Family Financial Stress,Parenting and Problem Behavior in Adolescents

Abstract

In this study, we examine how financial stress is associated with problem behavior in adolescents through the lives of their parents. Using an actor–partner interdependence model, we explore pathways within (actor) and between (partner) parents. Our data included 340 families, with both parents rating their financial stress, depressive symptoms, and interparental conflict, and with parents and adolescents rating parenting and problem behavior in adolescents. The results indicate that the association between financial stress and problem behavior in adolescents is mediated by depressive symptoms, interparental conflict, and positive parenting. Another finding is that financial stress has direct and indirect effects on interparental conflict. Furthermore, the impact of financial stress on positive parenting is greater for fathers than it is for mothers. Although actor effects are more prominent, there is also evidence of partner effects. Our results underscore the importance of including multiple family members in studies on family stress processes.

Keywords

family processes father-child relationship mother-child relationship adolescents dyadic relationship/quality/satisfaction

Introduction

During the past two decades, a large body of research has focused on the family stress model (Conger, Conger, & Martin, 2010), which examines family-based pathways through which financial stress is associated with negative child outcomes. These studies have shown that worries and insecurities about the family’s financial situation often generate psychological distress, possibly contributing to marital conflict and problems in parenting, in addition to exacerbating emotional and behavioral problems in children (Falconier & Epstein, 2010; Parke et al., 2004). Most research on the relationship between financial stress, parenting, and child development has focused on mothers (Barnett, 2008), even though fathers are becoming increasingly involved in the lives of their children (Lamb & Lewis, 2010). Further studies considering the interdependence of and mutual influence between mothers and fathers may therefore provide additional insight into family-based pathways through which financial stress affects the adjustment of children and adolescents.

The current study proceeds from the extended family stress model developed by Conger and colleagues (Conger & Conger, 2002; Conger et al., 2010). It expands upon previous studies on family stress processes by including data from both parents and an adolescent, as well as by studying separate pathways through which financial stress experienced by parents might affect the psychological distress, interparental experience, and parenting of parents and their partners, which might subsequently affect the behavioral adjustment of adolescents. To test our theoretical model (shown in Figure 1), we applied the actor–partner interdependence model (APIM) approach (Kenny, Kashy, & Cook, 2006). The APIM approach to examining family functioning is relatively recent, and it was designed to estimate the impact of the independent variables of individuals on their own dependent variables (actor effects), as well as on the dependent variables of their partners (partner effects). The model assumes that the two members of the dyad influence each other in the form of partner effects, thus creating interdependence between members (Ledermann & Macho, 2009).

Figure 1.

The proposed financial stress model.

The Economic Context of Belgium

Our study was conducted in the Dutch-speaking part of Belgium (Flanders), approximately 3 years after the onset of the financial crisis (i.e., in February–March 2012). It should therefore be viewed in this context. The financial crisis has affected Belgium since mid-2008, when two of the country’s largest banks started to face severe problems. The value of their stocks plummeted and infected the stock prices of other Belgian banks and companies. By the end of September 2008, both banks had been bailed out by the Belgian government (Eichler & Hielscher, 2012). Shortly after the relatively costly and unpopular bank rescues, the Belgian government had to address the economy. The financial crisis had a serious impact on the economy, on the labor market, and on families (Kickert, 2012). In December 2008, the Belgian government took several economic recovery measures, which ultimately proved effective (Kickert, 2012). The economic recovery in 2010 was faster and stronger than expected, partly due to the unexpectedly strong rebound in German exports (Lebrun, 2011).

The steady performance of the Belgian economy nevertheless stands in sharp contrast to the subjective evaluation of the financial crisis by the citizens of Belgium. According to the Eurobarometer, a cross-national comparative survey containing data from 27 European Union Member States, in November 2011, only 30% of all Belgian respondents considered the situation of the national economy to be good, as compared to 80% or more of citizens in other countries (e.g., Sweden, Luxembourg and Germany). Approximately 49% of the Belgian respondents expected that the economic situation of Belgium would be worse within the next 12 months, and only 23% believed that the impact of the economic crisis on the job market had reached its peak (EC, 2011). These findings are interesting from the perspective of family stress, as previous studies have suggested that the effects of the subjective experience of economic disadvantage on the lives of parents and children might be stronger than those related to the objective experience of being poor (Barnett, 2008; Conger & Donnellan, 2007; Mistry, Biesanz, Taylor, Burchinal, & Cox, 2004).

The Family Stress Model

Conger and colleagues developed the family stress model in order to study the effects of the farm economic crisis in the United States in the 1980s (Conger & Conger, 2002; Conger, Ge, Elder, Lorenz, & Simons, 1994). The model suggests that the perception of financial strain mediates the relationship between income and psychological distress in parents. The model further predicts that when financial strain or distress is high, individuals are at increased risk for psychological distress (e.g., depression or anxiety), which increases the likelihood of marital conflict. Rather than assuming that financial strain has a direct negative effect on the marital relationship, the model suggests that financial stress has only an indirect effect on relationship satisfaction through the psychological distress that it elicits in the parents. Several studies conducted in the United States (e.g., Conger & Conger, 2002; Parke et al., 2004) and in other countries (Aytac & Rankin, 2009; Kinnunen & Feldt, 2004; Kwon, Rueter, Lee, Koh, & Ok, 2003) have supported this supposition. Nevertheless, evidence from outside the United States suggests that financial stress is more likely to have both direct and mediating effects on marital relationships. This finding might reflect differences in cultural and/or economic traditions, and it therefore merits further investigation (Conger et al., 2010; Kwon et al., 2003).

In its expanded form, the family stress model predicts that financial strain or stress influences the development of children and adolescents primarily through the experiences of parents (Conger et al., 2010). As such, psychological distress caused by financial stress influences interparental conflict and problems in parenting, and disruptive parenting subsequently mediates or explains the influence of parental distress and interparental conflict on problem behavior or other child and adolescent outcomes (see Figure 1). In the past decade, studies have provided increasing support for these predictions (Benner & Kim, 2010; Linver, Brooks-Gunn, & Kohen, 2002; Mistry et al., 2004; Solantaus, Leinonen, & Punamaki, 2004), despite other evidence that interparental conflict has a direct effect on child maladjustment (Parke et al., 2004).

The current understanding of family stress processes is nevertheless subject to several limitations (Barnett, 2008; Conger et al., 2010). First, most studies that apply the family stress model to couples use aggregated constructs for financial stress, psychological distress, interparental relationship, and parenting (Leinonen, Solantaus, & Punamaki, 2002). These studies thus neglect the interdependence of the two parents and the mutual influence that they have on each other. In other words, they focus exclusively on actor effects. Our review of prior research on extended family stress models (i.e., studies including parenting and child outcomes as study variables) revealed only two studies that use a dyadic approach (Leinonen et al., 2002; Parke et al., 2004). In a study conducted in Finland, Leinonen and colleagues (2002) use a dyadic approach to investigate pathways from economic pressure to parenting through various aspects of mental health and interparental interaction. The study does not include child behavior as an outcome variable (Leinonen et al., 2002), however, and it therefore does not explore mediating pathways from financial stress in parents to child outcomes. In the other study, Parke and colleagues (2004) investigate the impact of financial stress in parents on child adjustment problems, through depressive symptoms and interparental problems. This study employs a dyadic approach for all of the constructs examined, with the exception of the construct for interparental problems. Given that the scores of mothers and fathers on the interparental problem construct are combined into a single score and used as a mediator between depressive symptoms and parenting, the authors are unable to assess effects on parenting within and between partners (Parke et al., 2004). The present study addresses this gap by discriminating between financial stress, depressive symptoms, interparental conflict, and parenting in mothers and fathers.

Another limitation in the current literature is that it pays little attention to possible gender differences in the pathways from stress to parenting (Barnett, 2008; Ponnet, Mortelmans, et al., 2013). This stems from the fact that early parenting research focused almost exclusively on mothers, partly due to the common assumption that mothers played a central role in child development (Adamsons & Buehler, 2007). Nevertheless, some evidence suggests that the nature of the relationship between stress and parenting may vary for mothers and fathers (Barnett, Deng, Mills-Koonce, Willoughby, & Cox, 2008). According to the fathering-vulnerability hypothesis (Cummings, Goeke-Morey, & Raymond, 2004), fathering and father-child relationships might be more vulnerable to stress than is the case with mothering and mother-child relationships. One possible explanation for the increased vulnerability of fathering is that the roles of fathers are less clearly defined by social conventions than are the roles of mothers, thus making fathering more sensitive to external influences (Coiro & Emery, 1998). Nevertheless, findings from the few APIM studies on determinants of parenting are inconclusive with regard to gender differences between parents. To the best of our knowledge, only five studies have used an APIM approach to investigate the influence of the interparental relationship (Leinonen et al., 2002; Nelson, O’Brien, Blankson, Calkins, & Keane, 2009; Ponnet, Mortelmans, et al., 2013) or depressive symptoms (Malmberg & Flouri, 2011; Nelson et al., 2009; Ponnet, Wouters, et al., 2013) on parent-child relationships. The studies by Nelson and colleagues (2009) and by Leinonen and colleagues (2002) report several gender-specific pathways, thus lending support to the fathering-vulnerability hypothesis. The other studies provide no evidence of gender differences (Malmberg & Flouri, 2011; Ponnet, Mortelmans, et al., 2013; Ponnet, Wouters, et al., 2013), however, thus underlining the need for further research (Ponnet, Wouters, et al., 2013). To improve insight into the ways in which mothers and fathers respond differently to financial stress, we explicitly test for gender differences in all pathways within and between parents.

Finally, most previous research was conducted in the United States. Belgium is quite different from the United States in terms of economic and social security. Relative to the United States, a larger fraction of the gross domestic product in Belgium is spent on pensions and other social security benefits. Furthermore, social insurance programs in the United States are usually organized and provided by employers. In Belgium, they are operated at the national level on a nearly universal basis and not through the employer (Garfinkel, Rainwater, & Smeeding, 2010). This study thus broadens existing understandings by investigating how two-parent families respond to financial stress within a welfare state in which families are buffered by social security.

Aims of the Study

The objective of this study is to replicate and extend the family stress model in a Belgian sample, thus broadening the model’s applicability by testing it in a differential contextual setting. As shown in Figure 1, we propose an APIM approach to generate additional insight into family stress processes. We have three aims.

The first aim is to understand how financial stress is related to problem behavior in adolescents through the depressive symptoms, experiences of interparental conflict, and positive parenting behavior in both parents. Our model discriminates between the responses of mothers and fathers, thereby enabling us to test for actor and partner effects. Based on the findings of other studies on family stress (e.g., Conger et al., 1994, 2002), we expect that financial stress has actor effects on depressive symptoms, with greater financial stress associated with higher levels of depressive symptoms. Furthermore, we expect positive actor effects between depressive symptoms and interparental conflict, as well as negative actor effects between interparental conflict and positive parenting. With regard to partner effects, our hypotheses are based on the findings of studies that use a dyadic approach to investigate the association between financial stress and marital relationships (e.g., Falconier & Epstein, 2010; Kinnunen & Feldt, 2004), as well as on the few APIM studies about determinants of parenting (e.g., Nelson et al., 2009; Ponnet, Mortelmans, et al., 2013). We expect that increased levels of financial stress experienced by one parent exacerbate depressive symptoms in the other parent and that depressive symptoms are subsequently associated with the other parent’s relationship adjustment. Furthermore, we expect partner effects from interparental conflicts to parenting.

The second aim of this study is to examine gender differences in the actor and partner pathways. Based on the fathering–vulnerability hypothesis (Cummings et al., 2004), we expect the parenting of fathers to be more susceptible to deterioration in the face of stress than is the parenting of mothers.

Third, this study involves a formal test for evidence of mediation. Consistent with the findings of other studies on family stress processes outside the United States (e.g., Aytac & Rankin, 2009; Kinnunen & Feldt, 2004; Kwon et al., 2003), we hypothesize that financial stress experienced by mothers and fathers has both direct and indirect effects on interparental conflict. Given that prior research on family processes has provided little evidence of direct effects of financial stress on problem behavior in adolescents (Conger et al., 2010), we expect depressive feelings, interparental conflicts, and parenting to mediate this relationship.

Method

This study is part of a larger study (entitled Relationships between mothers, fathers and children; RMFC) of families with adolescents between 11 and 17 years of age. The RMFC study used a multi-actor approach, focusing on families with different income levels. Given that many economically disadvantaged families are notoriously difficult to access in any systematic way (Faugier & Sargeant, 1997), and given the high rate of nonresponse associated with the collection of multiactor data (Kalmijn & Liefbroer, 2011), the RMFC project employed a nonprobabilistic sampling design. The study was approved by the Ethics Committee of the University of Antwerp (Belgian registration number: B300201215397).

Sample

The sample for the analyses includes 340 two-parent families with a target adolescent between 11 and 17 years of age (i.e., attending secondary school). Families were recruited from five regions of the Dutch-speaking part of Belgium with assistance from undergraduate students from two institutes of higher education. The students were instructed to recruit low-income, middle-income, and high-income two-parent families. Only nondivorced parents were eligible to participate. Students received course credit for their recruitment efforts. It is important to note that the average age of the students was 34.65 (SD = 1.24) and that most were working in the social services. As such, the study took advantage of the social networks of the students in order to obtain a large set of potential respondents. Families were sent a letter explaining the research purpose, and they were subsequently contacted and asked to participate. In all, 456 packages of envelopes and questionnaires were distributed, 359 (78.9%) of which were returned by post in February/March 2012. Target participants were instructed to complete the booklets individually and not to discuss the content of the questionnaire with one another. The booklets were returned to the first author in a closed envelope. Mothers and fathers were also asked to sign and return a written consent form by post in a separate envelope.

Only families in which the father, the mother, and the adolescent returned their questionnaires were included. Twelve families were excluded because the parents reported that their adolescents had pervasive developmental disorders and lived at home only during weekends. Seven additional families were excluded because one of the family members failed to complete the questionnaire adequately. The sample for these analyses therefore consists of 340 families (1020 individuals): 93.8% (n = 319) married and 6.2% (n = 21) cohabiting couples. The average age for fathers is 46.78 years (SD = 4.74, range = 29), and the average age for mothers is 44.68 years (SD = 4.12, range = 23). A paired t test reveals a significantly higher mean age for fathers: t(339) = −12.12, p < .001). Education is measured as the highest level of education achieved. The educational level of fathers is significantly different from that of mothers: χ²(9) = 140.85, p < .001. Within our sample, 13.7% of the fathers and 8.5% of the mothers had completed less than nine years of education (lower secondary); 29.2 % of the fathers and 25.2% of the mothers had completed secondary education; 22.8% of the fathers and 38.3 % of the mothers had completed at least three years of higher education; and 34.3% of the fathers and 28% of the mothers had completed more than three years of higher education. With regard to employment, 97% of the fathers (n = 327) and 87% of the mothers (n = 295) were working either full time or part time. Furthermore, three-person households account for 12.1% of the sample, with four-person households accounting for 41.2%, five-person households accounting for 32.6%, six-person households accounting for 10.3%, and households of seven or more persons accounting for 3.9%. Using the modified OECD equivalence scale (Haagenars, de Vos, & Zaidi, 1994), the average household income of our sample (M = 1,655.22€, SD = 649.92) is almost identical to the average household income of the Dutch-speaking part of Belgian households with at least one adolescent between 11 and 17 years old (M = 1,672.88€, SD = 749.90, authors’ own calculations based on EU-SILC 2010 UDB). In the present sample, 10.32% of the households had an equivalized monthly income below €1000 (8.87% in the population), 67.85% between €1000 and €2000 (58.19% in the population), 17.70% between €2000 and €3000 (28.61% in the population) and 4.13% above €3000 (6.03% in the population). This distribution suggests that middle-income families are slightly overrepresented in this sample and that higher-income families are slightly underrepresented.

The average age of the target adolescents is 14.24 years (SD = 1.80, range = 6), with 40.1% boys (n = 137) and 59.9% girls (n = 203). Univariate analysis of variance (ANOVA) reveals no between-group differences for age: F(1, 339) < 1.

Measures

Household income

Respondents were asked to report the monthly net income of the household (including wages, interests, child support, supplemental income, and other sources) ranging from 1 (less than €249) to 16 (€9000 or more). Consistent with other studies (Clark & Senik, 2010; McBride, 2001), we transformed the categorical variable into a continuous variable according to the midpoint of the categories (with €9000 as the midpoint of the last category). Because the ratings of mothers and fathers did not differ, t(315) = .20, ns, we used the average of both ratings in our analyses. The average household income in this sample is €4170.73 (SD = 1518.67).

Financial stress

The financial stress construct consists of three measures: financial need, financial insecurity, and financial burden. Financial need is measured using a three-item scale developed for this study. Two items are adapted from a study by Blau (1994): It is difficult to afford much more than the basics with our current income and I feel that our current income allows me to maintain a desirable standard of living (reverse-scored). The third item is With our current income, it is difficult to make the ends meet. Both mothers and fathers were asked to rate the items along a seven-point Likert scale ranging from 1 = strongly disagree to 7 = strongly agree. Principal component analyses (PCA) reveal high factor loadings for mothers (ranging from .82 to .88, R² = 74.34) and fathers (ranging from .82 to .86, R² = 71.18). The mothers in this sample perceived more financial need than did the fathers: paired t(335) = 2.50, p < .05. For financial insecurity, mothers and fathers were asked to rate the four self-constructed items. Sample items include I am worried that I will not be able to pay my bills in the near future and I think that I will have to scale down my living standards in the following months. All of the items are scored along a seven-point Likert scale ranging from 1 = strongly disagree to 7 = strongly agree. Factor loadings (PCA) range from .63 to .88 (R² = 60.99) for mothers and from .74 to .84 (R² = 63.43) for fathers. A paired t test reveals no differences between mothers and fathers, t(335) = 1.29. The financial burden items are adapted from the EU Statistics on Income and Living Conditions (EU–SILC) instrument (Eurostat, 2008). Parents rated the extent to which five sources of costs (e.g., medical; car/fuel; childcare or other child-related costs; house-related costs; mortgages, loans) posed a financial burden or struggle for their households. The items are scored along a four-point Likert scale ranging from 1 = not a burden/struggle to 4 = a heavy burden/struggle. The results of PCA reveal high factor loadings for mothers (ranging from .69 to .88, R² = 64.26) and for fathers (ranging from .74 to .88, R² = 67.91). The analysis reveals no significant differences between mothers and fathers: t(336) = 1.28.

Depressive symptoms

To measure depressive symptoms, the short-form 11-item version of the Center for Epidemiologic Studies—Depression scale (CES–D) was administered (Kohout, Berkman, Evans, & Cornoni-Huntley, 1993). Respondents were asked to think about the past week and to indicate how often they had felt or behaved in a certain way (e.g., felt depressed or felt that everything was an effort). All of the items are scored along a four-point Likert scale ranging from 1 = rarely or none of the time to 4 = most or all of the time. The results of PCA reveal high factor loadings for mothers (ranging from .42 to .79, R² = 37.44) and for fathers (ranging from .46 to .81, R² = 42.14). Mothers reported more depressive symptoms than fathers did: t(335) = 1.99, p < .05.

Interparental conflict

The interparental conflict construct includes three measures: overt hostility, verbal aggression, and stress within the relationship. Overt hostility is measured using the O’Leary-Porter scale (OPS, Johnson & O’Leary, 1987; Porter & O’Leary, 1980). The OPS is a widely used scale designed to assess the extent to which parents argue openly in the presence of their children. The scale consists of 10 items that are scored along a five-point Likert scale ranging from 1 = never to 5 = very often. All items are scored positively, with the exception of item 10 (regarding displays of affection), which is coded negatively. An example item is How often do you complain to your spouse about his/her personal habits in front of your child? Higher scores on the OPS indicate a greater level of overt hostility. The results of PCA reveal adequate factor loadings (ranging from .37 to .68 for mothers and from .36 to .77 for fathers), with the exception of the negatively worded item (.18 for mothers and .20 for fathers). The latter item was therefore omitted (R² = 42.33 for mothers and 42.72 for fathers). The results indicate no differences between mothers and fathers: t(332) = .39.

To measure verbally aggressive acts perpetrated by respondents and their partners, we administered a 10-item inventory similar to the verbal aggression subscale of the Conflicts and Problem Solving Strategies questionnaire (Kerig, 1996). The items concern such behaviors as yelling, accusing, insulting, cursing, raising the voice, and interrupting. Items are scored along a five-point Likert scale, ranging from 1 = never to 5 = very often. The results of PCA reveal high factor loadings for mothers (ranging from .58 to .83, R² = 52.94) and for fathers (ranging from .51 to .80, R² = 52.21). The results indicate no significant differences between mothers and fathers: t(332) = -.07.

To assess the perceived level of stress within the relationship, we use the Multidimensional Stress Questionnaire for Couples (Bodenmann, Ledermann, & Bradbury, 2007). Mothers and fathers were asked to use a five-point Likert scale to indicate how stressful/straining 10 situations (e.g., disturbing habits of the partner, different attitudes concerning the relationship and life) had been within the relationship during the past 12 months. The results of PCA reveal high factor loadings for mothers (ranging from .67 to .81, R² = 55.36) and for fathers (ranging from .67 to .79, R² = 56.08). The mothers in this sample perceived more stress than the fathers did: t(332) = 3.56, p < .001.

Parenting

Mothers, fathers, and adolescents independently rated the parenting behavior using the positive parenting subscale of the Parental Behavior Scale, short version (PBS–S; Van Leeuwen, Vermulst, Kroes, De Meyer, & Veerman, 2011), which is based on previous studies by Van Leeuwen and Vermulst (2004) and by Patterson (1982). The scale consists of eight items in the form of affirmatives (e.g., I make time to listen to my child, when he/she wants to tell me something). The items are scored along a four-point Likert scale ranging from 1 = never to 5 = always. The results of PCA indicate high factor loadings for mothers (ranging from .59 to .80, R² = 47.34) and for fathers (ranging from .61 to .81, R² = 53.92). The mothers in this sample perceived their parenting of their adolescents significantly more positively than the fathers did: t(325) = 9.81, p < .001. Adolescents completed scales on parenting twice, once for mothering and once for fathering. The results of PCA indicate high factor loadings on the adolescents’ reports of positive parenting in the mother–adolescent relationship (.60–.80, R² = 54.05) and in the father–adolescent relationship (.72–.83, R² = 60.29). No significant differences were found between the two adolescent reports: t(323) = -.21.

Problem behavior in adolescents

Mothers, fathers, and adolescents independently rated the adolescent’s problem behavior using the Externalizing Problems scale of the Child Behavior Checklist—Parent Report and Youth Self-Report (CBCL, Achenbach, 1991). The CBCL is a widely used questionnaire, consisting of a series of statements that might describe the youth during the previous 6 months. Responses are coded as 0 (not true), 1 (somewhat or sometimes true) or 2 (very true or often true). Examples of items include I lie or cheat (youth report)/My child lies or cheats (parent report) and I disobey at school (youth report)/My child disobeys at school (parent report). Items are summed and divided by the number of items. Alpha scores for mothers, fathers, and adolescents are .87, .87, .81, respectively. The adolescents in this sample reported more problem behavior (M = .28, SD = .18) than did their mothers, M = .18, SD = .16; paired t(336) = 9.42, p < .001, or their fathers, M = .18, SD = .16; paired-t(334) = 9.31, p < .001.

Analytic Strategy

We performed structural equation modeling (SEM) using Mplus (Muthén & Muthén, 2010) with maximum likelihood estimation to examine relationships between household income, financial stress, parental depressive symptoms, interparental conflict, parenting, and problem behavior in adolescents. Missing values on endogenous, exogenous and control variables were excluded from the analyses using listwise deletion. The final sample consists of 326 families. The analyses were performed as follows. First, we constructed a measurement model and investigated the fit. The latent constructs financial stress, interparental conflict, and positive parenting were created using the factor scores of the measures. The latent construct for problem behavior in adolescents was created using the standardized scores of the mother, the father, and the adolescent. Given that the adolescents rated the same parenting behavior twice (once for each parent), correlated error terms were allowed across the two constructs in order to rule out response bias (see also Lavee, McCubbin, & Olson, 1987; Leinonen et al., 2002). Second, we constructed structural equation models using depressive symptoms, interparental conflict, and positive parenting as mediators between financial stress and problem behavior in adolescents. These models were used to explore for actor and partner pathways between the study variables. To test for gender differences, we generated nested models by constraining pathways to be equal and comparing these models to the baseline (unconstrained) models. Finally, we performed a formal test for evidence of mediation, with depressive symptoms, interparental conflict, and parenting as mediators.

Results

Bivariate Correlations

The correlations among the variables in the path model are presented in Table 1. As expected, there are significant correlations between household income and financial stress in mothers and fathers. Our results indicate no significant associations between household income and the other variables, with the exception of a significant association between income and depressive symptoms in mothers. At the actor level, financial stress is significantly associated with depressive symptoms and experiences of interparental conflict in mothers and in fathers, as well as with positive parenting by fathers. At the partner level, financial stress is significantly associated with depressive symptoms in mothers as well as with feelings of interparental conflict on the part of fathers. With regard to the association between financial stress and problem behavior, nine out of the 18 correlations are significant. Furthermore, although depressive symptoms are significantly associated with experiences of interparental conflict for the respondent and the partner, there are hardly any significant associations between depressive symptoms and positive parenting. Finally, our results reveal significant actor and partner associations among interparental conflict, positive parenting, and problem behavior on the part of the adolescent.

Table 1.

Correlations Among the Variables.

Variables		1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22
1	Household income
2	Financial need MR	−.46**
3	Financial insecurity MR	−.31**	.68**
4	Financial burden MR	−.36**	.64**	.61**
5	Financial need FR	−.51**	.63**	.44**	.46**
6	Financial insecurity FR	−.33**	.46**	.52**	.48**	.60**
7	Financial burden FR	−.39**	.53**	.45**	.55**	.64**	.61**
8	Depression MR	−.12*	.22**	.29**	.18**	.18**	.22**	.17**
9	Depression FR	−.05	.07	.09	.10	.11*	.27**	.19**	.22**
10	Verbal aggression MR	−.01	.18**	.20**	.18**	.11*	.19**	.08	.26**	.16**
11	Overt hostility MR	−.02	.16**	.18**	.15**	.08	.17**	.08	.23**	.17**	.73**
12	Relationship stress MR	−.07	.19**	.16**	.17**	.09	.18**	.09	.41**	.31**	.46**	.52**
13	Verbal aggression FR	.06	.04	.05	.03	.09	.16**	.06	.12*	.28**	.60**	.47**	.34**
14	Overt hostility FR	.02	.14*	.10	.11*	.10	.19**	.09	.20**	.35**	.57**	.58**	.48**	.68**
15	Relationship stress FR	−.03	.16**	.17**	.15**	.21**	.28**	.20**	.31**	.43**	.32**	.32**	.51**	.48**	.55**
16	Positive mothering CR	−.04	.04	.06	.08	−.01	.03	.00	.00	−.07	−.11	−.12*	−.13*	−.01	−.09	−.01
17	Positive mothering MR	.04	.01	.04	.13*	.04	.05	−.01	.03	−.07	−.15**	−.14**	−.06	−.08	−.06	−.04	.42**
18	Positive fathering CR	−.02	−.03	−.02	−.07	−.12*	−.15**	−.16**	.04	−.09	−.13*	−.19**	−.19**	−.07	−.12*	−.09	.66**	.27**
19	Positive fathering FR	.06	−.03	.04	.01	−.16**	−.15**	−.11*	.06	−.17**	−.13*	−.14*	−.17**	−.14**	−.12*	−.12*	.25**	.29**	.46**
20	Externalizing MR	−.09	.15**	.09	.11*	.12*	.06	.09	.12*	.05	.30**	.22**	.18**	.20**	.16**	.15**	−.14**	−.15**	−.13*	−.11*
21	Externalizing PR	−.02	.06	.03	.05	.131*	.12*	.15**	.05	.06	.26**	.11*	.10	.28**	.14*	.14**	−.12*	−.15**	−.11	−.18**	.67**
22	Externalizing CR	−.01	.09	.08	.10	.19**	.14*	.19**	.03	.09	.15**	.05	.05	.14**	.09	.15**	−.23**	−.13*	−.18**	−.14*	.35**	.31**
	M	4170.73	2.71	3.05	1.85	2.56	2.96	1.81	1.57	1.51	2.06	1.86	2.09	1.92	1.81	1.94	3.50	4.09	3.13	3.72	18	.18	28
	SD	1518.67	1.30	1.18	.71	1.21	1.81	.72	.32	.31	.63	.50	.78	.58	.53	.71	.79	.83	.85	.61	.16	.16	.18

Note. FR = father report; MR = mother report; CR = child report.

p < .05. **p < .01.

Measurement Models

The initial measurement model provides a good fit to the data: χ²(130) = 321.125, p < .001; χ²/df = 2.47, CFI = .93, RMSEA = .066 (CI: .057–.075), SRMR = .046. All factor loadings are greater than .41. Our results reveal interdependencies between financial stress (r = .79, p < .001), experiences of interparental conflicts (r = .79, p < .001), and positive parenting (r = .60, p < .001) in mothers and in fathers. Because the objectives of our research require the inclusion of separate scores for the latent constructs of mothers and fathers, we conducted a test for each construct in order to determine whether such a distinction was warranted. We compared models in which maternal and paternal constructs are modeled separately to produce a model in which both constructs are combined into a single latent construct. The results of χ² difference tests indicate that combining the constructs decreases the fit significantly for financial stress, χ²(1) = 81.52, p < .001; interparental conflict, χ²(1) = 123.96, p < .001; and positive parenting, χ²(1) = 28.98, p < .001. As such, all latent constructs are modeled separately in our analyses.

We subsequently included the sociodemographic variables as covariates in the analyses and examined the relationships between the age and gender of the adolescent; the age, educational level, and financial responsibility of mothers and fathers; and all latent variables. The age of the adolescent is significantly associated with the parenting of the mother, β = −.17, S.E = .07, p < .05, and of the father, β = −.25, S.E = .06, p < .001. Furthermore, the educational level of fathers is significantly associated with their level of financial stress, β = −.30, S.E = .05, p < .001. In addition, financial responsibility is associated with financial stress in mothers, β = −.14, S.E = .06, p < .05. Our structural model (as presented below) has been adjusted for the influence of these variables.

Structural model: Financial stress relating to externalizing problem behaviors through parental depressive symptoms, interparental conflict, and parenting.

The results of the structural model are presented in Figure 2. The results of the fit statistics indicate an adequate model fit: χ²(249) = 543.46, p < .001, χ²/df = 2.18, CFI = .90, RMSEA = .06, SRMR = .06. The test results for our model are consistent with the family stress model, in that depressive symptoms, interparental conflict, and parenting (for each parent) are identified as mediators between financial stress in parents and problem behavior in adolescents. More specifically, financial stress is positively related to depressive symptoms and to interparental conflict, with interparental conflict being negatively associated with positive parenting and positively associated with problem behavior in adolescents. As expected, positive parenting is negatively associated with problem behavior in adolescents (see Figure 2). Contrary to our expectations, the results indicate no significant partner effects between financial stress and depressive symptoms, or between interparental conflict and positive parenting. Nevertheless, the actor and partner effects from depressive symptoms to interparental conflict are significant.

Figure 2.

Financial stress relating to problem behavior in adolescents via depressive symptoms, interparental conflicts, and positive parenting.

One-by-one comparisons of the constrained model to the baseline model reveal a gender difference in the actor effects from financial stress to positive parenting: χ²(1) = 10.85, p < .001. This result indicates that financial stress in fathers has a negative effect on positive parenting (β = −.16, p < .05), whereas there is no significant actor effect between financial stress in mothers and their positive parenting (β = .08, ns). Our analysis reveals no other gender differences in the actor and partner pathways.

As a formal test for evidence of mediation, we used the INDIRECT command in Mplus to estimate the value and significance of the product of the indirect pathways by which financial stress influences problem behavior in adolescents. Only the significant pathways were included in our analyses (see Figure 2). First, we tested the pathways between financial stress and interparental conflict. The indirect effect of financial stress on interparental conflict is significant (indirect β = .07, p < .001). As such, the results provide evidence that depressive symptoms partially mediate the relationship between financial stress and interparental conflict. Second, we tested the indirect pathway between interparental conflict to problem behavior in adolescents through parenting, yielding a significant result (indirect β = .03, p < .05). Third, we tested the mediation pathways between financial stress and positive parenting. As expected, there is a significant indirect effect from financial stress in mothers to their positive parenting, through their depressive symptoms and experiences of interparental conflict (indirect β = −.04, p < .05). The indirect pathway from financial stress in fathers to their parenting, through their depressive symptoms and experiences of interparental conflict is significant (indirect β = −.01, p < .05), as is the indirect pathway through interparental conflict only (indirect β = −.02, p < .05), as shown in Figure 2. Finally, our analyses reveal that the total indirect effects from financial stress in mothers to problem behavior in adolescents (indirect β = .04, p < .01) and from financial stress in fathers to problem behavior in adolescents (indirect β = .06, p < .001) are significant. The various indirect pathways provide evidence for the proposed family stress model (see Table 2).

Table 2.

Indirect Effects of Financial Stress on Child Outcome.

	Standardized estimate	SE
Financial stress MR to child outcome
Total indirect	.037	.011**
Fin. stress MR → conflict MR→ child outcome	.017	.008*
Fin. stress MR → depression MR → conflict MR → child outcome	.009	.003**
Fin. stress MR → depression MR → conflict FR → child outcome	.004	.002*
Fin. stress MR → conflict MR → parenting MR → child outcome	.003	.002*
Fin. stress MR → depression MR → conflict MR → parenting MR → child outcome	.002	.001*
Fin. stress MR → depression MR → conflict FR → parenting FR → child outcome	.001	.000†
Financial stress FR to child outcome
Total indirect	.064	.016***
Fin. stress FR → conflict FR → child outcome	.016	.007*
Fin. stress FR → parenting FR → child outcome	.030	.013*
Fin. stress FR → depression FR → conflict MR → child outcome	.004	.002*
Fin. stress FR → depression FR → conflict FR → child outcome	.009	.003**
Fin. stress FR → conflict FR → parenting FR → child outcome	.003	.002†
Fin. stress FR → depression FR → conflict MR → parenting MR → child outcome	.001	.000†
Fin. stress FR → depression FR → conflict FR → parenting FR → child outcome	.002	.002*

Note. Fin. stress = financial stress; FR = father report; MR = mother report.

†

p < .1. *p < .05. **p < .01. ***p < .001.

Discussion

In this study, we examined family stress processes in which financial stress experienced by Belgian couples is associated with problem behavior in adolescents. We draw upon the extended family stress model proposed by Conger and colleagues (2002), but expanded it by including data from both parents within the same family and by analyzing the data using an actor-partner interdependence approach. The current study had three objectives. First, we examined how the financial stress of parents is associated with the problem behavior of adolescents, through the lives of parents, exploring for pathways both within and between mothers and fathers. Second, we examined whether the pathways differed between parents. Third, we conducted a more formal test for evidence of mediation.

With respect to the first objective, the results support the extended stress model by Conger and colleagues (1994, 2002) and add to findings from previous studies (e.g., Leinonen et al., 2002) indicating that when parents are stressed about their current and future financial situation, the effects of this stress are related to increased depressive symptoms, which subsequently increase interparental conflict. The results further indicate that parenting partially mediates the association between interparental conflicts and problem behavior in adolescents.

By focusing on effects within and between family members, we estimated whether relational effects occurred as well (Kenny et al., 2006). Although actor effects were more prominent, we found partner effects between depressive symptoms and interparental conflict, suggesting that more depressive symptoms result in more relationship adjustment of the partner. This finding is consistent with the family system approach which emphasizes that the family is a complex, integrated whole whereby problems in the family system (e.g., depressive symptoms) tend to have negative effects on other family members. Our findings underscore the importance of treating parental depressive symptoms and their ramifications at the family level. Although parents can have depressive symptoms that exist independently of the spouse, these depressive symptoms have implications for other family members. Clinicians should therefore teach individuals to cope not only with their own depressive symptoms, but also with those of their partner.

Contrary to our expectation, we found no partner effects between financial stress and depressive symptoms. This contradicts findings from studies by Falconier and Epstein (2010, 2011), who found partner effects from financial stress to psychological distress. In the latter studies, however, psychological distress was operationalized as demand/withdraw (Falconier & Epstein, 2011) and psychological aggression/positive behavior toward the partner (Falconier & Epstein, 2010). These constructs specifically refer to the relationship. Our results agree with those obtained in previous dyadic studies that have focused on depressive symptoms or anxiety as a response to financial stress (e.g., Parke et al., 2004), constructs that refer to individual states.

With respect to our second objective, we investigated whether the strength of the pathways differed between mothers and fathers. Our results suggest that family stress processes are gendered to some extent. Although mothers experienced slightly more financial stress than fathers did, only the financial stress experienced by fathers had a direct effect on the parenting. In a way, this finding is consistent with the hypothesis that fathering and father–adolescent relationships are more vulnerable to stress than are mothering and mother–adolescent relationships (Cummings et al., 2004). It is interesting to note, however, that we found no gender differences in the relationship between interparental conflict and parenting. Either way, our findings demonstrate that too little attention has been paid to the role of fathers. Given that mothers seem to cope better with financial stress than fathers do, it might be interesting to investigate strategies that parents use to cope with financial stress.

With respect to the third objective, the results of our study underscore the importance of testing for family mediators when examining the association between financial stress and problem behavior in adolescents. In line with family stress studies conducted outside the United States (Aytac & Rankin, 2009; Kinnunen & Feldt, 2004; Kwon et al., 2003), we found that financial stress has both direct and indirect effects on interparental conflicts. In contrast, family stress studies conducted in the United States provide evidence only for indirect effects. Some scholars have attributed this difference to contrasting cultural traditions (e.g., gender role ideology). For example, in a Turkish study by Aytac and Rankin (2009) and a Korean study by Kwon and colleagues (2003), the authors suggest that the traditional gender roles of a society might lead husbands to interpret financial strain as a failure to perform in the male-breadwinner role, thus causing direct marital turmoil regardless of the level of emotional distress (Aytac & Rankin, 2010, p. 763). We believe that this explanation cannot be generalized to the present sample. Findings from the large-scale European Values Study (EVS, 2010) reveal that approximately 82% of all Belgians agree that men should take as much responsibility as women for the home and children, and approximately 91% of Belgian citizens agree that both the husband and wife should contribute to the household income. Another explanation, which is also suggested in a study by Kwon et al., is that we focused on parents with an adolescent target child, whereas other studies focused on parents with younger children (e.g., Conger et al., 2002; Leinonen et al., 2002). The financial demands of raising and educating older children are higher than those associated with younger children, and they may directly lead to marital friction (Kwon et al., 2003). Furthermore, the age of the adolescents might also explain why positive parenting only partially mediated the link between interparental conflicts and problem behavior in adolescents. For adolescents, peers and the experience of group belonging play an important role in protecting them from externalizing problems (Newman, Lohman, & Newman, 2007), and these are even more important than parenting behaviors (Buehler, 2006). Therefore, it is assumable that positive parenting serves only partly as a protective factor for problem behavior in adolescents. It might be interesting for future studies to examine family stress processes in families with children of varying ages.

Although we believe that our study makes a significant contribution to the current body of research on the mediating pathways between family financial stress and problem behavior in adolescents, our findings should be interpreted in light of some limitations. One major limitation of this study stems from the cross-sectional nature of the data: Causal relationships can be inferred only theoretically. However, from a theoretic perspective and based on results from the available longitudinal studies on family stress processes (Kiernan & Huerta, 2008; Linver et al., 2002; Mistry, Lowe, Benner, & Chien, 2008), we can assume that financial stress has an influence on the outcomes of children through the lives of their parents. Second, this study was based on nondivorced families with at least one adolescent in secondary school (i.e., between 11 and 17 years old), thereby excluding single or remarried parents. Although the sample was in many respects heterogeneous, parents with low educational attainment and parents with higher household incomes were underrepresented, which may limit the generalizability of the findings.

Within these limitations, the present study demonstrated successfully that the family stress model applies to a Belgian sample. Our study expands upon previous studies on the extended family stress model by examining actor and partner effects between the depressive symptoms of mothers and fathers, feelings of interparental conflict, and parenting. Although actor effects were more prominent, we also found evidence of partner effects. In addition, we found that the parenting of fathers was more affected by financial stress than was the parenting of mothers. As such, our findings underscore the importance of including all family members in future studies on family stress processes.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded with support from the FWO Methusalem Fund [41/FA040100/FFB2998] and the Research Fund (BOF) of University of Antwerp [41/FA040300/5/5628]. The authors would like to thank Bea Cantillon, PhD, for her support.

References

Achenbach

(1991). Manual for the child behavior checklist. Burlington: University of Vermont, Department of Psychiatry.

Adamsons

Buehler

(2007). Mothering versus fathering versus parenting: Measurement equivalence in parenting measures. Parenting: Science and Practice, 7, 271–303.

Aytac

Rankin

(2009). Economic crisis and marital problems in Turkey: Testing the family stress model. Journal of Marriage and Family, 71, 756–767.

Barnett

M. A.

(2008). Economic disadvantage in complex family systems: Expansion of family stress models. Clinical Child and Family Psychology Review, 11, 145–161.

Barnett

Deng

Mills-Koonce

Willoughby

Cox

(2008). Interdependence of parenting in mothers and fathers of infants. Journal of Family Psychology, 22, 561–573.

Benner

Kim

(2010). Understanding Chinese American adolescents’ developmental outcomes: Insights from the family stress model. Journal of Research on Adolescence, 20, 1–12.

Blau

(1994). Testing a 2-dimensional measure of job search behavior. Organizational Behavior and Human Decision Processes, 59, 288–312.

Bodenmann

Ledermann

Bradbury

(2007). Stress, sex, and satisfaction in marriage. Personal Relationships, 14, 551–569. doi:10.1111/j.1475-6811.2007.00171.x

Buehler

(2006). Parents and peers in relation to early adolescent problem behavior. Journal of Marriage and Family, 68, 109–124. doi:10.1111/j.1741-3737.2006.00237.x

10.

Clark

Senik

(2010). Who compares to whom? The anatomy of income comparisons in Europe. Economic Journal, 120(544), 573–594. doi:10.1111/j.1468-0297.2010.02359.x

11.

Coiro

Emery

(1998). Do marriage problems affect fathering more than mothering? A quantitative and qualitative review. Clinical Child and Family Psychology Review, 1, 23-40.

12.

Conger

(2002). Resilience in Midwestern families: Selected findings from the first decade of a prospective, longitudinal study. Journal of Marriage and Family, 64, 361–373.

13.

Conger

Martin

(2010). Socioeconomic status, family processes, and individual development. Journal of marriage and the family, 72, 685–704. doi:10.1111/j.1741-3737.2010.00725.x

14.

Conger

Donnellan

(2007). An interactionist perspective on the socioeconomic context of human development. Annual Review of Psychology, 58, 175–199.

15.

Conger

Elder

Lorenz

Simons

(1994). Economic-stress, coercive family process, and developmental problems of adolescents. Child Development, 65, 541–561.

16.

Conger

Wallace

Sun

Simons

McLoyd

Brody

(2002). Economic pressure in African American families: A replication and extension of the family stress model. Developmental Psychology, 38, 179–193. doi:10.1037//0012-1649.38.2.179

17.

Cummings

Goeke-Morey

Raymond

(2004). Fathers in family context: Effects of marital quality and marital conflict. In Lamb

(Ed.), The role of the father in child development (4th ed., pp. 196–221). New York, NY: Wiley.

18.

Eichler

Hielscher

(2012). Does the ECB act as a lender of last resort during the subprime lending crisis? Evidence from monetary policy reaction models. Journal of International Money and Finance, 31, 552–568.

19.

European Commission. (2011). Eurobarometer 76. Public opinion in the European Union. http://ec.europa.eu/public_opinion/index_en.htm.

20.

Eurostat. (2008). Survey on income and living conditions (SILC) questionnaire manual 2008. http://www.cso.ie/en/surveysandmethodology.

21.

EVS. (2010). European Values Study 2008: Belgium (EVS 2008): GESIS data archive, Cologne, Germany. ZA4759 Data file Version 1.1.0, doi:10.4232/1.10156.

22.

Falconier

Epstein

(2010). Relationship satisfaction in Argentinean couples under economic strain: Gender differences in a dyadic stress model. Journal of Social and Personal Relationships, 27, 781–799. doi:10.1177/0265407510373260

23.

Falconier

Epstein

(2011). Female-demand/male-withdraw communication in Argentinian couples: A mediating factor between economic strain and relationship distress. Personal Relationships, 18, 586–603. doi:10.1111/j.1475-6811.2010.01326.x

24.

Faugier

Sargeant

(1997). Sampling hard to reach populations. Journal of Advanced Nursing, 26, 790–797.

25.

Garfinkel

Rainwater

Smeeding

(2010). Wealth and Welfare States: Is America a Laggard or Leader? Oxford, England: Oxford University Press.

26.

Haagenars

de Vos

Zaidi

(1994). Poverty statistics in the late 1980s: Research based on micro-data. Luxembourg: Office for Official Publications of the European Communities.

27.

Johnson

O’Leary

(1987). Parental behavior patterns and conduct disorders in girls. Journal of Abnormal Child Psychology, 15, 573–581.

28.

Kalmijn

Liefbroer

(2011). Nonresponse of secondary respondents in multi-actor surveys: Determinants, consequences, and possible remedies. Journal of Family Issues, 32, 735–766. doi:10.1177/0192513x10390184

29.

Kenny

Kashy

Cook

(2006). Dyadic data analysis. New York, NY: Guilford.

30.

Kerig

(1996). Assessing the links between interparental conflict and child adjustment: The conflicts and problem-solving scales. Journal of Family Psychology, 10, 454–473.

31.

Kickert

(2012). State responses to the fiscal crisis: Belgium. Public Money & Management, 32, 303–310.

32.

Kiernan

Huerta

(2008). Economic deprivation, maternal depression, parenting and children’s cognitive and emotional development in early childhood. British Journal of Sociology, 59, 783–806.

33.

Kinnunen

Feldt

(2004). Economic stress and marital adjustment among couples: Analyses at the dyadic level. European Journal of Social Psychology, 34, 519–532. doi:10.1002/ejsp.213

34.

Kohout

Berkman

Evans

Cornoni-Huntley

(1993). Two shorter forms of the CES-D Depression Symptoms Index. Journal of Aging and Health, 5, 179–193.

35.

Kwon

Rueter

Lee

Koh

(2003). Marital relationships following the Korean economic crisis: Applying the family stress model. Journal of Marriage and Family, 65, 316–325. doi:10.1111/j.1741-3737.2003.00316.x

36.

Lamb

Lewis

(2010). The development and significance of father-child relationships in two-parent families. The role of the father in child development. M. Lamb. New Jersey, John Wiley & Sons, Inc.:94-153.

37.

Lavee

McCubbin

Olson

(1987). The effect of stressful life events and transitions on family functioning and well-being. Journal of Marriage and the Family, 49, 857–873. doi:10.2307/351979

38.

Lebrun

(2011). What has been the damage of the financial crisis to Belgian GDP? Working paper 8–11. Brussels, Belgium: Federal Planning Bureau. Economic Analyses and Forecasts.

39.

Ledermann

Macho

(2009). Mediation in dyadic data at the level of the dyads: A structural equation modeling approach. Journal of Family Psychology, 23, 661–670.

40.

Leinonen

Solantaus

Punamaki

(2002). The specific mediating paths between economic hardship and the quality of parenting. International Journal of Behavioral Development, 26, 423–435. doi:10.1080/01650250143000364

41.

Linver

Brooks-Gunn

Kohen

(2002). Family processes as pathways from income to young children’s development. Developmental Psychology, 38, 719–734.

42.

Malmberg

Flouri

(2011). The comparison and interdependence of maternal and paternal influences on young children’s behavior and resilience. Journal of Clinical Child and Adolescent Psychology, 40, 434–444.

43.

McBride

(2001). Relative-income effects on subjective well-being in the cross-section. Journal of Economic Behavior & Organization, 45, 251–278. doi:10.1016/s0167-2681(01)00145-7

44.

Mistry

Biesanz

Taylor

Burchinal

Cox

(2004). Family income and its relation to preschool children’s adjustment for families in the NICHD study of early child care. Developmental Psychology, 40, 727–745.

45.

Mistry

Lowe

Benner

Chien

(2008). Expanding the family economic stress model: Insights from a mixed-methods approach. Journal of Marriage and Family, 70, 196–209.

46.

Muthén

(2010). Mplus user’s guide. Version 6.0. Los Angeles, CA: Author.

47.

Nelson

O’Brien

Blankson

Calkins

Keane

(2009). Family stress and parental responses to children’s negative emotions: Tests of the spillover, crossover, and compensatory hypotheses. Journal of Family Psychology, 23, 671–679.

48.

Newman

Lohman

Newman

(2007). Peer group membership and sense of belonging: Their relationship to adolescent behavior problems. Adolescence, 42(166), 241–263.

49.

Parke

Coltrane

Duffy

Buriel

Dennis

Powers

. . .Widaman

(2004). Economic stress, parenting, and child adjustment in Mexican American and European American families. Child Development, 75, 1632–1656. doi:10.1111/j.1467-8624.2004.00807.x

50.

Patterson

(1982). Coercive family process. A social interactional approach (Vol. 3). Eugene, OR: Castalia.

51.

Ponnet

Mortelmans

Wouters

Van Leeuwen

Bastaits

Pasteels

(2013). Parenting stress and marital relationship as determinants of mothers’ and fathers’ parenting. Personal Relationships, 20, 259–276. doi:10.1111/j.1475-6811.2012.01404.x

52.

Ponnet

Wouters

Mortelmans

Pasteels

De Backer

Van Leeuwen

Van Hiel

(2013). The influence of mothers’ and fathers’ parenting stress and depressive symptoms on own and partner’s parent–child communication. Family Process, 52, 312–324. doi:10.1111/famp.12001

53.

Porter

O’Leary

(1980). Marital discord and childhood behavior problems. Journal of Abnormal Child Psychology, 8, 287–295. doi:10.1007/bf00916376

54.

Solantaus

Leinonen

Punamaki

(2004). Children’s mental health in times of economic recession: Replication and extension of the family economic stress model in Finland. Developmental Psychology, 40, 412–429. doi:10.1037/0012-1649.40.3.412

55.

Van Leeuwen

Vermulst

(2004). Some psychometric properties of the Ghent Parental Behavior Scale. European Journal of Psychological Assessment, 20, 283–298. doi:10.1027/1015-5759.20.4.283

56.

Van Leeuwen

Vermulst

Kroes

De Meyer

Veerman

(2011). De verkorte schaal voor ouderlijk gedrag. VSOG voor ouders van jeugdigen van 4 t/m 18 jaar. [Abbreviated Scale of Parental Behavior: For parents of juveniles between 4 and 18 years old.] Nijmegen, the Netherlands: Katholieke Universiteit Leuven/Praktikon bv.