover, it turned out that the sample size was too small to execute the analyses for two separate groups (boys and girls).
In the extended models, however, it was possible to include gender and age as separate variables. These models all led to a good fit. One interpretation might be that all models are a good reflection of reality and represent the functioning of the
processes in families: child psychopathology influences parenting stress and vice versa, but also influence each other at the same time, as was presented in the final model. No differences were discovered between mothers and fathers, which could be a confirmation of the suggestion that mothers and father’s stress levels are equally affected by child psychopathology. Furthermore, the influence of internalizing and externalizing problems on parenting stress seems to be the same. Although it is
suggested that externalizing problems are easier to detect and, therefore, are more like to heighten parenting stress levels, this was not confirmed here.
An interesting finding was that the age of the child and externalizing problems primarily affected stress as experienced by mother, as opposed to the stress levels of fathers that were not affected directly by any of the variables. It should be noted that gender had direct effects on child psychopathology as reported by fathers, where being a boy resulted in higher levels of psychopathology. Similarly, age affected father report of internalizing problems: the older the child the more internalizing problems were reported. Indirectly, the stress levels of fathers might be more strongly influenced by gender, whereas the stress levels of mothers are more likely to be affected by the age of the child.
Although different effects were significant for mothers and fathers, there appeared to be no significant difference regarding the effects of internalizing and externalizing problems on stress. Also for mother and fathers, these effects were the same. Nevertheless, these results should be carefully interpreted. Firstly, the results may be correct and provide a good representation of the real world: there is just no difference between mothers and fathers and the influence of internalizing versus externalizing problems. Yet, secondly, an alternative explanation may be that there was too little power in this research. In structural equation modelling large samples are required in order to be sure that the results are not purely based on chance. This investigation might give different results when executed with larger groups.
Nonetheless, a start has been made here with the inclusion of a large group of fathers. Up until now it seems to be problematic for researchers to include fathers or they are not convinced (which is undeserved) that it is a necessary condition for the results to be generalized. The results that no difference exists between mothers and fathers and
the suggestion that this is evidence that fathers are as important as are mothers, should be a stimulus for others to include fathers in future research.
On the other hand, it may be argued that it is sufficient to include only mothers in research, as parents experience similar levels of stress. Research often uses mother or female reports primarily, but it might be interpreted that the conclusions can be applied to both parents. Regarding parenting stress, this may be warranted. Yet, in this investigation the results suggest that parenting stress is different for mothers and fathers when child gender, age, and psychopathology are considered. At least, therefore, researchers should be cautious with their conclusions and should not generalize their results to fathers directly.
Limitations and Recommendations
A number of other limitations have to be mentioned, while at the same time these may lead to recommendations for future research as well. Firstly, the inclusion of several other or extended measures. First of all, a limited measure of parenting stress was used: the competence scale of the NOSI, consisting of 13 items. This is a small questionnaire that solely focuses on one aspect of parenting stress. The concept parenting stress probably includes more factors that should be considered in future research. An extended measure of parenting stress may lead to different results. In addition, the result that no relationship existed between internalizing problems in girls and parenting stress might be explained by the fact that stress was measured on a competence scale. It could be that stress experienced as a consequence of internalizing problems in girls would be manifested when another measure for parenting stress was used. For example, inclusion of other scales of the NOSI like ‘Depression’ or
feelings of competence seem to be an important factor in the experiences of parenting stress, and the consequences of stress (Noh, Dumas, Wolf, & Fisman, 1989). Yet, it should be noted that the subdivision of the sample into boys and girls, and hence smaller subgroups, can lead to a smaller power. It could, therefore, also be that the relationship between internalizing problems and parenting stress is no longer significant in the smaller subgroup of girls.
Next to the extension of the parenting stress measure, in future research it might be important to give more attention to the cognitive dysfunctional thoughts of parents. Such thoughts can follow from feelings of incompetence which in turn can be related to parenting stress. This may lead to insight in thoughts and antecedents/consequences of parenting stress. As cognitive dysfunctional thoughts are a good target for
treatment of parent’s stress and consequently the problem behaviours of their child (Beck, 1995).
Also, the consideration of involvement in child care might be an interesting, and necessary factor to include. This might be different for mothers and fathers, and depending on their involvement, parents may experience different levels of stress. The results in this study showed that the majority of fathers worked fulltime, as opposed to mother that worked primarily part-time. It would be very interesting to investigate if this is indeed a mediating factor on the relationship between parenting stress and child psychopathology.
One final important variable to include would be a measure of ‘coping’. How do parents cope with the difficult behaviour of their child? Do different coping patterns mediate/affect the associations between child psychopathology and parenting stress? With the inclusion of more variables, an extended model could be built which is a
better representation of real life, and might give better insights in the processes between parents and children.
Secondly, the generalizability of the study is limited, because a clinical population was used and no control group was included. The comparison with a general sample would have been useful to generalize the findings on a wider population. However, for clinical practice it is very important to know what the relationships between parenting stress and problem behaviour look like in the clinical population. Most importantly, the inclusion of a control group was not within the scope of this research, and it is therefore rather a recommendation for future research than a limitation of this study.
Finally, like former research, results are solely based on correlational data. It was therefore not possible to draw clear conclusions about causality. Several
theories/models exist showing the bidirectionality between parenting stress and child problem behaviour. There seems to be no clear answer to the question “what comes first?” In this research decisions about the direction of the effects were based on previous research to find out what model fitted best on the data of this sample. In the future more longitudinal research is recommended to test these models, and more variables should be included. Next to the correlational nature of the data, it should be noted that all data were parent reports. This means the measures were subjective assessments of parents about themselves and their children, and this makes results more dependent on the parent’s frame of reference. For future research it is recommended that more objective measures are used.
In sum, in this clinical population sample no differences were found between stress levels of mothers and fathers. There appeared to be no differences between the influence of internalizing versus externalizing problems on parenting stress, for
neither mothers nor fathers. However, analyses of variance revealed different results for mothers. With the use of structural equation modelling it was found that models with effects in all directions fitted the data very well, and it is assumed that the effects are bidirectional. However, the underlying mechanisms/processes between child psychopathology and parenting stress seem to be complicated and the models should be extended with the inclusion of more variables. An additional finding was that levels of parenting stress have strong associations with child psychopathology, and that different associations for mothers and fathers came to light depending on child gender in combination with the type of problem behaviour. Overall this study made a good start concerning the inclusion of fathers in research and showed that it is
References
Achenbach, T. M., & Rescorla, L. A. (2000). Manual for the ASEBA Preschool Forms
& Profiles. Burlington, VT: University of Vermont, Research Center for Children,
Youth, & Families.
Achenbach, T. M., & Rescorla, L. A. (2001). Manual for the ASEBA School-Age
Forms & Profiles. Burlington, VT: University of Vermont, Research Center for
Children, Youth, & Families.
Achenbach, T. M., & Rescorla, L.A. (2003). Manual for the ASEBA Adult Forms &
Profiles. Burlington, VT: University of Vermont, Research Center for Children,
Youth, and Families.
Abidin, R. R. (1995). Parenting Stress Index: Manual, 3rd ed. In D. B. Mitchel & P. Hauser-Cram (2010), Early childhood predictors of mothers’ and fathers’
relationships with adolescents with developmental disabilities. Journal of
Intellectual Disability Research, 54 (6), 487-500.
Abidin, R. R., & Wilfong, E. (1989). Parenting stress and its relationship to child health care. Children’s Health Care, 18 (2), 114-116.
Anastopoulos, A. D., Guevremont, D. C., Shelton, T. L., & DuPaul, G. J. (1992). Parenting stress among families of children with attention deficit hyperactivity disorder. Journal of Abnormal Child Psychology, 20, 503-520.
Anastopoulos, A. D., Hennis Rhoads, L., & Farley, S. E. (2006). Counselling and training parents. In E. J. Mash & R. A. Barkley (Eds.) (2006). Treatment of
childhood disorders (3rd ed.). New York: The Guilford Press.
Anthony, L. G., Anthony, B. J., Glanville, D. N., Naiman, D. Q., Waanders, C., & Shaffer, S. (2005). The relationship between parenting stress, parenting behaviour,
and pre-schoolers’ social competence and behaviour problems in the classroom. Infant and Child Development, 14, 133-154.
Baker, D. B. (1994). Parenting stress and ADHD: a comparison of mothers and fathers. Journal of Emotional and behavioral disorders, 2 (1), 46-50.
Barkley, R. A. (1997). Defiant children: A clinician’s manual for assessment and
parent training (2nd ed.). New York: The Guilford Press.
Barkley, R. A., & Murphy, K. R. (2006). Attention-deficit hyperactivity disorder: a
handbook for diagnosis and treatment (3rd ed.). New York: The Guilford Press. Beck, J. S. (1995). Cognitive therapy: basics and beyond. New York: The Guilford Press.
Bögels, S., & Phares, V. (2008). Father’s role in the etiology, prevention and
treatment of child anxiety: a review and new model. Clinical Psychology Review, 28,
539-558.
Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley. Brock. A. J. L. L. de, Vermulst, A. A., Gerris, J. R. M., & Abidin, R. R. (1992). Nijmeegse Ouderlijke Stress Index. Lisse: Swets en Zeitlinger.
Buchner, A., Erdfelder, E., & Faul, F. (1997). How to Use G*Power [WWW document]. Downloaded: January 27, 2010:
http://www.psycho.uni-duesseldorf.de/aap/projects/gpower/how_to_use_gpower.html
Cohen, S. & Wills, T. A. (1985). Stress, social support and the buffering hypothesis.
Psychological Bulletin, 98, 310-357.
Connell, A. M., & Goodman, S. H. (2002). The association between psychopathology in fathers versus mothers and children’s internalizing and externalizing behavior problems: a meta-analysis. Psychological Bulletin, 128 (5), 746-773.
childhood psychopathology: an examination of specificity to internalizing and externalizing symptoms. Journal of Psychopathology and Behavioral Assessment,
28 (2), 113-122.
Creasy, G., & Reese, M. (1996). Mothers’ and fathers’ perceptions of parenting hassles: associations with psychological symptoms, nonparenting hassles, and child behavior problems. Journal of Applied Developmental Psychology, 17, 393-406. Crnic, K., & Low, C. (2002). Everyday stresses and parenting. In Bornstein, M. H. (Ed.), Handbook of parenting: Vol. 5. Practical issues in parenting. New Jersey: Lawrence Erlbaum Associates.
Crnic, K. A., Gaze, C., & Hoffman, C. (2005). Cumulative parenting stress across the preschool period: relations to maternal parenting and child behaviour at age 5. Infant and Child Development, 14, 117-132.
Deater-Deckard, K. (2004). Parenting Stress. New Haven: Yale University Press. Deater-Deckard, K., & Scarr, S. (1996). Parenting stress among dual-earner mothers and fathers: Are there gender differences? Journal of Family Psychology, 10, 45-59. Dorsselaer, S. van, Looze, M. de, Vermeulen-Smit, E., Roos, S. de, Verdurmen, J., Bogt, T. ter, & Vollebergh, W. (2010). Gezondheid, welzijn en opvoeding van
jongeren in Nederland. Utrecht: Trimbos-instituut.
Duhig, A. M., Phares, V., & Birkeland, R. W. (2002). Involvement of fathers in therapy: a survey of clinicians. Professional Psychology: Research and Practice, 33
(4), 389-395.
Gottlieb, B. H. (1983). Social support strategies. Guidelines for mental health
practice. Beverly Hills: Sage.
Hill, R. (1949). Families under stress. In G. W. Peterson, & C. B. Hennon (2005). Conceptualizing parental stress with family stress theory. In P. C. McKenry & S. J.
Price. Families and change: Coping with stressful events and transitions (3rd ed.). Thousand Oaks: Sage Publications, p. 25-48.
Hill, R. (1958). Social stresses on the family: Generic features of families under stress. In G. W. Peterson, & C. B. Hennon (2005). Conceptualizing parental stress with family stress theory. In P. C. McKenry & S. J. Price. Families and change:
Coping with stressful events and transitions (3rd ed.). Thousand Oaks: Sage
Publications, p. 25-48.
Johnston, C., & Mash, E. J. (2001). Families of children with attention-
deficit/hyperactivity disorder: review and recommendations for future research. Clinical Child and Family Psychology Review, 4 (3), 183-207.
Kahn, R. S., Brandt, D., & Whitaker, R. C. (2004). Combined effect of mothers’ and fathers’ mental health symptoms on children’s behavioral and emotional well-being. Archives of Pediatrics & Adolescent Medicine, 158, 721-729.
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York: The Guilford Press.
Lamb, M. E., & Tamis-Lemonda, C. S. (2004). The role of the father: an introduction. In M. E. Lamb (Ed.), The role of the father in child development (4th ed., pp. 1-31).
New Jersey: Wiley.
Lamb, M. E. (2010). How do fathers influence children’s development? Let me count the ways. In M. E. Lamb (Ed.), The role of the father in child development (5th ed., pp. 1-26). New Jersey: Wiley.
Lecavalier, L., Leone, S., & Wiltz, J. (2006). The impact of behavior problems on caregiver stress in young people with autism spectrum disorders. Journal of
Intellectual Disability Research, 50, 172-183.
girls: a meta-analysis. Psychological Bulletin, 109, 267-296.
McBride, B. A. (1989). Stress and fathers’ parental competence: implications for family life and parent educators. Family Relations, 38 (4), 385-389.
Mesman, J., & Koot, H. M. (2000), Common and specific correlates of preadolescent internalizing and externalizing psychopathology. Journal of Abnormal Psychology,
109, 428-437.
Morgan, J., Robinson, D., & Aldridge, J. (2002). Parenting stress and externalizing child behavior. Child and Family Social Work, 7, 219-225.
Noh, S, Dumas, J. E., Wolf, L. C., & Fisman, S. N. (1989). Delineating sources of stress in parents of exceptional children. Family Relations, 38 (4), 456-461. Paquette, D. (2004). Theorizing the father-child relationship: mechanisms and developmental outcomes. Human Development, 47, 193-219.
Peterson, G. W. & Hennon, C. B. (2005). Conceptualizing parental stress with family stress theory. In P. C. McKenry & S. J. Price. Families and change: Coping with
stressful events and transitions (3rd ed.). Thousand Oaks: Sage Publications, p. 25-
48.
Phares, V., Lopez, E., Fields, S., Kamboukos, D., & Duhig, A. M. (2005). Are fathers involved in pediatric psychology research and treatment? Journal of Pediatric
Psychology, 30 (8), 613-643.
Plant, E. A., Hyde, J. S., Keltner, D., & Devine, P. G. (2000). The gender stereotyping of emotions. Psychology of Women Quarterly, 24, 81-92.
Pomerleau, A., Bolduc, D., Malcuit, G., & Cossette, L. (1990). Pink or blue:
environmental gender stereotypes in the first two years of life. Sex Roles, 22, 359- 367.
(1998). Parenting stress in mothers of young children with oppositional defiant disorder and other severe behavior problems. Child Study Journal, 28 (2), 93-110. Teti, D. M., & Gelfand, D. M., (1991). Behavioral competence among mothers of infants in the first year: the mediational role of maternal self-efficacy. Child
