Family resources as mediators in the relation between divorce and children's school engagement

G Model SOCSCI-1168; No. of Pages 16

ARTICLE IN PRESS The Social Science Journal xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

The Social Science Journal journal homepage: www.elsevier.com/locate/soscij

Family resources as mediators in the relation between divorce and children’s school engagement Nele Havermans, Sarah Botterman ∗ , Koen Matthijs Centre for Sociological Research, Family and Population Studies, University of Leuven, Belgium

a r t i c l e

i n f o

Article history: Received 27 November 2012 Received in revised form 7 April 2014 Accepted 8 April 2014 Available online xxx

Keywords: Family resources Divorce School engagement

a b s t r a c t Children are increasingly growing up in non-intact families. Because the family is a vital developmental part of growing-up, parental divorce can have far-reaching effects on children. This article investigates whether divorce interferes with children’s engagement in school. According to the deprivation perspective, the effect of a parental divorce on children is mediated through the availability of family resources. Structural equation models are performed on the Leuven’s Adolescent and Family Study. We conclude that the parental divorce effect on school engagement is mediated by the parent-child relationship, parental conflict, and financial problems at home. © 2014 Published by Elsevier Inc. on behalf of Western Social Science Association.

1. Introduction Demographic and social evolutions have made family structures more complex and diverse (Kalmijn, 2007). Rising divorce rates have resulted in a declining prevalence of traditional nuclear families (Wu, Hou, & Schimmele, 2008). These societal changes intersect with the day-to-day life experiences of children; as a result, more and more children live in single-parent and stepfamily households. Growing up with divorced parents is related to negative child outcomes, such as lower levels of well-being and lower academic achievement (Amato, 2010; Amato & James, 2010). The effect of divorce on children’s educational outcomes is especially important to examine, as education is one of the most important social cleavages in society (Berlin, Furstenberg, & Waters, 2010). Individuals with a higher educational degree generally have more opportunities in the labor market, whereas those with a lower educational

∗ Corresponding author. Tel.: +32 16 32 31 78. E-mail addresses: [email protected] (N. Havermans), [email protected] (S. Botterman), [email protected] (K. Matthijs).

level often attain a lower occupational status and income level (OECD, 2013). Previous research has primarily focused on the school performance of children after divorce; they generally have lower grades and a higher risk of early school drop-out than children of married parents (Cavanagh, Schiller, & Riegle-Crumb, 2006; Ham, 2003; McLanahan & Percheski, 2008; Sun & Li, 2001). A number of recent studies look into the effects of divorce on children’s non-cognitive educational outcomes (Brown, 2006; Cavanagh et al., 2006; Garg, Melanson, & Levin, 2006; Breivik & Olweus, 2008; Tillman, 2007). Non-cognitive educational outcomes relate to attitudinal, personal, and behavioral qualities at school. There is growing recognition of the importance of these non-cognitive outcomes for children, such as academic achievement, success in higher education, and employability (Heckman & Kautz, 2012; Johnson, Crosnoe, & Elder, 2001). This study focuses on the non-cognitive outcome of school engagement. The concept of school engagement not only covers behavioral aspects, such as school attendance, homework, and participation in class, it also covers more emotional and cognitive aspects, such as interest in school, motivation to study, and development of learning strategies (Dee & West, 2011). School engagement is

http://dx.doi.org/10.1016/j.soscij.2014.04.001 0362-3319/© 2014 Published by Elsevier Inc. on behalf of Western Social Science Association.

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model SOCSCI-1168; No. of Pages 16

ARTICLE IN PRESS N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

2

strongly linked to dropout rates (Finn, 1989), entrance into post-secondary education, and labor market participation in adult life (Finn & Owings, 2006). Furthermore, school engagement can be considered part of children’s well-being (Pollard & Lee, 2003). Looking into the mechanisms that affect children’s school engagement can help researchers design targeted interventions to improve children’s school experiences and decrease drop-out behavior (Fredricks, Blumenfeld, & Paris, 2004). As a consequence, the focus in this study is not only on the association between divorce and children’s school engagement, but also on potential mediators of the negative effects. Some researchers claim parental divorce by itself is the most important and direct cause of negative child outcomes (Kelly & Emery, 2003). Stress and crisis theories argue that during a parental divorce children experience divorce-related stress, which leads to different negative outcomes. According to these theories, the psychological impact of a divorce itself causes lower levels of wellbeing and educational performance (Amato, 2000). There is, however, a need to investigate the mechanisms through which a divorce can affect child outcomes (Amato, 2010; Lansford, 2009). Identifying the processes that mediate the relation between divorce and child outcomes has particular importance for the development of interventions for children, parents, and families as a unit. In what follows, we focus on the resource deprivation perspective, which considers the loss of parental resources after divorce as the most important cause of negative post-divorce child outcomes. The main research question is: “Does the availability of family resources mediate the association between parental divorce and children’s school engagement?” This study differs from previous research in a number of ways. Firstly, the resources deprivation perspective is explicitly tested by means of structural equation modeling. This technique allows us to estimate the indirect effects of divorce through different types of family resources. Also, relationships between different types of family resources are specified. To our knowledge, this has not been investigated in the same manner yet. Secondly, this research is conducted using a broad sample of secondary school pupils in Flanders, Belgium. Belgium has one of the highest crude divorce rates in Europe with 2.9 divorces per 1000 inhabitants in 2011 (Eurostat, 2012). Based on National Registry figures, Lodewijckx (2005) estimates that a minimum of 20% and a maximum of 24% of the 0–17 year olds have experienced a divorce of their parents. The crude divorce rate of Belgium is comparable to the one in the United States (Eurostat, 2012). Belgium combines this high divorce rate with significantly lower levels of school engagement than the OECD average (UNICEF, 2010). 2. Literature review 2.1. Family resources and children’s school engagement The availability of family resources is supposed to be a strong predictor of the educational outcomes of children. This perspective is strongly influenced by sociologists such as Bourdieu (1977) and Coleman (1988). Coleman’s

framework has been the most influential so far largely due to his emphasis on interfamily relations (e.g. McLanahan & Sandefur, 1994; Schlee, Mullis, & Schriner, 2009; Sun & Li, 2009). Following Coleman’s interpretation of family resources, we distinguish between financial, human and social family resources. Financial family resources – the economic means of the family, usually seen as family income – influence children’s school engagement in a number of ways. First of all, financial resources have an impact on physical environment at home. Low-income families tend to have inferior housing (Amato, 2005); a high-quality, safe living environment has been demonstrated to improve learning activities of children (Brooks-Gunn, Klebanov, & Liaw, 1995). Secondly, low income households have fewer financial means to invest in the cognitive stimulation of children. For instance, lowincome families generally have less reading material at home than families with more financial means. Thirdly, family income can affect a child’s health. Poor children tend to have more health problems which can interfere with their engagement and performance in school. Finally, financial problems at home can deteriorate the relationships between family members. These dynamics have an important impact on children’s engagement and performance in school (Hakvoort, Bos, van Balen & Hermanns; Perdue, Manzeske, & Estel, 2009). Human family resources encompass the skills and capabilities acquired by parents, typically their educational level. Parental educational attainment has a strong impact on the cognitive environment in the family (Raviv, Kessenich, & Morrisson, 2004). Higher educated parents tend to stimulate their children more to do well in school, as they comprehend the potential gains of education more than less educated parents (Astone & McLanahan, 1991), and they are more able to help their children with school work (Guo & Harris, 2000; Harper, Marcus, & Moore, 2003). Social family resources, the third type of family resources, refers to the relationships between parents and children and the relationship between parents. These relationships are considered vital for children’s academic success and engagement in school (Bartle-Haring, Younkin, & Day, 2012; Bowen, Rose, Powers, & Glennie, 2008; Brewster & Bowen, 2004; Perdue et al., 2009). Social family resources depend on both the quantity and quality of contact between parents and children. The amount of contact between parents and children is crucial for the transmission of human and financial resources to children, but this is not the only form of social family resources that affect children’s school engagement (Coleman, 1988). The strength of parent-child relationships and parental involvement with their children’s school work influences children’s academic success and engagement in school (Brewster & Bowen, 2004; Perdue et al., 2009; Rumberger, Ghatak, Poulos, Ritter, & Dornbusch, 1990). The quality of the relationship between parents can also influence children’s school engagement. A strained relationship between parents, which mostly results in parental conflict, often coincides with lower parental involvement, less effective parenting, and a worse parent-child relationship (Kalmijn, 2013; Lansford, 2009; Musick & Meier, 2010). Furthermore, parental conflict is highly stressful for children. Through

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model SOCSCI-1168; No. of Pages 16

ARTICLE IN PRESS N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

3

Fig. 1. Conceptual model.

these different pathways, parental conflict can interfere with children’s school engagement (Amato & Loomis, 1994; Lansford, 2009; Musick & Meier, 2010). Financial, human, and social family resources are interconnected by identifiable causal paths between them. Human family resources can be a predictor of financial resources, as higher academic attainment tends to lead to more successful, higher income employment (OECD, 2013). Secondly, families with high financial means generally have less parental conflict and better parent-child relationships (R.D. Conger, K.J. Conger, & Martin, 2010). Economic hardship can put a strain on the relationship between parents, which can have a spill-over effect on the parent-child relationship (Conger et al., 2010). Furthermore, in low-income families, parents tend to adopt less engaged, more ineffective parenting styles (Kiernan & Huerta, 2008; Martinez & Forgatch, 2002) with poorer parent-child relationships on average (Conger et al., 2010; Kalmijn, 2013). 2.2. The resources deprivation perspective The resource deprivation perspective assumes that lower availability of resources in families with divorced parents leads to lower success and engagement in school for children (Amato, 2005; Brown, 2004). Lower resource availability especially harms the financial and social resources in the family because post-divorce families tend to have fewer financial resources. The poverty risk of nonintact families is significantly higher than the poverty risk of families with continuously married parents (Aassve et al., 2007; McManus & DiPrete, 2001). Post-divorce economic hardship can have a negative effect on the outcomes of children. Additionally, children in post-divorce families generally have less contact with one or both parents, which hinders transmission of family resources to the child. Moreover, less contact erodes the parent-child relationship and lowers parental involvement with the child (Astone & McLanahan, 1991; Sun & Li, 2009). Furthermore, divorce rarely ends the existing conflicts between parents. Divorced couples are still at risk of conflict because they still need to make decisions together regarding, among others things, custody arrangements, alimony, and child rearing (Dronkers, 1999; Musick & Meier, 2010).

There are also fewer human resources available in nonintact families. The loss of contact with one or both of the parents lowers the availability of knowledge, skills, and capabilities to the child of divorced parents in general. On the other hand, human resources in the family is also considered a predictor of divorce. Recent studies have shown that less educated parents have a higher divorce risk in Flanders than parents with higher education levels (Härkönen & Dronkers, 2006). This negative educational gradient of divorce can be explained in part by better communication skills and better partner matching among higher educated couples which enhances the stability of the partnership (Härkönen & Dronkers, 2006). 3. Conceptual model The conceptual framework (Fig. 1) stipulates that the effect of divorce on children’s school engagement is mediated by the lower availability of financial and social family resources. For the human family resources, we specify a bidirectional path. Not only is availability of human resources in the family affected by divorce, availability can also function as a predictor of divorce. Consequently, this study makes no causal assumptions. Estimating the indirect effects reveals which path is the most informative for explaining children’s negative post-divorce outcomes. Furthermore, we include the causal patterns between the different types of resources. A first causal pattern concerns the effect of human resources on financial resources, as families with higher educated parents tend to have more financial means than families with lower educated parents (OECD, 2013). The second causal pattern runs from financial to social family resources. Past research shows that families with a lower socioeconomic position tend to have more problematic family relations (Chen & Kaplan, 2001; Conger et al., 2010). 4. Methods 4.1. Data and sample Four data rounds of the Leuven’s Adolescents and Family Study (LAFS, www.soc.kuleuven.be/lago) data are used. Surveys were conducted in 2008, 2009, 2010, and 2011

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model SOCSCI-1168; No. of Pages 16

ARTICLE IN PRESS N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

4

in Flanders by the Family and Population Studies research team of the University of Leuven. A two-phase sampling strategy is implemented. First, schools were selected in two phases by a disproportional quota sampling technique to increase the response rate. In the first phase, sociology master students were asked to contact their former secondary schools to increase the response rate because schools are more helpful toward ex-pupils than toward external people. This procedure, however, also leads to an overrepresentation of private secondary schools which only provide the general educational track, and are only located in certain Flemish provinces. This overrepresentation is corrected by a disproportionate stratified sample in the second phase. The entire Flemish school population is divided in subgroups (strata) on the basis of public versus private, the province versus municipality. Within each stratum, preference is given to schools that provide several educational tracks. The aim of the disproportional stratified sampling technique is a school sample that is representative for school type and educational track (Appendix 1). Pupils from all grades and educational tracks were surveyed in the selected schools: 7,035 pupils in 44 secondary schools responded for a total of about 1% of the Flanders school population. The distribution of gender, year, and educational track in the LAFS data (Appendix 2) strongly resembles the total Flemish pupil population (Vanassche, Sodermans, & Matthijs, 2011). The distribution of the LAFS pupil sample and the Flemish pupil population are presented in Appendix 2. 4.2. Dependent variable 4.2.1. School engagement School engagement is measured via twelve items (Brutsaert, 1993). Children are asked on a five-point scale ranging from ‘totally disagree’ to ‘totally agree’ whether they agree with statements regarding the behavioral, emotional, and cognitive dimensions of school engagement (Fredricks et al., 2004). Behavioral items relate to the conduct and involvement of a student in school, such as making the effort to complete homework and school tasks, and being persistent and eager to learn. Emotional items are linked to school and study, such as showing interest or disinterest in school, and being focused or distracted in class. Cognitive items of school engagement consider the psychological investment in learning, such as looking up additional information on subject material. The exact questions are reported in Appendix 3. 4.3. Independent variables 4.3.1. Parental divorce The marital status of children’s parents is used to construct a dummy variable. Children are asked whether their parents are (1) married and cohabiting; (2) cohabiting, but not married; (3) divorced after marriage; (4) separated after cohabitation; (5) married, but not cohabiting; (6) single, never cohabited. Children with one or both deceased parents are excluded from the analyses. Children whose parents are divorced, separated, or not cohabiting are coded one on the dummy variable “parental divorce.”

Research suggests that the consequences for children when parents separate after cohabitation are very similar divorce, especially within Europe (Amato, 2004). The other categories are coded zero. One quarter of the children (26.50%) experienced a parental divorce. This corresponds with the estimation that between one fifth and one fourth of the children in Flanders above the age of 12 live in a dissolved family (Lodewijckx, 2005). 4.3.2. Family resources We distinguish between three different types of family resources: financial, human, and social. For children of divorce, we consider the resources available in children’s primary residence. If children spend most of the time with one parent, we only take this parent’s resources into account. If children spend considerable time with both parents in co-residence (24.20% of children with divorced parents), we consider the situation of both parents. Financial family resources are measured via a question regarding the frequency of financial problems within the family. The frequency is indicated on a four-point scale (1 = never, 2 = seldom, 3 = sometimes, 4 = regularly). The most problematic financial situation of both parents is considered in case of co-residence. Because we question children and not their parents, it is not possible to include other measurements, such as the family income. The EUSILC data shows that 10% of Flemish children between 0 and 17 years old live below the poverty line. This increases to 17% in reports of financial problems at home (Research Centre of the Flemish Government, 2013). Within our sample, 14% report living in a family that often or always has financial difficulties. This proportion is thus close to official estimations of child poverty. Second, human family resources are measured via the educational level of mother and father. There are three categories: low (no degree of higher secondary school), medium (degree of higher secondary school) and high educational level (degree of higher education). The variables are recoded into two dummy variables, such as low and high educational level with medium educational level as the reference category. Third, social family resources are measured via three indicators: the quality of the relationship between children and their mother; the quality of the relationship between children and their father; and the conflict between parents at the moment of the interview. The quality of the relationship between children and their parents is made operational via the network resource identifier or NRIscale including nine items (Furman & Burmester, 1985). The Cronbach’s alpha is .89 for the NRI-scale with mother and .91 for the NRI-scale with father. Parental conflict is measured via the Conflict Awareness Scale (Grych & Fincham, 1993), which measures the intensity of parental conflict at the moment of the interview. Cronbach’s alpha for this scale is .79. The exact question wording and items are reported in Appendix 3. 4.3.3. Control variables We include gender, age, and educational track as control variables. A dummy variable measures gender: girls are coded one; boys are coded zero. Boys are less often engaged

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model SOCSCI-1168; No. of Pages 16

ARTICLE IN PRESS N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

5

Fig. 2. Confirmatory factor analysis of school engagement. Note: Entries are standardized factor loadings. Marker indicators are presented by◦ . Full arrows denote significant effects (p < .05). Error terms are available upon request.

at school than girls (Mo & Singh, 2008). Age is included as a continuous variable centered around its mean; school engagement often starts to decline in early adolescence (Simons-Morton & Chen, 2009). We also control for educational track. There are three educational tracks in Flanders: general, technical, and vocational. The general track is aimed at academically oriented pupils. Most children start in this track and are directed to the technical and vocational track when they do not perform well enough at school. This may cause pupils in technical and vocational tracks to feel more detached from school (Demanet & Van Houtte, 2012). The descriptive measurements and correlation matrix of the variables are presented in Appendix 4. 4.4. Analytical strategy Structural equation models (SEM) are estimated in two steps. First, measurement models are constructed and tested. Confirmatory factor analyses are performed to measure the latent constructs of school engagement, the quality of the relationship between children and mother, the quality of the relationship between children and father, and the parental conflict. We perform a confirmatory factor analysis in which all constructs are tested simultaneously, the so-called null model. Second, structural models are constructed and tested. Relations between the observed variables and the latent constructs are specified. Full Information Maximum Likelihood (FIML) estimations are produced to deal with missing values. FIML assumes that the missing elements are completely or at least partially random with a multivariate normal distribution (Brown, 2010). For each observation, a function is estimated using those variables that have a non-missing value. These functions are accumulated and maximized across the entire sample (Arbuckle, 1996). Monte Carlo simulations show that, all else equal, FIML produces unbiased and more efficient estimations than other missing data methods, such as pairwise deletion and multiple imputations (Enders & Bandalos, 2001).

The models are evaluated based on several fit indices. Because we have a rather large sample size, we do not rely on the 2 and its p-value. Instead, we study the more appropriate Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR). The CFI and TLI are two comparative fit indices. They can take on each value between zero and one. In general, values of .90 and higher are considered as indicators of a good model fit (Byrne, 2011). The RMSEA index penalizes for poor model parsimony and takes the error of approximation in the population into account. Values smaller than .08 suggest an adequate model fit and values smaller than .05 a good model fit (Browne & Cudeck, 1993). The SRMR assesses the difference between the implied correlation matrix and the correlation matrix in the sample. Values close to or below .08 indicates a good model fit (Brown, 2006). In case of an insufficient model fit, modification indices and parameter estimates are assessed to modify the model. 5. Results 5.1. Measurement model We conduct confirmatory factor analyses for every latent construct. A first confirmatory factor analysis is performed for the items of school engagement. After inspection of the correlation matrix (Appendix 4), item 6 (‘not important for future’) is excluded from the analysis because of a low correlation, less than .25, with six out of the eleven other items. Based on the modification indices, the constraint on the correlation between the error terms of three item pairs is freed. The standardized factor loadings of the final model are all larger than .400 and significant (p < .01). The fit indices also indicate a good model fit. The final factor solution is presented in Fig. 2. A second confirmatory factor analysis is conducted for the nine items measuring the quality of the relationship between children and their fathers, and for the nine items

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model SOCSCI-1168; No. of Pages 16

6

ARTICLE IN PRESS N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

Fig. 3. Confirmatory factor analysis of the quality of the relationship with mother and father. Note: Entries are standardized factor loadings. Marker indicators are presented by◦ . Full arrows denote significant effects (p < .05). Error terms are available upon request.

measuring the quality of the relationship between children and their mothers. After examination of the modification indices, error correlations between equivalent items of both scales are included. The two latent constructs are measured via successive quasi-identical items; therefore, children have the tendency to give highly similar answers to these questions. This is referred to as a method effect (Brown, 2006). After freeing the constraint on the correlations between these error terms, the model fit is sufficient (CFI = 931; TLI = 916; RMSEA = 087; SRMR = 060). The standardized loadings of the items are higher than .400 and significant (p < .01). The factor solution for both latent constructs is presented in Fig. 3. A third confirmatory factor analysis is conducted for parental conflict. Because this scale only consists of three items, the model is just-specified (CFI = 1.000; TLI = 1.000; RMSEA = .000; SRMR = .000). The standardized loadings of the items are higher than .400 and significant (p < .01). The factor solution is presented in Fig. 4. In the null model, the measurement models of all latent constructs – school engagement, the quality of the relationship between children and mother, the quality of the relationship between children and father, and parental conflict – are included simultaneously and allowed to correlate with each other. The fit indices indicate this measurement model is well-specified (CFI = .923; TLI = .914; RMSEA = .055; SRMR = .048). Furthermore, all standardized factor loadings are higher than .400 and significant (p < .01). After determining the measurement of the

latent constructs, we specify the relations between the observed variables and latent constructs and examine a structural model in which the effect of a parental divorce on school engagement is partially mediated through family resources. In this manner, we analyze the direct and indirect effects of parental divorce on school engagement mediated through the family resources. Not all indices meet the criteria to conclude that this is a good model fit (CFI = .887; TLI = .877; RMSEA = .054; SRMR = .096). Nevertheless, we proceed with this model because it corresponds the most to the theoretical framework of the resources perspective and because all fit indices are very close to the cut-off criteria. 5.2. Direct relations between parental divorce and family resources The direct relations between divorce and family resources are presented in Table 1. Parental divorce has a negative significant association on financial and social family resources. Children who live in non-intact families experience more financial problems at home and live in families with more strained relations. Children with divorced parents generally have lower educated parents than children who live in intact families. Furthermore, the results show that financial problems in the family are also negatively related to the parent-child relationship and parental conflict. The effects of parental divorce and financial problems are particularly strong for parental conflict.

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model SOCSCI-1168; No. of Pages 16

ARTICLE IN PRESS N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

7

Fig. 4. Confirmatory factor analysis of parental conflict. Note: Entries are standardized factor loadings. Marker indicators are presented by◦ . Full arrows denote significant effects (p < .05). Error terms are available upon request.

Parents’ educational level is significantly related to financial problems: a high educational level of mother and father is related to fewer financial problems in comparison to a medium educational level. Also, there is a significant correlation between parents’ educational level and having divorced parents. Children with higher educated parents are less likely to have divorced parents than children with medium educated parents. The opposite holds true for children with low educated parents: they are more likely to have divorced parents than children whose parents have a medium educational level.

and children and among parents themselves, feel more engaged in school. Financial and human family resources, and parental divorce are not directly related to school engagement. With regard to the control variables, girls, younger children, and children in the general and technical educational tracks are more engaged in school than boys, adolescents, and children in vocational tracks. 5.4. Indirect relations between parental divorce, family resources, and school engagement Indirect relations between parental divorce, financial and human family resources, and school engagement are presented in Table 3. Children with divorced parents have significantly worse relationships with their parents and a higher risk of disengagement from school. Additionally, children with divorced parents report a significantly higher frequency of conflict between their parents. The association with parental conflict is the strongest indirect pathway to

5.3. Direct relations between divorce, family resources and school engagement The three indicators of social family resources have strong and significant positive associations with school engagement (Table 2). Children who live in families with good interpersonal relationships between parents

Table 1 Unstandardized and standardized estimates, standard errors and significance levels for structural equation model of family resources. Estimate Dependent variable: relation with mother Parental divorcea −0.080 Financial problems −0.118 Dependent variable: relation with father a −0.454 Parental divorce Financial problems −0.168 Dependent variable: parental conflict 0.209 Parental divorcea Financial problems 0.632 Dependent variable: financial problems a 0.578 Parental divorce Educational level motherb Low 0.159 High −0.167 Educational level fatherb Low 0.025 −0.136 High Covariance with divorce Educational level motherb Low 0.003 High −0.019 Educational level fatherb Low 0.004 High −0.022

S.E.

Standardized estimate

Sign.

0.022 0.012

−0.048 −0.141

***

0.025 0.013

−0.236 −0.173

***

0.012 0.024

0.398 0.261

***

0.024

0.291

***

0.049 0.027

0.049 −0.095

***

0.045 0.027

0.008 −0.078

0.002 0.003

0.027 −0.085

*

0.002 0.003

0.034 −0.102

**

***

***

***

***

***

***

***

Source: LAFS 2008–2011. a Parental divorce: 0 = no, 1 = yes. b Educational level: reference category = medium. * p < .05. ** p < .01. *** p < .001.

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model

ARTICLE IN PRESS

SOCSCI-1168; No. of Pages 16

N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

8

Table 2 Unstandardized and standardized estimates, standard errors and significance levels for direct relations in the structural equation model of school engagement. Estimate Parental divorce Family resources Financial problems Educational level mother Low High Educational level father Low High Relation with mother Relation with father Parental conflict Control variables Age Gender Educational track General Technical

S.E.

Standardized estimate

0.023

0.022

0.017

0.011

0.011

0.015

0.036 −0.030

0.035 0.021

0.016 −0.024

−0.001 0.026 0.187 0.067 −0.096

0.034 0.020 0.013 0.012 0.017

−0.001 0.021 0.224 0.092 −0.109

−0.012 −0.141

0.004 0.016

−0.037 −0.115

0.128 0.079

0.021 0.021

0.103 0.059

Sign.

*** *** ***

** ***

*** ***

Source: LAFS 2008–2011. ** p < .01. *** p < .001.

Table 3 Unstandardized and standardized estimates, standard errors and significance levels for parental divorce’s indirect relations in the structural equation model of school engagement. Estimate

S.E.

Standardized estimate

Parental divorce via Financial problems Relation with mother Relation with father Parental conflict Financial problems and relation with mother Financial problems and relation with father Financial problems and parental conflict

0.006 −0.015 −0.030 −0.060 −0.013 −0.006 −0.012

0.006 0.004 0.006 0.011 0.002 0.001 0.002

0.004 −0.011 −0.022 −0.043 −0.009 −0.005 −0.008

Total of indirect effects

−0.130

0.014

−0.094

Sign.

a

*** *** *** *** *** *** ***

Source: LAFS 2008–2011. a Parental divorce: 0 = no, 1 = yes. *** p < .001.

school engagement. The mediating effect of financial family resources runs through social family resources. Children with divorced parents perceive more financial problems at home, which is related to more problematic relationships between parents, and between the child and the parent. Lower availability of social family resources is negatively associated with children’s school engagement. The indirect relations between family resources and school engagement are presented in Table 4. Financial problems have a significant negative relation with children’s school engagement via the availability of social family resources. The educational level of mother and father is also indirectly related to school engagement, but these estimates are smaller than the indirect relation between financial problems and school engagement. The results show that having a highly educated mother is positively related to school engagement in an indirect manner, in comparison to having a medium educated mother. Having a low educated mother is indirectly related to lower school engagement compared to having a medium

educated mother. These effects are mediated by financial resources and social family resources. Only the indirect relation via financial problems is insignificant. The father’s educational level is significant only in relation to a highly educated father and strong school engagement versus a medium educated father. This factor is also mediated by financial problems and the three measures of social family resources. 6. Discussion The aim of this article is to disentangle the effects of family life on children’s school engagement. More precisely, the focus is on the effect of parental divorce on school engagement, taking into account the mediating effect of family resources in this relationship. We find evidence that the negative relation between divorce and children’s school engagement runs through family resources. This confirms the findings of previous research, though this is the first research using structural equation

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model

ARTICLE IN PRESS

SOCSCI-1168; No. of Pages 16

N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

9

Table 4 Unstandardized and standardized estimates, standard errors and significance levels for family resources’ indirect relations in the structural equation model of school engagement.

Financial problems via Relation with mother Relation with father Parental conflict Total of indirect effects High educational level of mothera via Financial problems Financial problems and relation with mother Financial problems and relation with father Financial problems and parental conflict Total of indirect effects Low educational level of mothera via Financial problems Financial problems and relation with mother Financial problems and relation with father Financial problems and parental conflict Total of indirect effects Low educational level of fathera via Financial problems Financial problems and relation with mother Financial problems and relation with father Financial problems and parental conflict Total of indirect effects High educational level of fathera via Financial problems Financial problems and relation with mother Financial problems and relation with father Financial problems and parental conflict Total of indirect effects

Estimate

S.E.

Standardized estimate

Sign.

−0.022 −0.011 −0.020 −0.053

0.003 0.002 0.004 0.005

−0.032 −0.016 −0.028 −0.076

***

−0.002 0.004 0.002 0.003 0.007

0.002 0.001 0.000 0.001 0.002

−0.001 0.003 0.002 0.003 0.006

0.002 −0.004 −0.002 −0.003 −0.007

0.002 0.001 0.001 0.001 0.003

0.001 −0.002 −0.001 −0.001 −0.003

0.000 −0.001 0.000 0.000 −0.001

0.001 0.001 0.000 0.001 0.002

0.000 0.000 0.000 0.000 0.000

−0.001 0.003 0.002 0.003 0.006

0.002 0.001 0.000 0.001 0.002

−0.001 0.002 0.001 0.002 0.005

*** *** ***

*** *** *** ***

** ** ** *

*** *** *** **

Source: LAFS 2008–2011. a Educational level: reference category = medium. * p < .05. ** p < .01. *** p < .001.

models to test the indirect effect of divorce and takes into account the indirect paths between family resources. Lower availability of financial, human, and social resources in non-intact families largely explains why children with divorced parents experience more problems at school. These children appear to have fewer available family resources. The indirect effect of a parental divorce on children’s school engagement runs mostly through the social resources in the family. The decline in the quality of relationships between parents and children, and high levels of parental conflict appear to be the most important mediators of school engagement. These relations are also mediated by a two-step indirect relation: the lower availability of financial resources in non-intact families is related to social family resources, which is in turn related to lower school engagement of children. The present study confirms previous findings and contributes additional evidence that suggests that the indirect paths between family resources are important to explain the consequences of divorce on children. The generalizability of these results is subject to some limitations. First of all, human and financial family resources are measured via single indicators, respectively the educational level of parents and the presence of financial problems at home. Because children are surveyed and not their parents, data is unavailable regarding other socio-economic indicators, such as the family income or

the occupational status of parents. It is possible that the effect of a parental divorce on school engagement would diminish when accounting for these other indicators of family resources. A second caveat relates to the crosssectional character of the LAFS data. No causal relations can be inferred using cross-sectional data and, therefore, the direction of the relations may be reversed. Based on the literature, we assume that the specified direction of the relations is the most theoretically appropriate. Thirdly, we focus on the effect of family resources on children. Other processes, such as divorce-related stress or other psychological processes, are left out of the analysis, though they may manifest as part of the so-called direct effect of parental divorce. Consequently, interpreting the direct relation between divorce and children’s school engagement is very likely confounded with unmeasured indirect processes. A fourth limitation concerns the transferability of the results to other countries. There is very little European research comparing determinants of school engagement between countries. A study on school belonging and participation shows that the impact of family socioeconomic status and family dissolution on school engagement is significant for almost all European countries (Willms, 2003). Parental involvement, which is closely related to parent-child relationship (Simpkins, Weiss, McCartney, Kreider & Dearing) and parenting style influences educational outcomes significantly in almost all

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model

ARTICLE IN PRESS

SOCSCI-1168; No. of Pages 16

N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

10

European countries (Borgonovi & Montt, 2012). Although these are indications that the results of this study are not country-specific and may be transferred to other developed countries, more cross-national research is necessary to get a good grasp on the generalizability of the Flemish study. This research provokes several questions in need of further investigation. A possible direction is to include resources of several socialization agents when studying school engagement. In this study, focus was placed on the family resources, whereas peers, school, and neighborhood resources may come into play as well. A second possible direction for future research is the influence of post-divorce family configurations on children’s school engagement. For children in non-intact families the availability of family resources can be influenced by custody arrangements or the arrival of a stepparent in the family. Investigating heterogeneity within the group of children with divorced parents by including the presence of a stepparent and contact with the non-residential parent into the study can provide valuable additional information on divorce-related child outcomes. The research results can be used to develop targeted interventions aimed at remedying school disengagement. One policy approach is to tackle school disengagement by emphasizing relational family resources, as a supplement to the more tangible family resources that indicate socioeconomic inequalities between families. School projects directed at students with problematic family relations should be encouraged. For instance, the Children of Divorce Intervention Program (Pedro-Carroll, 2005) aims at exchanging experiences related to problematic family situations. Quasi-experimental and experimental studies have demonstrated that these group sessions increase children’s coping skills and subsequently produce positive outcomes (Armistead et al., 1990; Pedro-Carroll, 2005). A different policy approach is directed at parents. Marriage counseling and divorce mediation programs can decrease

conflict between parents, which is demonstrably harmful to children’s school engagement (Hawkins, Blanchard, Baldwin, & Fawcett, 2008). Thirdly, policy makers can also opt to work with the family as a unit. This can improve the parent-child relationships in non-intact families which are, according to the results of this study, crucial for children’s educational outcomes (Cornett & Bratton, 2013; Miller & McLeod, 2001). For instance, the New Beginnings Program (NBP) is aimed at improving mother-child relationship and parenting of single mothers by working with both children and mothers. Experimental studies on the NBP data indicates that families who participate in this program report better mother-child relationships, more disciplined parenting, and better child outcomes (Sigal, Wolchik, Tein, & Sandle, 2012; Zhou, Sandler, Millsap, Wolchik & Dawson-McClure 2008). An important limitation with this program is the absence of fathers; the results of this study indicate the lower quality of the father-child relationship has especially negative consequences for children. To summarize, this study shows that parental divorce is not entirely to blame for negative school outcomes. Social family resources appear to be more important for children’s school engagement. This finding supports Coleman (1988), who claimed that social family resources are a necessary prerequisite for children to benefit from the human and financial resources in the family. Financial means and the educational level of parents are not directly related to school engagement in the analyses when controlling for social family resources; rather, these relations run through the quality of the parent-child relationship and the frequency of parental conflict.

Appendix 1. Lafs school population Tables A.1 and A.2.

Table A.1 The school and pupil population in Flanders and LAFS according to school type (%). Flanders

School type Public school system Subsidized private schools Subsidized public schools

LAFS

Schools (N = 960)

Pupils (N = 429,745)

Schools (N = 44)

Pupils (N = 7,035)

23 70 7

17 75 8

27 66 7

25 67 8

Source: Vanassche et al. (2011). Table A.2 The school and pupil population in Flanders and LAFS according to school type and educational track (%).

School type Public school system Subsidized private schools Subsidized public schools

Flanders (N = 290,517)

LAFS (N = 4,803)

Educational type

Educational type

General

Technical/arts

Vocational

General

Technical/arts

Vocational

38 43 12

26 34 44

36 22 44

37 56 21

30 31 40

33 13 21

Source: Vanassche et al. (2011). Notes: The proportions are calculated a limited sample of pupils in the third to sixth year of secondary school.

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model

ARTICLE IN PRESS

SOCSCI-1168; No. of Pages 16

N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

11

Table A.3 The pupil population in Flanders and LAFS: gender, educational track and year (%). Year

Educational track

Boys

Girls

Total

Boys

Girls

Total

First and second

General, technical and arts track Vocational track Total

13.1 3.2 16.3

13.4 2.5 15.9

26.5 5.7 32.2

11.0 1.5 12.5

11.6 2.1 13.7

22.6 3.6 26.2

Third and fourth

General Technical Arts Vocational Total

6.7 5.8 0.2 4.0 16.7

7.9 4.3 0.4 3.3 15.9

14.6 10.1 0.7 7.3 32.6

9.3 6.1 0.4 2.6 18.3

9.9 5.2 0.9 4.2 20.2

19.2 11.3 1.3 6.7 38.6

Fifth, sixth and seventh

General Technical Arts Vocational Total

5.6 6.5 0.3 5.4 17.8

7.1 5.1 0.5 4.7 17.4

12.7 11.7 0.8 10.1 35.2

6.2 4.9 0.3 3.6 14.9

9.0 5.2 0.8 5.4 20.3

15.2 10.0 1.0 9.0 35.3

50.8

49.1

100.0

45.7

54.3

100.0

Total

Flanders (N = 425,316)

LAFS (N = 6,919)

Source: Vanassche et al. (2011).

Appendix 2. LAFS pupil population Table A.3. Appendix 3. Question wording of scales Question wording of school engagement items Please give your own opinion on the following positions regarding your study. Do you totally disagree, disagree, are you uncertain, do you agree, do you totally agree? 1. The more I learn, the more eager I get to learn even more (curiosity) 2. When I get home, I mostly feel like I have learned something (knowledge improvement) 3. Studying mostly seems like a waste of time (loss of time) 4. There are many things in life, which I feel are more important than studying (not that important) 5. It seems to me that studying is important, because I feel it helps my development (development) 6. I don’t understand why studying is of any importance for the things I want to do later in my life (not important for future) 7. Studying in itself is a waste of time (useless) 8. I do not like to study (don’t like to study) 9. When I don’t immediately find a solution for a problem relating to my study, I keep on searching until I find a solution (persistent looking for solution) 10. I like to search for additional information on school topics (extra documentation) 11. I often daydream in class (daydreaming) 12. If I would know in advance that a subject would not be examined, I would not study for it (study if necessary for test)

Answer next questions about your biological (or adoptive) mother and your biological (or adoptive) father. [Answer categories are: not or little; a little bit, much, very much, most possible] 1. Does your mother respect you? 2. And your father? 3. How often do you spend time with, make fun with your mother? 4. And your father? 5. Do you share secrets and feelings with your mother? 6. And your father? 7. How much do you care about your mother? 8. And your father? 9. Does your mother care about you? 10. And your father? 11. Does your mother appreciate what you do? 12. And your father? 13. Does your mother think it is worth talking to you? 14. And your father? 15. Does your mother think you have good ideas? 16. And your father? 17. Does your mother think she can learn from you? 18. And your father? Question wording of parental conflict items Answer next questions about the level of conflict between your parents, according to your current situation. [answer categories are: never; seldom; sometimes; often; continuously] 1. How often do your parents fight or argue about money? 2. How often do your parents absolutely disagree with each other? 3. How often do your parents have severe conflicts? Appendix 4. Descriptive values and correlations

Question wording of relationship quality between child and mother/father items

Tables A.4–A.9.

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model SOCSCI-1168; No. of Pages 16

12

ARTICLE IN PRESS N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

Table A.4 Descriptive measurements: mean/proportion, standard deviation, range and proportion missings. Mean/proportion Dependent variable school engagement 1. Curiosity 2. Knowledge improvement 3. Loss of time 4. Not that important 5. Development 6. Not important for future 7. Useless 8. Don’t like to study 9. Persistent looking for solution 10. Extra documentation 11. Daydreaming 12. Study if necessary for test Control variables Gendera Age Educational track General Technical Vocational Parental divorceb Financial family resources Human family resources (mother) Low educational level mother Medium educational level mother High educational level mother Human family resources (father) Low educational level father Medium educational level father High educational level father Social resources Parental conflict 1. Fight money 2. Disagree 3. Severe conflict Relationship with mother 1. Respect 2. Spend time, have fun 3. Share secrets 4. Care for mother 5. Care for child 6. Appreciate 7. Worth talking to 8. Good ideas 9. Learn from Relationship with father 1. Respect 2. Spend time, have fun 3. Share secrets 4. Care for father 5. Care for child 6. Appreciate 7. Worth talking to 8. Good ideas 9. Learn from

2.65 2.86 2.59 3.58 3.31 1.98 2.29 3.61 3.17 2.34 3.28 3.68 .46 15.13

SD 1.09 1.03 1.10 1.07 0.97 1.00 1.07 1.16 1.02 1.01 1.11 1.07

1.94 0.01

.45 .30 .24

Proportion missings

1–5 1–5 1–5 1–5 1–5 1–5 1–5 1–5 1–5 1–5 1–5 1–5

0.01 0.01 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02

0–1 11–25

0.00 0.00

0–1 0–1 0–1

.27 1.70

Range

.88

0–1

0.03

1–4

0.14 0.11

.07 .31 .50

0–1 0–1 0–1

.08 .34 .47

0–1 0–1 0–1

0.12

1.92 2.63 2.10

1.05 1.04 1.08

1–5 1–5 1–5

0.15 0.16 0.16

3.72 3.00 2.34 4.39 4.27 3.57 3.79 3.30 2.75

1.00 1.08 1.25 0.87 0.93 1.01 1.05 1.02 1.10

1–5 1–5 1–5 1–5 1–5 1–5 1–5 1–5 1–5

0.01 0.01 0.01 0.02 0.02 0.02 0.02 0.02 0.02

3.52 2.77 1.83 4.17 4.07 3.39 3.45 3.12 2.52

1.10 1.13 1.02 1.07 1.10 1.10 1.17 1.09 1.12

1–5 1–5 1–5 1–5 1–5 1–5 1–5 1–5 1–5

0.04 0.03 0.03 0.04 0.04 0.04 0.04 0.04 0.04

Source: LAFS 2008–2011. a Gender: 0 = girls, 1 = boys. b Parental divorce: 0 = no, 1 = yes.

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

General track

−.02

.02 −.34***

Technical track −.06 .16*** −.60***

Vocational track .04*** .23** −.52*** −.38***

Parental divorce −.05*** .04** −.13*** .11*** .04**

Financial problems −.10*** .08*** −.13*** .09*** .06*** .30***

Education mother: low −.03* .03* −.12*** .11*** .02 .03* .10***

Education mother: medium −.04** .10*** −.18*** .13*** .07*** .07*** .13*** −.22***

Education mother: high .06** −.11*** .24*** −.19*** −.08*** −.09*** −.17*** −.34*** −.85***

Education father: low

Education father: medium −.03* .09*** −.14*** .09*** .07*** .08*** .12*** .00 .41*** −.40*** −.24***

Extra documentation

Daydreaming

.04** −.08*** .20*** −.14*** −.08*** −.10*** −.16*** −.22*** −.41*** .51*** −.33*** −.84***

Source: LAFS 2008–2011. * p < .05. ** p < .01. *** p < .001.

Table A.6 Correlation matrix for school engagement items.

Curiosity Knowledge improvement Loss of time Not that important Development Not important for future Useless Don’t like to study Persistent looking for solution Extra documentation Daydreaming

Knowledge improvement

Loss of time

Not that important

.50***

−.33*** −.35***

−.28*** −.28*** .49***

Development

.44*** .40*** −.43*** −.35***

Not important for future −.18*** −.16*** .45*** .28*** −.38***

Useless

Don’t like to study

−.27*** −.26*** .58*** .36*** −.40*** .57***

−.43*** −.36*** .49*** .42*** −.38*** .31*** .42***

Persistent looking for solution .35*** .36*** −.27*** −.22*** .35*** −.20*** −.24*** −.27***

.42*** .38*** −.22*** −.27*** .32*** −.04*** −.16*** −.33*** .39***

−.24*** −.32*** .36*** .36*** −.27*** .22*** .29*** .40*** −.28*** −.25***

Study if necessary for test −.26*** −.33*** .36*** .35*** −.25*** .15*** .27*** .42*** −.24*** −.33*** .41***

Source: LAFS 2008–2011. Notes: *p < .05; **p < .01. *** p < .001.

ARTICLE IN PRESS

−.01 −.00 −.11*** .10*** .02 .03* .08***  .40** .02 −.23***

Education father: high

G Model

SOCSCI-1168; No. of Pages 16

Gender Age General track Technical track Vocational track Parental divorce Financial problems Education mother: low Education mother: medium Education mother: high Education father: low Education father: medium

Age

N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx 13

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

Table A.5 Correlation matrix for directly observed variables.

G Model

ARTICLE IN PRESS

SOCSCI-1168; No. of Pages 16

N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

14

Table A.7 Correlation matrix for parental conflict items.

Conflict money Disagree

Disagree

Severe conflict

.53***

.53*** .68***

Source: LAFS 2008–2011. Notes: *p < .05; **p < .01. *** p < .001. Table A.8 Correlation matrix for mother-child relationship items.

Respect Spend time, have fun Share secrets Care for mother Care for child Appreciate Worth talking to Good ideas

Spend time, have fun

Share secrets

Care for mother

Care for child

Appreciate

Worth talking to

Good ideas

Learn from

.55***

.40*** .55***

.59*** .46*** .34***

.59*** .51*** .39*** .75***

.60*** .51*** .40*** .54*** .55***

.58*** .50*** .40*** .59*** .56*** .63***

.51*** .48*** .37*** .45*** .45*** .59*** .61***

.41*** .42*** .34*** .33*** .34*** .48*** .46*** .58***

Source: LAFS 2008–2011. Notes: *p < .05; **p < .01. *** p < .001. Table A.9 Correlation matrix for father-child relationship items.

Respect Spend time, have fun Share secrets Care for father Care for child Appreciate Worth talking to Good ideas

Spend time, have fun

Share secrets

Care for mother

Care for child

Appreciate

Worth talking to

Good ideas

Learn from

.59***

.40*** .52***

.67*** .51*** .33***

.67*** .54*** .35*** .81***

.65*** .55*** .41*** .63*** .62***

.64*** .55*** .44*** .64*** .62*** .68***

.56*** .51*** .41*** .54*** .53*** .65*** .65***

.44*** .45*** .39*** .38*** .38*** .49*** .48*** .58***

Source: LAFS 2008–2011. Notes: *p < .05; **p < .01. *** p < .001.

References Aassve, A., Betti, G., Mazzuco, S., & Mencarini, L. (2007). Marital disruption and economic well-being: A comparative analysis. ISER Working Paper 2006-07. Colchester: University of Essex. Amato, P. (2000). The consequences of divorce for adults and children. Journal of Marriage and Family, 62(4), 1269–1287. Amato, P. R. (2004). Reconciling divergent perspectives: Judith Wallerstein, quantitative family research, and children of divorce. Family Relations, 52(4), 332–339. Amato, P. R. (2005). The impact of family formation change on the cognitive, social, and emotional well-being of the next generation. Future of Children, 15(2), 75–96. Amato, P. R. (2010). Research on divorce: Continuing trends and new developments. Journal of Marriage and Family, 72(3), 650–666. Amato, P. R., & James, S. (2010). Divorce in Europe and the United States: Commonalities and differences across nations. Family Science, 1(1), 2–13. Amato, P., & Loomis, L. (1994). Parental divorce, marital conflict, and offspring well-being during early adulthood. Social Forces, 73(3), 895–916. Arbuckle, J. L. (1996). Full information estimation in the presence of incomplete data. In G. A. Marcoulides, & R. E. Schumacker (Eds.), Advanced structural equation modeling: Issues and techniques. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Armistead, L., McCombs, A., Forehand, R., Wierson, M., Long, N., & Fauber, R. (1990). Coping with divorce: A study of young adolescents. Journal of Clinical Child Psychology, 19, 79–84.

Astone, N. M., & McLanahan, S. S. (1991). Family structure, parental practices and high school completion. American Sociological Review, 56(3), 309–320. Bartle-Haring, S., Younkin, F. L., & Day, R. (2012). Family distance regulation and school engagement in middle-school-aged children. Family Relations, 61(2), 192–206. Berlin, G., Furstenberg, F. F., & Waters, M. C. (2010). Introducing the issue. Future of Children, 20(1), 3–18. Borgonovi, F., & Montt, G. (2012). Parental Involvement in Selected PISA Countries and Economies [OECD Education Working Papers No 73]. Paris: OECD. Bourdieu, P. (1977). Cultural reproduction and social reproduction. In A. H. Halsey, & J. Kaubel (Eds.), Power and ideology in education (pp. 487–511). New York: Oxford University Press. Bowen, G. L., Rose, R. A., Powers, J. D., & Glennie, E. J. (2008). The joint effects of neighborhoods, schools, peers, and families on changes in the school success of middle school students. Family Relations, 57(4), 504–516. Breivik, K., & Olweus, D. (2008). Adolescent’ s adjustment in four postdivorce family structures. Journal of Divorce & Remarriage, 44(3), 99–124. Brewster, A. B., & Bowen, G. L. (2004). Teacher support and the school engagement of Latino middle and high school students at risk of school failure. Child and Adolescent Social Work Journal, 21(1), 47–67. Brooks-Gunn, J., Klebanov, P. K., & Liaw, F. (1995). The learning, physical, and emotional environment of the home in the context of poverty: The

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model SOCSCI-1168; No. of Pages 16

ARTICLE IN PRESS N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

Infant Health and Development Program. Children and Youth Services Review, 17(1-2), 251–276. Brown, S. L. (2004). Family structure and child well-being: The significance of parental cohabitation. Journal of Marriage and Family, 66(2), 351–367. Brown, S. L. (2006). Family structure transitions and adolescent wellbeing. Demography, 43(3), 447–461. Brown, S. L. (2010). Marriage and child well-being: Research and policy perspectives. Journal of Marriage and the Family, 72(5), 1059–1077. Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen, & J. S. Long (Eds.), Testing structural equation models (pp. 136–162). Beverly Hills, CA: Sage. Brutsaert, H. (1993). (School, family and well-being: Sixth-graders and their social environment) School, gezin en welbevinden: zesdeklassers en hun sociale omgeving. Leuven/Apeldoorn: Garant. Byrne, M. (2011). Structural equation modeling with Mplus: Basic concepts, applications, and programming. London: Routledge. Cavanagh, S., Schiller, K. S., & Riegle-Crumb, C. (2006). Marital transitions, parenting, and schooling: Exploring the link between family-structure history and adolescents’ academic status. Sociology of Education, 79(4), 329–354. Chen, Z., & Kaplan, H. (2001). Intergenerational transmission of constructive parenting. Journal of Marriage and Family, 63(1), 17–31. Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94(1), 95–120. Conger, R. D., Conger, K. J., & Martin, M. J. (2010). Socioeconomic status, family processes, and individual development. Journal of Marriage and Family, 72(3), 685–704. Cornett, N., & Bratton, S. C. (2013). Examining the impact of Child Parent Relationship Therapy (CPRT) on family functioning. Journal of Marital and Family Therapy (forthcoming) Dee, T. S., & West, M. R. (2011). The non-cognitive returns to class size. Educational Evaluation and Policy Analysis, 33(1), 23–46. Demanet, J., & Van Houtte, M. (2012). School belonging and school misconduct: The differing role of teacher and peer attachment. Journal of Youth and Adolescence, 41(4), 499–514. Dronkers, J. (1999). The effects of parental conflicts and divorce on the well-being of pupils in Dutch secondary education. European Sociological Review, 15(2), 195–212. Enders, C., & Bandalos, D. (2001). The relative performance of full information maximum likelihood estimation for missing data in structural equation models. Structural Equation Modeling: A Multidisciplinary Journal, 8(3), 430–457. Eurostat. (2012). Marriage and divorce statistics. Retrieved from http://epp.eurostat.ec.europa.eu/statistics explained/index.php/ Marriage and divorce statistics# Finn, J. D. (1989). Withdrawing from school. Review of Educational Research, 59(2), 117–142. Finn, J. D., & Owings, J. (2006). The adult lives of at-risk students: The roles of attainment and engagement in high school: Statistical analysis. Washington D.C.: National Center for Education Statistics. Fredricks, J. A., Blumenfeld, P. C., & Paris, A. H. (2004). School engagement: Potential of the concept, state of the evidence. Review of Educational Research, 74(1), 59–109. Furman, W., & Burmester, D. (1985). Children’s perceptions of the personal relationships in their social network. Developmental Psychology, 21, 1016–1024. Garg, R., Melanson, S., & Levin, E. (2006). Educational aspirations of male and female adolescents from single-parent and two biological parent families: A comparison of influential factors. Journal of Youth and Adolescence, 36(8), 1010–1023. http://dx.doi.org/10.1007/s10964-006-9137-3 Grych, J. H., & Fincham, F. D. (1993). Children’s appraisals of marital conflict: Initial investigations of the cognitive-contextual framework. Child Development, 64, 215–230. Guo, G., & Harris, K. M. (2000). The mechanisms mediating the effects of poverty on children’s intellectual development. Demography, 37(4), 431–447. Ham, B. (2003). The effects of divorce on the academic achievement of high school seniors. Journal of Divorce & Remarriage, 38(3–4), 167–185. Hakvoort, E. M., Bos, H. M. W., van Balen, F., & Hermanns, J. M. (2010). Family relationships and the psychosocial adjustment of school-aged children in intact families. The Journal of Genetic Psychology, 171(2), 182–201. Härkönen, J., & Dronkers, J. (2006). Stability and change in the educational gradient of divorce. A comparison of seventeen countries. European Sociological Review, 22(5), 501–517. Harper, C., Marcus, R., & Moore, K. (2003). Enduring poverty and the conditions of childhood: Lifecourse and intergenerational poverty transmissions. World Development, 31(3), 535–554.

15

Hawkins, A. J., Blanchard, V. L., Baldwin, S. A., & Fawcett, E. B. (2008). Does marriage and relationship education work? A meta-analytic study. Journal of Consulting and Clinical Psychology, 76(5), 723–734. Heckman, J., & Kautz, T. (2012). Hard evidence on soft skills. Labour Economics, 19(4), 451–464. Johnson, M. K., Crosnoe, R., & Elder, G. H. J. (2001). Students’ attachment and academic engagement: The role of race and ethnicity. Sociology of Education, 74(4), 318–340. Kalmijn, M. (2007). Explaining cross-national differences in marriage, cohabitation, and divorce in Europe, 1990–2000. Population Studies, 61, 243–263. Kalmijn, M. (2013). Long-term effects of divorce on parent-child relationships: Within-family comparisons of fathers and mothers. European Sociological Review, 29(5), 888–898. Kelly, J., & Emery, R. (2003). Children’s adjustment following divorce: Risk and resilience perspectives. Family Relations, 52(4), 352–362. Kiernan, K. E., & Huerta, M. C. (2008). Economic deprivation, maternal depression, parenting and children’s cognitive and emotional development in early childhood. The British Journal of Sociology, 59(4), 783–806. Lansford, J. E. (2009). Parental divorce and children’s adjustment. Perspectives on Psychological Science, 4(2), 140–152. http://dx.doi.org/10.1111/j.1745-6924.2009.01114.x Lodewijckx, E. (2005). (Children and parental divorce in the Flemish Region. An analysis on National Register data) Kinderen en scheiding bij hun ouders in het Vlaamse Gewest. Een analyse op basis van Rijksregistergegevens. Brussel: Research Centre of the Flemish Government. Martinez, C. R., & Forgatch, M. S. (2002). Adjusting to change: Linking family structure transitions with parenting and boys’ adjustment. Journal of Family Psychology, 16(2), 107–117. McLanahan, S., & Percheski, C. (2008). Family structure and the reproduction of inequalities. Annual Review of Sociology, 34(1), 257–276. McLanahan, S., & Sandefur, G. (1994). Growing up with a single parent: What hurts, what helps. Cambridge. MA: Harvard University Press. McManus, P. A., & DiPrete, T. A. (2001). Losers and winners: The financial consequences of separation and divorce for men. American Sociological Review, 66(2), 246–268. Miller, L. D., & McLeod, E. (2001). Children as participants in family therapy: Practice, research, and theoretical concerns. Family Journal, 9(4), 375–383. Mo, Y., & Singh, K. (2008). Parents’ relationships and involvement: Effects on students’ school engagement and performance. Research in Middle Level Education Online, 31(10), 1–11. Musick, K., & Meier, A. (2010). Are both parents always better than one? Parental conflict and young adult well-being. Social Science Research, 39(5), 814–830. OECD. (2013). Education at a glance 2013: OECD indicators. Paris: OECD Publishing. Pedro-Carroll, J. L. (2005). Fostering resilience in the aftermath of divorce. The role of evidence-based programs for children. Family Court Review, 43, 52–64. Perdue, N. H., Manzeske, D. P., & Estell, D. B. (2009). Early predictors of school engagement: Exploring the role of peer relationships. Psychology in the Schools, 46(10), 1084–1097. Pollard, E., & Lee, P. (2003). Child well-being: A systematic review of the literature. Social Indicators Research, 61(1), 59–78. Research Centre of the Flemish Government. (2013). (Child poverty in Flanders) Kinderarmoede in Vlaanderen. Brussels: Research Centre of the Flemish Government. Raviv, T., Kessenich, M., & Morrison, F. J. (2004). A mediational model of the association between socioeconomic status and three-year-old language abilities: The role of parenting factors. Early Childhood Research Quarterly, 19(4), 528–547. Rumberger, R. W., Ghatak, R., Poulos, G., Ritter, P. L., & Dornbusch, S. M. (1990). Family influences on dropout behavior in one California high school. Sociology of Education, 63(4), 283–299. Schlee, B., Mullis, A., & Shriner, M. (2009). Parents social and resource capital: Predictors of academic achievement during early childhood. Children and Youth Services Review, 31(2), 227–234. Sigal, A. B., Wolchik, S. A., Tein, J.-Y., & Sandler, I. N. (2012). Enhancing youth outcomes following parental divorce: A longitudinal study of the effects of the New Beginnings Program on educational and occupational goals. Journal of Clinical Child and. Adolescent Psychology, 41(2), 150–165. Simpkins, S. D., Weiss, H. B., McCartney, K., Kreider, H. M., & Dearing, E. (2006). Mother-child relationship as a moderator of the relation between family educational involvement and child achievement. Parenting: Science and Practice, 6, 49–57. Simons-Morton, B., & Chen, R. (2009). Peer and parent influences on school engagement among early adolescents. Youth & Society, 41(1), 3–25.

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001

G Model SOCSCI-1168; No. of Pages 16

16

ARTICLE IN PRESS N. Havermans et al. / The Social Science Journal xxx (2014) xxx–xxx

Sun, Y., & Li, Y. (2001). Marital disruption, parental investment, and children’s academic achievement: A prospective analysis. Journal of Family Issues, 22(1), 27–62. Sun, Y., & Li, Y. (2009). Parental divorce, sibship size, family resources, and children’s academic performance. Social Science Research, 38(3), 622–634. Tillman, K. H. (2007). Family structure pathways and academic disadvantage among adolescents in stepfamilies. Sociological Inquiry, 77(3), 383–424. UNICEF (2010). The children left behind: A league table of inequality in child well-being in the world’s rich countries. Innocenti Report Card 9. Florence: Innocenti Research Center. Vanassche, S., Sodermans, A. K., & Matthijs, K. (2011). Methodologische documenten Leuvens Adolescenten- en Gezinnenonderzoek. Versie 1.0.

Centrum voor Sociologisch Onderzoek (CeSO), Onderzoeksdomein Gezin en Bevolking, CeSO/GB/2011-3, Leuven: K.U. Leuven. Centrum voor Sociologisch Onderzoek (CeSO). Willms, J. D. (2003). Student engagement at school: A sense of belonging and participation. In Results from PISA 2000. Paris: OECD. Wu, Z., Hou, F., & Schimmele, C. M. (2008). Family structure and children’s psychosocial outcomes. Journal of Family Issues, 29(12), 1600– 1624. Zhou, Q., Sandler, I. N., Millsap, R. E., Wolchik, S. A., & Dawson-McClure, S. R. (2008). Mother-child relationship quality and effective discipline as mediators of the 6-year effects of the New Beginnings Program for children from divorced families. Journal of Consulting and Clinical Psychology, 76(4), 579–594.

Please cite this article in press as: Havermans, N., et al. Family resources as mediators in the relation between divorce and children’s school engagement. The Social Science Journal (2014), http://dx.doi.org/10.1016/j.soscij.2014.04.001