Occasional Paper No. 46
Parental marital status and children’s wellbeing
LIXIA QU AND RUTH WESTON AUSTRALIAN INSTITUTE OF FAMILY STUDIES
A strong and fair society for all Australians
Research OP46_FINAL.indd 1 27/06/12 2:03 PM
© Commonwealth of Australia 2012
ISSN 1839-2334
ISBN 978-1-921975-67-7
All material presented in this publication is provided under a Creative Commons CC-BY Attribution 3.0 Australia (http://creativecommons.org/licenses/by/3.0/au/deed.en) licence.
For the avoidance of doubt, this means this licence only applies to material as set out in this document.
With the exception of the Commonwealth Coat of Arms (for terms of use, refer to <http://www.itsanhonour.gov.au/coat-arms/index.cfm>), the details of the relevant licence conditions are available on the Creative Commons website (accessible using the links provided) as is the full legal code for the CC-BY 3.0 AU licence
(http://creativecommons.org/licenses/by/3.0/au/legalcode).
Acknowledgements This research paper makes use of data from Growing Up in Australia: the Longitudinal Study of Australian Children (LSAC). LSAC is conducted in partnership between the Department of Families, Housing, Community Services and Indigenous Affairs (FaHCSIA), the Australian Institute of Family Studies (AIFS) and Australian Bureau of Statistics (ABS), with advice provided by a consortium of leading researchers.
The authors would like to thank colleagues at AIFS and FaHCSIA for their valuable comments on previous versions of this report.
The views expressed in this publication are those of individual authors and may not reflect those of AIFS.
For more information Research Publications Unit Research and Analysis Branch Australian Government Department of Families, Housing, Community Services and Indigenous Affairs PO Box 7576 Canberra Business Centre ACT 2610 Phone: (02) 6146 8061 Fax: (02) 6293 3289 Email: [email protected]
Research OP46_FINAL.indd 2 27/06/12 2:03 PM
iii
Contents
ContentsExecutive summary v
1 Introduction—research background 1 1.1 Trends in couple formation 1 1.2 Trends in ex-nuptial births and associated partnership circumstances 1 1.3 Related societal attitudes 2 1.4 Institutionalisation of cohabitation: International patterns 2 1.5 Characteristics of married and cohabiting families and sole-mother families 3 1.6 Implications for children’s wellbeing 4 1.7 Research questions 6
2 Sample, measures and analytical approaches 7 2.1 Parental characteristics and family financial circumstances 8 2.2 Parenting self-efficacy and practices 8 2.3 Parents’ psychological distress 9 2.4 Quality of inter-parental relationships 9 2.5 Child wellbeing measures 9
3 Results 10 3.1 Family structure 10 3.2 Child, parent and family characteristics 13 3.3 Children’s wellbeing 23 3.4 Cohabitation pathways and child wellbeing 33
4 Summary and discussion 37 4.1 Research questions and findings 38 4.2 Final word 40
Appendix: Measures used and tables 41Parenting practice measures 41Argumentative relationship scale 42
List of shortened forms 77
Endnotes 78
References 80
Research OP46_FINAL.indd 3 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 46iv
List of tablesTable 1: Family form by survey wave, K cohort children 10Table 2: Change in family form from Wave 1 to Wave 3, K cohort children 12Table 3: Selected characteristics of children by (unchanged) family form, K cohort, Wave 1(a) 13Table 4: Selected parental characteristics by unchanged family form, K cohort(a) 16Table 5: Mean scores of parenting approach scales, fathers and mothers, by unchanged family form,
Waves 1 to 3 20Table 6: Mean scores of psychological distress scale and quality of parental relationship of fathers
and mothers by unchanged family form, Waves 1 to 3 22Table 7: Child outcomes by family form, primary caregiver reports, Waves 1 to 3 24Table 8: Child SDQ outcomes by unchanged family form, teacher reports, Waves 1 to 3 25Table 9: Regressions of child outcomes, base model 28Table 10: Regressions of child outcomes, full model 31Table 11: Parents who were cohabiting in Wave 1: Parental psychological distress and the quality of
inter-parental relationships in Wave 1, by parental cohabitation pathways by Wave 3 34Table 12: Cohabiting families in Wave 1: Children’s developmental progress, by parental cohabitation
pathways by Wave 3 35
Appendix tablesTable A1: Sample attritions for three family types at Wave 1, K cohort 42Table A2: Children’s characteristics by family types, waves 1 to 3 (cross-sectional analysis) 43Table A3: Characteristics of fathers and mothers by family type, waves 1 to 3 (cross-sectional analysis) 44Table A4: Mean scores of parenting practice scales of fathers and mothers by family type, Waves 1 to 3
(cross-sectional analysis) 47Table A5: Mean scores of psychological distress scale and quality of parental relationship of fathers and
mothers by family type, Waves 1 to 3 (cross-sectional analysis) 48Table A6: Child outcome by family type, parents’ reports, Waves 1 to 3 (cross-sectional analysis) 49Table A7: Child outcome by family type, teachers’ reports, Waves 1 to 3 (cross-sectional analysis) 50Table A8: Regressions of child outcomes, as reported by parents (cross-sectional analysis), base model 51Table A9: Regressions of child outcomes, as reported by teachers (cross-sectional analysis), base model 54Table A10: Regressions of child outcomes, as reported by parents (cross-sectional analysis), full model 56Table A11: Regressions of child outcomes, as reported by teachers (cross-sectional analysis), full models 61Table A12: Regressions of child outcomes, as reported by parents (unchanged family form), base models 64Table A13: Regressions of child outcomes, as reported by teachers (unchanged family form), base models 66Table A14: Regressions of child outcomes, as reported by parents (unchanged family form), full models 68Table A15: Regressions of child outcomes, as reported by teachers (unchanged family form), full models 74
List of figuresFigure 1: Experience of one or more financial hardships, by (unchanged) family form, primary
carers’ reports, Waves 1 to 3 18Figure 2: Ratings of families’ financial situation, by (unchanged) family form, primary carers’ reports,
Waves 1 to 3 18Figure 3: Mothers’ ratings of self as a parent, by (unchanged) family form, Waves 1 to 3 19Figure 4: Fathers’ ratings of self as a parent by (unchanged) family form, Waves 1 to 3 20
Appendix figuresFigure A1: Proportions of families having experienced one or more listed financial hardships by family
type, as reported by primary carers, Waves 1 to 3 46Figure A2: Ratings of financial situation by family type, as reported by primary carers, Waves 1 to 3 46Figure A3: Mothers’ ratings of self as a parent compared with other parents by family type, Waves 1 to 3 46Figure A4: Fathers’ ratings of self as a parent compared with other parents by family type, Waves 1 to 3 47
Research OP46_FINAL.indd 4 27/06/12 2:03 PM
v
exeCutive summary
Executive summary
As in many Western countries, the rise in cohabitation in Australia represents one of the most striking of the many changes to family life that have emerged since the 1970s. Not only is cohabitation now the normative pathway to marriage, but it is also increasingly used as a context for raising a family.
While there is considerable evidence suggesting that cohabiting couples are more likely to separate than married couples, much of this research has not assessed whether this applies where children have been born of the cohabiting relationship. Some overseas studies have suggested that children with biological married parents have higher wellbeing than those whose biological parents are cohabiting. Nevertheless, a less favourable picture for those in cohabiting rather than married families has not been apparent across all wellbeing measures examined or in all studies. Based on the data of the first three waves of the Longitudinal Study of Australian Children, this paper focused on the wellbeing of young Australian children who were living with their cohabiting or married biological parents or with their sole mother over a four-year period (Wave 1 to Wave 3).
Key findings
Prevalence and stability of different family forms over a four-year period
◗◗ Not surprisingly, most of the children were living with both parents who were married to each other (here called ‘married families’) and sole-mother families were more commonly represented than families in which the child lived with both biological parents who were cohabiting (here called ‘cohabiting families’).
◗◗ Most married, cohabiting and sole-mother families in Wave 1 were in the same family form in Wave 3, with levels of stability being considerably greater for married families than for the other two forms.
◗◗ Although more than one in 10 children who lived in married families did not experience a change in family form, those who did experience a change tended to be living with their sole mother by Wave 3.
◗◗ Children in cohabiting families were just as likely to experience parental marriage as to experience parental separation.
◗◗ Not surprisingly, any change for those in sole-mother families tended to involve the introduction of a stepfather, although a few experienced their biological parents moving in together.
Socio-demographic characteristics, parenting practices and quality of the inter-parental relationship in different family forms
◗◗ Compared with married parents, cohabiting parents and sole mothers were younger, more likely to identify as Indigenous, had a lower level of education, and were less likely to be employed.
◗◗ The proportion of cohabiting families experiencing financial hardship was higher than that of married families but lower than that of sole-mother families.
◗◗ Cohabiting parents and sole mothers appeared to be less consistent in their parenting than was the case for married parents.
◗◗ Compared with married parents, cohabiting parents tended to describe the quality of their inter-parental relationship in more negative terms and to indicate higher emotional distress.
Research OP46_FINAL.indd 5 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 46vi
The wellbeing of children living in different family forms
◗◗ Compared with children with married parents, those living with their sole mother appeared to fare less well in terms of social–emotional, learning and physical development, as reported by their primary caregivers (typically their mothers) and teachers.
◗◗ The reports of primary caregivers indicated that the children with cohabiting parents were doing less well than those with married parents in some areas (e.g. overall social–emotional development) and general development, but they were doing better than those in sole-mother families.
◗◗ The family-related factors that were examined accounted for virtually all of the differences apparent between children with married parents and those with cohabiting parents.
◗◗ The differences apparent in Wave 1 in the wellbeing of children in married families and those in sole mother families could be largely explained by the family-related factors that were examined. Nevertheless, by Wave 3, the differences between the two groups in some wellbeing measures still remained after taking account of the family characteristics.
Research OP46_FINAL.indd 6 27/06/12 2:03 PM
1
introduCtion—researCh baCkground
1 Introduction—research background
As Benjamin Disraeli observed, ‘Change is inevitable. Change is a constant’. This is well demonstrated in families, the basic unit of society. It is reflected, for example, in the pathways to family formation, childbearing trends, patterns of family functioning, parental (or partnership) separation, and the prevalence of different family forms (see Hayes, Qu, Weston & Baxter 2011; Hayes, Weston, Qu & Gray 2010).
The family forms of key interest in this report are those comprising couples with children (in which the child who forms the focus of the present analysis—the ‘study child’—was born to the couple), and the parents are legally married to each other or are living together outside legal marriage (here called ‘cohabitation’).
This study explores the following questions: to what extent is the distinction between married and cohabiting couples relevant in terms of settings for raising children? That is, how similar or different are the parental/family characteristics and parenting practices of cohabiting and married families in which the children of interest were born? How do children in these two forms of families fare? To help put similarities and differences between these two groups in perspective, the analyses also extend to sole-mother families. Data from Growing Up in Australia: The Longitudinal Study of Australian Children (LSAC) are used in this study (see Section 2 for details about LSAC).
The discussion in this section first outlines trends in couple formation, in childbearing within and outside legal marriage, and in related societal attitudes. This is followed by a discussion of proposed ‘stages’ that countries go through as cohabitation increases in prevalence and the apparent process of convergence of the two institutions. Next, ways in which the two institutions still appear to differ are explored, with particular attention given to factors that may affect children’s wellbeing. Finally, previous research into the wellbeing of children in these two family forms is discussed—along with comparative studies of children in sole-mother families as a frame of reference.
1.1 Trends in couple formation
The establishment of a committed ‘living together’ relationship between two adults is typically a central milestone in the family formation process. Throughout most of the 20th century, this new way of life almost always began with marriage, but this is no longer the case, for many couples now live together outside marriage in a cohabiting relationship. Indeed, it is fairly unusual these days for couples to marry without having lived together first. Of all couples who married in 2009, 77 per cent had cohabited beforehand, compared with 31 per cent in 1981 (Australian Bureau of Statistics [ABS] 2002; 2010b). Some cohabiting couples never marry, while others end up separating, and for some other couples, cohabitation is a prelude to marriage.
Not surprisingly, cohabitation is the most common form of union for couples under the age of 25 years, although even among those in their late twenties who are living together, around four in 10 are cohabiting. And while the predominance of marriage increases with age, the proportion of couples who are cohabiting has increased across all age groups (Weston & Qu 2007).
1.2 Trends in ex-nuptial births and associated partnership circumstances
There is also evidence that cohabitation has increasingly become a setting for raising children. This is apparent both in surveys of families and in official birth statistics. For example, surveys conducted by the ABS suggest that, among intact families with children under the age of 18 years, the proportion of parents who
Research OP46_FINAL.indd 1 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 462
were cohabiting rather than married increased from 6 per cent in 1997 to 9 per cent in 2006–07.1 In addition, the proportion of infants born outside marriage has increased markedly in recent decades, from 8 per cent of registered births in 1970 to 34 per cent of births registered in 2009 (ABS 2010a). These latter statistics, which are based on birth registrations, are not accompanied by information concerning whether the mother was cohabiting with the father at the time of birth. However, perinatal statistics published by Australian Institute of Health and Welfare (AIHW), based on data collected from midwives or other attendants at births, suggest that the proportion of infants born to single women has remained stable since the early 1990s. For example, these data suggest that, among all women who gave birth, the proportion who were ‘single’ (that is, not living with a partner) at the time of the birth was 12 per cent in 1991 (Lancaster, Huang & Pedisich 1994), and 13 per cent in both 1999 and 2005 (reported by Nassar & Sullivan 2001,2 and Laws, Abeywardana, Walker & Sullivan 2007, respectively). Of these single women, over 90 per cent had never been married. In other words, the rise in the proportion of all babies born outside marriage can be attributed to the increase in the proportion of babies born to cohabiting couples.
1.3 Related societal attitudes
The increasing prevalence of cohabitation both fuels and is reinforced by an increasing social acceptance of such lifestyles. Changed views about such lifestyles were highlighted by Qu and Weston (2008a), who found that around three-quarters of adult Australians in a national survey conducted in 2005 agreed with the statement, ‘It is alright for an unmarried couple to live together even if they have no intention of marrying’, while around 70 per cent rejected the notion that ‘marriage is an outdated institution’.3 To gauge some idea of the level of attitudinal change that has occurred, Qu and Weston also referred to the results of a 1971 survey of married women in Melbourne, conducted by the Australian National University (ANU). Two-thirds of these women said that they would be ‘extremely horrified’, ‘considerably upset’ or would ‘consider that they had failed as a parent in his upbringing’, should their son indicate that he intended to live with a woman without marrying her first. In addition, 52 per cent said that they would be more horrified if their daughter reported such intentions, while 2 per cent said that they would be less upset, with the remaining 44 per cent indicating that their reactions would not differ according to their child’s sex.4
Not only is cohabitation now commonly accepted, but cohabiting couples with children are also generally considered to be ‘family’, although a small minority of Australians do not hold this view. This issue was captured in the Australian Survey of Social Attitudes 2003. Respondents were asked which of four different living arrangements they would consider to be a family: a married couple without children, an unmarried couple without children, a married couple with children, and an unmarried couple with children. While married couples with children were universally considered to be families, cohabiting couples with children were also considered so by around eight in 10 respondents (77 per cent of men and 82 per cent of women) (Evans & Gray 2005).5 It would be interesting to see if the gap has narrowed since this survey was conducted.
1.4 Institutionalisation of cohabitation: International patterns
Many Western countries have experienced an increase in cohabitation, although the timing and extent of such trends have varied. According to Kiernan (2001, 2002), countries go through four stages in relation to cohabitation. In the initial stage, cohabitation represents a deviant or avant-garde phenomenon and applies to a small minority of the population (for example, Italy, Greece and Spain were at this stage at the time of Kiernan’s analysis). The second stage involves cohabitation as a prelude to marriage or a trial marriage, with marriage remaining the setting for childbearing (for example, the Netherlands, Switzerland and West Germany, according to Kiernan). Subsequently, cohabitation becomes a socially accepted alternative to marriage, with childbearing no longer restricted to marriage (examples being Austria, Great Britain and Ireland). In the fourth stage, cohabitation and marriage become indistinguishable (apparent in Sweden and Denmark, with France and East Germany ‘hovering around this stage’). However, Kiernan noted that in most cases, couples who had a child outside marriage tended to marry within five years.
Research OP46_FINAL.indd 2 27/06/12 2:03 PM
3
introduCtion—researCh baCkground
The increase in the proportion of babies born to cohabiting couples and the social acceptance of cohabitation in its own right suggests that Australia would be in the third stage, although around one-quarter of adults either reject or express uncertainty about whether it is all right for couples to live together without intending to marry.
Related to the issue of proposed stages through which Western countries progress, Cherlin (2004) maintained that, in the United States, marriage has been de-institutionalised while cohabitation is becoming increasingly institutionalised, from both a legal and a social perspective. Such trends would accompany and possibly narrow differences in the everyday meaning of the two arrangements.
The institutionalisation of cohabitation is also apparent in Australia, as evidenced by the introduction of the Family Law Amendment (De Facto Financial Matters and Other Measures) Act 2008. This Act enables cohabiting couples, including those of the same sex, to be treated the same way as married couples in relation to the settlement of financial matters. Even before this amendment to the Family Law Act 1975, cohabiting parents had been treated in the same way as married parents in the sense that their parenting matters were dealt with in the federal sphere. However, prior to the enactment of the amendment, property matters were dealt with in state and territory jurisdictions. In relation to assessments regarding entitlements to government benefits, such as family payment, cohabiting and married parents have been treated in the same manner.
The increasing prevalence of cohabitation has also led to classifications in research that discriminate between the types of living-together union. For instance, the ABS (for example, 2005) uses a measure called ‘social marital status’, which ascertains whether the couple is married or cohabiting.
1.5 Characteristics of married and cohabiting families and sole-mother families
Cherlin (2004) maintained that, although marriage has very much lost its practical meaning, it remains a symbol of status, prestige and achievement in the United States. He noted, for instance, that married couples tend to have higher socioeconomic status than cohabiting couples and that many people who are cohabiting want to get married, but are waiting until they have saved for an expensive wedding or have overcome other difficulties, including any problems in their relationship.
Studies in the United States also suggest that children living with cohabiting parents have a different family environment from those in married families. For example, cohabiting parents in the United States are less likely to participate in paid work and thus, unsurprisingly, tend to experience poorer financial circumstances (Artis 2007; Brown 2004). Some research also suggests that cohabiting mothers are more likely than married mothers to experience mental health problems (Artis 2007; Brown 2000). In addition, it appears that compared with married parents, cohabiting parents (including those with children) are likely to face greater uncertainty concerning the future of their union, with previous research suggesting that cohabiting relationships are considerably more likely than marriages to end in separation (Ermish 2002; Osborne, Manning & Smock 2007; Qu & Weston 2008b).
While much of this research has failed to take into account whether the children were living with a step-parent rather than with both biological (that is, natural) parents, in their study based in the United States Osborne et al. (2007) found that children born to cohabiting parents had a substantially greater risk of their parents separating compared to those born to married parents. These authors found that the difference was more marked for white children than for African–American or Mexican children.
1.6 Implications for children’s wellbeing
Little research has been undertaken into the implications for children’s wellbeing of growing up with both biological parents who are cohabiting. As Brown and Manning (2009) pointed out, the social marital status
Research OP46_FINAL.indd 3 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 464
of parents in couple families has been largely overlooked in studies of family structure. If these cohabiting families are more fragile, then the circumstances and associated family functioning processes that are linked with their fragility may well have negative consequences for children’s developmental progress and wellbeing (see Magnuson & Berger 2009; Pryor & Rodgers 2001). However, as will be seen below, a 2008 Canadian study challenges this assumption (Wu, Hou & Schimmele 2008). The following discussion provides a summary of findings derived from some of the limited relevant research that has been undertaken to date.
Research by Artis (2007) in the United States has suggested that kindergarten children who lived with their biological cohabiting parents had poorer cognitive and psychosocial development indicators than their counterparts with married parents. Similarly, in another United States study, Osborne, McLanahan and Brooks-Gunn (2003) found that children aged 3 years old who were living with cohabiting parents were more likely than their counterparts with married parents to express aggressive, withdrawn and anxious/depressive behaviour three years later. Hofferth (2006) likewise observed more behavioural problems in a sample of children (aged 3 to 12 years) in the United States who were living with their biological cohabiting parents, compared with their counterparts who were living with married parents, but no differences were apparent in relation to cognitive achievement. Focusing on United States school-aged children and adolescents, Brown (2004) reported similar trends relating to psychological wellbeing (for example, in terms of behavioural and emotional problems and children’s engagement in school—such as completing homework and caring about doing well at school), with children who were living with their biological cohabiting parents faring no better than those living in sole-mother families.
Using data from a large-scale longitudinal study of children in the United Kingdom, Goodman and Greaves (2010) found that children aged 3 years who were born to married parents performed better in both cognitive development and social and emotional development than their counterparts who were born to cohabiting parents. In addition, they found that the gap in cognitive development between the two groups of children appeared to have increased by the time they had reached 5 years of age, while the difference in social and emotional development remained the same.
A study of psychosocial behaviours of Canadian children aged 10 to 15 years generated mixed findings between those who continued to live with their married parents over a five-year period and their counterparts who continued to live with both biological parents who remained unmarried over a two-year period (Wu et al. 2008). Specifically, children who continued to live with both biological parents who were cohabiting were less likely to exhibit prosocial behaviour, but also less likely to exhibit emotional problems compared with the other group of children. Their scores on the conduct disorder measure did not differ significantly. The findings from this study seem counter-intuitive in suggesting no harmful effects.
Finally, Bjorklund, Ginther and Sundstrom (2007) examined the grade point average at age 16 of 130,000 children who were born in Sweden between 1972 and 1987, and found that children whose parents had married before their birth performed significantly better than those whose parents were cohabiting.
All the above research has entailed attempts to explain the observed differences between the two groups of children by controlling for the effects of various family environment factors. Mixed results have emerged from such analyses.
Artis (2007) managed to explain differences that emerged for two of the three cognitive measures and for both measures of psychosocial development in terms of systematic differences between the families of these children in such matters as financial resources, mothers’ depressive symptoms and parenting practices. Similarly, Goodman and Greaves (2010) found that the differences in children’s cognitive, social and emotional development were very much explained by differences in parental characteristics, socioeconomic status, parental relationship quality and the history of parental relationship stability.
Osborne et al. (2003) reported that the effect of cohabitation on children’s aggressive and withdrawn behaviours disappeared once parental characteristics and economic and parental resources were controlled. However, the link between the parents’ relationship status and children’s anxious and depressive behaviours weakened but was still apparent.
Research OP46_FINAL.indd 4 27/06/12 2:03 PM
5
introduCtion—researCh baCkground
Hofferth (2006), on the other hand, found that the elevated risk of behavioural problems apparent for those living with cohabiting parents approached statistical significance when such factors as fathers’ education and parental income were controlled for (p < 0.10). Furthermore, with these controls in place, the children who were living with their cohabiting biological parents performed better in terms of cognitive achievement than those who were living with their married biological parents.
In their study of fathering in Australia, which was based on the same longitudinal study as that used in the present report, Baxter and Smart (2010) found that children of cohabiting fathers fared better in terms of cognitive development than children of married fathers. In the United States, Brown (2004) indicated that differences in economic and parental resources between cohabiting families and married families could not fully explain the differences in children’s wellbeing that she examined.
The Canadian study by Wu et al. (2008) suggested that the poorer prosocial behaviour of the children who were living with their cohabiting parents continued to hold when the effects of various child- and parent-related demographic characteristics, along with the family’s financial status and indicators of family dysfunction and parental nurturance, were controlled for. However, the significantly higher emotional disorder scores of children with cohabiting rather than married parents could be explained by child- and parent-related demographic characteristics alone. This study also examined the progress of children who experienced the parental divorce or separation of their cohabiting parents (called ‘cohabitation separation’). The latter children appeared not to be progressing as well as those whose parents stayed together, while parental divorce appeared to have a significantly damaging effect on the children.
While Björklund et al. (2007) reported that the positive association of children’s educational performance with parental marriage was largely due to other differences in family characteristics, they found that marriage appeared to have a positive effect on educational performance for boys but not for girls. Bjorklund et al. concluded that marriage does not provide an advantage for Swedish children in terms of educational progress, but they did not rule out any positive effect linked with marriage for children in countries where cohabitation is less prevalent than in Sweden.
In short, there is some evidence that children living with both biological parents who are cohabiting have a higher risk of poorer developmental progress than children who are living with their married parents, but the results are not entirely consistent, and the family-related characteristics and processes that explain such differences remain uncertain. Some evidence has suggested that children living with their cohabiting parents may be no better off in terms of developmental progress or wellbeing than those in sole-mother families. However, parental divorce may have a more harmful impact on children than cohabitation separation. Reasons for this are unclear. Wu et al. (2008) suggested that, compared with separation from a cohabiting relationship, divorce may entail a more drawn-out process that prolongs children’s distress, although these authors also acknowledged that the results they observed may be a function of sampling problems.
Regarding sole-mother families, a great deal of evidence has amassed suggesting that children in these families are at an elevated risk of displaying a broad range of emotional and behavioural adjustment problems, and low school achievement. Factors explaining such trends include the poorer financial circumstances of sole-mother families, poorer maternal health, fewer effective parenting practices, and more limited social support networks (see Mackay 2005; Tobias, Kokaua, Gerritsen & Templeton 2010; Wu et al. 2008). In most cases, sole-mother families result from parental separation. Stressful circumstances—including inter-parental conflict leading to, and following, such separation—also emerge as important factors contributing to children’s elevated risk of diminished wellbeing (see Magnuson & Berger 2009; Pryor & Rodgers 2001). However, there has been little research investigating whether the differences observed in the wellbeing of children in sole-mother families and those in intact couple families are largely a reflection of differences between the former group and children living with two married, rather than cohabiting, parents, given that the parents of the vast majority of children in intact families are married to each other. Brown’s (2004) research would suggest this to be the case.
These limited studies of children in cohabiting and married families are important in that they raise policy-relevant hypotheses about factors that increase the risk of children displaying poor developmental progress. However, generalising these findings (particularly across countries) is difficult for at least three reasons.
Research OP46_FINAL.indd 5 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 466
Firstly, as Brown (2004) has pointed out, much of the research tends to rely on cross-sectional data; therefore, the relationships between parental cohabitation and poorer child wellbeing may be spurious. Secondly, in the absence of life history data for the children, it has not been possible to explore potential long-term or cumulative differential impacts on children of their living in cohabiting or married families. Thirdly, the circumstances experienced by children in cohabiting families are likely to differ in different countries. For example, in countries in which childbearing in cohabitation is unusual and largely ‘frowned upon’, families are likely to differ markedly (for instance, in terms of family resources and parenting) from those in countries in which such behaviour is widespread and sanctioned by the vast majority of people.
1.7 Research questions
The analyses outlined in this report focus on Australian families, use longitudinal data, and rely on multiple measures of children’s developmental progress, derived from three sources (the primary caregiver, a teacher and the child). While the research interest largely concerns the progress and experiences of children living with married and cohabiting biological parents, in order to place any observed differences in perspective, data concerning children living in sole-mother families were also examined. The analyses were designed to address the following questions in relation to these three groups of families with young children (aged 4 to 5 years when first examined):
◗◗ How common and how stable are these family forms over a four-year period and, of those that change, what are their most likely pathways? (For example, are cohabiting parents more likely to marry or separate?)
◗◗ To what extent do these family forms differ in terms of the financial circumstances of the family, selected socio-demographic characteristics of the parents and children, and indicators of parenting practices and parental wellbeing?
◗◗ To what extent do the children in these different family forms differ in terms of their developmental progress?
◗◗ To what extent, if at all, do any differences in child wellbeing between the groups diverge or diminish over the four-year period?
◗◗ If differences in child wellbeing are apparent, to what extent can they be explained by any observed systematic differences in child and family characteristics or parenting practices and parental wellbeing?
◗◗ Of all the children in cohabiting families in the initial survey wave, to what extent, if at all, does their wellbeing vary (at the outset and subsequently) according to whether their parents marry, separate or continue to cohabit over the four-year period?
For succinctness, the families in which the biological parents of the study child were married to each other are called ‘married biological families’, while those in which the parents were cohabiting are called ‘cohabiting biological families’. Some of the study children were also living with a half-sibling (6 per cent in the married families, 19 per cent in the cohabiting families, and 20 per cent in the sole-mother families).
Research OP46_FINAL.indd 6 27/06/12 2:03 PM
7
sample, measures and analytiCal approaChes
2 Sample, measures and analytical approaches
This report uses data derived from the first three waves (2004, 2006 and 2008) of Growing Up in Australia: The Longitudinal Study of Australian Children, funded by the Australian Government through the Department of Families, Housing, Community Services and Indigenous Affairs.6 LSAC focuses on two cohorts of children: those born between March 2003 and February 2004 (infants in the first survey wave) and those born between March 1999 and February 2000 (children aged 4 to 5 years in the first survey wave). A nationally representative sample of around 5000 children was drawn for each age cohort (for details see Soloff, Lawrence & Johnstone 2005). The analysis on which the present report is based focuses on the elder of these cohorts (often referred to as the K cohort). These children were aged 6 to 7 years in Wave 2, and 8 to 9 years in Wave 3.
This report directs attention to children living with both biological parents and those living with their single mother. Depending on the marital status of the parents in couple families, these families are referred to as ‘married families’ or ‘cohabiting families’. (As will be shown below, only small proportions of children lived in other family forms.) The primary caregiver’s views of the focus child’s wellbeing, along with those of the child’s teacher, form the basis of almost all the analyses of the child’s wellbeing. The mother was the primary caregiver in 98 per cent of families.
Two sets of analyses were carried out for each wave of data, with the comparisons covering three forms of families—married families, cohabiting families and sole-mother families. The first set of analyses is cross-sectional in nature, covering the family forms in each of the three waves, taken separately (‘cross-sectional family groups’). The second set focuses on the three family groups that had not changed family form across the survey waves (‘unchanged family groups’). As mentioned above, cohabiting couples are more likely than married couples to separate, so restricting attention to the three unchanged family groups removes possible differential family conditions that contribute to any separation that, in turn, may affect children’s wellbeing. These two approaches assess the extent to which the three family groups differ systematically in terms of a selection of family characteristics, indicators of family processes and children’s developmental progress.
Multivariate analyses were then carried out to assess the extent to which any differences in children’s wellbeing could be attributed to any existing systematic differences in family characteristics, as well as children’s characteristics (such as age, gender, country of birth, and presence of older or younger siblings in the household). As noted above, much of this work involved analyses based on: (a) members of the three family forms apparent in each survey wave, regardless of whether some members held a different family form status in another survey wave (cross-sectional family groups); and (b) members of the three family forms who remained in the same family form in each survey wave (unchanged family groups). Because the results based on the cross-sectional family groups turned out to be similar to those based on the three unchanged groups, the results of the latter sets of analyses (based on the unchanged family groups) are outlined in the main body of this report, while those based on the three cross-sectional family groups are presented in the Appendix.
A final set of analyses focused exclusively on children who, in Wave 1, had been living with their biological cohabiting parents. The developmental progress of those whose parents subsequently married, separated or continued to live together was examined. Although limited in scope by the small number of children who experienced either transition,7 the key aim of this approach was to gauge whether there was any evidence that the children whose parents eventually separated might be negatively affected by this experience and, if so, whether some negative effects were apparent before the separation occurred. Given the small number of children who experienced the marriage or separation of their initially cohabiting parents, only bivariate analyses were undertaken.
Research OP46_FINAL.indd 7 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 468
2.1 Parental characteristics and family financial circumstances
The following parental characteristics were examined: personal age, country of birth, Indigenous status, educational attainment, and employment status. These characteristics were derived for mothers in all three family forms and for fathers in the married and cohabiting families. Two measures were used as indicators of the family’s financial circumstances, based on the reports of the primary caregivers (almost always the mother): an overall evaluation of their financial status, and experiences of financial hardship in the previous 12 months.8
In relation to the first of these measures, the primary caregivers were asked whether, given their current needs and financial responsibilities, they would describe their circumstances as ‘prosperous’, ‘very comfortable’, ‘reasonably comfortable’, ‘just getting by’, ‘poor’ or ‘very poor’. To tap specific financial difficulties, these respondents were asked whether, because of financial hardship: (a) they had been unable to pay gas, electricity or telephone bills on time; (b) they had been unable to pay the mortgage or rent on time; (c) adults or children (in the family) had gone without meals; (d) they had been unable to heat or cool their home; (e) they had pawned or sold something; and (f ) they had sought assistance from a welfare or community organisation.9
2.2 Parenting self-efficacy and practices
The measure of parenting self-efficacy was based on mothers’ and fathers’ ratings of their competence relative to other parents’ (a single-item measure). Four measures of parenting practices were derived, again based on the self-reports of fathers and mothers: ‘parental warmth’, ‘hostile parenting’, ‘inductive reasoning’ and ‘consistent parenting’.
The ‘parental warmth’ measure focused on such issues as how often the respondents expressed affection, listened to their children, had a warm and close time together with their children, and felt close to their children. On the other hand, the ‘hostile parenting’ scale gauged such matters as how often they praised their children or expressed disapproval of their children’s behaviour, how often they felt angry when they punished their children, and how often they had problems in managing their children. The ‘inductive reasoning’ measure tapped the frequency with which parents provided an explanation when they corrected their children’s behaviour or reasoned with their children when they misbehaved. The ‘consistent parenting’ scale assessed such issues as how often parents felt their children ‘got away’ with things when they should have been punished, how often the parent made sure that the children obeyed directions, and how often their children ignored any punishment they received.
All the items in the parenting practice scales were rated on a scale ranging from 1: ‘never/almost never’ to 5: ‘all the time’. Each parent’s mean rating for each scale was derived. To the extent that respondents provided accurate accounts of their behaviour, a high score on the parental warmth scale suggests that the parent often expressed warmth towards the children, while a high score on the hostile parenting scale suggests that the parent often expressed harsh attitudes towards the children. In relation to the scales concerning parenting consistency and use of reasoning, high scores suggest that the parent tended to be consistent in dealing with the children’s behaviour and often engaged in attempts to reason with them. Further details of these measures are presented in the Appendix.
2.3 Parents’ psychological distress
The six-item Kessler-6 screening scale for psychological distress was used (Kessler et al. 2002). Fathers and mothers were asked how often in the past four weeks they had felt: (a) nervous, (b) hopeless, (c) restless or fidgety, (d) worthless, and (e) so sad that nothing could cheer them up, and (f ) how often they felt that everything was an effort. These items were designed to detect parental mood and anxiety disorders. Scores
Research OP46_FINAL.indd 8 27/06/12 2:03 PM
9
sample, measures and analytiCal approaChes
on this scale ranged from 1 to 5, with higher scores suggesting a higher level of distress in the present report. Each parent’s average rating for the six items was derived.
2.4 Quality of inter-parental relationships
Two measures relating to the quality of inter-parental relationships were derived from both parents in cohabiting and married families: relationship happiness and arguments.
Parents were asked to rate their overall happiness in the relationship on a scale ranging from 1 (extremely unhappy) to 7 (perfectly happy). The measure on arguments between the father and mother was derived from parents’ responses to five questions about how often they and their partner engaged in disagreements, arguments and physical confrontation, and how often they felt stress and hostility towards each other. Each parent’s average rating for these five items was derived. Higher scores suggest poorer relationships.
2.5 Child wellbeing measures
One of the strengths of the LSAC data is its wide range of child wellbeing measures, based on assessments provided by the primary caregiver and a teacher, and on measures administered to the child (for details on these measures, see Sanson, Misson and the LSAC Outcome Index Working Group 2005). The wellbeing measures cover components of the child’s physical, social–emotional and cognitive developmental progress. As explained by Sanson et al., an overall ‘outcome index’ (or overall measure of developmental progress) was derived, based on three other outcome indices covering developmental progress in the physical, social/emotional and learning domains. As their name implies, these four outcome indices are composite measures and provided in standardised scores, with a mean at 100 and standard deviation at 10. Higher scores indicate better outcomes (or developmental progress). These indices are comparable across waves.
The present report focuses on the overall outcome index, the three component indices, and one sub-component of the learning index—a measure of the child’s receptive language skills (using the Peabody Picture Vocabulary Test [PPVT-III Form IIA]). PPVT scores for the K cohort of children ranged from 28 to 106 (with a mean of 72 and a standard deviation of 8).
It is common for research on the clinical assessment of psychological state to generate different results based on different informants (for example, reports of children’s behaviours from parents, teachers, clinical observations), and there is no evidence available regarding who make accurate and reliable informants (see De Los Reyes 2011). For example, children do not behave in the same way across different social contexts, and inconsistencies in reporting children’s behaviours among informants (for example, parents, teachers) may reveal useful information about variations of children’s behaviours in different settings (De Los Reyes 2011). Thus, data derived from five multi-item scales based on both parents’ and teachers’ responses to the Strengths and Difficulties Questionnaire (SDQ) were used in the analyses. That is, two scores were derived for each of the five scales—one based on parents’ reports, and the other on teachers’ reports. The five scales represent components of the social–emotional index and cover: (a) prosocial behaviour (that is, sharing easily, being considerate, being kind to younger children); (b) peer problems (that is, having difficulties in developing favourable relationships with other children); (c) emotional symptoms (that is, frequency of exhibiting negative emotional states); (d) hyperactivity (covering such behaviour as fidgeting and concentration span); and (e) conduct problems (exhibited when interacted with others). All scores ranged from 1 to 5, with higher scores reflecting better outcomes for the prosocial behaviour measure, and poorer outcomes for the other four scales. All measures except the conduct problems scale were comparable across the survey waves.
Research OP46_FINAL.indd 9 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4610
3 Results
The results of the various analyses are set out according to the research question they were designed to address. These cover: (a) how common and stable the various family forms were over the four-year period; (b) the extent to which the three family forms of interest differ in character (that is, financial wellbeing, child and parent socio-demographic characteristics), parenting practices and parental wellbeing; (c) the extent to which children in the three family forms differ in terms of their developmental progress; (d) whether there was any evidence that differences between the three family groups diverged or diminished over the four-year period; (e) the extent to which any such differences can be explained by the measures tapping child and family characteristics, parenting practices and parental wellbeing; and (f ) whether the wellbeing of children living with cohabiting parents varied (at the outset and subsequently) according to whether their parents married, separated or continued to cohabit over the four-year period.
3.1 Family structure
How common are these family forms?
Table 1 shows the proportion of children in the various family forms in each survey wave taken separately (2004, 2006 and 2008). As noted above, the children’s ages were 4 to 5 years in Wave 1, 6 to 7 years in Wave 2, and 8 to 9 years in Wave 3. The family forms were based on whether children were living with both parents or one parent, whether the child had a step-parent in the household, and the social marital status (married or cohabiting) of the parents.
Table 1: Family form by survey wave, K cohort children
4 to 5 years(Wave 1)
6 to 7 years(Wave 2)
8 to 9 years(Wave 3)
%
Two biological, married parents 72.6 71.8 68.8
Two biological, cohabiting parents 9.3 7.4 6.3
One biological parent, married stepfamily 0.8 1.3 2.9
One biological parent, cohabiting stepfamily 2.0 2.2 4.2
Sole biological mother 14.2 15.8 15.6
Sole biological father 0.6 0.9 1.1
Neither biological parent in the household 0.4 0.4 0.5
Other (a) 0.0 0.2 0.6
Total 100.0 100.0 100.0
No. of respondents 4983 4464 4331
(a) ‘Other’ covers families for which information about the relationship between members was incomplete.
Note: The K cohort refers to children who were aged between 4 and 5 years in Wave 1 (2004). Data have been weighted. The number of respondents in the table is unweighted. There were only nine adopted children in the entire sample. These were therefore treated as biological children. Their inclusion does not change the results at all.
Research OP46_FINAL.indd 10 27/06/12 2:03 PM
11
results
Across the three waves, most of the children (69 to 73 per cent) were living in married biological families, with only 6 to 9 per cent living in cohabiting biological families. In total, then, 75 to 82 per cent of children were living with both biological parents in each survey wave (taken separately). A substantial minority were living with one biological parent, usually the mother (14 to 16 per cent were in sole-mother families and around 1 per cent were in sole-father families). A small proportion of children in each survey wave were living with one biological parent and one step-parent (3 to 7 per cent). Of those in stepfamilies, the vast majority were living with their biological mother and stepfather (96 per cent) (data not shown).
The parents in stepfamilies were as likely to be cohabiting as to be married (2 to 4 per cent of all children were in cohabiting stepfamilies; 1 to 3 per cent were in married stepfamilies). Few children in any of the survey waves were living with neither biological parent (fewer than 1 per cent).10 In a small proportion of cases (fewer than 1 per cent), the family form in which the child was living could not be determined.
Table 1 also shows that the proportion of children in married and cohabiting biological families declined slightly over the three waves (married: from 73 per cent in Wave 1 to 72 per cent in Wave 2 and 69 per cent in Wave 3; cohabiting: from 9 per cent in Wave 1, to 7 per cent in Wave 2 and 6 per cent in Wave 3). On the other hand, there was a marginal increase in the proportion of children living with one parent (from 15 per cent in Wave 1 to 17 per cent in two subsequent waves), and in the proportion living with one biological and one step-parent (from 3 per cent to 7 per cent).
Overall, there were 519 fewer children represented in the second than first survey (representing an attrition rate of only 10 per cent) and 652 fewer children represented in the third than first survey (representing an attrition rate of 13 per cent). In Wave 3, 84 per cent of the original sample of K cohort children were followed up in both subsequent survey waves, and 3 per cent were followed up in Wave 3 but not in Wave 2.
How stable were these family forms over a four-year period and, of those that change family form, what are their most likely pathways?
As living systems, families are by nature dynamic. They change in characteristics and in functioning as children are born, as members grow older, and as children gain independence and leave home. Various factors, such as external pressures and opportunities, health issues, financial issues, employment circumstances and relationship dynamics can also combine to bring about further changes, with some parents separating, some re-partnering, others reconciling, and some cohabiting parents marrying.
Table 2 focuses on three family forms in Wave 1—married and cohabiting biological families and sole-mother families—and shows the extent to which they had changed their form by Wave 3 (four years later). This table is restricted to the families of K cohort children who were represented in both these survey waves.11
Married biological families represented the most stable of the three family forms: more than nine in 10 had not changed status from Wave 1 to 3, with 6 per cent of children living with one biological parent, typically the mother, and only 1 per cent living with their mother and a stepfather. It is worth noting that, for a very small proportion of children (less than 1 per cent) in married biological families in Wave 1, the recorded marital status of parents had changed from married to cohabiting by Wave 3. While this may have resulted from mis-recording, it is also possible that the parents had divorced then reconciled during the four-year interval between the two surveys.
Research OP46_FINAL.indd 11 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4612
Table 2: Change in family form from Wave 1 to Wave 3, K cohort children
Wave 3
Wave 1
Two biological, married parents
Two biological, cohabiting parents
Sole mother
%
Two biological, married parents 92.5 15.7 2.4
Two biological, cohabiting parents 0.4 65.0 3.2
Step-parent family 1.1 3.5 28.4
Sole mother 5.5 14.3 63.2
Sole father 0.5 1.5 1.4
Other 0.0 0.0 1.4
Total 100.0 100.0 100.0
No of observations 3349 343 497
Note: Unweighted data.
Two-thirds of the children from cohabiting biological families in Wave 1 were in the same living arrangement in Wave 3. In other words, consistent with previous research outlined above, children from these families were more likely to experience parental separation between Wave 1 and Wave 3 (19 per cent) than those from married biological families (7 per cent). Thus, children from cohabiting biological families were more likely than those from married families to experience a switch to living in lone-parent families or stepfamilies some four years after the first survey. By Wave 3, 14 per cent of the children whose parents were cohabiting in Wave 1 were living with a sole mother, fewer than 2 per cent were living with a sole father, and around 4 per cent were in stepfamilies (almost always involving a stepfather). On the other hand, rates of subsequent marriage of cohabiting parents (16 per cent) by Wave 3 were similar to rates of separation (19 per cent).
Much the same proportion of children from cohabiting biological families and sole-mother families experienced unchanged family forms (65 per cent and 63 per cent respectively). In most cases, the change from a sole-mother family came about because the mother had re-partnered (applying to 28 per cent of all children in sole-mother families in Wave 1). Nevertheless, 6 per cent of sole mothers appeared to have reconciled with their child’s father, comprising over 3 per cent who were apparently cohabiting with him in Wave 3 and over 2 per cent who had apparently remarried him. It is also possible that some of these sole mothers may have lived apart from their child’s father for practical reasons (for example, work-related locational issues) and had begun living together by Wave 3. Just over 1 per cent of the children had moved from living with their sole mother to living with their sole father.12
Despite the many advantages of longitudinal studies, a key difficulty relates to sample attrition. By Wave 3, 13 per cent of the original K cohort of children were not in the study. Table A1 in the Appendix shows that, of the three family groups (the two biological couple families and sole-mother families), sample attrition was highest for the sole-mother families and lowest for the married biological families. Put another way, the following proportions of children were represented in all three survey waves: in Wave 1, 88 per cent who were living with both biological married parents; 75 per cent who were living with both biological cohabiting parents, and 72 per cent who were living with their sole mothers.
As explained in the LSAC 2008–09 Annual Report (LSAC 2009), families that did not participate in the later waves differed somewhat from those who continued to participate. For example, parents of the former group were more likely to have a non-English-speaking background, to identify as Indigenous, and to have lower incomes. Despite the socioeconomic differences between families that remained in the study and those who dropped out, it is difficult to ascertain the extent to which parental separation contributed to discontinuation in the study, especially for cohabiting parents.
Research OP46_FINAL.indd 12 27/06/12 2:03 PM
13
results
The best that can be achieved at this stage is to examine separation rates among those who participated in Waves 1 and 3, but not Wave 2 (reflecting temporary discontinuation). There were 16 children who were in cohabiting biological families in Wave 1 whose families did not participate in Wave 2 but returned to the study in Wave 3. In three of these cases, the parents had married by Wave 3; in another three, the parents had separated; and in the remaining 10, the parents continued to cohabit. Of the 86 married biological families who participated in Waves 1 and 3 but not Wave 2, only 11 had separated by Wave 3. While these numbers are too small to provide any firm evidence, it is possible that parental separation was not the key reason behind temporary discontinuation in the study.
3.2 Child, parent and family characteristics
To what extent do the family forms differ in terms of the financial circumstances of the family, selected socio-demographic characteristics of the parents and children, and indicators of parenting practices and parental wellbeing?
This section focuses on children in the three unchanged forms of families of key interest: married and cohabiting biological families and sole-mother families. The child, parent and family characteristics of these groups are compared (bivariate analysis).
Child’s characteristics
Table 3 shows selected characteristics of children in Wave 1 across the three unchanged family groups. The children in each group were similar in age and gender. In Wave 1, they were 4.8 years old on average and 51 to 53 per cent were boys. While few children were born overseas, those in married biological families were more likely to have been born outside Australia (5 per cent) than those in the other two groups (1 to 2 per cent). Although uncommon across all groups, the children in married families were the least likely to have an Indigenous background (2 per cent as opposed to 8 to 10 per cent). Only a minority spoke a language other than English at home, with children in married biological families being the most likely to do so (16 per cent) and those in cohabiting biological families being the least likely to do so (5 per cent).
While much the same proportion of children in each group had an elder sibling (40 to 44 per cent), children in sole-mother families were less likely to have a younger sibling (32 per cent) than those who were living with both biological parents (49 to 50 per cent).
Table 3: Selected characteristics of children by (unchanged) family form, K cohort, Wave 1 (a)
Married families Cohabitating families Sole-mother families
Age (mean years) 4.80 4.80 4.81
% % %
Sex
Boy 51.0 52.2 52.6
Girl 49.0 47.8 47.4
Total 100.0 100.0 100.0
Country of birth
Australia 94.8 98.6 98.8
Other country 5.2 1.5 ** 1.2 ***
Total 100.0 100.0 100.0
Research OP46_FINAL.indd 13 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4614
Married families Cohabitating families Sole-mother families
Indigenous status
No 98.3 89.7 *** 91.7 ***
Yes 1.7 10.3 8.4
Total 100.0 100.0 100.0
Language at home
English 84.1 94.8 88.7
Other 15.9 5.3 *** 11.3 **
Total 100.0 100.0 100.0
Any older siblings in household
No 39.8 43.6 42.9
Yes 60.2 56.4 57.1
Total 100.0 100.0 100.0
Any younger siblings in household
No 51.1 50.5 67.7 ***
Yes 48.9 49.5 32.3
Total 100.0 100.0 100.0
No. of respondents 3702 436 656
(a) The family forms in this table remained the same across three waves and comprised those in which the children lived with their biological parents. Children in cohabiting and sole-mother families were separately compared with those in married families.
Note: Statistical differences (chi-square test for categorical variables; one-way analysis of variance for continuous variables) are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted.
Parental characteristics
Table 4 presents information about biological mothers’ characteristics for all three unchanged family groups and biological fathers’ characteristics for the two forms of couple families. Some characteristics, such as country of birth and Indigenous status, are fixed and are therefore only presented in Wave 1. Parents’ age is also only presented for Wave 1, whereas their employment status and level of education during each survey wave are shown.
Compared with married biological mothers, cohabiting biological mothers and sole mothers were marginally, but significantly, younger (in Wave 1: mean = 34 years as opposed to 35 years), less likely to have a degree or higher qualification, and more likely to have no qualification at all. For example, 15 per cent of the cohabiting and sole mothers had a degree or higher in Wave 1, compared with 31 per cent of the married mothers. Not surprisingly, this pattern changed little across time.13
Table 3: Selected characteristics of children by (unchanged) family form, K cohort, Wave 1 (a) (continued)
Research OP46_FINAL.indd 14 27/06/12 2:03 PM
15
results
The vast majority of mothers in all three groups were non-Indigenous and most were born in Australia. Nevertheless, compared with the married group, the cohabiting and sole mothers (especially the latter) were slightly more likely to identify as Indigenous (3 to 6 per cent as opposed to 1 per cent) and to be born in Australia (80 to 84 per cent as opposed to 74 per cent).
While sole mothers were significantly less likely than married mothers to be in paid work (Wave 1: 49 per cent as opposed to 60 per cent; Wave 2: 54 per cent as opposed to 67 per cent; Wave 3: 63 per cent as opposed to 75 per cent), there was no significant difference in the employment status of the cohabiting and married mothers except in Wave 3, where the cohabiting mothers were more likely to be non-employed and less likely to be working part time. The employment status of the cohabiting and sole mothers was more alike in Wave 3 than in the other two waves. There was an overall similar employment pattern across the first and second waves, with cohabiting mothers’ employment rates lying between those of married and sole mothers; for example, in Wave 2, 33 per cent of married mothers, 39 per cent of cohabiting mothers, and 46 per cent of sole mothers were not employed. In Wave 3, the employment rates of cohabiting mothers were similar to those of sole mothers, but were lower compared with those of married mothers.
Table 4 (on page 16) shows that cohabiting biological fathers differed from married biological fathers in a number of ways. Specifically, the cohabiting fathers had a lower level of education, were more likely to identify as Indigenous, and were less likely to be employed. For example, 30 per cent of the married fathers and only 15 per cent of the cohabiting fathers had a degree or higher qualification in Wave 1; around 1 per cent of the married fathers and 4 per cent of the cohabiting fathers had an Indigenous background; and 90 to 91 per cent of the married fathers in each survey wave were working full time, compared with 82 to 85 per cent of the cohabiting fathers.
Very similar trends in parental characteristics also emerged when analyses were based on the cross-sectional family groups (that is, when the focus included families that changed their form over the period investigated). The latter results are presented in Appendix Table A2.
Family financial circumstances
The financial circumstances of the three groups of families that did not change in form differed considerably. In each survey wave, primary carers of children (mostly mothers) were asked whether they had experienced any of six financial hardships in the previous 12 months, such as the inability to pay utility bills, or inability to pay the mortgage or rent (see Section 2.1 for complete list). In addition, they were asked whether they were in a ‘comfortable’ or ‘prosperous’ position, whether they were ‘just getting by’, or whether they were ‘poor’ or ‘very poor’. Responses to these two sets of questions are presented in Figure 1 and Figure 2 respectively.
As shown in Figure 1, married biological families were the least likely of the three unchanged family groups to report having experienced any of the six indicators of financial hardship, while sole-mother families were the most likely to report such experiences. This pattern of results was apparent in all the three survey waves. Specifically, at least one financial hardship was reported by 22 per cent of the married respondents in Wave 1, compared with 40 per cent of the cohabiting respondents and 56 per cent of the sole mothers. By Wave 3, the respective proportions were 13 per cent, 17 per cent and 46 per cent. On this measure, then, it would seem that the cohabiting biological families were better off than the sole-mother families, but worse off than the married families.
Research OP46_FINAL.indd 15 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4616
Tabl
e 4:
Se
lect
ed p
aren
tal c
hara
cter
istic
s by
unc
hang
ed fa
mily
form
, K c
ohor
t (a)
Wav
e 1
Wav
e 2
Wav
e 3
Mar
ried
fam
ilies
Coha
bitin
g fa
mili
esSo
le-m
othe
r fa
mili
esM
arrie
d fa
mili
esCo
habi
ting
fam
ilies
Sole
-mot
her
fam
ilies
Mar
ried
fam
ilies
Coha
bitin
g fa
mili
esSo
le-m
othe
r fa
mili
es
Mot
hers
’ cha
ract
eris
tics
Age
(mea
n ye
ars)
35
.434
.3**
34.1
**
%%
%
Educ
atio
n
Deg
ree
or h
ighe
r
31.1
14.7
***
15.2
***
30.7
14.0
***
12.6
***
30.4
13.7
***
12.9
***
Oth
er p
ost-s
choo
l qu
alifi
catio
n
36.0
38.4
41.8
39.1
40.3
49.3
**41
.845
.753
.8**
*
No
qual
ifica
tion
32
.946
.9**
*43
.0**
*30
.245
.7**
*38
.1*
27.8
40.6
**33
.3
Tota
l
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
0
Coun
try
of b
irth
Aust
ralia
73
.580
.083
.8**
*
Oth
er c
ount
ry
26.5
20.0
16.2
***
Tota
l
100.
010
0.0
100.
0
Indi
geno
us s
tatu
s
No
99
.096
.9**
93.8
***
Yes
1.
03.
1**
6.2
***
Tota
l
100.
010
0.0
100.
0
Empl
oym
ent s
tatu
s
Full
time
15.3
15.4
11.8
19.0
15.5
15.1
23.2
20.2
21.7
Part
tim
e45
.040
.336
.9**
48.4
46.0
39.2
**51
.543
.4*
41.0
***
Not
em
ploy
ed39
.744
.451
.4**
*32
.638
.545
.8**
*25
.336
.4**
37.3
***
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
No. o
f obs
erva
tions
3010
208
280
3020
209
284
3020
209
284
Research OP46_FINAL.indd 16 27/06/12 2:03 PM
17
results
Wav
e 1
Wav
e 2
Wav
e 3
Mar
ried
fam
ilies
Coha
bitin
g fa
mili
esSo
le-m
othe
r fa
mili
esM
arrie
d fa
mili
esCo
habi
ting
fam
ilies
Sole
-mot
her
fam
ilies
Mar
ried
fam
ilies
Coha
bitin
g fa
mili
esSo
le-m
othe
r fa
mili
es
Fath
ers’
cha
ract
eris
tics
Age
(mea
n ye
ars)
37.9
37.2 %
%%
Educ
atio
n
Deg
ree
or h
ighe
r30
.415
.4**
*30
.314
.5**
*30
.714
.4**
*
Oth
er p
ost-s
choo
l qu
alifi
catio
n45
.854
.1*
46.9
56.3
*47
.957
.9*
No
qual
ifica
tion
23.8
30.6
22.8
29.3
21.4
27.8
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
0
Coun
try
of b
irth
Oth
er c
ount
ry27
.521
.9
Aust
ralia
72.5
78.1
Tota
l10
0.0
100.
0
Indi
geno
us s
tatu
s
No
99.3
96.2
***
Yes
0.7
3.8
Tota
l10
0.0
100.
0
Empl
oym
ent s
tatu
s
Full
time
89.6
81.9
**89
.784
.3*
90.8
84.5
**
Part
tim
e5.
57.
05.
77.
45.
38.
2
Not
em
ploy
ed4.
911
.1**
*4.
68.
3*
3.8
7.3
*
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
0
No. o
f res
pond
ents
3003
207
3018
209
3020
209
(a)
The
fam
ily fo
rms
in th
is ta
ble
rem
aine
d th
e sa
me
acro
ss th
e th
ree
wav
es a
nd c
ompr
ised
thos
e in
whi
ch th
e ch
ildre
n liv
ed w
ith th
eir b
iolo
gica
l par
ents
. For
bot
h m
othe
rs a
nd fa
ther
s,
coha
bitin
g pa
rent
s an
d so
le m
othe
rs w
ere
com
pare
d w
ith m
arrie
d pa
rent
s.
Not
e: S
tatis
tical
diff
eren
ces
(chi
-squ
are
test
for c
ateg
oric
al v
aria
bles
and
one
-way
AN
OVA
for c
ontin
uous
var
iabl
es) a
re s
how
n as
: * p
< 0.
05, *
* p
< 0.0
1, *
** p
< 0.
001.
Dat
a ha
ve b
een
wei
ghte
d.
Tabl
e 4:
Se
lect
ed p
aren
tal c
hara
cter
istic
s by
unc
hang
ed fa
mily
form
, K c
ohor
t (a) (
cont
inue
d)
Research OP46_FINAL.indd 17 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4618
Figure 1: Experience of one or more financial hardships, by (unchanged) family form, primary carers’ reports, Waves 1 to 3
0%
20%
40%
60%
80%
Marriedfamilies
Cohabitatingfamilies
Wave 1
Sole-motherfamilies
Marriedfamilies
Cohabitatingfamilies
Wave 2
Sole-motherfamilies
Marriedfamilies
Cohabitatingfamilies
Wave 3
Sole-motherfamilies
Note: The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents.
Consistent with the patterns concerning financial hardship, married biological respondents were the most likely of the three groups to describe their financial circumstances as being ‘reasonably comfortable’ or ‘prosperous’, and sole mothers were the least likely to do so. Conversely, sole mothers were the most likely to indicate that they were ‘just getting by’, and married respondents were the least likely to state this. While few respondents considered that they were ‘poor’ or ‘very poor’, sole mothers more commonly made such claims than did cohabiting or married respondents. For example, in Wave 1, 42 per cent of sole mothers, 64 per cent of cohabiting respondents and 69 per cent of married respondents described their circumstances as being ‘reasonably comfortable’, ‘comfortable’ or ‘prosperous’, while the view that they were ‘just getting by’ was expressed by 48 per cent, 32 per cent, and 30 per cent respectively, and reports that they were ‘poor’ or ‘very poor’ were provided by 10 per cent, 4 per cent and 2 per cent respectively. Much the same results emerged when the analyses were based on the cross-sectional family groups (see Appendix Table A3).
Figure 2: Ratings of families’ financial situation, by (unchanged) family form, primary carers’ reports, Waves 1 to 3
0%
20%
40%
60%
100%
80%
Marriedfamilies
Cohabitatingfamilies
Wave 1
Sole-motherfamilies
Marriedfamilies
Cohabitatingfamilies
Sole-motherfamilies
Marriedfamilies
Cohabitatingfamilies
Sole-motherfamilies
Wave 2 Wave 3
Comfortable to prosperous Getting by Poor or very poor
Note: The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents.
Research OP46_FINAL.indd 18 27/06/12 2:03 PM
19
results
Parenting self-efficacy
Parents were asked to indicate how good a parent they were, compared with other parents. This item was used as an indicator of parenting self-efficacy. The patterns of responses of the married, cohabiting and sole mothers in the unchanged family forms are depicted in Figure 3, while those of married and cohabiting fathers are depicted in Figure 4.14
Regardless of family form, mothers most commonly considered themselves to be ‘very good’ or ‘better than average’ parents, with 51 to 73 per cent holding this view across the three survey waves. Most of the other mothers considered themselves to be ‘average’ parents, and only a very small percentage rated themselves as ‘not very good’ parents or ‘had some trouble’ being parents.
However, systematic differences in sense of parenting efficacy were apparent across the three unchanged family groups. Favourable self-assessments (that is, that they were a very good or better than average parent) were provided by a marginally higher proportion of married than cohabiting mothers (65 per cent as opposed to 61 per cent in Wave 1), and by a lower proportion of sole mothers (55 per cent in Wave 1). Similar trends were apparent in the other two survey waves. Conversely, a slightly higher proportion of sole mothers than married or cohabiting mothers in each survey wave provided the most unfavourable self-assessments (that is, they were not very good parents or had some trouble being parents). For example, in Wave 1, this was reported by 7 per cent of sole mothers compared to 3 to 4 per cent of couple mothers.
Figure 3: Mothers’ ratings of self as a parent, by (unchanged) family form, Waves 1 to 3
0%
20%
40%
60%
100%
80%
Marriedmothers
Cohabitatingmothers
Wave 1
Solemothers
Marriedmothers
Cohabitatingmothers
Solemothers
Marriedmothers
Cohabitatingmothers
Solemothers
Wave 2 Wave 3
A very good or better than average parent An average parent Not very good or had some trouble being a parent
Note: The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents.
Married and cohabiting fathers were also inclined to hold favourable views of themselves as parents, with 47 to 70 per cent seeing themselves as very good or better than average parents, and 28 to 46 per cent believing that they were average parents across the three survey waves. Nevertheless, cohabiting fathers provided consistently less positive self-assessments compared with married fathers. For example, 68 per cent of married fathers in Wave 1 felt that they were very good or better than average parents, compared with 47 per cent of cohabiting fathers. While 3 per cent of married fathers provided negative self-assessments, 7 per cent of cohabiting fathers shared this view.
Research OP46_FINAL.indd 19 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4620
Figure 4: Fathers’ ratings of self as a parent by (unchanged) family form, Waves 1 to 3
0%
20%
40%
60%
100%
80%
Marriedfathers
Cohabitatingfathers
Wave 1
Marriedfathers
Cohabitatingfathers
Wave 2
Marriedfathers
Cohabitatingfathers
Wave 3
A very good or better than average parent An average parent
Not very good or had some trouble being a parent
Note: The family forms in this table remained the same across the three waves and comprised the biological parents of the child.
Overall, then, married mothers and fathers provided similar (and favourable) self-assessments, with 65 to 68 per cent believing that they were very good or above average parents in Wave 1. However, cohabiting fathers were less likely than cohabiting mothers to hold such views (47 per cent as opposed to 61 per cent in Wave 1).
Parenting practices
As outlined earlier, parents were asked a series of questions designed to tap their tendency to ‘parent’ their children in a warm manner; parent their children in a hostile manner; reason with their child when they corrected them or when their child misbehaved; and handle their child’s behaviour in a consistent manner (for example, in dealing with misbehaviour and in ensuring that their child followed their directions). Each of these measures was based on ratings of the frequency with which the parents engaged in the behaviour in question (on a five-point scale). Table 5 depicts the mean scores of mothers in the three unchanged family forms and those of fathers in the two unchanged couple family forms.
Table 5: Mean scores of parenting approach scales, fathers and mothers, by unchanged family form, Waves 1 to 3
Mothers Fathers
Married mothers
Cohabiting mothers
Sole mothersMarried fathers
Cohabiting fathers
Wave 1
Warm parenting a 4.43 4.42 4.47 4.09 4.02
Hostile parenting b 2.17 2.23 2.20 2.28 2.33
Use of reasoning c 4.25 4.25 4.27 3.98 3.91
Consistency of parenting d 4.10 3.92 *** 3.84 *** 3.99 3.75 ***
Wave 2
Warm parenting a 4.44 4.43 4.46 4.12 4.14
Hostile parenting b 2.16 2.20 2.28 2.08 2.10
Research OP46_FINAL.indd 20 27/06/12 2:03 PM
21
results
Mothers Fathers
Married mothers
Cohabiting mothers
Sole mothersMarried fathers
Cohabiting fathers
Use of reasoning c 4.22 4.18 4.21 3.95 3.91
Consistency of parenting d 4.17 4.01 *** 3.95 *** 4.09 3.91 **
Wave 3
Warm parenting a 4.34 4.29 4.37 * 4.06 4.12
Hostile parenting b 2.13 2.21 2.23 2.16 2.18
Use of reasoning c 4.14 4.08 4.13 3.91 4.09 **
Consistency of parenting d 4.20 3.93 *** 3.93 *** 4.10 3.99
No. of observations 3020 208 284 3020 208
Note: All scores are on a five-point scale. a Higher scores = greater warmth. b Higher scores = higher hostility.
c Higher scores = higher use of reasoning. d Higher scores = greater consistency. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. For both mothers and fathers, cohabiting parents and sole mothers were separately compared with married parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted. Missing data result in slight variations in numbers of observations across waves and scales.
Across the three waves, mothers reported high warmth, low hostility, high use of reasoning and also generally high consistency in the way they parented their child. These patterns were also apparent among the reports of fathers. Across the three survey waves, mean scores across all groups of mothers and fathers respectively varied as follows:15
◗◗ warmth: 4.29 to 4.47 and 4.02 to 4.14
◗◗ hostility: 2.13 to 2.28 and 2.08 to 2.33
◗◗ reasoning: 4.08 to 4.27 and 3.91 to 4.09
◗◗ consistency: 3.84 to 4.20 and 3.75 to 4.10.
Table 5 also shows that, regardless of the survey wave, there was little difference across the three groups of mothers in their reports on the frequency with which they expressed warmth or hostility and the frequency with which they used reasoning. However, in all three waves the cohabiting mothers and sole mothers reported significantly lower consistency in their parenting. Mean scores on this measure were 4.10 in Wave 1, 4.17 in Wave 2 and 4.20 in Wave 3 for married mothers, 3.92 in Wave 1, 4.01 in Wave 2 and 3.93 in Wave 3 for cohabiting mothers, and 3.83 in Wave 1, 3.95 in Wave 2 and 3.93 in Wave 3 for sole mothers.
The reports of cohabiting and married (biological) fathers were similar in relation to the frequency with which they adopted a warm or hostile approach. However, cohabiting fathers were more likely than married fathers to indicate frequent use of reasoning in Wave 3 (mean scores: 4.09 as opposed to 3.91), with ratings on this measure not varying significantly in Waves 1 and 2. Across each survey wave, cohabiting fathers indicated that they were less likely to adopt a consistent approach to parenting compared with married fathers (Wave 1: 3.75 as opposed to 3.99; Wave 2: 3.91 as opposed to 4.09; Wave 3: 3.99 as opposed to 4.10), though the difference was not statistically significant in Wave 3. The results based on the cross-sectional family groups are presented in Appendix Table A4.
Table 5: Mean scores of parenting approach scales, fathers and mothers, by unchanged family form, Waves 1 to 3 (continued)
Research OP46_FINAL.indd 21 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4622
Parental psychological wellbeing and quality of couple relationships
The measure of parental emotional distress indicates how often, during the previous four weeks, parents had felt nervous, hopeless, restless/fidgety, worthless, and so sad that nothing could cheer them up, and how often they felt that everything was an effort (that is, six ratings). Each parent’s mean rating (ranging from 1 to 5) was derived, and higher mean scores reflected higher average distress. Table 6 shows that the overall mean scores for all groups of mothers and fathers were below the midpoint of the scale, suggesting generally limited experiences of distress.
Across the three waves, sole mothers indicated significantly higher psychological distress than married mothers (mean scores: Wave 1, 1.96 as opposed to 1.62; Wave 2, 1.85 as opposed to 1.49; Wave 3, 1.86 as opposed to 1.54), while cohabiting mothers indicated significantly higher distress than married mothers in Wave 1 only (mean scores: 1.73 as opposed to 1.62). Similarly, cohabiting fathers indicated significantly higher distress than married fathers in Wave 1 (mean scores: 1.69 as opposed to 1.55), but not in subsequent waves.
Both mothers and fathers tended to indicate that they had favourable inter-parental relationships (as measured by overall happiness and the frequency with which they tended to relate to each other in argumentative, hostile ways). Nevertheless, cohabiting parents provided significantly less favourable assessments than married parents. Mean ratings on the relationship happiness scale were 4.88 in Wave 1, 4.83 in Wave 2 and 4.75 in Wave 3 for cohabiting mothers, and 5.39 in Wave 1, 5.37 in Wave 2 and 5.24 in Wave 3 for married mothers. Mean ratings were 4.85 in Wave 1, 4.81 in Wave 2 and 4.79 in Wave 3 for cohabiting fathers, while they were 5.47 in Wave 1, 5.36 in Wave 2 and 5.44 in Wave 3 for married fathers. Mean scores on the argumentative relationship scale were 2.21 in Wave 1, 2.03 in Wave 2 and 2.12 in Wave 3 for cohabiting mothers, and 1.98 in Wave 1, 1.82 in Wave 2 and 1.89 in Wave 3 for married mothers. For cohabiting fathers, mean scores were 2.29 in Wave 1, and 2.07 in the two later waves, while for married fathers mean scores were 1.96 in Wave 1, 1.87 in Wave 2 and 1.84 in Wave 3. Again, the same patterns emerged when the analyses were based on the cross-sectional family groups (see Appendix Table A5).
Table 6: Mean scores of psychological distress scale and quality of parental relationship of fathers and mothers by unchanged family form, Waves 1 to 3
Mothers Fathers
Married mothers
Cohabiting mothers
Sole mothers
Married fathers
Cohabiting fathers
Wave 1
Psychological distress (Kessler 6) a 1.62 1.73 * 1.96 *** 1.55 1.69 *
Degree of happiness in relationship with the partner b 5.39 4.88 *** 5.47 4.85 ***
Argumentative relationship scale c 1.98 2.21 *** 1.96 2.29 ***
Wave 2
Psychological distress (Kessler 6) a 1.49 1.57 1.85 *** 1.52 1.59
Degree of happiness in relationship with the partner b 5.37 4.83 *** 5.36 4.81 ***
Argumentative relationship scale c 1.82 2.03 *** 1.87 2.07 **
Research OP46_FINAL.indd 22 27/06/12 2:03 PM
23
results
Mothers Fathers
Married mothers
Cohabiting mothers
Sole mothers
Married fathers
Cohabiting fathers
Wave 3
Psychological distress (Kessler 6) a 1.54 1.65 1.86 *** 1.52 1.62
Degree of happiness in relationship with the partner b 5.24 4.75 *** 5.44 4.79 ***
Argumentative relationship scale c 1.89 2.12 *** 1.84 2.07 **
No of observations 3020 208 284 3020 208
Note: a Scale of 1 to 5, higher scores = higher distress. b Scale of 1–7, 1 = extremely unhappy, 7 = perfectly happy.
c Scale of 1 to 5, higher scores = more prone to argument. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. For both mothers and fathers, cohabiting parents and sole mothers were compared with married parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted. Missing data result in slight variations in numbers of observations across waves and scales.
3.3 Children’s wellbeing
This section first compares the wellbeing or developmental progress of children in the three unchanged family forms. The analyses focus on the range of indicators of wellbeing outlined in Section 2.5. One of the measures tapped the children’s receptive language skills (based on the PPVT that was administered to the children); nine were based on the assessments of the primary caregiver; and assessments for five of these nine were also provided by one of the child’s teachers. The extent to which any differences in the wellbeing of the children in the three family forms can be explained by systematic differences between the family forms (noted in Section 3.1) is then examined. Once again, attention is directed to the children in the three unchanged family forms, while the results for the ‘cross-sectional’ group (some of whom experienced changes in family form) appear in Appendix Tables A6 and A7.
Wellbeing of children in each family form
To what extent do the children in the different family forms differ in terms of their developmental progress?
To what extent, if at all, do any differences in child wellbeing between the groups diverge or diminish over the four-year period?
Table 7 provides the children’s mean scores on the PPVT, along with various assessments relating to the children’s wellbeing provided by the primary caregiver (usually the mother). These assessments formed part of the outcome indices covering the child’s physical, social–emotional, learning and overall development. In addition, the respondents’ assessments relating to five aspects of the children’s social–emotional development (sub-scales of the SDQ) are presented. The latter five assessments provided by teachers are summarised in Table 8.
Overall, the primary caregiver in married biological families tended to provide the most favourable assessments of children’s wellbeing, while sole mothers tended to provide the least favourable assessments. These differences were statistically significant with the following exceptions: the SDQ sub-scale for prosocial behaviour in Wave 1 and the outcome index for physical development in Wave 2. Across the three waves, the PPVT average scores for children in sole-mother families were also significantly lower than those for children in married biological families, with the difference being most marked in Wave 3.
Table 6: Mean scores of psychological distress scale and quality of parental relationship of fathers and mothers by unchanged family form, Waves 1 to 3 (continued)
Research OP46_FINAL.indd 23 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4624
Table 7: Child outcomes by family form, primary caregiver reports, Waves 1 to 3
Married familiesCohabiting
familiesSole-mother
families
Wave 1
PPVT score a 64.5 64.3 63.2 *
Outcome indices
Physical development index a 100.4 98.9 98.4 **
Social–emotional development index a 101.2 98.6 ** 96.8 ***
Learning index a 101.4 100.5 98.1 ***
Overall development index a 101.4 99.0 ** 96.8 ***
SDQ social–emotional sub-scales
Prosocial behaviour a 7.8 7.6 7.6
Peer problems b 1.6 1.8 * 2.1 ***
Emotional problems b 1.6 1.8 2.1 ***
Hyperactivity problems b 3.4 3.7 4.0 ***
Conduct problems b, c 2.3 3.1 *** 3.2 ***
Wave 2
PPVT score a 73.8 73.9 73.0 *
Outcome indices
Physical development index a 100.3 98.9 99.4
Social–emotional development index a 101.2 98.8 ** 96.0 ***
Learning index a 101.2 99.5 * 97.2 ***
Overall development index a 101.3 98.6 ** 96.4 ***
SDQ social–emotional sub-scales
Prosocial behaviour a 8.3 8.0 7.9 **
Peer problems b 1.4 1.8 ** 2.1 ***
Emotional problems b 1.6 1.8 2.2 ***
Hyperactivity problems b 3.2 3.6 4.0 ***
Conduct problems b, c 1.4 1.7 * 2.1 ***
Wave 3
PPVT score 78.2 78.3 77.1 *
Outcome indices
Physical development index a 100.7 99.0 * 97.6 ***
Social–emotional development index a 101.2 99.2 * 95.5 ***
Research OP46_FINAL.indd 24 27/06/12 2:03 PM
25
results
Married familiesCohabiting
familiesSole-mother
families
Learning index a 101.0 99.0 * 97.1 ***
Overall development index a 101.3 98.7 ** 95.4 ***
SDQ social–emotional sub-scales
Prosocial behaviour a 8.3 8.5 8.0 *
Peer problems b 1.4 1.8 ** 2.1 ***
Emotional problems b 1.5 1.8 * 2.4 ***
Hyperactivity problems b 3.1 3.4 4.1 ***
Conduct problems b, c 1.2 1.6 ** 1.8 ***
Note: a Higher score = better outcome. b Higher score = worse outcome. c Items that formed this scale were the same in Wave 2 and Wave 3, but differed in Wave 1. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. For both mothers and fathers, cohabiting parents and sole mothers were compared with married parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted. Missing data result in slight variations in numbers of observations across waves and scales.
As shown in Table 8, the teachers also tended to assess the developmental progress of children in sole-mother families less favourably than that of children who were living in married biological families. The differences in assessment were statistically significant for four of the five measures examined in each survey wave. The exception concerned the peer problems measure, where differences in ratings remained non-significant in each survey wave.
Table 8: Child SDQ outcomes by unchanged family form, teacher reports, Waves 1 to 3
Married familiesCohabiting
familiesSole-mother
families
Wave 1
Prosocial behaviour a 7.3 7.2 6.9 **
Peer problems b 2.5 2.7 3.0
Emotional problems b 1.3 1.5 1.6 **
Hyperactivity problems b 2.9 3.3 3.6 ***
Conduct problems b, c 1.7 1.9 2.2 ***
Wave 2
Prosocial behaviour a 7.7 7.5 6.9 ***
Peer problems b 1.2 1.4 1.9
Emotional problems b 1.2 1.4 1.7 **
Hyperactivity problems b 2.7 2.8 3.8 ***
Conduct problems b, c 0.7 0.9 1.4 ***
Table 7: Child outcomes by family form, primary caregiver reports, Waves 1 to 3 (continued)
Research OP46_FINAL.indd 25 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4626
Married familiesCohabiting
familiesSole-mother
families
Wave 3
Prosocial behaviour a 7.7 7.6 6.9 ***
Peer problems b 1.2 1.8 2.0
Emotional problems b 1.1 1.6 * 1.9 ***
Hyperactivity problems b 2.7 3.0 3.8 ***
Conduct problems b, c 0.7 0.9 1.4 ***
Note: a Higher score = better outcome. b Higher score = worse outcome. c Items that formed this scale were the same in Wave 2 and Wave 3, but differed in Wave 1. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. For both mothers and fathers, cohabiting parents and sole mothers were compared with married parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted. Missing data result in slight variations in numbers of observations across waves and scales.
The children who were living with cohabiting biological parents fared less well in some domains than their counterparts whose parents were married, according to the primary caregivers’ assessments (Table 7). Differences were statistically significant for four measures in Wave 1 (social–emotional and overall development indices and the SDQ sub-scales concerning peer problems and conduct problems), five measures in Wave 2 (social–emotional, learning and overall development indices, and the peer problems and conduct problems sub-scales), and seven measures in Wave 3 (all four development indices, along with the sub-scales covering peer, emotional and conduct problems). However, the PPVT scores of the children in cohabiting and married families did not differ significantly in any of the three survey waves. While teachers’ assessments appeared to be slightly less favourable for children in cohabiting than married biological families, only one difference was statistically significant (emotional problems in Wave 3) (Table 8). Therefore, by Wave 3, the children with married parents appeared to outperform the children with cohabiting parents on eight measures, seven of which were based on the primary caregivers’ assessments.
In summary, both the primary caregivers’ and teachers’ reports suggest that children in sole-mother families were progressing less well in virtually all measures in each survey wave. Especially in Wave 3, the PPVT measure of receptive language skills administered to the children themselves also suggests that those in sole-mother families were not doing as well as those living with their married parents.
In addition, the children in cohabiting biological families were faring less well than those in married families in some areas, although this trend derived almost exclusively from the assessments provided by the primary caregivers and was not apparent in the children’s PPVT scores or from most teacher assessments. Overall, the children in married biological families outperformed those in cohabiting biological families on an increasing number of measures over the four-year period (with significant differences emerging for four measures in Wave 1, five in Wave 2, and eight in Wave 3).
On the whole, children from cohabiting families with two biological parents seemed to be progressing better than those in sole-mother families.16 In Wave 1, children in cohabiting families appeared to fare significantly better than those in sole-mother families in the following areas: social–emotional developmental progress, learning, and peer relationships, as reported by the primary caregiver. In Wave 2, apparent superior progress of children in cohabiting families emerged in relation to the primary caregivers’ assessments of learning; social–emotional and overall developmental progress; peer relationships; emotional expression and conduct; and in relation to teachers’ assessments of prosocial behaviour, peer relationships, level of hyperactivity and conduct issues. In Wave 3, the children in cohabiting families continued to outperform those in sole-mother families, as reflected in: (a) parents’ assessments of learning, social–emotional and overall developmental indices and all SDQ sub-scales except the peer problems measure; (b) teachers’
Table 8: Child SDQ outcomes by unchanged family form, teacher reports, Waves 1 to 3 (continued)
Research OP46_FINAL.indd 26 27/06/12 2:03 PM
27
results
assessments of three of the five SDQ sub-scales; and (c) the children’s PPVT test scores. In other words, children living with their cohabiting parents appeared to be progressing better than those in sole-mother families, with differences being more apparent in Waves 2 and 3 than in Wave 1. This suggests a process of divergence over the four-year period.
Explanations for differences in children’s wellbeing: Multivariate analysis
If differences in child wellbeing are apparent, to what extent can they be explained by any observed systematic differences in child and family characteristics, or parenting practices and parental wellbeing?
The above analysis suggests that the developmental progress of children who were living with both married biological parents tended to be superior to that of children in the other two groups, especially those living with their sole mother. Furthermore, it seems that children who were living with both biological parents tended to outperform those in sole-mother families across a range of areas, especially in Waves 2 and 3. Overall, differences were apparent between the children in sole-mother and married families in most areas examined in all survey waves, while those between children with married and cohabiting parents appeared to become more pronounced over the four-year period.
However, these three groups of children differed systematically in terms of their family environments, which appeared to be most favourable for children in married parent families and least for those in sole-mother families. This appeared to relate to issues such as the families’ apparent financial circumstances, and the resident parents’ educational attainment level and apparent consistency in ensuring that their children behaved appropriately. In regard to these issues, the family environments of cohabiting families seemed to lie between the two other family types.
This section shows the extent to which such systematic variation in the family environments of the children in the three family forms explained their apparent differences in developmental progress. Once again, attention is directed to children who did not experience a change in family form across the survey waves. The ‘cross-sectional’ results (covering all children, regardless of whether they had experienced a change in family form during the four-year period) appear in Appendix Tables A8 to A11. These results were very similar to those outlined for the children who did not experience any change in family form.
As for the above-mentioned bivariate analysis, 15 measures of children’s developmental progress were examined: the PPVT administered to the child, nine measures (outcome indices and SDQ sub-scales) based on primary caregivers’ reports, and five SDQ sub-scale measures based on teachers’ reports. For each of these measures, two regressions were performed, with married families representing the reference group:
◗◗ base model—where the explanatory variables included the child’s characteristics (age, sex, country of birth, Indigenous status, and presence of older siblings or younger siblings [assessed separately])
◗◗ full model—where the explanatory variables included the child’s characteristics from the base model, along with demographic characteristics of the mother (age, education, country of birth, Indigenous status, and employment status); apparent financial circumstances of the family (experience of any of the six financial hardships and level of prosperity, as reported by the primary caregiver); mother’s self-efficacy as a parent; her self-assessed parenting practices; and the level of her emotional distress. The two measures of the quality of the inter-parental relationship were not included in this model, given that these measures were not applicable to sole mothers.
Results for the base model
The regression results relating to the base model (which controlled for the children’s characteristics only) are presented in Table 9, with the top panel focusing on the primary caregivers’ assessments and the child’s PPVT test scores, and the lower panel focusing on the teachers’ assessments. The full set of results appears in Appendix Tables A12 and A13.
Research OP46_FINAL.indd 27 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4628
Tabl
e 9:
Re
gres
sion
s of
chi
ld o
utco
mes
, bas
e m
odel
PPVT
sco
re
Out
com
e in
dice
s (a
)SD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
velo
pmen
t in
dex
Soci
al–
emot
iona
l de
velo
pmen
t in
dex
Lear
ning
in
dex
Ove
rall
deve
lopm
ent
inde
x
Pros
ocia
l be
havi
our
Peer
pr
oble
ms
Emot
iona
l pr
oble
ms
Hype
ract
ivity
pr
oble
ms
Cond
uct
prob
lem
s (a
)
Prim
ary
care
give
rs’ r
epor
ts
Wav
e 1
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es–0
.469
–1.2
35–2
.376
**–0
.601
–2.1
02*
–0.1
510.
249
*0.
148
0.24
80.
834
***
Sole
-mot
her
fam
ilies
–1.5
93**
*–1
.79
*–4
.277
***
–3.2
24**
*–4
.428
***
–0.2
130.
447
***
0.46
**0.
69**
*0.
961
***
Wav
e 2
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es–0
.124
–0.9
38–2
.405
**–1
.83
*–2
.546
**–0
.219
0.34
4**
0.23
70.
345
0.34
3*
Sole
-mot
her
fam
ilies
–1.0
29**
–0.4
3–5
.448
***
–4.0
95**
*–4
.806
***
–0.3
75**
0.67
5**
*0.
677
***
0.68
3**
*0.
806
***
Wav
e 3
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es0.
121
–1.2
68–2
.053
*–2
.198
*–2
.547
**0.
226
0.40
4**
0.31
60.
41*
0.36
6*
Sole
-mot
her
fam
ilies
–0.9
83*
–3.3
5**
*–5
.783
***
–3.3
46**
*–5
.703
***
–0.3
69**
0.70
3**
*0.
892
***
0.92
6**
*0.
66**
*
Research OP46_FINAL.indd 28 27/06/12 2:03 PM
29
results
PPVT
sco
re
Out
com
e in
dice
s (a
)SD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
velo
pmen
t in
dex
Soci
al–
emot
iona
l de
velo
pmen
t in
dex
Lear
ning
in
dex
Ove
rall
deve
lopm
ent
inde
x
Pros
ocia
l be
havi
our
Peer
pr
oble
ms
Emot
iona
l pr
oble
ms
Hype
ract
ivity
pr
oble
ms
Cond
uct
prob
lem
s (a
)
Teac
hers
’ rep
orts
Wav
e 1
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es–0
.031
0.07
50.
186
0.22
60.
088
Sole
-mot
her
fam
ilies
–0.3
010.
462
**0.
365
**0.
500
**0.
547
***
Wav
e 2
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es–0
.185
0.17
50.
145
–0.0
080.
151
Sole
-mot
her
fam
ilies
–0.7
63**
*0.
695
***
0.47
1**
0.94
0**
*0.
681
***
Wav
e 3
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es–0
.196
0.53
5**
0.48
4*
0.25
80.
193
Sole
-mot
her
fam
ilies
–0.6
77**
*0.
725
***
0.72
0**
*1.
137
***
0.72
3**
*
(a)
Thes
e m
easu
res
wer
e on
ly g
ener
ated
bas
ed o
n pa
rent
s’ re
port
s.N
ote:
Ite
ms
that
form
ed th
is s
cale
wer
e th
e sa
me
in W
ave
2 an
d W
ave
3, b
ut d
iffer
ed in
Wav
e 1.
The
fam
ily fo
rms
in th
is ta
ble
rem
aine
d th
e sa
me
acro
ss th
e th
ree
wav
es a
nd c
ompr
ised
thos
e in
w
hich
the
child
ren
lived
with
thei
r bio
logi
cal p
aren
ts..
Stat
istic
al d
iffer
ence
s ar
e sh
own
as: *
p <
0.05
, **
p < 0
.01,
***
p <
0.00
1. D
ata
have
bee
n w
eigh
ted.
Tabl
e 9:
Re
gres
sion
s of
chi
ld o
utco
mes
, bas
e m
odel
(con
tinue
d)
Research OP46_FINAL.indd 29 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4630
When the children’s characteristics were controlled, those in cohabiting families appeared to be progressing less well than those in married families, according to four measures in Wave 1, five measures in Wave 2, and eight measures in Wave 3. On the other hand, the developmental progress of children in sole-mother families appeared to be worse across all 15 measures, with two exceptions: prosocial behaviour as assessed by teachers in Wave 1 and physical development (which only the primary caregiver has assessed) in Wave 2. These patterns of results were consistent with those based on the bivariate analysis discussed above.
Results for full model
Table 10 presents the results of the regression analyses based on the full model, which, as noted above, controlled not only for children’s characteristics, but also for parental characteristics, apparent family financial circumstances, and mothers’ psychological distress and parenting practices.
For nine measures in Wave 1, five in Wave 2 and four in Wave 3, the differences in the progress between children with married parents and those with sole mothers were not statistically significant, according to primary caregivers’ assessments. Nevertheless, compared with the children with married parents, those in sole-mother families appeared to be progressing less well in relation to: (a) conduct in Wave 1, as assessed by the primary caregiver; (b) eight measures in Wave 2 (five assessed by the primary caregiver and three by a teacher); and (c) 11 measures in Wave 3 (six assessed by the primary caregiver and all five assessed by a teacher).
On the other hand, differences in the progress of children in cohabiting families and married families had largely disappeared. For parents’ reports, significant differences continued to exist for only one measure in Wave 1 and two measures in Wave 3. For teachers’ reports, no measures remained statistically significant in Waves 1 or 2 and only one measure was statistically significant in Wave 3. Specifically, compared with the children with married parents, those with cohabiting parents were reported by the primary caregiver to fare less well in Wave 1 only in terms of conduct. No significant differences between these groups emerged from primary caregivers’ reports in Wave 2. Furthermore, while they differed significantly on three measures in Wave 3 (PPVT score, primary caregivers’ assessments of prosocial behaviour, and teachers’ assessments of peer relationships), for two of these measures (the child’s PPVT score and the SDQ prosocial sub-scale, as assessed by the primary caregiver), the children with cohabiting parents appeared to be doing better than those with married parents, but worse in terms of peer relationships, according to the teachers.
Further analysis suggested that the better performance of children with cohabiting than married parents regarding prosocial behaviour (as assessed by the primary caregiver) and receptive language skills was not apparent when the demographic characteristics of both the children and mothers were controlled (see Appendix Tables A16 and A17 for results). Once parenting approaches were introduced, children with cohabiting parents were better in receptive language and prosocial behaviours. (The early analysis on parenting approaches shows that cohabiting parents were less consistent in their parenting than married parents.)
In summary, the measures of family environment (including parenting practices) appeared to explain most of the Wave 1 developmental progress differences between the children in married families on the one hand, and those in cohabiting and lone-parent families on the other. However, these control measures were increasingly ‘unable’ to explain differences between those living with married parents and those living with a sole mother across the four-year period. Few differences were apparent between those with married and cohabiting parents net of the family environment factors than that assessed, and in Wave 3, the children with cohabiting parents appeared to be doing better, net of the factors, in receptive language skills and prosocial behaviour, although the teachers’ assessments of their peer relations suggest that they were not doing as well as those in married families.
Research OP46_FINAL.indd 30 27/06/12 2:03 PM
31
results
Tabl
e 10
: Re
gres
sion
s of
chi
ld o
utco
mes
, ful
l mod
el
PPVT
sco
re
Out
com
e in
dice
s (a
)SD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
velo
pmen
t in
dex
Soci
al–
emot
iona
l de
velo
pmen
t in
dex
Lear
ning
in
dex
Ove
rall
deve
lopm
ent
inde
x
Pros
ocia
l be
havi
our
Peer
pr
oble
ms
Emot
iona
l pr
oble
ms
Hype
ract
ivity
pr
oble
ms
Cond
uct
prob
lem
s a
Prim
ary
care
give
rs’ r
epor
ts
Wav
e 1
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es0.
159
–0.4
96–0
.666
0.43
4–0
.425
–0.0
560.
054
–0.0
48–0
.098
0.57
0**
*
Sole
-mot
her
fam
ilies
–0.5
150.
240
–1.0
12–1
.378
–1.0
76–0
.045
0.07
00.
030
0.10
60.
498
**
Wav
e 2
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es0.
486
0.03
2–0
.913
–0.4
13–0
.721
–0.0
980.
186
0.06
60.
079
0.16
5
Sole
-mot
her
fam
ilies
–0.0
561.
963
**–1
.765
**–1
.462
*–0
.672
–0.1
510.
267
*0.
158
0.09
50.
428
***
Wav
e 3
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es0.
777
*–0
.042
–0.5
40–0
.595
–0.5
460.
357
**0.
203
0.15
20.
123
0.16
8
Sole
-mot
her
fam
ilies
–0.1
12–0
.641
–2.5
78**
*–0
.916
–1.8
76*
–0.2
010.
286
*0.
462
**0.
350
*0.
232
*
Research OP46_FINAL.indd 31 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4632
PPVT
sco
re
Out
com
e in
dice
s (a
)SD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
velo
pmen
t in
dex
Soci
al–
emot
iona
l de
velo
pmen
t in
dex
Lear
ning
in
dex
Ove
rall
deve
lopm
ent
inde
x
Pros
ocia
l be
havi
our
Peer
pr
oble
ms
Emot
iona
l pr
oble
ms
Hype
ract
ivity
pr
oble
ms
Cond
uct
prob
lem
s a
Teac
hers
’ rep
orts
Wav
e 1
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es0.
085
–0.0
210.
064
0.01
–0.0
85
Sole
-mot
her
fam
ilies
–0.0
960.
242
0.14
0.09
60.
261
Wav
e 2
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es–0
.068
0.10
10.
047
–0.2
030.
051
Sole
-mot
her
fam
ilies
–0.4
53*
0.49
6**
*0.
188
0.38
20.
398
**
Wav
e 3
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es–0
.034
0.43
7*
0.37
60.
004
0.05
7
Sole
-mot
her
fam
ilies
–0.3
88*
0.52
4**
0.49
1*
0.63
1*
0.43
1*
(a)
Thes
e m
easu
res
wer
e on
ly g
ener
ated
bas
ed o
n pa
rent
s’ re
port
s.N
ote:
Ite
ms
that
form
ed th
is s
cale
wer
e th
e sa
me
in W
ave
2 an
d W
ave
3, b
ut d
iffer
ed in
Wav
e 1.
The
fam
ily fo
rms
in th
is ta
ble
rem
aine
d th
e sa
me
acro
ss th
e th
ree
wav
es a
nd c
ompr
ised
thos
e in
w
hich
the
child
ren
lived
with
thei
r bio
logi
cal p
aren
ts. S
tatis
tical
diff
eren
ces
are
show
n as
: * p
< 0.
05, *
* p
< 0.0
1, *
** p
< 0.
001.
Dat
a ha
ve b
een
wei
ghte
d.
Tabl
e 10
: Re
gres
sion
s of
chi
ld o
utco
mes
, ful
l mod
el (c
ontin
ued)
Research OP46_FINAL.indd 32 27/06/12 2:03 PM
33
results
Significant predictors
The results of the multivariate analysis (presented in Appendix Tables A14 and A15) suggest that the boys were progressing less well than the girls, net of all measures in the full model. These patterns were consistent across the survey waves, though fewer differences were statistically significant in Wave 3 than in the previous two waves. Children whose mother had a degree or higher qualification appeared to be progressing significantly better than other children across most measures for all three waves. Not surprisingly, better family financial circumstances were associated with superior child progress: in each survey wave, the children in families with more favourable financial circumstances (as measured by one or both indicators) appeared to be significantly outperforming other children in terms of around half the measures. It is also not surprising that mothers’ level of emotional distress was negatively associated with their child’s developmental progress. Specifically, lower psychological distress of mothers was linked with more superior child development across 11 measures in Wave 1, 10 measures in Wave 2, and seven in Wave 3.
Mothers’ perceptions about themselves as parents were also significantly associated with the children’s developmental progress. In Wave 1, for seven measures, the more confident that mothers felt as parents, the better their child fared in terms of those seven measures. However, this association appeared to weaken as the child matured, for significant differences emerged for only three measures in Waves 2 and 3. It is important to point out that any causal connection between mothers’ confidence as parents and their children’s developmental progress may be bi-directional. That is, less confident mothers may be less competent in handling their children’s behaviour, which in turn may compromise their children’s developmental progress, while at the same time, relatively poor developmental progress may also make mothers feel less confident than otherwise.
Mothers’ self-reported parenting practices were also significantly associated with children’s developmental progress across all three waves. Specifically, the children appeared to fare better where mothers said that they frequently displayed warmth, and worse where their mothers reported a hostile approach to parenting. In addition, children of mothers who reported greater use of reasoning and consistency in parenting tended to fare better than other children. However, of these four measures of parenting behaviour (warmth, hostility, consistency and reasoning), those concerning hostility and consistency were more strongly linked with children’s developmental progress. Specifically, mothers’ reported warmth was statistically significant for only two to five developmental progress measures in each wave, while their reported use of reasoning was statistically significant for one to three measures. In contrast, statistically significant relationships emerged between most measures of children’s developmental progress and mothers’ reports regarding their tendency to display ‘hostile parenting’ and consistency.
3.4 Cohabitation pathways and child wellbeing
Of all the children in cohabiting families in the initial survey wave, to what extent, if at all, does their wellbeing vary (at the outset and subsequently) according to whether their parents marry, separate or continue to cohabit over the four-year period?
The final set of analyses focused on three groups of children who were living with their cohabiting parents in Wave 1: those who continued to experience this arrangement (n = 223), those whose parents had married by Wave 3 (n = 54) and those whose parents had separated by Wave 3, resulting in the children living with their sole mother (n = 49). Given the small number of children who experienced either of the transitions, the results outlined in this section may best be treated as a means of generating hypotheses for subsequent research.
It was predicted that the cohabiting parents who had married by Wave 3 would have provided the most favourable assessments of their inter-parental relationship in Wave 1, while those who had separated by Wave 3 would have the least favourable assessments. Table 11 presents the mean scores of mothers and fathers in the three groups regarding the two measures tapping the quality of their inter-parental
Research OP46_FINAL.indd 33 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4634
relationships, along with their mean scores regarding psychological distress. The analyses compared the means for married parents with those for each of the other two groups, taken separately.
As predicted, from the perspective of both fathers and mothers, those who subsequently married had significantly better inter-parental relationships in Wave 1 than those who continued to cohabit and those who separated. This was apparent for each of the two measures tapping relationship quality (happiness in the relationship and argumentativeness). In addition, among the couples who married, the fathers appeared to be less distressed in Wave 1 than those who continued to cohabit. Although the mean distress score of fathers who married was also lower than that of fathers who separated, this difference was not statistically significant.
Table 11: Parents who were cohabiting in Wave 1: Parental psychological distress and the quality of inter-parental relationships in Wave 1, by parental cohabitation pathways by Wave 3
Parental psychological distress and relationships, Wave 1
Family types, Wave 3
MarriedContinued cohabiting
Separated, with sole mothers
Mothers’ reports
Psychological distress (Kessler 6) a 1.59 1.69 1.77
Degree of happiness in relationship with the partner b 5.62 4.96 *** 4.13 ***
Argumentative relationship scale c 1.99 2.19 * 2.43 ***
Fathers’ reports
Psychological distress (Kessler 6) a 1.47 1.67 * 1.68
Degree of happiness in relationship with the partner b 5.47 4.92 ** 4.69 **
Argumentative relationship scale c 1.96 2.23 ** 2.37 **
Number of observations 54 223 49
Note: a Scale of 1 to 5, higher scores = higher distress. b Scale of 1 to 7, 1 = extremely unhappy, 7 = perfectly happy. c Scale of 1 to 5, higher scores = more prone to argument. The parents represented here were the biological parents of the study child. The reference group comprised Wave 1 cohabiting parents who were married by Wave 3. * p < 0.05, ** p < 0.01, *** p < 0.001. Missing data result in slight variations in numbers of observations across waves and scales.
Table 12 presents the mean scores for all 15 measures of developmental progress and highlights differences that were statistically significant. Once again, the scores for the children whose parents married were compared (separately) with those for children whose parents continued to cohabit or were separated (with the child living with the sole mother).
None of the differences in mean scores of children whose parents eventually married or continued to cohabit differed significantly. On face value, the direction of most results would suggest that the children whose parents separated were not faring as well as the children whose parents married. However, in Wave 1, none of the differences reached the conventional 5 per cent level of statistical significance, although two approached significance (p < 0.05)—the caregivers’ assessment of their child’s overall developmental progress and emotional problems. By Wave 3, six of the 15 mean scores for children in these two groups differed significantly, with results suggesting poorer progress for the children who were living with their sole mothers. The children who were living with their sole mothers appeared to be doing worse in terms of:
◗◗ overall social–emotional progress
◗◗ the caregivers’ assessments of prosocial behaviour
Research OP46_FINAL.indd 34 27/06/12 2:03 PM
35
results
◗◗ the caregivers’ and teachers’ assessments of emotional problems
◗◗ the teachers’ assessments of peer problems and conduct problems.
In summary, the cohabiting parents who subsequently married tended to evaluate the quality of their relationship more favourably than those who continued to cohabit and those who had separated (with the child living with the sole mother), though none of the differences between the children with married parents and those whose parents continued to cohabit were statistically significant. While the Wave 1 mean scores of the children who ended up living with their sole mother tended to be lower than those of children whose parents were married, the differences did not reach statistical significance (though two approached significance). By Wave 3, the children who were living with their sole mothers appeared to be progressing less well according to six of the 15 measures, but better according to one. Some of these measures were based on the primary caregivers’ reports and some on the teachers’ reports.
Table 12: Cohabiting families in Wave 1: Children’s developmental progress, by parental cohabitation pathways by Wave 3
MarriedContinued cohabiting
Separated, with sole mothers
Wave 1
PPVT score a 63.65 64.35 64.9
Primary caregiver’s assessments
Outcome indices
Physical development index a 101.01 99.34 98.0
Social–emotional development index a 100.75 99.18 97.3
Learning index a 100.15 100.17 98.6
Overall development index a 100.82 99.28 97.0 #
SDQ social–emotional sub-scales
Prosocial behaviour a 7.89 7.53 7.3
Peer problems b 1.56 1.73 1.9
Emotional problems b 1.37 1.71 2.0 #
Hyperactivity problems b 3.89 3.60 3.8
Conduct problems b, c 2.89 3.02 3.2
Teacher’s assessments
Prosocial behaviour a 7.33 7.40 7.1
Peer problems b 2.69 2.72 3.0
Emotional problems b 1.17 1.45 1.3
Hyperactivity problems b 3.30 3.20 3.4
Conduct problems b, c 2.00 1.82 2.5
Research OP46_FINAL.indd 35 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4636
MarriedContinued cohabiting
Separated, with sole mothers
Wave 3
PPVT a score 78.36 78.34 77.60
Primary caregiver’s assessments
Outcome indices
Physical development index a 99.59 99.27 98.69
Social–emotional development index a 101.51 99.59 96.03 *
Learning index a 99.71 99.36 99.02
Overall development index a 100.76 99.19 97.34
SDQ social–emotional sub-scales
Prosocial behaviour a 8.69 8.42 7.81 *
Peer problems b 1.59 1.72 2.00
Emotional problems b 1.17 1.72 2.17 **
Hyperactivity problems b 3.55 3.27 3.86
Conduct problems b, c 1.43 1.62 1.95
Teacher’s assessments
Prosocial behaviour a 7.39 7.54 6.53
Peer problems b 1.33 1.69 2.21 *
Emotional problems b 1.04 1.49 1.89 *
Hyperactivity problems b 3.28 2.76 4.18
Conduct problems b, c 0.89 0.96 2.08 *
Number of respondents 54 223 49
Note: a Higher score = better outcome. b Higher score = worse outcome. c Items that formed this scale were the same in Wave 2 and Wave 3, but differed in Wave 1. All children represented here were living with biological parents only. The reference group comprised Wave 1 cohabiting parents who were married by Wave 3. # p < 0.1, * p < 0.05, ** p < 0.01. Missing data result in slight variations in numbers of observations across waves and scales.
Table 12: Cohabiting families in Wave 1: Children’s developmental progress, by parental cohabitation pathways by Wave 3 (continued)
Research OP46_FINAL.indd 36 27/06/12 2:03 PM
37
summary and disCussion
4 Summary and discussion
As in many Western countries, the rise in cohabitation in Australia represents one of the most striking of the many changes to family life that have emerged since the 1970s. At first, the rise largely represented a new period between leaving the parental home and marrying. However, cohabitation is increasingly being used as a context for childbearing, although the vast majority of couples postpone childbearing until they are married.
The increase in cohabitation rates has fuelled and has been influenced by a growing social acceptance of this choice in lifestyle, and policies have changed to remove areas of discrimination in the opportunities available to cohabiting and married families. As Cherlin (2004) observed, marriage has become less institutionalised while cohabitation has become increasingly institutionalised.
Nevertheless, Australia does not appear to have reached a situation whereby cohabitation and marriage are indistinguishable (see Kiernan 2001). For example, as noted above, most people wait until they are married before they have children, and there is evidence that cohabiting couples are more likely to separate than married couples. However, much of this research has not taken into account the presence or absence of children. While many studies over the years have examined differences in the wellbeing of children in different family forms, very few have taken into account differences in the wellbeing of children between those whose biological parents are married to each other or those whose parents are cohabiting. The few studies that have been conducted have, for the most part, suggested that the children with biological married parents had higher wellbeing than those whose biological parents were cohabiting (for example, Artis 2007; Goodman & Greaves 2010; Wu et al. 2008). In fact, one study suggested that the latter group was faring no better than children who were living in sole-mother families (and many studies suggest that children in sole-mother families fare relatively poorly across a range of measures) (Brown 2004).
However, a less favourable picture for those in cohabiting rather than married families has not been apparent across all measures examined or in all studies. For example, in a Canadian study of children aged between zero and 15 years whose parents had been cohabiting or married, the children with cohabiting parents appeared to be progressing better than those with married parents in one area (display of emotional problems), poorer in another (prosocial behaviour), and did not differ in a third area examined (conduct disorders) (Wu et al. 2008).
To date, the reasons for children who live with cohabiting biological parents faring worse than those who live with married parents remain uncertain, although the various studies outlined in this report tend to suggest that the poorer financial resources of cohabiting families contribute to poorer developmental progress. In fact, one study suggested that children in cohabiting families would display superior cognitive skills were it not for their families’ more difficult circumstances, associated with such factors as fathers’ education and mothers’ earnings (Hofferth 2006). Wu and colleagues (2008), on the other hand, were unable to explain the poorer social–emotional behaviour of children with cohabiting parents in terms of the joint effects of a range of circumstances covering characteristics of child and parent, financial resources, and aspects of family functioning and parental nurturance.
All of the above-mentioned research was conducted in either the United States or Canada. The present study focused on the developmental progress of young Australian children who were living with their cohabiting or married biological parents or with their sole mother over a four-year period, with assessments being made at the start, middle and end of this period. The present study was designed to address several research questions. These and their related findings are outlined below.17
Research OP46_FINAL.indd 37 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4638
4.1 Research questions and findings
How common and how stable are these family forms over a four-year period?
Not surprisingly, most of the children were living with both parents who were married to each other (here called ‘married families’) and sole-mother families were more commonly represented than families in which the child lived with both biological parents who were cohabiting (here called ‘cohabiting families’). The cohabiting families were as common as, if not more common than, stepfamilies (with rates varying according to survey wave).
Most married, cohabiting and sole-mother families in Wave 1 were in the same family form in Wave 3, with levels of stability being considerably greater for married families than for the other two forms. Indeed, the level of stability of cohabiting and sole-mother families was very similar (63 to 65 per cent were in the same family form in Wave 3). However, this assessment treated change in social marital status (from cohabitation to marriage) as a form of instability. If stability is measured in terms of whether families separated or re-partnered, then the married families continue to be the most stable and the sole-mother families are the least stable.
Where change in family form occurs, what are their most likely pathways?
Although 93 per cent of children who lived in married families did not experience a change in family form, those who did experience a change tended to be living with their sole mother by Wave 3. Children in cohabiting families were just as likely to experience parental marriage as to experience parental separation. Not surprisingly, any change for those in sole-mother families tended to involve the introduction of a stepfather, although a few experienced their parents moving in together.
To what extent do these family forms differ in terms of the financial circumstances of the family, and selected socio-demographic characteristics of the parents and children, indicators of parenting practices and parental wellbeing?
Systematic differences were evident in the experiences of children in the three family forms in relation to these matters. Similar results were apparent for each survey wave. For example, compared with married parents, cohabiting parents and sole mothers were younger, more likely to identify as Indigenous, had a lower level of education, and were less likely to be employed. Cohabiting families and sole-mother families were more likely than married families to experience financial hardships and the parents tended to have less favourable views about their financial situation. It also seems that the two former groups of parents were less consistent in their parenting than was the case for married parents. Finally, compared with married parents, those who were cohabiting tended to report poorer inter-parental relationship quality and higher emotional distress. In other words, some of the characteristics of cohabiting families appeared to lie between those of married families and sole-mother families. For example, the proportion of cohabiting families experiencing financial hardship was higher than that of married families but lower than that of sole-mother families.
To what extent do the children in the three family forms differ in terms of their developmental progress?
To what extent, if at all, do any differences in child wellbeing between the groups diverge or diminish over the four-year period?
Compared with children with married parents, those living with their sole mother appeared to be progressing less well in terms of receptive language skills and in terms of social–emotional, learning and physical development, as reported by their primary caregiver. Teachers, likewise, tended to provide less favourable assessments of the children in sole-mother families than those in married families. These trends were consistent across the survey waves.
The children with cohabiting parents appeared to be progressing less well than those with married parents in some areas only. The relative underperformance of children of cohabiting parents was not apparent in the test scores of receptive language skills administered to the children and emerged in only one of the teachers’
Research OP46_FINAL.indd 38 27/06/12 2:03 PM
39
summary and disCussion
assessments (emotional problems in Wave 3). Based on the primary caregivers’ reports, the children with cohabiting parents were not doing as well as those with married parents in terms of overall social–emotional development and general development (derived from physical, social–emotional and learning indices), peer relationships and conduct problems, with these trends being apparent in all three survey waves.
It should be pointed out that, while children with cohabiting parents appeared to fare less well than those with married parents, they were doing better than those in sole-mother families, based on primary caregivers’ reports. This trend was apparent in all three waves.
If differences in child wellbeing are apparent, to what extent can they be explained by any observed systematic differences in child and family characteristics or parenting practices and parental wellbeing?
The above-noted differences between the groups regarding family environment appeared to explain most of the variations between the groups in the initial survey wave. However, by Wave 3, the poorer progress of children in sole-mother families compared to married families could not be fully explained by family characteristics. This is not surprising, given that peer and school influences will strengthen as the children mature. Edwards (2005), for instance, has shown that children in poorer neighbourhoods tend to progress less well in social–emotional and learning domains.
The family-related factors examined in this research accounted for virtually all of the differences apparent between children with married parents and those with cohabiting parents. However, one surprising trend emerged. The bivariate analysis yielded no significant differences between these two groups in scores on the receptive language measure and the primary caregivers’ assessments of the children’s prosocial behaviour. The multivariate analysis, on the other hand, suggested that the children of cohabiting parents would have displayed superior performance in these areas were it not for their differential family environment experiences. This may reflect differences between the two types of families in characteristics that could not be captured in this study. The greater apparent consistency of married mothers than sole mothers also appeared to contribute to the differences in the developmental progress of children in these two family forms.
Other factors explaining other differences in the developmental progress of children with married parents and those with cohabiting parents and sole mothers included the family’s financial circumstances, and mothers’ educational attainment, emotional distress, and confidence in parenting. Again, building effective parenting capacity should help boost mothers’ confidence as parents. As noted earlier, mothers’ confidence may facilitate the developmental progress of their children, while children’s poor behaviour may also undermine mothers’ confidence. Regardless of the direction of these trends, effective parenting skills should promote mothers’ confidence directly and also indirectly, through improving the children’s behaviour. The results also suggest that measures that improve mothers’ educational attainment and financial circumstances would also narrow the gap in the developmental progress of children in these three family forms.
Of all children in cohabiting families in the initial survey wave, to what extent, if at all, does their wellbeing vary (at the outset and subsequently) according to whether their parents marry, separate or continue to cohabit over the four-year period?
The parents who married tended to evaluate the quality of their relationship more favourably than those who continued to cohabit and those who separated (with the child living with the sole mother). While the Wave 1 mean scores of the children who ended up living with their sole mother tended to be lower than those of children whose parents were married by Wave 3, the differences did not reach statistical significance (with two approaching significance). By Wave 3, the children who were living with their sole mothers appeared to be progressing less well according to six of the 15 measures, but better according to one. Some of these measures were based on the primary caregivers’ reports and some on the teachers’ reports. These findings should be interpreted with caution given that the analysis was based on a small sample size.
Research OP46_FINAL.indd 39 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4640
4.2 Final word
Currently, children with cohabiting parents are more likely to experience parental separation, and there is some evidence to suggest that those whose parents continue to cohabit fall behind those whose parents remain married. But as Australia progresses towards cohabitation becoming more normative, any differences in the children’s development are likely to diminish further. In the meantime, the two groups of families appear to differ in various ways and the children in cohabiting families do not appear to be faring as well as those in married families in some areas. Children in sole-mother families fare much worse. This report also shows that family characteristics such as socioeconomic circumstances, the quality of parental relationship, and parenting practices are important factors explaining differences between the groups. In other words, the results outlined suggest that policies that manage to effectively address such factors (for example, financial disadvantage, inadequate parenting skills and problematic inter-parental relationships) would represent important means of enhancing children’s wellbeing across the family forms.
Research OP46_FINAL.indd 40 27/06/12 2:03 PM
41
appendix: measures used and tables
Appendix: Measures used and tables
Parenting practice measures
Parental warmth is based on six items:
◗◗ How often do you express affection by hugging, kissing and holding this child?
◗◗ How often do you hug or hold this child for no particular reason?
◗◗ How often do you tell this child how happy he/she makes you?
◗◗ How often do you have warm, close times together with this child?
◗◗ How often do you enjoy listening to this child and doing things with him/her?
◗◗ How often do you feel close to this child both when he/she is happy and when he/she is upset?
Four items formed the hostile parenting scale:
◗◗ Of all the times that you talk to this child about his/her behaviour, how often is this praise?
◗◗ Of all the times that you talk to this child about his/her behaviour, how often is this disapproval?
◗◗ How often are you angry when you punish this child?
◗◗ How often do you feel you are having problems managing this child in general?
Inductive reasoning includes two items:
◗◗ How often do you explain to this child why he/she is being corrected?
◗◗ How often do you talk it over and reason with this child when he/she misbehaves?
The consistency parenting scale consists of five items:
◗◗ When you give this child an instruction or make a request to do something, how often do you make sure that he/she does it?
◗◗ If you tell this child he/she will get punished if he/she doesn’t stop doing something, but he/she keeps doing it, how often will you punish him/her?
◗◗ How often does this child get away with things that you feel should have been punished?
◗◗ How often is this child able to get out of punishment when he/she sets his/her mind to it?
◗◗ When you discipline this child, how often does he/she ignore the punishment?
Response options for all these parenting practice items are from 1 ‘never/almost never’ to 5 ‘always/almost always’. For each scale, the mean of responses to the items involved for each parent is the score of the scale.
Research OP46_FINAL.indd 41 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4642
Argumentative relationship scale
This scale includes five items:
◗◗ Thinking about any disagreements (between you and your partner):
◗– how often do you and your partner disagree about basic child-rearing issues?
◗– how often is your conversation awkward or stressed?
◗– how often do you disagree?
◗– how often is there anger or hostility between you?
◗◗ How often do you have arguments with your partner that end up with people pushing, hitting, kicking or shoving?
Response options range from 1 to 5 (respectively indicating Never, Rarely, Sometimes, Often, Always). The mean of the responses for each parent is derived.
Table A1: Sample attritions for three family types at Wave 1, K cohort
Family type at Wave 1
Two biological, married parents
Two biological, cohabiting parents Sole mother
Number
Participated in three waves 3263 327 472
Missing at Wave 2 only 86 16 25
Missing at Wave 3 only 150 39 64
Missing at both Waves 2 & 3 203 54 95
No. of observations 3702 436 656
%
Participated in three waves 88.1 75.0 72.0
Missing at Wave 2 only 2.3 3.7 3.8
Missing at Wave 3 only 4.1 8.9 9.8
Missing at both Waves 2 & 3 5.5 12.4 14.5
Total 100.0 100.0 100.0
Note: Percentages may not total 100.0% due to rounding.
Research OP46_FINAL.indd 42 27/06/12 2:03 PM
43
appendix: measures used and tables
Tabl
e A2
: Ch
ildre
n’s
char
acte
ristic
s by
fam
ily ty
pes,
wav
es 1
to 3
(cro
ss-s
ectio
nal a
naly
sis)
Wav
e 1
Wav
e 2
Wav
e 3
Mar
ried
fam
ilies
Coha
bitin
g fa
mili
esSo
le-m
othe
r fa
mili
esM
arrie
d fa
mili
esCo
habi
ting
fam
ilies
Sole
-mot
her
fam
ilies
Mar
ried
fam
ilies
Coha
bitin
g fa
mili
esSo
le-m
othe
r fa
mili
esAg
e: m
ean
4.8
4.8
4.8
6.9
6.9
6.9
**8.
98.
98.
9**
%%
%
Sex Bo
y51
.052
.252
.651
.752
.750
.251
.751
.448
.9
Girl
49.0
47.8
47.4
48.3
47.3
49.8
48.3
48.6
51.1
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
Coun
try
of b
irth
Oth
er c
ount
ry5.
21.
5**
1.2
***
5.5
1.3
**2.
0**
5.3
1.0
**2.
5*
Aust
ralia
94.8
98.6
98.8
94.5
98.7
98.0
94.7
99.0
97.5
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
Indi
geno
us s
tatu
s
No
98.3
89.7
***
91.7
***
98.5
89.4
***
92.9
***
98.4
91.2
***
92.2
***
Yes
1.7
10.3
8.4
1.6
10.7
7.2
1.6
8.8
7.8
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
Lang
uage
at h
ome
Oth
er15
.95.
3**
*11
.3**
16.8
6.0
***
12.1
*16
.17.
8**
10.8
**
Engl
ish
84.1
94.8
88.7
83.2
94.0
87.9
83.9
92.2
89.2
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
Any
olde
r sib
lings
in h
ouse
hold
No
39.8
43.6
42.9
40.3
47.7
*44
.040
.549
.0*
44.7
Yes
60.2
56.4
57.1
59.7
52.3
56.1
59.5
51.0
55.3
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
Any
youn
ger s
iblin
gs in
hou
seho
ld
No
51.1
50.5
67.7
***
46.6
44.6
60.9
***
44.9
43.9
55.4
***
Yes
48.9
49.5
32.3
53.4
55.4
39.1
55.1
56.1
44.6
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
No. o
f obs
erva
tions
3702
436
656
3333
304
616
3164
251
575
Not
e: T
he fa
mily
form
s in
this
tabl
e re
mai
ned
the
sam
e ac
ross
the
thre
e w
aves
and
com
pris
ed th
ose
in w
hich
the
child
ren
lived
with
thei
r bio
logi
cal p
aren
ts. C
hild
ren
in c
ohab
iting
and
sol
e-m
othe
r fa
mili
es w
ere
com
pare
d w
ith th
ose
in m
arrie
d fa
mili
es. S
tatis
tical
diff
eren
ces
are
show
n as
: * p
< 0.
05, *
* p
< 0.0
1, *
** p
< 0.
001.
Dat
a ha
ve b
een
wei
ghte
d.
Research OP46_FINAL.indd 43 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4644
Tabl
e A3
: Ch
arac
teris
tics
of fa
ther
s an
d m
othe
rs b
y fa
mily
type
, wav
es 1
to 3
(cro
ss-s
ectio
nal a
naly
sis)
Wav
e 1
Wav
e 2
Wav
e 3
Mar
ried
fam
ilies
Coha
bitin
g fa
mili
esSo
le-m
othe
r fa
mili
esM
arrie
d fa
mili
esCo
habi
ting
fam
ilies
Sole
-mot
her
fam
ilies
Mar
ried
fam
ilies
Coha
bitin
g fa
mili
esSo
le-m
othe
r fa
mili
es
Mot
hers
’ cha
ract
eris
tics
Age
35.3
33.2
***
32.5
***
37.3
35.0
***
35.3
***
39.2
37.6
***
37.6
***
%%
%
Educ
atio
n
Deg
ree
or h
ighe
r29
.211
.4**
*11
.2**
*29
.310
.9**
*14
.3**
*30
.112
.4**
*14
.4**
*
Oth
er p
ost-s
choo
l qu
alifi
catio
n36
.239
.337
.839
.738
.745
.1*
41.9
43.7
48.7
**
No
qual
ifica
tion
34.6
49.3
***
51.0
***
31.0
50.5
***
40.6
***
28.1
43.9
***
36.9
***
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
Coun
try
of b
irth
Oth
er c
ount
ry28
.719
.3**
*18
.7**
*29
.717
.3**
*21
.6**
*29
.320
.0*
19.8
***
Aust
ralia
71.3
80.7
***
81.3
***
70.3
82.7
***
78.4
***
70.7
80.0
*80
.2**
*
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
Indi
geno
us s
tatu
s
No
99.0
92.6
***
93.3
***
99.0
92.4
***
94.5
***
99.0
95.2
***
93.6
***
Yes
1.1
7.5
***
6.7
***
1.0
7.7
***
5.5
***
1.0
4.8
***
6.4
***
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
Empl
oym
ent s
tatu
s
Full
time
15.6
16.4
11.3
*19
.415
.316
.723
.319
.620
.8
Part
tim
e43
.534
.4**
*29
.7**
*47
.042
.138
.1**
*51
.043
.9*
39.4
***
Not
em
ploy
ed40
.949
.3**
59.1
***
33.7
42.6
**45
.2**
*25
.736
.4**
*39
.8**
*
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
100.
010
0.0
No. o
f res
pond
ents
3702
436
656
3333
304
616
3164
251
575
Research OP46_FINAL.indd 44 27/06/12 2:03 PM
45
appendix: measures used and tables
Wav
e 1
Wav
e 2
Wav
e 3
Mar
ried
fam
ilies
Coha
bitin
g fa
mili
esSo
le-m
othe
r fa
mili
esM
arrie
d fa
mili
esCo
habi
ting
fam
ilies
Sole
-mot
her
fam
ilies
Mar
ried
fam
ilies
Coha
bitin
g fa
mili
esSo
le-m
othe
r fa
mili
es
Fath
ers’
cha
ract
eris
tics
Age
37.8
36.2
***
39.9
37.8
***
41.8
40.8
%%
%
Educ
atio
n
Deg
ree
or h
ighe
r29
.611
.7**
*29
.412
.0**
*30
.513
.4**
*
Oth
er p
ost-s
choo
l qu
alifi
catio
n46
.156
.7**
*47
.255
.5*
48.1
57.6
**
No
qual
ifica
tion
24.2
31.6
**23
.332
.5**
21.5
29.0
*
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
0
Coun
try
of b
irth
Oth
er c
ount
ry29
.421
.1**
30.3
21.1
**29
.721
.4*
Aust
ralia
70.6
79.0
**69
.778
.9**
70.4
78.6
*
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
0
Indi
geno
us s
tatu
s
No
99.1
93.2
***
99.3
93.6
***
99.3
94.9
***
Yes
0.9
6.8
***
0.7
6.4
***
0.7
5.1
***
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
0
Empl
oym
ent s
tatu
s
Full
time
88.4
75.2
***
88.8
76.6
***
90.6
81.1
***
Part
tim
e6.
08.
8*
6.0
9.3
*5.
48.
6*
Not
em
ploy
ed5.
616
.1**
*5.
214
.1**
*4.
110
.3**
*
Tota
l10
0.0
100.
010
0.0
100.
010
0.0
100.
0
No. o
f res
pond
ents
3702
43
6
3333
30
4
3164
25
1
Not
e: T
he fa
mily
form
s in
this
tabl
e re
mai
ned
the
sam
e ac
ross
the
thre
e w
aves
and
com
pris
ed th
ose
in w
hich
the
child
ren
lived
with
thei
r bio
logi
cal p
aren
ts. F
or b
oth
mot
hers
and
fath
ers,
co
habi
ting
pare
nts
and
sole
mot
hers
wer
e co
mpa
red
with
mar
ried
pare
nts.
Sta
tistic
al d
iffer
ence
s ar
e sh
own
as: *
p <
0.05
, **
p < 0
.01,
***
p <
0.00
1. D
ata
have
bee
n w
eigh
ted.
Mis
sing
dat
a re
sult
in s
light
var
iatio
ns in
num
bers
of o
bser
vatio
ns a
cros
s va
riabl
es.
Tabl
e A3
: Ch
arac
teris
tics
of fa
ther
s an
d m
othe
rs b
y fa
mily
type
, wav
es 1
to 3
(cro
ss-s
ectio
nal a
naly
sis)
(con
tinue
d)
Research OP46_FINAL.indd 45 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4646
Figure A1: Proportions of families having experienced one or more listed financial hardships by family type, as reported by primary carers, Waves 1 to 3
0%
20%
40%
60%
80%
Marriedfamilies
Cohabitatingfamilies
Wave 1
Sole-motherfamilies
Marriedfamilies
Cohabitatingfamilies
Wave 2
Sole-motherfamilies
Marriedfamilies
Cohabitatingfamilies
Wave 3
Sole-motherfamilies
Note: Family type can change between waves, and families across the three waves are not the same families.
Figure A2: Ratings of financial situation by family type, as reported by primary carers, Waves 1 to 3
0%
20%
40%
60%
100%
80%
Marriedfamilies
Cohabitatingfamilies
Wave 1
Sole-motherfamilies
Marriedfamilies
Cohabitatingfamilies
Sole-motherfamilies
Marriedfamilies
Cohabitatingfamilies
Sole-motherfamilies
Wave 2 Wave 3
Comfortable to prosperous Getting by Poor or very poor
Note: Families across the three waves are not the same families.
Figure A3: Mothers’ ratings of self as a parent compared with other parents by family type, Waves 1 to 3
Note: Families across the three waves are not the same.
0%
20%
40%
60%
100%
80%
Marriedmothers
Cohabitatingmothers
Wave 1
Solemothers
Marriedmothers
Cohabitatingmothers
Solemothers
Marriedmothers
Cohabitatingmothers
Solemothers
Wave 2 Wave 3
A very good or better than average parent An average parent Not very good or had some trouble being a parent
Research OP46_FINAL.indd 46 27/06/12 2:03 PM
47
appendix: measures used and tables
Figure A4: Fathers’ ratings of self as a parent compared with other parents by family type, Waves 1 to 3
0%
20%
40%
60%
100%
80%
Marriedfathers
Cohabitatingfathers
Wave 1
Marriedfathers
Cohabitatingfathers
Wave 2
Marriedfathers
Cohabitatingfathers
Wave 3
A very good or better than average parent An average parent
Not very good or had some trouble being a parent
Note: Families across the waves are not the same families.
Table A4: Mean scores of parenting practice scales of fathers and mothers by family type, Waves 1 to 3 (cross-sectional analysis)
Mothers Fathers
Married mothers
Cohabiting mothers Sole mothers Married
fathersCohabiting
fathers
Wave 1
Warm parenting a 4.43 4.43 4.49 * 4.09 4.04
Hostile parenting b 2.17 2.26 ** 2.24 * 2.28 2.36 *
Use of reasoning c 4.25 4.22 4.28 3.99 3.93
Consistency of parenting d 4.07 3.88 *** 3.84 *** 3.98 3.78 ***
No. of observations 3702 435 655 3702 435
Wave 2
Warm parenting a 4.45 4.45 4.46 4.13 4.09
Hostile parenting b 2.16 2.24 2.23 * 2.08 2.12
Use of reasoning c 4.23 4.20 4.27 3.96 3.88
Consistency of parenting d 4.15 3.98 *** 3.98 *** 4.07 3.92 *
No. of observations 3333 303 615 3333 303
Wave 3
Warm parenting a 4.34 4.27 4.35 4.06 4.07
Hostile parenting b 2.13 2.17 2.20 2.17 2.18
Use of reasoning c 4.14 4.09 4.14 3.91 4.03
Consistency of parenting d 4.20 3.97 *** 3.98 *** 4.09 3.96 *
No. of observations 3164 250 575 3164 250
Note: All scores are on a five-point scale. a Higher scores = greater warmth. b Higher scores = higher hostility.
c Higher scores = higher use of reasoning. d Higher scores = greater consistency. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. For both mothers and fathers, cohabiting parents and sole mothers were compared with married parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted. Missing data result in slight variations in numbers of observations across waves and scales.
Research OP46_FINAL.indd 47 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4648
Table A5: Mean scores of psychological distress scale and quality of parental relationship of fathers and mothers by family type, Waves 1 to 3 (cross-sectional analysis)
Mothers Fathers
Married mothers
Cohabiting mothers
Sole mothers
Married fathers
Cohabiting fathers
Wave 1
Psychological distress (Kessler 6) a 1.65 1.74 * 1.97 *** 1.56 1.68 **
Degree of happiness in relationship with the partner b 5.32 4.83 *** 5.42 5.01 ***
Argumentative relationship scale c 2.00 2.21 *** 1.98 2.20 ***
No. of observations 3672 429 651 2962 287
Wave 2
Psychological distress (Kessler 6) a 1.50 1.58 1.86 *** 1.53 1.59
Degree of happiness in relationship with the partner b 5.33 4.84 *** 5.33 4.86 ***
Argumentative relationship scale c 1.84 2.06 *** 1.88 2.06 ***
No. of observations 3233 298 593 2673 179
Wave 3
Psychological distress (Kessler 6) a 1.55 1.63 1.92 *** 1.52 1.61
Degree of happiness in relationship with the partner b 5.25 4.76 *** 5.43 4.88 ***
Argumentative relationship scale c 1.89 2.09 *** 1.85 2.02 *
No. of observations 2785 194 472 2351 138
Note: a Scale of 1 to 5, higher scores = higher distress. b Scale of 1–7, 1 = extremely unhappy, 7 = perfectly happy.
c Scale of 1 to 5, higher scores = more prone to argument. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. For both mothers and fathers, cohabiting parents and sole mothers were compared with married parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted. Missing data result in slight variations in numbers of observations across waves and scales.
Research OP46_FINAL.indd 48 27/06/12 2:03 PM
49
appendix: measures used and tables
Table A6: Child outcome by family type, parents’ reports, Waves 1 to 3 (cross-sectional analysis)
Married familiesCohabiting
familiesSole-mother
families
Wave 1
PPVT score a 64.3 63.5 * 62.5 ***
Outcome indices
Physical development index a 100.4 98.9 * 98.6 ***
Social–emotional development index a 101.0 98.2 *** 96.6 ***
Learning index a 101.0 98.3 *** 96.7 ***
Overall development index a 101.2 97.8 *** 96.2 ***
SDQ social–emotional sub-scales
Prosocial behaviour a 7.8 7.6 ** 7.5 **
Peer problems b 1.6 1.9 ** 2.1 ***
Emotional problems b 1.7 1.9 * 2.1 ***
Hyperactivity problems b 3.4 3.9 *** 4.1 ***
Conduct problems b, c 2.3 3.1 *** 3.2 ***
Wave 2
PPVT score a 73.7 73.5 72.8 ***
Outcome indices
Physical development index a 100.3 99.4 99.1 **
Social–emotional development index a 101.1 98.6 *** 96.6 ***
Learning index a 101.0 98.4 *** 97.2 ***
Overall development index a 101.2 98.2 *** 96.5 ***
SDQ social–emotional sub-scales
Prosocial behaviour a 8.3 8.0 ** 8.0 *
Peer problems b 1.5 1.7 ** 2.1 ***
Emotional problems b 1.6 1.8 2.1 ***
Hyperactivity problems b 3.2 3.7 ** 3.9 ***
Conduct problems b, c 1.4 1.7 ** 2.0 ***
Wave 3
PPVT score a 78.2 77.8 77.2 **
Outcome indices
Physical development index a 100.6 98.6 * 98.0 ***
Social–emotional development index a 101.2 99.0 ** 96.5 ***
Learning index a 101.0 97.8 *** 97.9 ***
Overall development index a 101.2 97.9 *** 96.5 ***
SDQ social–emotional sub-scales
Research OP46_FINAL.indd 49 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4650
Married familiesCohabiting
familiesSole-mother
families
Prosocial behaviour a 8.3 8.5 8.0 *
Peer problems b 1.4 1.8 ** 2.0 ***
Emotional problems b 1.5 1.9 * 2.3 ***
Hyperactivity problems b 3.1 3.5 * 3.8 ***
Conduct problems b, c 1.2 1.6 ** 1.8 ***
Note: a Higher score = better outcome. b Higher score = worse outcome. c Items that formed this scale were the same in Wave 2 and Wave 3, but differed in Wave 1. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. For both mothers and fathers, cohabiting parents and sole mothers were compared with married parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted. Missing data result in slight variations in numbers of observations across waves and scales.
Table A7: Child outcome by family type, teachers’ reports, Waves 1 to 3 (cross-sectional analysis)
Married families Cohabiting families
Sole-mother families
Wave 1
Prosocial behaviour a 7.3 7.2 7.0 **
Peer problems b 2.5 2.9 ** 3.1 ***
Emotional problems b 1.3 1.5 * 1.7 ***
Hyperactivity problems b 3.0 3.5 *** 3.7 ***
Conduct problems b, c 1.8 2.2 *** 2.4 ***
Wave 2
Prosocial behaviour a 7.7 7.4 * 7.0 **
Peer problems b 1.3 1.4 1.8 ***
Emotional problems b 1.2 1.4 1.7 ***
Hyperactivity problems b 2.8 2.9 3.7 ***
Conduct problems b, c 0.8 1.0 * 1.4 ***
Wave 3
Prosocial behaviour a 7.7 7.6 7.0 ***
Peer problems b 1.2 1.7 ** 2.0 ***
Emotional problems b 1.1 1.6 ** 1.9 ***
Hyperactivity problems b 2.7 3.1 3.7 ***
Conduct problems b, c 0.7 1.0 * 1.4 ***
Note: a Higher score = better outcome. b Higher score = worse outcome. c Items that formed this scale were the same in Wave 2 and Wave 3, but differed in Wave 1. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. For both mothers and fathers, cohabiting parents and sole mothers were compared with married parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted. Missing data result in slight variations in numbers of observations across waves and scales.
Table A6: Child outcome by family type, parents’ reports, Waves 1 to 3 (cross-sectional analysis) (continued)
Research OP46_FINAL.indd 50 27/06/12 2:03 PM
51
appendix: measures used and tables
Tabl
e A8
: Re
gres
sion
s of
chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(cro
ss-s
ectio
nal a
naly
sis)
, bas
e m
odel
PPVT
sco
re
Out
com
e in
dice
sSD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
velo
pmen
t in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ng
Ove
rall
deve
lopm
ent
inde
x
Pros
ocia
l be
havi
ours
Peer
pr
oble
ms
Emot
iona
l pr
oble
ms
Hype
ract
ivity
pr
oble
ms
Cond
uct
prob
lem
s a
Wav
e 1
Age
of c
hild
(yea
rs)
3.83
***
–0.4
330.
943
0.69
90.
589
0.21
1–0
.217
0.08
5–0
.337
*–0
.064
Sex
of c
hild
: Boy
–0.9
46**
*–1
.055
***
–2.3
83**
*–4
.402
***
–3.6
8**
*–0
.486
***
0.21
***
0.02
70.
768
***
0.27
3**
*
Child
: Bor
n in
Au
stra
lia1.
029
0.34
60.
214
–0.5
80.
091
0.21
5–0
.227
0.02
90.
194
0.17
6
Child
: Ind
igen
ous
–1.1
54–0
.935
–2.7
92*
–3.4
81**
*–3
.348
**–0
.029
0.37
1*
0.31
50.
598
**0.
489
*
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.3
77**
*0.
193
0.23
9–1
.552
***
–0.6
36–0
.175
*–0
.043
–0.1
29–0
.149
0.05
4
Child
: H
ad y
oung
er
sibl
ing(
s) in
HH
–0.8
73**
*–0
.068
-1.3
09**
–0.6
53–1
.407
***
–0.1
270.
043
0.32
8**
*0.
016
0.13
9
Fam
ily fo
rm (r
ef. =
Mar
ried
fam
ilies
)
Coha
bitin
g fa
mili
es–0
.922
*–1
.368
*–2
.709
***
–2.1
26**
*–3
.038
***
–0.2
12*
0.27
6**
0.19
0.40
1**
0.76
5**
*
Sole
mot
hers
–2.1
89**
*–1
.662
**–4
.448
***
–4.4
49**
*–5
.099
***
–0.2
54**
0.51
***
0.47
3**
*0.
723
***
0.90
3**
*
Cons
tant
46.9
76**
*10
3.06
1**
*98
.211
***
102.
129
***
101.
535
***
6.98
3**
*2.
726
***
1.10
24.
507
***
2.17
4**
r20.
048
0.01
0.04
90.
085
0.07
90.
024
0.02
20.
023
0.04
80.
04
No. o
f res
pond
ents
3636
3998
3997
3978
3977
3997
3997
3997
3997
3997
Wav
e 2
Age
of c
hild
2.08
4**
*–0
.069
–0.3
99–0
.37
–0.4
640.
095
0.35
**0.
26*
0.3
0.08
4
Sex
of c
hild
: Boy
0.49
4**
–0.9
44**
–2.8
55**
*–0
.893
*–2
.233
***
–0.6
02**
*0.
227
***
–0.0
361.
022
***
0.35
2**
*
Research OP46_FINAL.indd 51 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4652
PPVT
sco
re
Out
com
e in
dice
sSD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
velo
pmen
t in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ng
Ove
rall
deve
lopm
ent
inde
x
Pros
ocia
l be
havi
ours
Peer
pr
oble
ms
Emot
iona
l pr
oble
ms
Hype
ract
ivity
pr
oble
ms
Cond
uct
prob
lem
s a
Child
: Bor
n in
Au
stra
lia0.
008
0.65
40.
85–1
.073
0.07
40.
185
–0.1
05–0
.217
0.04
70.
06
Child
: Ind
igen
ous
–0.7
22–2
.141
*–2
.31
*–5
.464
***
–4.7
73**
*–0
.546
*0.
508
**0.
047
0.25
90.
123
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.7
57**
*1.
386
***
–0.0
97–2
.41
***
–0.6
12–0
.085
–0.0
35–0
.155
*–0
.095
0.30
7**
*
Child
: Had
you
nger
si
blin
g(s)
in H
H–0
.822
***
0.56
9–1
.333
***
–0.7
5–0
.768
*–0
.137
*0.
035
0.26
2**
*–0
.09
0.3
***
Fam
ily fo
rm (r
ef. =
Mar
ried
fam
ilies
)
Coha
bitin
g fa
mili
es–0
.406
–0.5
31–2
.303
***
–2.4
35**
*–2
.592
***
–0.2
34*
0.24
1*
0.18
50.
45**
0.35
2**
*
Sole
mot
her
–1.1
75**
*–1
.002
*–4
.731
***
–3.8
24**
*–4
.653
***
–0.2
2*
0.60
7**
*0.
588
***
0.68
4**
*0.
712
***
Cons
tant
60.7
23**
*99
.516
***
105.
389
***
107.
134
***
106.
355
***
7.87
***
–0.9
95–0
.067
0.68
10.
2
r20.
041
0.01
0.05
80.
049
0.05
60.
038
0.03
40.
027
0.06
50.
055
No. o
f res
pond
ents
3947
4073
4071
4061
4059
4069
4068
4068
4068
4068
Wav
e 3
Age
of c
hild
1.92
3**
*–1
.515
–0.4
67–0
.624
–1.2
750.
115
0.12
40.
154
0.01
90.
040
Sex
of c
hild
: Boy
0.37
4–0
.697
–3.4
27**
*–0
.849
*–2
.263
***
–0.7
41**
*0.
228
***
0.00
91.
166
***
0.45
3**
*
Child
: Bor
n in
Au
stra
lia0.
789
0.91
9–1
.066
–1.0
53–0
.562
0.20
00.
107
0.23
00.
359
*0.
178
Child
: Ind
igen
ous
–1.6
46**
–1.9
13–4
.000
**–5
.299
***
–5.1
78**
*–0
.369
0.25
10.
625
*0.
551
0.62
7**
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.7
16**
*1.
003
*–0
.088
–2.3
13**
*–0
.705
–0.0
91–0
.052
–0.1
00–0
.162
0.27
2**
*
Tabl
e A8
: Re
gres
sion
s of
chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(cro
ss-s
ectio
nal a
naly
sis)
, bas
e m
odel
(con
tinue
d)
Research OP46_FINAL.indd 52 27/06/12 2:03 PM
53
appendix: measures used and tables
PPVT
sco
re
Out
com
e in
dice
sSD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
velo
pmen
t in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ng
Ove
rall
deve
lopm
ent
inde
x
Pros
ocia
l be
havi
ours
Peer
pr
oble
ms
Emot
iona
l pr
oble
ms
Hype
ract
ivity
pr
oble
ms
Cond
uct
prob
lem
s a
Child
: Had
you
nger
si
blin
g(s)
in H
H–0
.964
***
0.72
7–0
.825
–1.2
03**
–0.6
57–0
.209
**–0
.014
0.20
7*
–0.2
61**
0.19
7**
Fam
ily fo
rm (r
ef. =
Mar
ried
fam
ilies
)
Coha
bitin
g fa
mili
es–0
.183
–1.6
73–2
.008
*–3
.052
***
–3.1
05**
*0.
249
*0.
400
**0.
326
*0.
417
*0.
344
**
Sole
mot
her
–0.8
78**
–2.8
25**
*–4
.680
***
–2.4
48**
*–4
.556
***
–0.2
85**
0.56
3**
*0.
767
***
0.66
2**
*0.
543
***
Cons
tant
61.9
43**
*11
2.77
3**
*10
8.72
1**
*11
0.45
6**
*11
5.09
9**
*7.
620
***
0.12
9–0
.172
2.18
70.
204
r20.
041
0.01
90.
069
0.03
40.
055
0.05
40.
025
0.03
70.
083
0.05
5
No. o
f res
pond
ents
3385
3413
3411
3406
3404
3411
3410
3411
3411
3411
(a)
Item
s th
at fo
rmed
this
sca
le w
ere
the
sam
e in
Wav
e 2
and
Wav
e 3,
but
diff
ered
in W
ave
1.
Not
e: H
H =
hou
seho
ld. T
he fa
mily
form
s in
this
tabl
e re
mai
ned
the
sam
e ac
ross
the
thre
e w
aves
and
com
pris
ed th
ose
in w
hich
the
child
ren
lived
with
thei
r bio
logi
cal p
aren
ts. S
tatis
tical
diff
eren
ces
are
show
n as
: * p
< 0.
05, *
* p
< 0.0
1, *
** p
< 0.
001.
Dat
a ha
ve b
een
wei
ghte
d.
Tabl
e A8
: Re
gres
sion
s of
chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(cro
ss-s
ectio
nal a
naly
sis)
, bas
e m
odel
(con
tinue
d)
Research OP46_FINAL.indd 53 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4654
Table A9: Regressions of child outcomes, as reported by teachers (cross-sectional analysis), base model
Prosocial behaviours Peer problems Emotional
problemsHyperactivity
problems
Wave 1
Age of child 0.271 –0.677 *** –0.077 –0.105
Sex of child: Boy –0.831 *** 0.197 * –0.005 0.932 ***
Child: Born in Australia 0.012 0.005 0.017 0.056
Child: Indigenous –0.347 0.726 * 0.206 0.971 ***
Child: Had older sibling(s) in HH 0.030 –0.082 -0.0004 –0.306 ***
Child: Had younger sibling(s) in HH
0.183 * –0.030 0.144 * –0.343 ***
Family form (ref. = Married families)
Cohabiting families –0.061 0.255 * 0.191 0.388 **
Sole mothers –0.218 0.561 *** 0.39 *** 0.512 ***
Constant 6.358 *** 5.656 *** 1.566 * 3.242 ***
r2 0.04 0.019 0.008 0.057
No. of respondents 3998 3998 3998 3998
Wave 2
Age of child –0.104 0.203 0.024 0.433
Sex of child: Boy –1.151 *** 0.246 *** –0.091 1.662 ***
Born in Australia –0.058 0.177 0.206 0.751 **
Being Indigenous –0.25 0.197 0.331 0.703
Had older sibling(s) in HH 0.092 –0.183 * –0.19 * –0.14
Had younger sibling(s) in HH 0.025 0.042 0.04 –0.206
Family form (ref. = Married families)
Cohabiting families –0.293 * 0.123 0.091 0.041
Sole mothers –0.651 *** 0.478 *** 0.484 *** 0.898 ***
Constant 8.98 *** –0.343 0.987 –1.616
r2 0.08 0.023 0.016 0.114
No. of respondents 3339 3338 3340 3341
Wave 3
Age of child –0.11 0.183 –0.065 0.096
Sex of child: Boy –1.352 *** 0.128 –0.117 1.689 ***
Born in Australia –0.136 0.149 0.054 –0.014
Research OP46_FINAL.indd 54 27/06/12 2:03 PM
55
appendix: measures used and tables
Prosocial behaviours Peer problems Emotional
problemsHyperactivity
problems
Being Indigenous 0.241 0.144 0.17 0.405
Had older sibling(s) in HH 0.045 –0.216 * –0.206 * –0.159
Had younger sibling(s) in HH 0.188 * –0.089 –0.004 –0.24 *
Family form (ref. = Married families)
Cohabiting families –0.101 0.492 ** 0.582 ** 0.197
Sole mothers –0.633 *** 0.646 *** 0.686 *** 0.968 ***
Constant 9.416 *** –0.463 1.828 1.096
r2 0.102 0.029 0.027 0.115
No. of respondents 2900 2901 2900 2902
Note: HH = household. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted.
Table A9: Regressions of child outcomes, as reported by teachers (cross-sectional analysis), base model (continued)
Research OP46_FINAL.indd 55 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4656
Tabl
e A1
0: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(cro
ss-s
ectio
nal a
naly
sis)
, ful
l mod
el
PPVT
sco
re
Out
com
e in
dice
sSD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
x
Pare
nt:
Pros
ocia
l be
havi
ours
Pare
nt: P
eer
prob
lem
s
Pare
nt:
Emot
iona
l pr
oble
ms
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s a
Wav
e 1
Fam
ily fo
rm (r
ef. =
Mar
ried
fam
ilies
)
Coha
bitin
g fa
mili
es–0
.186
–0.5
73–0
.731
–0.8
22–1
.085
*–0
.094
0.05
8–0
.02
–0.0
110.
436
***
Sole
mot
hers
–0.9
49*
0.18
6–1
.17
*–2
.297
***
–1.6
34**
–0.1
140.
129
0.05
60.
102
0.40
4**
*
Age
of c
hild
: 4.
03**
*–0
.231
1.17
0.62
30.
790.
213
–0.2
5*
0.03
8–0
.347
*–0
.093
Sex
of c
hild
: Bo
y–0
.976
***
–0.8
44**
–1.9
08**
*–4
.379
***
–3.3
65**
*–0
.433
***
0.17
6**
*–0
.018
0.68
3**
*0.
174
**
Child
: Bor
n in
Aus
tral
ia–0
.587
–0.1
53–0
.385
–1.0
42–0
.677
0.24
3–0
.077
0.17
10.
211
0.17
4
Child
: Ind
igen
ous
1.48
0.67
7–0
.212
0.72
70.
551
–0.1
50.
093
0.09
2–0
.219
–0.1
38
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.2
11**
*0.
464
0.74
*–1
.051
**–0
.014
–0.0
79–0
.109
–0.1
86**
–0.2
2**
0.04
4
Child
: Had
you
nger
si
blin
g(s)
in H
H–0
.635
*–0
.457
–0.3
47–0
.22
–0.5
64–0
.02
–0.0
280.
229
***
–0.1
7*
–0.0
29
Age
of m
othe
r0.
107
***
–0.0
070.
102
**0.
117
**0.
105
**–0
.004
–0.0
09–0
.008
–0.0
34**
*–0
.026
***
Mot
her:
Edu
catio
n at
tain
men
t (re
f. =
Deg
ree
or h
ighe
r)
Oth
er p
ost-s
choo
l qu
alifi
catio
n–1
.204
***
0.74
5*
–1.0
68**
*–1
.888
***
–1.0
75**
*0.
023
0.19
3**
0.05
90.
442
***
0.09
6
No
post
-sch
ool
qual
ifica
tion
–2.0
11**
*0.
64–2
.032
***
–3.4
03**
*–2
.414
***
–0.0
730.
324
***
0.15
9*
0.64
7**
*0.
173
*
Mot
her:
Bor
n in
Au
stra
lia2.
52**
*–0
.371
–0.3
520.
669
0.01
6–0
.224
***
–0.1
26*
–0.0
360.
075
0.19
**
Mot
her:
Indi
geno
us
stat
us–3
.164
*–1
.073
–0.7
1–4
.506
**–2
.81
0.37
0.08
60.
018
0.55
80.
364
Mot
her’s
em
ploy
men
t sta
tus
(ref
. = F
ull t
ime)
Part
tim
e0.
392
–0.1
750.
272
0.23
50.
221
–0.1
29–0
.202
*0.
022
–0.1
630.
007
Not
em
ploy
ed–0
.229
–0.3
89–0
.485
–0.9
39*
–0.7
71–0
.157
–0.0
660.
136
0.04
8–0
.035
Research OP46_FINAL.indd 56 27/06/12 2:03 PM
57
appendix: measures used and tables
PPVT
sco
re
Out
com
e in
dice
sSD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
x
Pare
nt:
Pros
ocia
l be
havi
ours
Pare
nt: P
eer
prob
lem
s
Pare
nt:
Emot
iona
l pr
oble
ms
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s a
Fina
ncia
l har
dshi
p (u
p to
six
type
s)–0
.214
–0.5
05*
–0.6
84**
*–0
.332
–0.7
3**
*0.
026
0.12
2**
*0.
111
**0.
101
*0.
088
*
Ratin
g of
fina
ncia
l pr
ospe
rity
(1 to
6)
0.01
20.
461
*0.
345
0.24
70.
446
*0.
051
–0.0
39–0
.036
–0.0
49–0
.04
Mot
hers
’ rat
ings
of
pare
ntin
g se
lf-effi
cacy
–0.2
91*
0.30
40.
723
***
0.15
20.
582
***
0.19
9**
*–0
.056
0.03
5–0
.21
***
–0.1
4**
*
Mot
hers
’ par
entin
g st
yle
War
m p
aren
ting
–0.0
290.
261
1.22
6**
–0.7
160.
572
0.52
1**
*–0
.176
**–0
.088
0.10
8–0
.011
Hos
tile
pare
ntin
g–0
.057
–1.5
3**
*–5
.043
***
0.05
3–3
.182
***
–0.5
85**
*0.
266
***
0.38
6**
*0.
963
***
1.27
6**
*
Indu
ctiv
e re
ason
ing
0.25
20.
417
0.33
10.
457
0.35
20.
259
***
–0.0
210.
02–0
.106
0.13
5*
Cons
iste
ncy
0.95
4**
*1.
361
***
2.28
9**
*1.
637
***
2.41
3**
*0.
224
***
–0.2
12**
*–0
.194
***
–0.4
33**
*–0
.469
***
Mot
hers
’ K–6
de
pres
sion
sca
le–0
.92
***
–2.3
8**
*–2
.43
***
–0.8
55**
–2.7
38**
*–0
.053
0.31
2**
*0.
519
***
0.19
6**
0.19
4**
*
Cons
tant
35.1
64**
*84
.077
***
74.6
28**
*89
.1**
*75
.32
***
2.96
7**
*6.
041
***
3.95
7**
*7.
031
***
3.30
9**
*
r20.
149
0.1
0.34
10.
151
0.30
60.
184
0.12
60.
135
0.22
30.
289
No. o
f res
pond
ents
3636
3998
3997
3978
3977
3997
3997
3997
3997
3997
Wav
e 2
Fam
ily fo
rm (r
ef. =
Mar
ried
fam
ilies
)
Coha
bitin
g fa
mili
es0.
304
0.42
1–0
.657
–0.7
36–0
.563
–0.1
070.
083
0.00
60.
123
0.13
8
Sole
mot
hers
–0.3
330.
965
–1.6
57**
*–1
.692
**–1
.218
**–0
.047
0.26
9**
0.12
20.
185
0.40
8**
*
Age
of c
hild
2.26
1**
*0.
249
–0.1
38–0
.187
–0.0
870.
085
0.25
8*
0.16
80.
319
*0.
130
Sex
of c
hild
: Boy
0.54
1**
–0.7
42*
–2.0
83**
*–0
.770
*–1
.694
***
–0.5
22**
*0.
178
***
–0.1
20*
0.87
1**
*0.
226
***
Child
: Bor
n in
Aus
tral
ia–0
.869
–0.1
92–1
.179
–1.1
86–1
.342
–0.0
310.
199
0.03
90.
235
0.25
9*
Child
: Ind
igen
ous
0.69
22.
869
*1.
198
–0.7
301.
562
–0.1
08–0
.029
–0.1
54–0
.760
*–0
.075
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.5
55**
*1.
813
***
0.71
3*
–1.9
13**
*0.
225
0.03
6–0
.124
*–0
.249
***
–0.1
94*
0.22
6**
*
Tabl
e A1
0: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(cro
ss-s
ectio
nal a
naly
sis)
, ful
l mod
el (c
ontin
ued)
Research OP46_FINAL.indd 57 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4658
PPVT
sco
re
Out
com
e in
dice
sSD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
x
Pare
nt:
Pros
ocia
l be
havi
ours
Pare
nt: P
eer
prob
lem
s
Pare
nt:
Emot
iona
l pr
oble
ms
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s a
Child
: Had
you
nger
si
blin
g(s)
in H
H–0
.422
1.22
1**
–0.2
06–0
.029
0.41
6–0
.003
–0.0
210.
138
*–0
.334
***
0.13
0**
Age
of m
othe
r0.
111
***
0.04
30.
063
*0.
153
***
0.11
6**
0.00
0–0
.005
0.00
0–0
.027
***
–0.0
16**
*
Mot
her:
Edu
catio
n at
tain
men
t (re
f. =
Deg
ree
or h
ighe
r)
Oth
er p
ost-s
choo
l qu
alifi
catio
n–1
.480
***
0.19
9–1
.179
***
–3.3
12**
*–2
.094
***
–0.0
510.
140
*0.
065
0.32
9**
*0.
210
***
No
post
-sch
ool
qual
ifica
tion
–1.8
26**
*–0
.040
–1.5
15**
*–3
.938
***
–2.6
51**
*–0
.073
0.17
1*
0.10
20.
553
***
0.16
6**
Mot
her:
Bor
n in
Au
stra
lia1.
288
***
–0.0
440.
221
–0.3
72–0
.088
–0.0
13–0
.219
**–0
.047
0.11
10.
078
Mot
her:
Indi
geno
us–1
.521
–6.0
28**
*–2
.840
–4.9
35**
–6.6
35**
*–0
.429
0.60
20.
077
1.00
5*
–0.0
45
Mot
her’s
em
ploy
men
t sta
tus
(ref
. = F
ull t
ime)
Part
tim
e0.
211
–0.3
110.
892
*0.
667
0.65
00.
063
–0.1
89**
0.05
9–0
.313
**–0
.130
*
Not
em
ploy
ed–0
.163
–0.9
77*
0.54
0–0
.140
–0.1
93–0
.020
–0.1
59*
0.12
5–0
.366
***
–0.1
04
Fina
ncia
l har
dshi
p (u
p to
six
type
s)0.
067
–0.6
29*
–0.6
52**
–0.1
51–0
.721
**–0
.054
0.05
70.
123
**0.
089
0.02
8
Ratin
g of
fina
ncia
l pr
ospe
rity
(1 to
6)
0.25
5*
0.80
7**
0.36
80.
567
*0.
801
***
–0.0
76–0
.082
*–0
.068
–0.1
08*
–0.0
20
Mot
hers
’ rat
ings
of
pare
ntin
g se
lf-effi
cacy
–0.3
05**
0.74
1**
*0.
234
–0.5
30*
0.24
80.
085
*–0
.007
0.02
7–0
.093
*–0
.049
Mot
hers
’ par
entin
g st
yle W
arm
par
entin
g–0
.178
0.85
6*
1.50
6**
*–0
.757
0.83
8*
0.59
0**
*–0
.108
–0.0
45–0
.180
*–0
.105
Hos
tile
pare
ntin
g–0
.104
–0.5
46–5
.339
***
–0.5
17–3
.138
***
–0.6
15**
*0.
250
***
0.42
1**
*1.
156
***
0.98
1**
*
Indu
ctiv
e re
ason
ing
0.06
30.
204
–0.2
75–0
.139
–0.2
480.
107
–0.0
270.
081
0.23
3**
*0.
093
*
Cons
iste
ncy
0.87
1**
*1.
356
***
2.37
0**
*1.
860
***
2.55
6**
*0.
311
***
–0.3
16**
*–0
.188
***
–0.1
90**
–0.3
31**
*
Tabl
e A1
0: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(cro
ss-s
ectio
nal a
naly
sis)
, ful
l mod
el (c
ontin
ued)
Research OP46_FINAL.indd 58 27/06/12 2:03 PM
59
appendix: measures used and tables
PPVT
sco
re
Out
com
e in
dice
sSD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
x
Pare
nt:
Pros
ocia
l be
havi
ours
Pare
nt: P
eer
prob
lem
s
Pare
nt:
Emot
iona
l pr
oble
ms
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s a
Mot
hers
’ K–6
de
pres
sion
sca
le–0
.230
–1.5
26**
*–3
.206
***
–0.7
89*
–2.6
66**
*–0
.144
*0.
428
***
0.70
4**
*0.
300
***
0.16
0**
*
Cons
tant
52.2
77**
*74
.037
***
81.1
98**
*95
.678
***
78.0
89**
*4.
252
***
3.48
8**
*3.
423
***
1.78
80.
727
r20.
118
0.07
20.
356
0.12
40.
260
0.19
90.
141
0.16
40.
225
0.31
1
No. o
f res
pond
ents
3947
4073
4071
4061
4059
4069
4068
4068
4068
4068
Wav
e 3
Fam
ily fo
rm (r
ef. =
Mar
ried
fam
ilies
)
Coha
bitin
g fa
mili
es0.
489
–0.4
81–0
.565
–1.4
05–1
.130
0.37
7**
0.18
90.
179
0.14
60.
161
Sole
mot
hers
–0.0
56–0
.206
–1.5
59**
–0.2
95–0
.924
–0.1
510.
116
0.32
1**
0.12
40.
165
*
Age
of c
hild
2.15
7**
*–1
.111
0.07
5–0
.178
–0.6
100.
129
0.01
10.
067
–0.0
570.
003
Sex
of c
hild
: Boy
0.49
2**
–0.3
68–2
.503
***
–0.6
08–1
.584
***
–0.6
33**
*0.
152
**–0
.080
0.98
4**
*0.
300
***
Child
: Bor
n in
Aus
tral
ia0.
244
0.22
8–1
.955
**–0
.758
–1.1
580.
135
0.30
1*
0.34
2*
0.43
2**
0.26
9*
Child
: Ind
igen
ous
–0.1
890.
153
–1.4
97–0
.829
–1.0
000.
054
0.22
20.
378
0.18
60.
090
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.5
49**
*1.
450
***
0.12
4–1
.837
***
–0.1
51–0
.054
–0.1
08–0
.125
–0.1
80*
0.27
8**
*
Child
: Had
you
nger
si
blin
g(s)
in H
H–0
.616
**1.
513
***
0.38
2–0
.505
0.60
9–0
.072
–0.1
44*
0.08
4–0
.481
***
0.02
3
Age
of m
othe
r0.
111
***
0.03
80.
068
*0.
182
***
0.13
2**
–0.0
05–0
.012
–0.0
03–0
.019
*–0
.019
***
Mot
her:
Edu
catio
n at
tain
men
t (re
f. =
Deg
ree
or h
ighe
r)
Oth
er p
ost-s
choo
l qu
alifi
catio
n–1
.547
***
–0.4
40–1
.089
**–3
.365
***
–2.2
98**
*–0
.147
*0.
149
*0.
020
0.27
7**
0.18
7**
*
No
post
-sch
ool
qual
ifica
tion
–1.9
28**
*–0
.635
–1.1
69**
–4.4
14**
*–2
.902
***
–0.1
370.
237
**0.
008
0.33
2**
0.14
2*
Mot
her:
Bor
n in
Au
stra
lia0.
808
***
–0.1
82–0
.405
–0.7
12–0
.587
–0.0
84–0
.087
0.04
10.
175
0.08
5
Mot
her:
Indi
geno
us–1
.559
–1.8
08–2
.525
–4.9
84*
–4.3
15*
–0.6
20–0
.164
0.22
40.
294
0.63
5
Tabl
e A1
0: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(cro
ss-s
ectio
nal a
naly
sis)
, ful
l mod
el (c
ontin
ued)
Research OP46_FINAL.indd 59 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4660
PPVT
sco
re
Out
com
e in
dice
sSD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
x
Pare
nt:
Pros
ocia
l be
havi
ours
Pare
nt: P
eer
prob
lem
s
Pare
nt:
Emot
iona
l pr
oble
ms
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s a
Mot
her’s
em
ploy
men
t sta
tus
(ref
. = F
ull t
ime)
Part
tim
e0.
540
*0.
811
*0.
578
1.26
7**
1.25
1**
–0.0
48–0
.215
***
0.07
5–0
.238
**–0
.152
**
Not
em
ploy
ed0.
274
–1.0
37–0
.391
0.52
6–0
.400
–0.1
020.
004
0.19
3*
–0.1
10–0
.069
Fina
ncia
l har
dshi
p (u
p to
six
type
s)–0
.081
–0.6
04–0
.895
**–0
.034
–0.7
30*
0.00
60.
137
*0.
151
*0.
124
*0.
127
*
Ratin
g of
fina
ncia
l pr
ospe
rity
(1 to
6)
0.20
01.
235
***
0.55
1**
0.89
6**
*1.
234
***
–0.0
13–0
.112
**–0
.083
–0.1
32**
–0.0
38
Mot
hers
’ rat
ings
of
pare
ntin
g se
lf-effi
cacy
–0.2
05*
0.43
2*
0.46
5*
–0.3
540.
253
0.07
6*
–0.0
34–0
.018
–0.1
08*
–0.0
89**
Mot
hers
’ par
entin
g st
yle
War
m p
aren
ting
–0.2
760.
338
0.73
7–0
.821
*0.
135
0.42
5**
*–0
.073
0.04
6–0
.101
0.00
8
Hos
tile
pare
ntin
g–0
.253
–1.2
81**
*–5
.761
***
–0.5
49–3
.486
***
–0.7
32**
*0.
381
***
0.54
1**
*1.
131
***
1.00
8**
*
Indu
ctiv
e re
ason
ing
0.20
1–0
.264
–0.4
09–0
.122
–0.3
690.
023
–0.0
220.
085
0.13
4*
0.08
6*
Cons
iste
ncy
0.71
0**
*1.
241
***
1.56
8**
*1.
879
***
2.15
6**
*0.
188
**–0
.180
***
–0.1
41*
–0.2
60**
*–0
.251
***
Mot
hers
’ K–6
de
pres
sion
sca
le–0
.082
–1.5
35**
*–2
.412
***
–0.1
53–1
.877
***
0.00
50.
396
***
0.53
0**
*0.
301
***
0.14
7**
Cons
tant
53.4
30**
*91
.524
***
91.2
79**
*94
.995
***
89.9
41**
*6.
409
***
4.30
2**
*2.
219
4.39
6**
0.91
1
r20.
110.
095
0.34
30.
115
0.24
40.
202
0.14
10.
152
0.24
90.
324
No. o
f res
pond
ents
3385
3413
3411
3406
3404
3411
3410
3411
3411
3411
(a)
Item
s th
at fo
rmed
this
sca
le w
ere
the
sam
e in
Wav
e 2
and
Wav
e 3,
but
diff
ered
in W
ave
1.
Not
e: H
H =
hou
seho
ld. T
he fa
mily
form
s in
this
tabl
e re
mai
ned
the
sam
e ac
ross
the
thre
e w
aves
and
com
pris
ed th
ose
in w
hich
the
child
ren
lived
with
thei
r bio
logi
cal p
aren
ts. S
tatis
tical
diff
eren
ces
are
show
n as
: * p
< 0.
05, *
* p
< 0.0
1, *
** p
< 0.
001.
Dat
a ha
ve b
een
wei
ghte
d.
Tabl
e A1
0: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(cro
ss-s
ectio
nal a
naly
sis)
, ful
l mod
el (c
ontin
ued)
Research OP46_FINAL.indd 60 27/06/12 2:03 PM
61
appendix: measures used and tables
Table A11: Regressions of child outcomes, as reported by teachers (cross-sectional analysis), full models
Teacher: Prosocial
behaviours
Teacher: Peer problems
Teacher: Emotional problems
Teacher: Hyperactivity
problems
Wave 1
Family form (ref. = Married families)
Cohabiting families 0.053 0.097 0.061 0.132
Sole mothers –0.038 0.273 * 0.129 0.104
Age of child 0.273 –0.7 *** –0.121 –0.154
Sex of child: Boy –0.79 *** 0.187 * –0.033 0.89 ***
Child: Born in Australia –0.042 0.293 0.175 0.254
Child: Indigenous –0.709 0.366 0.135 0.636
Child: Had older sibling(s) in HH 0.091 –0.12 –0.081 –0.373 ***
Child: Had younger sibling(s) in HH 0.246 ** –0.079 0.106 –0.395 ***
Age of mother 0 –0.014 0.01 –0.004
Mother: Education attainment (ref. = Degree or higher)
Other post-school qualification 0.002 0.174 0.09 0.216 *
No post-school qualification –0.116 0.23 * 0.209 ** 0.415 ***
Mother: Born in Australia 0.008 –0.334 *** –0.113 –0.175
Mother: Indigenous status 0.73 0.33 –0.06 0.111
Mother’s employment status (ref. = Full time)
Part time 0.053 –0.208 –0.03 –0.452 ***
Not employed –0.055 0.029 0.185 * –0.168
Financial hardship (up to six types) –0.045 0.075 0.098 * 0.176 ***
Rating of financial prosperity (1 to 6) 0.059 –0.028 –0.024 –0.025
Mothers’ ratings of parenting self-efficacy 0.195 *** 0.082 0.062 0.029
Mothers’ parenting style
Warm parenting 0.258 ** –0.029 0.01 0.046
Hostile parenting –0.313 *** 0.044 0.193 ** 0.433 ***
Inductive reasoning 0.116 0.031 –0.045 0.039
Consistency 0.131 * –0.177 * –0.102 * –0.248 ***
Mothers’ K–6 depression scale –0.06 0.276 *** 0.293 *** 0.151 *
Constant 3.616 *** 7.807 *** 2.506 ** 3.985 ***
r2 0.083 0.049 0.052 0.105
No. of respondents 3998 3998 3998 3998
Research OP46_FINAL.indd 61 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4662
Teacher: Prosocial
behaviours
Teacher: Peer problems
Teacher: Emotional problems
Teacher: Hyperactivity
problems
Wave 2
Family form (ref. = Married families)
Cohabiting families –0.165 0.027 0.000 –0.175
Sole mothers –0.398 ** 0.284 ** 0.239 0.433 **
Age of child –0.077 0.174 –0.009 0.396
Sex of child: Boy –1.093 *** 0.207 ** –0.116 1.565 ***
Child: Born in Australia –0.172 0.181 0.116 0.742 **
Child: Indigenous 0.728 –0.390 –0.291 –0.651
Child: Had older sibling(s) in HH 0.107 –0.225 ** –0.238 ** –0.189
Child: Had younger sibling(s) in HH 0.122 0.008 0.034 –0.363 **
Age of mother 0.016 0.001 0.013 –0.017
Mother: Education attainment (ref. = Degree or higher)
Other post-school qualification –0.061 0.066 0.161 0.195
No post-school qualification –0.044 0.154 0.103 0.215
Mother: Born in Australia –0.054 0.172 * 0.275 *** 0.377 **
Mother: Indigenous status –1.112 * 0.637 * 0.705 1.482 **
Mother’s employment status (Full time)
Part time 0.374 *** –0.309 *** 0.015 –0.414 **
Not employed 0.202 –0.207 * 0.012 –0.338 *
Financial hardship (up to six types) –0.099 0.001 0.084 0.152
Rating of financial prosperity (1 to 6) 0.053 –0.122 ** –0.144 ** –0.147 *
Mothers’ ratings of parenting self-efficacy 0.080 –0.003 –0.077 0.053
Mothers’ parenting style
Warm parenting –0.064 0.012 0.107 0.008
Hostile parenting –0.455 *** 0.305 *** –0.032 0.778 ***
Inductive reasoning –0.075 0.015 0.094 0.207 *
Consistency 0.203 ** –0.094 –0.090 –0.159
Mothers’ K–6 depression scale –0.059 0.133 * 0.200 ** 0.235 *
Constant 7.975 *** 0.612 1.905 –1.178
r2 0.117 0.057 0.039 0.168
No. of respondents 3339 3338 3340 3341
Table A11: Regressions of child outcomes, as reported by teachers (cross-sectional analysis), full models (continued)
Research OP46_FINAL.indd 62 27/06/12 2:03 PM
63
appendix: measures used and tables
Teacher: Prosocial
behaviours
Teacher: Peer problems
Teacher: Emotional problems
Teacher: Hyperactivity
problems
Wave 3
Family form (ref. = Married families)
Cohabiting families 0.077 0.369 * 0.476 ** –0.064
Sole mothers –0.357 ** 0.457 *** 0.484 ** 0.509 **
Age of child –0.086 0.174 –0.032 0.089
Sex of child: Boy –1.264 *** 0.087 –0.165 * 1.557 ***
Child: Born in Australia –0.079 0.130 –0.139 –0.175
Child: Indigenous 0.398 –0.375 –0.435 –0.988 *
Child: Had older sibling(s) in HH 0.043 –0.206 * –0.218 * –0.154
Child: Had younger sibling(s) in HH 0.327 ** –0.211 * –0.088 –0.425 **
Age of mother 0.027 ** –0.024 ** –0.001 –0.034 **
Mother: Education attainment (ref. = Degree or higher)
Other post-school qualification –0.308 ** 0.035 0.162 0.404 ***
No post-school qualification –0.257 * –0.051 0.083 0.338 **
Mother: Born in Australia –0.209 * 0.185 * 0.328 *** 0.408 **
Mother: Indigenous status 0.023 0.597 0.697 1.673 **
Mother’s employment status (ref. = Full time)
Part time 0.277 ** –0.209 * 0.149 –0.264 *
Not employed 0.195 0.053 0.207 –0.230
Financial hardship (up to six types) –0.107 –0.020 0.025 0.154
Rating of financial prosperity (1 to 6) 0.067 –0.080 –0.113 * –0.189 **
Mothers’ ratings of parenting self-efficacy 0.111 * –0.073 –0.067 –0.089
Mothers’ parenting style
Warm parenting –0.021 0.049 0.102 0.196
Hostile parenting –0.496 *** 0.206 ** 0.237 ** 0.719 ***
Inductive reasoning –0.057 0.034 0.117 * 0.104
Consistency 0.161 * –0.029 –0.024 –0.258 **
Mothers’ K–6 depression scale 0.055 0.116 0.074 –0.045
Constant 8.414 *** 1.090 1.093 1.611
r2 0.145 0.056 0.051 0.173
No. of respondents 2900 2901 2900 2902
Note: HH = household. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted.
Table A11: Regressions of child outcomes, as reported by teachers (cross-sectional analysis), full models (continued)
Research OP46_FINAL.indd 63 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4664
Tabl
e A1
2: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(unc
hang
ed fa
mily
form
), ba
se m
odel
s
PPVT
sco
re
Out
com
e in
dice
sSD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
xPr
osoc
ial
beha
viou
rsPe
er
prob
lem
sEm
otio
nal
prob
lem
sHy
pera
ctiv
ity
prob
lem
sCo
nduc
t pr
oble
ms (a
)
Wav
e 1
Age
of c
hild
(yea
rs)
3.44
5**
*–1
.207
0.52
60.
004
–0.2
300.
222
–0.1
140.
122
–0.1
89–0
.071
Sex
of c
hild
: Boy
–0.9
5**
*–1
.181
***
–2.3
74**
*–4
.359
***
–3.7
22**
*–0
.424
***
0.23
***
0.03
70.
793
***
0.26
9**
*
Child
: Bor
n in
Au
stra
lia1.
531
0.32
4–0
.349
–0.2
52–0
.017
0.13
9–0
.201
0.17
0.21
10.
192
Child
: Bei
ng a
n In
dige
nous
–1.2
420.
172
–1.0
82–2
.728
–1.5
190.
229
0.31
70.
135
0.38
20.
083
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.2
48**
*0.
281
0.46
–0.9
72*
–0.2
23–0
.174
*–0
.073
–0.1
75*
–0.1
470.
014
Child
: Had
you
nger
si
blin
g(s)
in H
H–0
.705
*–0
.655
–0.9
51*
–0.2
78–1
.063
*–0
.092
–0.0
230.
284
***
–0.0
340.
125
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es–0
.469
–1.2
35–2
.376
**–0
.601
–2.1
02*
–0.1
510.
249
*0.
148
0.24
80.
834
***
Sole
mot
hers
–1.5
93**
*–1
.79
*–4
.277
***
–3.2
24**
*–4
.428
***
–0.2
130.
447
***
0.46
**0.
69**
*0.
961
***
Cons
tant
48.3
18**
*10
6.69
9**
*10
0.52
1**
*10
4.86
5**
*10
5.3
***
6.93
8**
*2.
239
**0.
824
3.76
2**
*2.
191
*
r20.
035
0.01
0.03
60.
065
0.05
80.
018
0.01
50.
020.
041
0.03
2
No o
f res
pond
ents
2839
3092
3091
3077
3076
3091
3091
3091
3091
3091
Wav
e 2
Age
of c
hild
2.25
1**
*–0
.447
–0.4
44–0
.575
–0.7
40.
120.
434
***
0.23
10.
225
0.14
4
Sex
of c
hild
: Boy
0.66
3**
–1.2
08**
*–2
.857
***
–0.6
32–2
.244
***
–0.6
01**
*0.
252
***
–0.0
331.
033
***
0.30
5**
*
Child
: Bor
n in
Au
stra
lia–0
.197
–0.0
460.
327
–0.8
93–0
.47
0.07
–0.1
52–0
.24
0.34
2*
0.13
7
Child
: Ind
igen
ous
–0.6
32–2
.274
*–2
.538
*–5
.139
***
–4.8
18**
*–0
.64
*0.
463
*0.
104
0.30
20.
091
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.6
56**
*1.
731
***
0.05
4–2
.317
***
–0.3
3–0
.052
–0.0
74–0
.162
*–0
.099
0.31
4**
*
Research OP46_FINAL.indd 64 27/06/12 2:03 PM
65
appendix: measures used and tables
PPVT
sco
re
Out
com
e in
dice
sSD
Q s
ocia
l–em
otio
nal s
ub-s
cale
s
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
xPr
osoc
ial
beha
viou
rsPe
er
prob
lem
sEm
otio
nal
prob
lem
sHy
pera
ctiv
ity
prob
lem
sCo
nduc
t pr
oble
ms (a
)
Child
: Had
you
nger
si
blin
g(s)
in H
H–0
.722
**0.
879
*–1
.074
*–0
.533
–0.3
84–0
.106
–0.0
290.
247
**–0
.164
0.30
2**
*
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es–0
.124
–0.9
38–2
.405
**–1
.83
*–2
.546
**–0
.219
0.34
4**
0.23
70.
345
0.34
3*
Sole
mot
her
–1.0
29**
–0.4
3–5
.448
***
–4.0
95**
*–4
.806
***
–0.3
75**
0.67
5**
*0.
677
***
0.68
3**
*0.
806
***
Cons
tant
59.6
7**
*10
2.54
5**
*10
6.12
4**
*10
8.24
6**
*10
8.56
1**
*7.
79**
*–1
.518
0.16
0.93
8–0
.281
r20.
037
0.01
20.
055
0.03
70.
045
0.04
0.03
30.
025
0.06
30.
048
No o
f res
pond
ents
3265
3369
3368
3360
3359
3366
3366
3365
3367
3366
Wav
e 3
Age
of c
hild
1.90
8**
*–1
.924
*–0
.286
–0.9
7–1
.54
0.11
20.
175
0.05
60.
023
0.01
5
Sex
of c
hild
: Boy
0.41
2*
–1.0
08*
–3.3
75**
*–0
.927
*–2
.417
***
–0.7
35**
*0.
22**
*0.
023
1.16
9**
*0.
412
***
Child
: Bor
n in
Au
stra
lia1.
018
0.36
7–1
.076
–0.6
09–0
.612
0.18
50.
071
0.21
20.
397
*0.
206
Child
: Ind
igen
ous
–1.4
08*
–2.9
06–5
.236
***
–5.4
56**
*–6
.286
***
–0.3
390.
516
*0.
928
**0.
460.
778
**
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.8
47**
*1.
194
*0.
059
–2.5
34**
*–0
.648
–0.0
72–0
.112
–0.1
3–0
.119
0.27
5**
*
Child
: Had
you
nger
si
blin
g(s)
in H
H–1
.124
***
0.63
7–0
.72
–1.4
77**
–0.7
76–0
.189
*–0
.039
0.19
6*
–0.2
3*
0.17
1*
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es0.
121
–1.2
68–2
.053
*–2
.198
*–2
.547
**0.
226
0.40
4**
0.31
60.
41*
0.36
6*
Sole
mot
her
–0.9
83*
–3.3
5**
*–5
.783
***
–3.3
46**
*–5
.703
***
–0.3
69**
0.70
3**
*0.
892
***
0.92
6**
*0.
66**
*
Cons
tant
62.0
29**
*11
7.13
3**
*10
7.02
8**
*11
3.48
9**
*11
7.72
9**
*7.
634
***
–0.2
470.
715
2.05
40.
421
r20.
043
0.02
20.
073
0.03
50.
061
0.05
30.
030
0.03
80.
087
0.05
5
No. o
f res
pond
ents
3020
3040
3038
3036
3034
3038
3037
3038
3038
3038
(a)
Item
s th
at fo
rmed
this
sca
le w
ere
the
sam
e in
Wav
e 2
and
Wav
e 3,
but
diff
ered
in W
ave
1.
Not
e: H
H =
hou
seho
ld. T
he fa
mily
form
s in
this
tabl
e re
mai
ned
the
sam
e ac
ross
the
thre
e w
aves
and
com
pris
ed th
ose
in w
hich
the
child
ren
lived
with
thei
r bio
logi
cal p
aren
ts. S
tatis
tical
diff
eren
ces
are
show
n as
: * p
< 0.
05, *
* p
< 0.0
1, *
** p
< 0.
001.
Dat
a ha
ve b
een
wei
ghte
d.
Tabl
e A1
2: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(unc
hang
ed fa
mily
form
), ba
se m
odel
s (c
ontin
ued)
Research OP46_FINAL.indd 65 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4666
Table A13: Regressions of child outcomes, as reported by teachers (unchanged family form), base models
Prosocial behaviours
Peer problems
Emotional problems
Hyperactivity problems
Wave 1
Age of child 0.360 –0.553 ** –0.094 –0.075
Sex of child: Boy –0.800 *** 0.273 ** 0.018 1.063 ***
Child: Born in Australia 0.107 0.095 0.045 0.077
Child: Indigenous –0.446 0.917 ** 0.096 0.959 **
Child: Had older sibling(s) in HH 0.000 –0.127 –0.098 –0.305 **
Child: Had younger sibling(s) in HH 0.190 * –0.069 0.054 –0.358 **
Family form (unchanged; ref. = Married families)
Cohabiting families –0.031 0.075 0.186 0.226
Sole mothers –0.301 0.462 ** 0.365 ** 0.5 **
Constant 5.833 *** 4.919 *** 1.68 * 2.968 **
r2 0.039 0.016 0.006 0.062
No. of respondents 3092 3092 3092 3092
Wave 2
Age of child –0.075 0.239 0.017 0.584 *
Sex of child: Boy –1.135 *** 0.246 *** –0.076 1.659 ***
Child: Born in Australia –0.091 0.023 0.104 0.744 **
Child: Being an Indigenous –0.137 0.203 0.432 0.656
Child: Had older sibling(s) in HH 0.088 –0.174 * –0.248 ** –0.1
Child: Had younger sibling(s) in HH –0.015 0.031 –0.037 –0.226
Family form (unchanged; ref. = Married families)
Cohabiting families –0.185 0.175 0.145 –0.008
Sole mothers –0.763 *** 0.695 *** 0.471 ** 0.94 ***
Constant 8.818 *** –0.477 1.202 –2.664
r2 0.078 0.027 0.013 0.112
No. of respondents 2789 2787 2789 2790
Wave 3
Age of child –0.130 0.253 –0.122 0.171
Sex of child: Boy –1.362 *** 0.133 –0.074 1.721 ***
Child: Born in Australia –0.069 0.084 0.033 –0.16
Child: Indigenous 0.157 0.238 0.218 0.749 *
Research OP46_FINAL.indd 66 27/06/12 2:03 PM
67
appendix: measures used and tables
Prosocial behaviours
Peer problems
Emotional problems
Hyperactivity problems
Child: Had older sibling(s) in HH 0.047 –0.236 * –0.251 ** –0.19
Child: Had younger sibling(s) in HH 0.171 –0.077 –0.013 –0.28 *
Family form (unchanged; ref. = Married families)
Cohabiting families –0.196 0.535 ** 0.484 * 0.258
Sole mothers –0.677 *** 0.725 *** 0.72 *** 1.137 ***
Constant 9.543 *** –1.044 2.356 0.555
r2 0.101 0.029 0.023 0.122
No. of respondents 2591 2592 2590 2592
Note: HH = household. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted.
Table A13: Regressions of child outcomes, as reported by teachers (unchanged family form), base models (continued)
Research OP46_FINAL.indd 67 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4668
Tabl
e A1
4: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(unc
hang
ed fa
mily
form
), fu
ll m
odel
s
PPVT
sco
reO
utco
me
indi
ces
SDQ
soc
ial–
emot
iona
l sub
-sca
les
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
xPa
rent
: Pr
osoc
ial
beha
viou
rs
Pare
nt: P
eer
prob
lem
sPa
rent
: Em
otio
nal
prob
lem
s
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s (a)
Wav
e 1
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es0.
159
–0.4
96–0
.666
0.43
4–0
.425
–0.0
560.
054
–0.0
48–0
.098
0.57
0**
*
Sole
mot
hers
–0.5
150.
240
–1.0
12–1
.378
–1.0
76–0
.045
0.07
00.
030
0.10
60.
498
**
Child
: Age
3.69
0**
*–0
.888
0.77
3–0
.032
0.04
00.
219
–0.1
570.
071
–0.1
94–0
.100
Sex
of c
hild
: Boy
–0.8
63**
*–0
.868
**–1
.670
***
–4.2
08**
*–3
.191
***
–0.3
48**
*0.
169
**–0
.033
0.67
1**
*0.
138
*
Child
: Bor
n in
Au
stra
lia–0
.023
–0.2
22–0
.905
–0.4
51–0
.640
0.19
7–0
.066
0.31
4*
0.22
80.
205
Child
: Ind
igen
ous
1.28
01.
193
1.29
42.
936
2.53
9–0
.051
0.12
6–0
.094
–0.5
70–0
.567
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.0
81**
*0.
467
0.83
8*
–0.5
090.
259
–0.0
93–0
.132
–0.2
04**
–0.2
15*
0.01
2
Child
: Had
you
nger
si
blin
g(s)
in H
H–0
.528
–0.6
62–0
.387
0.07
6–0
.542
–0.0
34–0
.048
0.23
2**
–0.1
790.
016
Age
of m
othe
r0.
110
***
–0.0
280.
084
*0.
107
*0.
084
*–0
.007
–0.0
05–0
.012
–0.0
30**
–0.0
17*
Mot
her:
Edu
catio
n at
tain
men
t (re
f. =
Deg
ree
or h
ighe
r)
Oth
er p
ost-s
choo
l qu
alifi
catio
n–1
.056
***
0.91
9*
–1.1
26**
–1.6
30**
*–0
.917
**0.
021
0.15
6*
0.09
40.
459
***
0.10
4
No
post
-sch
ool
qual
ifica
tion
–1.8
88**
*0.
822
*–1
.977
***
–3.2
23**
*–2
.240
***
–0.0
720.
331
***
0.12
80.
662
***
0.17
3*
Mot
her:
Bor
n in
Au
stra
lia2.
393
***
–0.2
70–0
.230
0.19
7–0
.098
–0.2
22**
–0.1
17–0
.074
0.05
20.
181
*
Mot
her:
Indi
geno
us
stat
us–3
.108
*–0
.683
–1.1
12–6
.938
**–3
.732
0.59
3–0
.004
0.12
00.
884
0.58
1
Research OP46_FINAL.indd 68 27/06/12 2:03 PM
69
appendix: measures used and tables
PPVT
sco
reO
utco
me
indi
ces
SDQ
soc
ial–
emot
iona
l sub
-sca
les
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
xPa
rent
: Pr
osoc
ial
beha
viou
rs
Pare
nt: P
eer
prob
lem
sPa
rent
: Em
otio
nal
prob
lem
s
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s (a)
Mot
her’s
em
ploy
men
t sta
tus
(ref
. = F
ull t
ime)
Part
tim
e0.
529
0.08
40.
637
0.66
20.
751
–0.1
58–0
.259
**–0
.078
–0.1
42–0
.052
Not
em
ploy
ed0.
049
–0.0
940.
065
–0.5
66–0
.161
–0.1
11–0
.131
–0.0
020.
069
–0.0
61
Fina
ncia
l har
dshi
p (u
p to
six
type
s)–0
.300
–0.6
61*
–0.8
42**
*–0
.274
–0.8
56**
0.03
40.
135
**0.
127
**0.
162
**0.
119
*
Ratin
g of
fina
ncia
l pr
ospe
rity
(1 to
6)
0.14
90.
420
0.45
2*
0.26
20.
476
*0.
085
*–0
.070
–0.0
53–0
.030
–0.0
23
Mot
hers
’ rat
ings
of
par
entin
g se
lf-effi
cacy
–0.2
430.
380
0.83
8**
*0.
211
0.71
1**
*0.
192
***
–0.0
70*
–0.0
18–0
.182
***
–0.1
35**
*
Mot
hers
’ par
entin
g st
yle
War
m p
aren
ting
0.31
90.
513
1.28
4**
–0.5
130.
892
*0.
497
***
–0.2
14*
–0.0
600.
084
–0.0
66
Hos
tile
pare
ntin
g0.
116
–1.2
47**
*–4
.847
***
–0.0
99–3
.066
***
–0.6
05**
*0.
242
***
0.33
8**
*0.
948
***
1.24
2**
*
Indu
ctiv
e re
ason
ing
0.21
90.
286
0.43
00.
550
0.31
90.
234
**–0
.055
–0.0
23–0
.089
0.14
6*
Cons
iste
ncy
1.06
8**
*1.
603
***
2.59
6**
*1.
511
***
2.54
7**
*0.
284
***
–0.2
55**
*–0
.244
***
–0.4
33**
*–0
.477
***
Mot
hers
’ K–6
de
pres
sion
sca
le–0
.900
***
–2.3
38**
*–2
.329
***
–0.8
01*
–2.6
69**
*–0
.030
0.28
8**
*0.
499
***
0.20
4**
0.20
8**
Cons
tant
32.9
35**
*85
.450
***
74.2
28**
*90
.952
***
76.4
25**
*3.
052
**6.
142
***
4.52
3**
*6.
047
***
3.30
5**
*
r20.
136
0.09
50.
333
0.12
70.
284
0.18
0.12
20.
132
0.21
0.27
8
No. o
f res
pond
ents
2839
3092
3091
3077
3076
3091
3091
3091
3091
3091
Wav
e 2
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es0.
486
0.03
2–0
.913
–0.4
13–0
.721
–0.0
980.
186
0.06
60.
079
0.16
5
Tabl
e A1
4: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(unc
hang
ed fa
mily
form
), fu
ll m
odel
s (c
ontin
ued)
Research OP46_FINAL.indd 69 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4670
PPVT
sco
reO
utco
me
indi
ces
SDQ
soc
ial–
emot
iona
l sub
-sca
les
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
xPa
rent
: Pr
osoc
ial
beha
viou
rs
Pare
nt: P
eer
prob
lem
sPa
rent
: Em
otio
nal
prob
lem
s
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s (a)
Sole
mot
hers
–0.0
561.
963
**–1
.765
**–1
.462
*–0
.672
–0.1
510.
267
*0.
158
0.09
50.
428
***
Age
of c
hild
2.47
5**
*–0
.011
0.10
6–0
.354
–0.1
770.
139
0.31
3**
0.11
40.
207
0.13
7
Sex
of c
hild
: Boy
0.74
2**
*–0
.934
**–2
.067
***
–0.4
91–1
.647
***
–0.5
13**
*0.
194
***
–0.1
23*
0.88
5**
*0.
193
***
Child
: Bor
n in
Au
stra
lia–0
.911
–0.8
03–1
.177
–0.5
49–1
.348
–0.1
040.
097
–0.0
560.
468
**0.
270
*
Child
: Ind
igen
ous
0.21
72.
369
0.18
60.
273
1.33
8–0
.208
–0.0
34–0
.019
–0.5
800.
076
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.4
30**
*2.
035
***
0.70
2–1
.782
***
0.37
60.
045
–0.1
38*
–0.2
21**
–0.1
88*
0.22
5**
*
Child
: Had
you
nger
si
blin
g(s)
in H
H–0
.296
1.35
2**
–0.2
640.
259
0.59
50.
011
–0.0
460.
170
*–0
.347
***
0.17
1**
Age
of m
othe
r0.
131
***
0.05
10.
061
0.17
8**
*0.
130
***
0.00
0–0
.006
–0.0
04–0
.021
*–0
.010
Mot
her:
Edu
catio
n at
tain
men
t (re
f. =
Deg
ree
or h
ighe
r)
Oth
er p
ost-s
choo
l qu
alifi
catio
n–1
.382
***
0.22
6–1
.063
**–3
.343
***
–2.0
44**
*–0
.024
0.10
40.
076
0.35
9**
*0.
194
***
No
post
-sch
ool
qual
ifica
tion
–1.6
46**
*0.
041
–1.3
23**
*–3
.836
***
–2.4
86**
*–0
.009
0.13
80.
091
0.55
8**
*0.
164
**
Mot
her:
Bor
n in
Au
stra
lia1.
085
***
0.03
9–0
.152
–0.8
15–0
.447
–0.0
55–0
.187
*–0
.012
0.12
90.
143
*
Mot
her:
Indi
geno
us
stat
us–0
.762
–5.8
19**
–2.3
69–6
.240
**–6
.958
***
–0.5
750.
595
0.01
90.
889
–0.2
49
Mot
her’s
em
ploy
men
t sta
tus
(ref
. = F
ull t
ime)
Part
tim
e0.
084
–0.1
180.
926
*0.
418
0.64
5–0
.024
–0.2
21**
0.01
0–0
.315
**–0
.156
*
Not
em
ploy
ed–0
.185
–0.7
690.
722
–0.1
490.
021
–0.0
78–0
.210
*0.
049
–0.4
10**
*–0
.126
Tabl
e A1
4: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(unc
hang
ed fa
mily
form
), fu
ll m
odel
s (c
ontin
ued)
Research OP46_FINAL.indd 70 27/06/12 2:03 PM
71
appendix: measures used and tables
PPVT
sco
reO
utco
me
indi
ces
SDQ
soc
ial–
emot
iona
l sub
-sca
les
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
xPa
rent
: Pr
osoc
ial
beha
viou
rs
Pare
nt: P
eer
prob
lem
sPa
rent
: Em
otio
nal
prob
lem
s
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s (a)
Fina
ncia
l har
dshi
p (u
p to
six
type
s)0.
072
–0.6
60–0
.547
–0.4
69–0
.846
**0.
012
0.08
40.
110
0.06
90.
044
Ratin
g of
fina
ncia
l pr
ospe
rity
(1 to
6)
0.23
80.
891
**0.
523
*0.
616
*0.
932
***
–0.0
50–0
.110
**–0
.080
–0.1
28*
–0.0
17
Mot
hers
’ rat
ings
of
par
entin
g se
lf-effi
cacy
–0.2
81*
0.87
5**
*0.
274
–0.4
87*
0.35
30.
070
–0.0
230.
013
–0.0
97–0
.047
Mot
hers
’ par
entin
g st
yle
War
m p
aren
ting
–0.3
440.
872
*0.
974
*–0
.922
*0.
512
0.56
7**
*–0
.047
0.01
9–0
.056
–0.0
62
Hos
tile
pare
ntin
g–0
.053
–0.3
07–5
.482
***
–0.3
96–3
.037
***
–0.6
70**
*0.
245
***
0.40
6**
*1.
232
***
0.98
0**
*
Indu
ctiv
e re
ason
ing
0.11
90.
102
–0.0
64–0
.007
–0.1
580.
085
–0.0
610.
051
0.17
2**
0.08
0*
Cons
iste
ncy
1.05
5**
*1.
601
***
2.32
9**
*1.
706
***
2.54
1**
*0.
331
***
–0.3
19**
*–0
.190
**–0
.135
–0.3
18**
*
Mot
hers
’ K–6
de
pres
sion
sca
le–0
.365
–1.6
79**
*–3
.504
***
–0.7
93*
–2.8
86**
*–0
.150
*0.
444
***
0.74
5**
*0.
371
***
0.19
2**
*
Cons
tant
49.0
15**
*72
.901
***
79.5
75**
*95
.595
***
77.0
89**
*4.
186
***
3.41
2**
*4.
213
***
1.86
00.
350
r20.
115
0.07
60.
357
0.11
40.
249
0.20
60.
141
0.15
70.
227
0.30
5
No. o
f res
pond
ents
3265
3369
3368
3360
3359
3366
3366
3365
3367
3366
Wav
e 3
Fam
ily fo
rm (u
ncha
nged
; ref
. = M
arrie
d fa
mili
es)
Coha
bitin
g fa
mili
es0.
777
*–0
.042
–0.5
40–0
.595
–0.5
460.
357
**0.
203
0.15
20.
123
0.16
8
Sole
mot
hers
–0.1
12–0
.641
–2.5
78**
*–0
.916
–1.8
76*
–0.2
010.
286
*0.
462
**0.
350
*0.
232
*
Age
of c
hild
2.15
1**
*–1
.493
0.23
5–0
.528
–0.8
680.
119
0.07
1–0
.034
–0.0
49–0
.022
Sex
of c
hild
: Boy
0.54
0**
–0.6
69–2
.313
***
–0.6
74–1
.668
***
–0.6
13**
*0.
130
*–0
.084
0.96
5**
*0.
241
***
Tabl
e A1
4: R
egre
ssio
ns o
f chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(unc
hang
ed fa
mily
form
), fu
ll m
odel
s (c
ontin
ued)
Research OP46_FINAL.indd 71 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4672
PPVT
sco
reO
utco
me
indi
ces
SDQ
soc
ial–
emot
iona
l sub
-sca
les
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
xPa
rent
: Pr
osoc
ial
beha
viou
rs
Pare
nt: P
eer
prob
lem
sPa
rent
: Em
otio
nal
prob
lem
s
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s (a)
Child
: Bor
n in
Au
stra
lia0.
404
–0.0
43–1
.878
**–0
.390
–1.0
750.
157
0.26
60.
310
*0.
497
**0.
283
*
Child
: Ind
igen
ous
0.48
4–1
.419
–2.3
080.
629
–1.4
290.
449
0.62
80.
714
0.06
00.
155
Child
: Had
old
er
sibl
ing(
s) in
HH
–1.6
94**
*1.
525
**0.
093
–2.0
35**
*–0
.221
–0.0
46–0
.136
–0.1
24–0
.116
0.28
8**
*
Child
: Had
you
nger
si
blin
g(s)
in H
H–0
.791
**1.
253
*0.
255
–0.7
360.
315
–0.0
60–0
.120
0.11
8–0
.447
***
0.01
8
Age
of m
othe
r0.
112
***
0.02
80.
048
0.18
7**
*0.
119
**–0
.010
–0.0
10–0
.001
–0.0
21*
–0.0
13*
Mot
her:
Edu
catio
n at
tain
men
t (re
f. =
Deg
ree
or h
ighe
r)
Oth
er p
ost-s
choo
l qu
alifi
catio
n–1
.517
***
–0.3
23–1
.382
***
–3.3
32**
*–2
.365
***
–0.1
66*
0.16
1*
0.08
50.
304
***
0.22
8**
*
No
post
-sch
ool
qual
ifica
tion
–1.9
54**
*–0
.568
–1.2
31**
–4.0
41**
*–2
.728
***
–0.1
190.
228
**0.
048
0.33
6**
0.14
6*
Mot
her:
Bor
n in
Au
stra
lia0.
825
***
–0.5
82–0
.289
–0.8
61–0
.784
–0.1
08–0
.109
0.01
70.
124
0.07
9
Mot
her:
Indi
geno
us
stat
us–2
.257
–0.9
78–3
.321
–7.4
28**
*–5
.431
*–1
.130
*–0
.323
0.23
90.
357
0.74
8
Mot
her’s
em
ploy
men
t sta
tus
(ref
. = F
ull t
ime)
Part
tim
e0.
558
*0.
534
0.21
51.
087
*0.
881
*–0
.098
–0.1
82**
0.11
4–0
.172
–0.1
03
Not
em
ploy
ed0.
204
–1.2
97*
–0.4
610.
281
–0.6
65–0
.106
0.00
00.
183
–0.0
79–0
.031
Fina
ncia
l har
dshi
p (u
p to
six
type
s)0.
048
–0.8
51*
–0.9
89**
–0.0
66–0
.892
*0.
021
0.12
20.
180
*0.
136
0.16
3**
Ratin
g of
fina
ncia
l pr
ospe
rity
(1 to
6)
0.28
4*
0.94
8**
0.32
71.
008
***
1.05
7**
*–0
.002
–0.0
94*
–0.0
36–0
.079
–0.0
09
Tabl
e 14
: Re
gres
sion
s of
chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(unc
hang
ed fa
mily
form
), fu
ll m
odel
s (c
ontin
ued)
Research OP46_FINAL.indd 72 27/06/12 2:03 PM
73
appendix: measures used and tables
PPVT
sco
reO
utco
me
indi
ces
SDQ
soc
ial–
emot
iona
l sub
-sca
les
Phys
ical
de
vt in
dex
Soci
al–
emot
iona
l de
vt in
dex
Inde
x of
le
arni
ngO
vera
ll de
vt
inde
xPa
rent
: Pr
osoc
ial
beha
viou
rs
Pare
nt: P
eer
prob
lem
sPa
rent
: Em
otio
nal
prob
lem
s
Pare
nt:
Hype
ract
ivity
pr
oble
ms
Pare
nt:
Cond
uct
prob
lem
s (a)
Mot
hers
’ rat
ings
of
par
entin
g se
lf-effi
cacy
–0.1
870.
552
*0.
365
–0.3
410.
272
0.06
7–0
.022
0.00
4–0
.103
*–0
.082
*
Mot
hers
’ par
entin
g st
yle
War
m p
aren
ting
–0.4
26*
0.34
20.
693
–1.0
02*
0.03
30.
411
***
–0.0
580.
061
–0.1
710.
030
Hos
tile
pare
ntin
g–0
.384
*–1
.068
**–5
.934
***
–0.6
97*
–3.5
33**
*–0
.780
***
0.39
5**
*0.
543
***
1.13
9**
*1.
049
***
Indu
ctiv
e re
ason
ing
0.26
8–0
.148
–0.3
460.
035
–0.2
120.
010
–0.0
280.
074
0.13
9*
0.05
3
Cons
iste
ncy
0.67
0**
*1.
254
***
1.24
2**
*1.
891
***
2.01
8**
*0.
139
*–0
.156
**–0
.110
–0.2
10**
–0.1
94**
*
Mot
hers
’ K–6
de
pres
sion
sca
le–0
.119
–1.7
11**
*–2
.617
***
0.01
3–1
.973
***
0.00
50.
416
***
0.57
0**
*0.
359
***
0.15
9**
Cons
tant
53.6
91**
*95
.280
***
92.9
60**
*98
.628
***
94.0
85**
*7.
126
***
3.52
0*
2.77
4*
4.39
0**
0.48
8
r20.
115
0.09
40.
338
0.11
80.
240.
198
0.13
40.
147
0.25
20.
322
No. o
f res
pond
ents
3020
3040
3038
3036
3034
3038
3037
3038
3038
3038
(a)
Item
s th
at fo
rmed
this
sca
le w
ere
the
sam
e in
Wav
e 2
and
Wav
e 3,
but
diff
ered
in W
ave
1.
Not
e: H
H =
hou
seho
ld. T
he fa
mily
form
s in
this
tabl
e re
mai
ned
the
sam
e ac
ross
the
thre
e w
aves
and
com
pris
ed th
ose
in w
hich
the
child
ren
lived
with
thei
r bio
logi
cal p
aren
ts. S
tatis
tical
diff
eren
ces
are
show
n as
: * p
< 0.
05, *
* p
< 0.0
1, *
** p
< 0.
001.
Dat
a ha
ve b
een
wei
ghte
d.
Tabl
e 14
: Re
gres
sion
s of
chi
ld o
utco
mes
, as
repo
rted
by
pare
nts
(unc
hang
ed fa
mily
form
), fu
ll m
odel
s (c
ontin
ued)
Research OP46_FINAL.indd 73 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4674
Table A15: Regressions of child outcomes, as reported by teachers (unchanged family form), full models
Teacher: Prosocial
behaviours
Teacher: Peer problems
Teacher: Emotional problems
Teacher: Hyperactivity
problems
Wave 1
Family form (unchanged; ref. = Married families)
Cohabiting families 0.085 –0.021 0.064 0.010
Sole mothers –0.096 0.242 0.140 0.096
Age of child 0.358 –0.607 ** –0.131 –0.138
Sex of child: Boy –0.742 *** 0.244 ** –0.026 0.998 ***
Child: Born in Australia 0.035 0.408 * 0.192 0.256
Child: Indigenous –0.893 0.270 0.309 0.323
Child: Had older sibling(s) in HH 0.063 –0.144 –0.148 –0.362 ***
Child: Had younger sibling(s) in HH 0.241 * –0.035 0.061 –0.353 **
Age of mother 0.001 0.001 0.009 0.008
Mother: Education attainment (ref. = Degree or higher)
Other post-school qualification 0.003 0.084 0.103 0.187
No post-school qualification –0.077 0.141 0.216 ** 0.417 ***
Mother: Born in Australia 0.006 –0.367 *** –0.154 –0.193
Mother: Indigenous status 0.850 0.875 –0.447 0.626
Mother’s employment status (ref. = Full time)
Part time 0.002 –0.384 ** –0.085 –0.437 **
Not employed –0.174 –0.160 0.129 –0.191
Financial hardship (up to six types) –0.032 0.079 0.067 0.246 ***
Rating of financial prosperity (1 to 6) 0.050 –0.053 –0.045 –0.058
Mothers’ ratings of parenting self-efficacy
0.167 ** 0.046 –0.008 –0.008
Mothers’ parenting style
Warm parenting 0.255 * 0.024 0.096 0.006
Hostile parenting –0.301 *** 0.005 0.168 * 0.360 ***
Inductive reasoning 0.116 –0.068 –0.064 0.000
Consistency 0.217 ** –0.144 –0.114 –0.193 **
Mothers’ K–6 depression scale –0.094 0.221 ** 0.238 *** 0.150
Constant 2.752 * 7.038 *** 2.626 ** 3.916 **
r2 0.084 0.04 0.044 0.107
No. of respondents 3092 3092 3092 3092
Research OP46_FINAL.indd 74 27/06/12 2:03 PM
75
appendix: measures used and tables
Teacher: Prosocial
behaviours
Teacher: Peer problems
Teacher: Emotional problems
Teacher: Hyperactivity
problems
Wave 2
Family form (unchanged; ref. = Married families)
Cohabiting families –0.068 0.101 0.047 –0.203
Sole mothers –0.453 * 0.496 *** 0.188 0.382
Age of child –0.024 0.189 –0.030 0.532 *
Sex of child: Boy –1.067 *** 0.206 ** –0.107 1.548 ***
Child: Born in Australia –0.179 –0.003 –0.022 0.700 **
Child: Indigenous 0.630 –0.192 –0.139 –0.600
Child: Had older sibling(s) in HH 0.113 –0.209 ** –0.282 ** –0.162
Child: Had younger sibling(s) in HH 0.052 0.058 –0.020 –0.349 *
Age of mother 0.010 0.010 0.016 –0.008
Mother: Education attainment (ref. = Degree or higher)
Other post-school qualification –0.059 0.072 0.165 0.328 **
No post-school qualification –0.034 0.143 0.075 0.256
Mother: Born in Australia –0.080 0.168 * 0.317 *** 0.370 **
Mother: Indigenous status –0.856 0.406 0.628 1.409 *
Mother’s employment status (ref. = Full time)
Part time 0.393 *** –0.389 *** –0.042 –0.389 **
Not employed 0.226 –0.313 ** –0.015 –0.331 *
Financial hardship (up to six types) –0.119 0.037 0.093 0.242 *
Rating of financial prosperity (1 to 6) 0.089 –0.126 ** –0.134 * –0.172 *
Mothers’ ratings of parenting self-efficacy
0.074 –0.007 –0.076 0.027
Mothers’ parenting style
Warm parenting –0.050 0.038 0.084 –0.031
Hostile parenting –0.441 *** 0.276 *** –0.022 0.741 ***
Inductive reasoning –0.076 –0.004 0.084 0.211 *
Consistency 0.196 * –0.075 –0.086 –0.168
Mothers’ K–6 depression scale –0.081 0.100 0.229 ** 0.160
Constant 7.598 *** 0.188 2.351 –2.363
r2 0.113 0.058 0.037 0.165
No. of respondents 2789 2787 2789 2790
Table A15: Regressions of child outcomes, as reported by teachers (unchanged family form), full models (continued)
Research OP46_FINAL.indd 75 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4676
Teacher: Prosocial
behaviours
Teacher: Peer problems
Teacher: Emotional problems
Teacher: Hyperactivity
problems
Wave 3
Family form (unchanged; ref. = Married families)
Cohabiting families –0.034 0.437 * 0.376 0.004
Sole mothers –0.388 * 0.524 ** 0.491 * 0.631 *
Age of child –0.128 0.259 –0.084 0.192
Sex of child: Boy –1.270 *** 0.098 –0.123 1.579 ***
Child: Born in Australia –0.002 0.051 –0.176 –0.350
Child: Indigenous 0.602 –0.243 –0.385 –0.655
Child: Had older sibling(s) in HH 0.045 –0.223 * –0.263 ** –0.177
Child: Had younger sibling(s) in HH 0.311 ** –0.182 * –0.079 –0.457 **
Age of mother 0.028 ** –0.023 * 0.001 –0.034 **
Mother: Education attainment (ref. = Degree or higher)
Other post-school qualification –0.270 * 0.053 0.142 0.381 **
No post-school qualification –0.153 –0.104 0.042 0.208
Mother: Born in Australia –0.241 * 0.199 * 0.340 *** 0.456 **
Mother: Indigenous status –0.358 0.521 0.660 1.567 *
Mother’s employment status (ref. = Full time)
Part time 0.225 * –0.150 0.179 * –0.195
Not employed 0.161 0.027 0.173 –0.174
Financial hardship (up to six types) –0.114 0.036 0.090 0.206
Rating of financial prosperity (1 to 6) 0.063 –0.056 –0.107 –0.172 *
Mothers’ ratings of parenting self-efficacy
0.078 –0.091 –0.080 –0.085
Mothers’ parenting style
Warm parenting 0.000 0.039 0.046 0.160
Hostile parenting –0.517 *** 0.165 * 0.185 * 0.738 ***
Inductive reasoning –0.016 0.010 0.107 0.097
Consistency 0.165 * –0.017 –0.047 –0.241 *
Mothers’ K–6 depression scale 0.099 0.105 0.053 –0.077
Constant 8.791 *** 0.377 1.915 0.650
r2 0.141 0.053 0.046 0.179
No. of respondents 2591 2592 2590 2592
Note: HH = household. The family forms in this table remained the same across the three waves and comprised those in which the children lived with their biological parents. Statistical differences are shown as: * p < 0.05, ** p < 0.01, *** p < 0.001. Data have been weighted.
Table A15: Regressions of child outcomes, as reported by teachers (unchanged family form), full models (continued)
Research OP46_FINAL.indd 76 27/06/12 2:03 PM
77
list of shortened forms
List of shortened forms
ABS Australian Bureau of Statistics
AIFS Australian Institute of Family Studies
ANU Australian National University
ASSDA Australian Social Sciences Data Archives
FaHCSIA Department of Families, Housing, Community Services and Indigenous Affairs
HILDA Household, Income and Labour Dynamics in Australia
LSAC Longitudinal Study of Australian Children
PPVT Peabody Picture Vocabulary Test
SDQ Strengths and Difficulties Questionnaire
Research OP46_FINAL.indd 77 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4678
Endnotes
1. The authors derived the percentage for 1997 based on the ABS (1998) publication relating to its Family Characteristics Survey (Table 23). The authors derived the percentage for 2006–07 from the expanded, confidentialised unit record file for the Family Characteristics and Transitions Survey, Australia (ABS 2008). ‘Intact families’ refers to those where both parents were living with their biological children and with no stepchildren. However, it is possible for one or both parents to have married or lived with someone previously and it is also possible for one or both parents to have had children from a previous relationship(s) living elsewhere.
2. The 1999 data on the marital status of mothers for New South Wales were excluded due to methodological issues in the data collection (Nassar & Sullivan 2001).
3. The 2005 results were based on the Household, Income and Labour Dynamics in Australia (HILDA) survey. The HILDA project was initiated and is funded by the Australian Government Department of Families, Housing, Community Services and Indigenous Affairs (FaHCSIA) and is managed by the Melbourne Institute of Applied Economic and Social Research (Melbourne Institute). The findings and views reported, however, are those of Qu and Weston (2008a) and should not be attributed to either FaHCSIA or the Melbourne Institute.
4. The 1971 ANU survey was conducted by the Department of Demography. Relevant information about the project is available at the Australian Social Sciences Data Archives (ASSDA) website: <assda-nesstar.anu.edu.au/webview>.
5. Evans and Gray (2005) also reported that 61 to 64 per cent of men and women saw married couples without children as being ‘family’, but only 29 per cent of men and 36 per cent of women used this classification for cohabiting couples without children.
6. LSAC is conducted in partnership between FaHCSIA, the Australian Institute of Family Studies (AIFS) and the ABS. The findings and views reported in this analysis are those of the authors and should not be attributed to FaHCSIA or the ABS.
7. Of the 326 cohabiting biological parents in Wave 1, 54 had married and 49 had separated by Wave 3.8. While equivalised household income (that is, household income relative to costs associated with
household composition) may have been a superior indicator of objective financial circumstances, the first LSAC wave collected parental income but not household income, and the results for parental income were based on a low response rate (79 per cent).
9. Primary carers in Waves 1 and 3 were also asked whether, in the last 12 months, they had financial limits on the type of food they could buy. Because this question was not asked in Wave 2, the item was excluded in the analyses to achieve consistency of measurement across the three waves.
10. Children who were not living with a biological parent were living with grandparents, other relatives or foster parents.
11. Of the children in these three family groups, 85 per cent were represented in all three waves and 3 per cent were followed up in Wave 3 but not Wave 2.
12. Seven children had such a change in living arrangements, with two children experiencing a shared care-time arrangement in Wave 1.
13. The proportions of mothers who were recorded as having a degree or higher qualification for all three unchanged family groups fell very slightly from Wave 1 to Wave 3. This may reflect reporting errors or the possibility that in subsequent waves some mothers with a degree reported a more recently (and lower level) acquired qualification rather than their highest one.
14. Here, as elsewhere, married and cohabiting parents refer to those in couple families in which the study child was born to the couple.
15. Ranges reported here refer to mean scores for different groups of fathers and mothers across the three waves to avoid cumbersome reporting of each individual mean score, and they are not confidential intervals.
Research OP46_FINAL.indd 78 27/06/12 2:03 PM
79
endnotes
16. The discussion in this paragraph is based on separate analyses that assessed the statistical significance of any differences in the wellbeing scores of children in cohabiting families and sole-mother families. Results are available from the authors on request.
17. While some of discussions in this paper focus on stable family types in order to control the effect of change in family forms, the analyses based on cross-sectional data are also carried out and available in the Appendix and the results are in general similar to those based on stable family types.
Research OP46_FINAL.indd 79 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4680
References
Artis, J 2007, ‘Maternal cohabitation and child well-being among kindergarten children’, Journal of Marriage and Family, vol. 69, pp. 222–36.
Australian Bureau of Statistics 1998, Family characteristics Australia 1997, cat. no. 4442.0, ABS, Canberra.
—— 2002, Marriages and divorces 2001, cat. no. 3310.0, ABS, Canberra.
—— 2005, Family, household and income unit variables, 2005, cat. no. 1286.0, ABS, Canberra.
—— 2008, Family characteristics and transitions, Australia, 2006–07, cat. no. 4442.0, ABS, Canberra.
—— 2010a, Births Australia 2009, cat. no. 3301.0, ABS, Canberra.
—— 2010b, Marriages and divorces 2009, cat. no. 3310.0, ABS, Canberra.
Baxter, JA & Smart, D 2010, Fathering in Australia among couple families with young children, Occasional Paper No. 37, Department of Families, Housing, Community Services and Indigenous Affairs, Canberra.
Björklund, A, Ginther, D & Sundström, M 2007, Does marriage matter for children? Assessing the causal impact of legal marriage, IZA Discussion Paper No. 3189, Institute for the Study of Labour, Bonn.
Brown, SL 2000, ‘The effect of union type on psychological well-being: Depression among cohabitors versus married’, Journal of Health and Social Behaviour, vol. 41, pp. 244–55.
—— 2004, ‘Family structure and child well-being: The significance of parental cohabitation’, Journal of Marriage and Family, vol. 66, pp. 351–67.
—— Manning, WD 2009, ‘Family boundary ambiguity and the measurement of family structure: The significance of cohabitation’, Demography, vol. 46, no. 1, pp. 85–101.
Cherlin, A 2004, ‘The deinstitutionalization of American marriage’, Journal of Marriage and Family, vol. 66, pp. 848–61.
De Los Rayes, A 2011, ‘Introduction to the special edition: More than measurement error. Discovering meaning behind informant discrepancies in clinical assessment of children and adolescents’, Journal of Clinical Child & Adolescent Psychology, vol. 40, no. 1, pp. 1–9.
Edwards, B 2005, ‘Does it take a village? An investigation of neighbourhood effects on Australian children’s development’, Family Matters, vol. 72, pp. 36–43.
Ermisch, J 2002, Trying again: Repartnering after dissolution of a union, ISER Working Papers No. 2002–19, Institute for Social and Economic Research, University of Essex, Essex.
Evans, A & Gray, E 2005, ‘What makes an Australian family?’, in S Wilson, G Megaher, R Gibson, D Denemark & M Western (eds), Australian social attitudes: The first report , UNSW Press, Sydney.
Goodman, A & Greaves, E 2010, Cohabitation, marriage and child outcomes, Institute of Fiscal Studies, London.
Hayes, A, Qu, L, Weston, R & Baxter, J 2011, Families in Australia 2011: Sticking together in good and tough times, Australian Institute of Family Studies, Melbourne.
Hayes, A, Weston, R, Qu, L & Gray, M 2010, Families then and now 1980–2010, Fact sheet, Australian Institute of Family Studies, Melbourne.
Hofferth, S 2006, ‘Residential father family type and child well-being: Investment versus selection’, Demography, vol. 43, no. 1, pp. 53–77.
Research OP46_FINAL.indd 80 27/06/12 2:03 PM
81
referenCes
Kessler, RC, Andrews, G, Colpe, LJ, Hiripi, E, Mroczek, DK, Normand, SLT et al. 2002, ‘Short screening scales to monitor population prevalences and trends in nonspecific psychological distress’, Psychological Medicine, vol. 32, no. 6, pp. 959–76.
Kiernan, K 2001, ‘The rise of cohabitation and childbearing outside marriage in Western Europe’, International Journal of Law, Policy and the Family, vol. 15, pp. 1–21.
—— 2002, ‘Cohabitation in western Europe: Trends, issues, and implications’, in A Booth & AC Crouter (eds), Just living together: Implications of cohabitations on families, children and social policy , Lawrence Erlbaum Associates, Inc., Mahwah, New Jersey.
Lancaster, P, Huang, J & Pedisich, E 1994, Australia’s mothers and babies 1991, National Perinatal Statistics No. 1, Australian Institute of Health and Welfare, Canberra.
Laws, P, Abeywardana, S, Walker, J & Sullivan, E 2007, Australia’s mothers and babies 2005, National Perinatal Statistics No. 20, Australian Institute of Health and Welfare, Canberra.
Longitudinal Study of Australian Children 2009, 2008–09 annual report, Department of Families, Housing, Community Services and Indigenous Affairs and Australian Institute of Family Studies, Canberra.
Mackay, R 2005, ‘The impact of family structure and family change on child outcomes: A personal reading of the research literature’, Social Policy Journal of New Zealand, vol. 24, pp. 111–33. Retrieved from: <www.msd.govt.nz/documents/publications/mds/journal/issue24/24-pages111–113.pdf>.
Magnuson, K & Berger, LM 2009, ‘Family states and transitions: Associations with children’s well-being during middle childhood’, Journal of Marriage and Family, vol. 71, pp. 575–91.
Nassar, N & Sullivan, E 2001, Australia’s mothers and babies 1999, National Perinatal Statistics No. 11, Australian Institute of Health and Welfare, Canberra.
Osborne, C, McLanahan, S & Brooks-Gunn, J 2003, Young children’s behavioral problems in married and cohabiting families, Center for Research on Child Wellbeing Working Paper No. 03–09-FF, Center for Research on Child Wellbeing, Office of Population Research, Princeton University, Princeton, New Jersey.
Osbourne, C, Manning, WD & Smock, PJ 2007, ‘Married and cohabiting parents’ relationship stability: A focus on race and ethnicity’, Journal of Marriage and Family, vol. 69, pp. 1345–66.
Pryor, J & Rodgers, B 2001, Children in changing families: Life after parental separation, Blackwell, Oxford.
Qu, L & Weston, R 2008a, ‘Attitudes towards marriage and cohabitation’, Family Relationship Quarterly, vol. 8, pp. 5–10.
—— 2008b, Snapshots of family relationships, Australian Institute of Family Studies, Melbourne.
Sanson, A, Mission, S & the LSAC Outcome Index Working Group 2005, Summarising children’s wellbeing: The LSAC outcome index, LSAC Technical Paper No. 2, Australian Institute of Family Studies, Melbourne.
Soloff, C, Lawrence, D & Johnstone, R 2005, Sample design, LSAC Technical Paper No. 1, Australian Institute of Family Studies, Melbourne.
Tobias, T, Kokaua, J, Gerritsen, S & Templeton, R 2010, ‘The health of children in sole-parent families in New Zealand: Results of a population-based cross-sectional survey’, Australian and New Zealand Journal of Public Health, vol. 34, no. 3, pp. 274–80.
Weston, R & Qu, L 2007, ‘An update on partnership formation trends: What does the 2006 Census suggest?’, Family Relationships Quarterly, vol. 6, pp. 16–19.
Wu, Z, Hou, F & Schimmele, CM 2008, ‘Family structure and children’s psychosocial outcomes’, Journal of Family Issues, vol. 29, no. 12, pp. 1600–24.
Research OP46_FINAL.indd 81 27/06/12 2:03 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4682
Occasional papers
1. Income support and related statistics: a ten-year compendium, 1989–99 Kim Bond and Jie Wang (2001)
2. Low fertility: a discussion paper Alison Barnes (2001)
3. The identification and analysis of indicators of community strength and outcomes Alan Black and Phillip Hughes (2001)
4. Hardship in Australia: an analysis of financial stress indicators in the 1998–99 Australian Bureau of Statistics Household Expenditure Survey J Rob Bray (2001)
5. Welfare Reform Pilots: characteristics and participation patterns of three disadvantaged groups Chris Carlile, Michael Fuery, Carole Heyworth, Mary Ivec, Kerry Marshall and Marie Newey (2002)
6. The Australian system of social protection—an overview (second edition) Peter Whiteford and Gregory Angenent (2002)
7. Income support customers: a statistical overview 2001 Corporate Information and Mapping Services, Strategic Policy and Knowledge Branch, Family and Community Services (2003)
8. Inquiry into long-term strategies to address the ageing of the Australian population over the next 40 years Commonwealth Department of Family and Community Services submission to the 2003 House of Representatives Standing Committee on Ageing (2003)
9. Inquiry into poverty and financial hardship Commonwealth Department of Family and Community Services submission to the Senate Community Affairs References Committee (2003)
10. Families of prisoners: literature review on issues and difficulties Rosemary Woodward (2003)
11. Inquiries into retirement and superannuation Australian Government Department of Family and Community Services submissions to the Senate Select Committee on Superannuation (2003)
12. A compendium of legislative changes in social security 1908–1982 (2006)
13. A compendium of legislative changes in social security 1983–2000 Part 1 1983–1993, Part 2 1994–2000 Bob Daprè (2006)
14. Evaluation of Fixing Houses for Better Health Projects 2, 3 and 4 SGS Economics & Planning in conjunction with Tallegalla Consultants Pty Ltd (2006)
15. The ‘growing up’ of Aboriginal and Torres Strait Islander children: a literature review Professor Robyn Penman (2006)
16. Aboriginal and Torres Strait Islander views on research in their communities Professor Robyn Penman (2006)
17. Growing up in the Torres Strait Islands: a report from the Footprints in Time trials Cooperative Research Centre for Aboriginal Health in collaboration with the Telethon Institute for Child Health Research and the Department of Families, Community Services and Indigenous Affairs (2006)
Research OP46_FINAL.indd 82 27/06/12 2:04 PM
83
oCCasional papers
18. Costs of children: research commissioned by the Ministerial Taskforce on Child Support Paul Henman; Richard Percival and Ann Harding; Matthew Gray (2007)
19. Lessons learnt about strengthening Indigenous families and communities: what’s working and what’s not? John Scougall (2008)
20. Stories on ‘growing up’ from Indigenous people in the ACT metro/Queanbeyan region Cooperative Research Centre for Aboriginal Health in collaboration with the Telethon Institute for Child Health Research and the Department of Families, Housing, Community Services and Indigenous Affairs (2008)
21. Inquiry into the cost of living pressures on older Australians Australian Government Department of Families, Housing, Community Services and Indigenous Affairs submissions to the Senate Standing Committee on Community Affairs (2008)
22. Engaging fathers in child and family services: participation, perception and good practice Claire Berlyn, Sarah Wise and Grace Soriano (2008)
23. Indigenous families and children: coordination and provision of services Saul Flaxman, Kristy Muir and Ioana Oprea (2009)
24. National evaluation (2004–2008) of the Stronger Families and Communities Strategy 2004–2009 Kristy Muir, Ilan Katz, Christiane Purcal, Roger Patulny, Saul Flaxman, David Abelló, Natasha Cortis, Cathy Thomson, Ioana Oprea, Sarah Wise, Ben Edwards, Matthew Gray and Alan Hayes (2009)
25. Stronger Families in Australia study: the impact of Communities for Children Ben Edwards, Sarah Wise, Matthew Gray, Alan Hayes, Ilan Katz, Sebastian Misson, Roger Patulny and Kristy Muir (2009)
26. Engaging hard-to-reach families and children Natasha Cortis, Ilan Katz and Roger Patulny (2009)
27. Ageing and Australian Disability Enterprises Shannon McDermott, Robyn Edwards, David Abelló and Ilan Katz (2010)
28. Needs of clients in the Supported Accommodation Assistance Program Australian Institute of Health and Welfare (2010)
29. Effectiveness of individual funding approaches for disability support Karen R Fisher, Ryan Gleeson, Robyn Edwards, Christiane Purcal, Tomasz Sitek, Brooke Dinning, Carmel Laragy, Lel D’aegher and Denise Thompson (2010)
30. Families’ experiences of services Morag McArthur, Lorraine Thomson, Gail Winkworth and Kate Butler (2010)
31. Housing costs and living standards among the elderly Bruce Bradbury and Bina Gubhaju (2010)
32. Incentives, rewards, motivation and the receipt of income support Jacqueline Homel and Chris Ryan (2010)
33. Problem gamblers and the role of the financial sector The South Australian Centre for Economic Studies (2010)
34. Evaluation of income management in the Northern Territory Australian Institute of Health and Welfare (2010)
35. Post-diagnosis support for children with Autism Spectrum Disorder, their families and carers Kylie Valentine and Marianne Rajkovic, with Brooke Dinning and Denise Thompson; Marianne Rajkovic, Denise Thompson and kylie valentine (2011)
Research OP46_FINAL.indd 83 27/06/12 2:04 PM
Parental marital status and children’s wellbeing
Occasional Paper no. 4684
36. Approaches to personal money management The Social Research Centre and Data Analysis Australia (2011)
37. Fathering in Australia among couple families with young children Jennifer Baxter and Diana Smart (2011)
38. Financial and non-financial support to out-of-home carers Marilyn McHugh and kylie valentine (2011)
39. Community attitudes to people with disability: scoping project Denise Thompson, Karen R Fisher, Christiane Purcal, Chris Deeming and Pooja Sawrikar (2012)
44. Paid Parental Leave evaluation: Phase 1 Bill Martin, Belinda Hewitt, Marian Baird, Janeen Baxter, Alexandra Heron, Gillian Whitehouse, Marian Zadoroznyj, Ning Xiang, Dorothy Broom, Luke Connelly, Andrew Jones, Guyonne Kalb, Duncan McVicar, Lyndall Strazdins, Margaret Walter, Mark Western, Mark Wooden (2012)
Research OP46_FINAL.indd 84 27/06/12 2:04 PM