LSAC and LSIC (Longitudinal Study of Indigenous Children) Research Conference, 2013

Program abstracts

Early bird catches the worm: The causal impact of pre-school participation and teacher qualifications on Year 3 National NAPLAN cognitive tests

Diana Warren, John Haisken-DeNew
Melbourne Institute of Applied Economic and Social Research, The University of Melbourne, Melbourne

Using data from the Longitudinal Survey of Australian Children (LSAC), this is the first analysis for Australia to evaluate the impact of attendance at pre-school programs on matched Year 3 nation-wide NAPLAN test outcomes in the domains of Numeracy, Reading, Spelling, Writing and Grammar. We additionally disaggregate specific teacher qualifications on children's cognitive outcomes.

While one year of learning in Year 3 is represented by about 50 NAPLAN points, we find average pre-school domain effects as much as 10-15 points, mainly driven by the upper quantiles of the NAPLAN distribution. To address causality issues, we use Kernel matching, whereby the average treatment effects are of the magnitude 20 NAPLAN points, and are reduced only modestly to about 15 points with additional controls for ability. The highest increases in Year 3 NAPLAN scores are attained by children whose pre-school teacher had a Degree or Diploma qualification in Early Childhood Education or Child Care.

These results confirm the importance of high quality pre-school programs for later cognitive outcomes. The COAG agreement ensuring that all children have access to a high quality early childhood education program delivered by a degree-qualified early childhood teacher in the year before formal schooling, along with the introduction of the new National Quality Standard for early childhood education and care providers in Australia are likely to have substantial long-term benefits, particularly for children who would not have had the opportunity to attend a pre-school with a suitably qualified teacher if these reforms had not taken place.

Back to top

Attendance in primary school: Factors and consequences

Galina Daraganova, Killian Mullan, Ben Edwards
Australian Institute of Family Studies, Melbourne

Improving school attendance is a key to improving education outcomes for children in Australia (DEEWR, 2012). To date, most studies have focussed on adolescents in the secondary school and little is understood about school attendance in the primary school. Using the LSAC data on children aged from 4-5 up to 10-11 years old and their NAPLAN reading and numeracy scores, this paper has investigated possible factors that are associated with different attendance patterns in primary school as well as examined the effects of students' non-attendance on academic achievements in primary school.

The main findings were: family and parenting factors were more important for children attendance in the early primary school years and became less important as they got older; previous rates of school attendance became more influential, implying that the absenteeism process becomes more self-sustaining over time; and children with higher levels of school readiness at 4-5 years were less likely to be absent early on, and school readiness continued to be significantly associated with absenteeism 6 years later when children were 10-11 years.

It has also been found that higher levels of school non-attendance were associated with lower levels of numeracy particularly in the early primary school years. While numeracy achievement in the later primary years was not directly affected by non-attendance, there was indirect influence of high levels of non-attendance early on (measured prior to the first NAPLAN test) through numeracy academic achievement at age 8-9 years old on numeracy academic achievement at age 10-11 years old.

Back to top

Modelling the relationship between preschool competencies and school achievement in the Longitudinal Study of Australian Children

Kate Reid, Clare Ozolins, Sarah Buckley, Siek Toon Khoo
Australian Council for Educational Research, Melbourne

The focus on successful transitions from preschool contexts to the school environment has received substantial attention with recognition of the importance of early childhood environments in preparing children to achieve well at school. Children's' early number and language competencies are claimed to be highly predictive of school achievement, yet there are few studies which model these relationships longitudinally. There has also been a long standing recognition in developmental research that single indices of developmental outcomes offer poor explanatory power in comparison to interactions between combinations of predictors. Nonetheless, the tendency to focus on isolated processes and individual predictors is well-established in developmental research, although they tend to have only marginal predictive power for childhood outcomes.

The availability of complex modelling techniques provides an opportunity to move beyond single indicators to consider the role of growth trajectories as predictors of developmental outcomes. Longitudinal data are rare in developmental research and the Longitudinal Study of Australian Children provides a unique opportunity to study the relationships between preschool competence measures (and the growth in these measures) in multiple domains (e.g., cognitive, social) and school achievement. In this context, the current study aims to explore the growth in children's academic competencies across the transition from preschool to school. In addition, we aim to model the degree to which developmental transitions differ for subgroups of children (e.g., according to socio-economic factors).

Back to top

Intergenerational mobility in Australia: Has the link between parents' socio-economic status and their children's educational performance weakened since the 1970s?

Gerry Redmond1, Ilan Katz2, Bruce Bradbury2, Melissa Wong2
1Flinders University, Adelaide, 2University of NSW, Sydney

The rhetoric of politicians of all political persuasions, and of official policy documents, highlight the aspiration that every young person in Australia, irrespective of their socio-economic background, should have the opportunity to achieve to their fullest potential. This aspiration suggests a high level of intergenerational social mobility.

This paper presents new findings from a study that examines trends in one indicator of intergenerational mobility in Australia between 1975 and 2006 - the relationship between parents' socio-economic status and their children's academic outcomes. The study, based on the survey micro data from the Youth in Transition and Longitudinal Surveys of Australian Children, suggests that while overall levels of education have increased, there is little evidence of significant change in the relationship between parents' socio-economic status and their children's academic performance. However, the data also suggest a growing influence of average school-level socio-economic status on student outcomes.

Back to top

The impact of multiple major life events on children's social and emotional wellbeing

Deborah Kikkawa, Stuart Bell, Helene Shin, Helen Rogers, Fiona Skelton
Department of Social Services (DSS), Canberra

It is well-recognised in the Australian and international literature that experiencing disadvantage can create a range of difficulties for parents. The impact of multiple disadvantage-such as unemployment and financial stress-upon parents can be particularly detrimental to their wellbeing, with poor health and reduced economic security often-reported outcomes. A challenge is creating policy that draws on what is known about those children who, despite adversity, do not experience clinically significant social and emotional difficulties or those who recover quickly from such adversity.

Using data from multiple waves of the Longitudinal Study of Australian Children (LSAC) and the Longitudinal Study of Indigenous Children (LSIC), this presentation will explore how parents' experiences of multiple major life events affect their children. Data from the two studies, will be drawn upon to investigate how parents' experiences of job loss, financial stress and other major adverse life events over time relates to social and emotional wellbeing of their children at age 6/7. This presentation will examine the influence of factors associated with children who 'thrive' despite these times of adversity. It will then draw implications for policy.

Back to top

Family income and child cognitive and non-cognitive development: The possible pathways

Rasheda Khanam1, Hong Son Nghiem2
1University of Southern Queensland, Toowoomba, 2University of Queensland, Brisbane

This paper investigates the routes through which family income may affect children's cognitive and non-cognitive development by exploiting comprehensive information from the Longitudinal Study of Australian Children (LSAC). Our paper takes a new approach to combine economists' and psychologists' views in modelling the relationship between household income and child development outcomes.

Using a dynamic panel data framework, this paper reveals that when a basic set of covariates is used family income is strongly associated with child cognitive and behavioural outcomes. However, when indicators of parental investment, parental stress, parenting practice and neighbourhood characteristics are controlled for, the income coefficients are no longer significant for most children's outcomes. We also find that income has higher effects on children's cognitive development than upon their non-cognitive development. Our results suggest that the effect of income can be mediated by the family's ability to invest in materials, services and a home environment, parenting practice and neighbourhood characteristics. We find that parental mental health and parenting practice are particularly important for children's' behavioural and emotional development. When unobserved heterogeneity is controlled for using a random and fixed effect estimators, we did not find any significant association between family income and cognitive and emotional and behavioural development of children. We also find evidence of the dynamic nature of children's human capital investment that current cognitive and non-cognitive outcomes of a child are significantly related to previous outcomes.

Back to top

Symposium title: Predictors, pathways and outcomes of high childhood body mass index.
Presentation: Family and neighbourhood socioeconomic inequalities in developmental trajectories of overweight children: Results from the Longitudinal Study of Australian Children

Pauline W. Jansen1 ,2, Fiona K. Mensah2 ,3, Jan M. Nicholson2 ,4, Melissa Wake2 ,3
1Department of Child and Adolescent Psychiatry / Psychology, Erasmus MC-University Medical Centre, Rotterdam, The Netherlands, 2Murdoch Children's Research Institute, Melbourne, 3The Royal Children's Hospital, Melbourne, 4Parenting Research Centre, Melbourne

Background: Cross-sectional research has repeatedly shown socioeconomic inequalities in childhood overweight. Longitudinal research may offer a life-course viewpoint of how overweight and its socioeconomic gradient develop.

Objective: Determine timing and strength of the association between socioeconomic status (SES) and children's body mass index (BMI) in the pre- and primary school years, and examine socioeconomic differences in longitudinal trajectories of overweight.

Methods: Participants - LSAC Kindergarten cohort (n=4949). Predictor - Composite variables of family and neighbourhood SES, categorized as quintiles. Outcome - BMI measured at four biennial waves starting at age 4-5 years (in 2004).

Results: Cross-sectional analyses showed that socioeconomic differences in high BMI already present at age 4-5 years persisted and more than doubled by age 10-11 years; this reflected decreasing mean BMI among advantaged rather than increasing means among disadvantaged children. Latent class analysis identified four weight trajectories across childhood: stable normal (68% of children), persistent overweight (15%), late-onset overweight (14%), and resolving overweight (3%). The risks of persistent and late-onset childhood overweight were higher amongst families of lower SES (e.g. most disadvantaged quintile: OR persistent=2.51, 95%CI: 1.83-3.43), and only partly explained by birth weight and parental overweight. Neighbourhood SES relationships were weaker and attenuated fully on adjustment for family SES. No gradient was observed for resolving overweight.

Conclusions: Childhood appears to be a critical period when socioeconomic inequalities in overweight emerge and strengthen. Targeting disadvantaged children with early overweight must be a top priority, but the prevalence of overweight even among less-disadvantaged families suggests only whole-society approaches will eliminate overweight-associated morbidity.

Back to top

Symposium title: Predictors, pathways and outcomes of high childhood body mass index
Presentation: Bi-directional associations between mothers' and fathers' parenting consistency and child BMI from ages 4-10 years

Pauline W. Jansen1 ,2, Rebecca Giallo3, Elizabeth M. Westrupp2 ,3, Melissa Wake2 ,4, Jan M. Nicholson2 ,3
1Department of Child and Adolescent Psychiatry / Psychology, Erasmus MC-University Medical Centre, Rotterdam, The Netherlands, 2Murdoch Children's Research Institute,3Parenting Research Centre, Melbourne, 4The Royal Children's Hospital, Melbourne

Background: A small body of research suggests that general parenting dimensions and styles are associated with children's body mass index (BMI), but directionality in this relationship remains unknown. Moreover, there have been strong calls in the literature to consider not only mothers, but also the possible role of fathers.

Objective: To examine reciprocal relationships between maternal and paternal parenting inconsistency and child BMI.

Methods: Participants were 4002 children and their parents from the LSAC Kindergarten cohort. Parent's self-reported parenting consistency and children's BMI were measured at four biennial waves starting at age 4-5 years in 2004. Bi-directionality between parenting and child BMI was examined using regression analyses in cross-lagged models.

Results: The best-fitting models indicated a modest influence from parenting to child BMI. For mothers, higher levels of parenting consistency predicted lower BMI in children from Waves 1 to 2 and 3 to 4; e.g. for every standard deviation increase in mothers' parenting consistency at Wave 1, child BMI z-score fell by 0.025 in Wave 2 (95% confidence interval: -0.05; -0.003). For fathers, higher levels of parenting consistency were associated with lower child BMI from Waves 1 to 2 and 2 to 3.

Conclusions: Parenting inconsistency of mothers and fathers prospectively predicted small increases in offspring BMI over two year periods across middle childhood. However, child BMI did not appear to influence parenting behaviour. These findings support recent calls for expanding childhood overweight interventions to address the broad parenting context while involving both mothers and fathers.

Back to top

Symposium title: Predictors, pathways and outcomes of high childhood body mass index
Presentation: Health outcomes of diverse childhood BMI trajectories: The Longitudinal Study of Australian Children

Susan A. Clifford1, Pauline W. Jansen1 ,3, Fiona K. Mensah1 ,2, Melissa Wake1 2
1Murdoch Children's Research Institute, Melbourne, 2The Royal Children's Hospital, Melbourne, 3Department of Child and Adolescent Psychiatry / Psychology, Erasmus MC-University Medical Centre, Rotterdam, The Netherlands

Background: During the early primary school years, around 25% of Australian children are overweight or obese, but few experience concurrent health problems. Early identification of those most likely to develop later problems is vital for targeting treatment. This could become practical if 'typical' childhood BMI trajectories could be identified that are highly predictive of both good and poor obesity-related health outcomes in later childhood.

Objective: To determine how BMI trajectories from age 4-5 years predict physical health at age 10-11 years.

Methods: Participants - Kindergarten LSAC cohort (n=4164). Outcomes at age 10-11 years - Hypertension, global health, physical functioning, asthma, sleep problems, special healthcare needs and 6-year Medicare costs. Predictor - Four common BMI trajectories emerged from latent class analyses: persistent overweight (15% of sample), resolving overweight (3%), late onset overweight (14%) and persistent normal weight (68%). Analysis - Logistic and linear regressions.

Results: Compared to the persistent normal weight group, odds ratios (OR) were intermediate for the late onset overweight group and highest for the persistent overweight group of experiencing hypertension (OR 2.0 and 4.0, respectively), poor global health (OR 1.7 and 2.2) and poor physical functioning (OR 1.8 and 2.3). The persistently overweight group also accrued higher Medicare costs. Children in the resolving overweight group had similar health outcomes to those with persistent normal weight.

Conclusions: Hypertension and poor global and physical health congregate in persistently obese children. If these children were targeted and effectively treated, our findings for resolving overweight offer optimism than these later comorbidities could be obviated.

Back to top

Social and emotional wellbeing of Indigenous Australian mothers with infants and toddlers: What can we learn from LSIC?

Fabrizio D'Esposito, Rebecca Giallo, Jan Nicholson
Parenting Research Centre, Melbourne

Maternal mental health in the postnatal and early childhood period is an important determinant of children's short- and long-term wellbeing. However, the mental health and social and emotional wellbeing (SEWB) of Indigenous Australian mothers and the factors associated with this are not well understood.

The aims of this study were to report on the SEWB of Indigenous mothers with infants and toddlers (aged 6-24 months), and to identify factors associated with this. This study used wave one data from 498 Indigenous mothers whose children were enrolled in the "Baby Cohort" of the Longitudinal Study of Indigenous Children (LSIC). The modified Strong Souls Questionnaire used in LSIC was found to be a valid and reliable tool to assess the SEWB of the mothers in this study. More than 80% of mothers reported having strong social support and strategies to cope with difficulties, and the majority (78.5%) reported no or minimal symptoms of psychological distress.However 21.5% reported frequently experiencing two or more symptoms of distress. Unadjusted analyses revealed that SEWB is associated with maternal, family, and neighbourhood factors, with kinship and cultural factors, and with experiences of grief and trauma.In the adjusted analyses low income, stressful life events, neighbourhood factors, geographical isolation and experiencing discrimination were found to be risk factors for poor SEWB above and beyond all other factors explored in this study. This suggests that broad structural and societal investments are needed to provide a physically and culturally safe environment and to promote the SEWB of new Indigenous mothers.

Back to top

Risk factors associated with trajectories of mothers' psychological distress across the postnatal and early parenting period

Rebecca Giallo1 ,2, Amanda Cooklin1, Jan Nicholson1 ,2
1Parenting Research Centre, Melbourne, 2Murdoch Childrens Research Institute

Women are at increased risk of psychological distress in the first year postpartum. Few studies have examined the course, persistence and associated risk factors of distress beyond this time.

The aims of the study were to:

  1. Report on the course of distress symptoms over the early parenting period (6-7 years postpartum);
  2. Identify distinct groups (classes) of women defined by their trajectory of symptoms over time; and
  3. Identify antenatal and early postnatal risk factors associated with persistent symptoms.

Data from 4879 women participating in the Longitudinal Study of Australian Children were analysed. Latent growth class analysis was conducted to identify classes defined by distinct trajectories of distress symptoms, and logistic regression conducted to identify risk factors associated with class membership. For the overall sample, distress symptoms were highest during the first year postpartum and then gradually decreased across the early parenting period. Two distinct classes were identified with the majority of women (84%) reporting minimal distress symptoms over time, and 16% experiencing persistently high symptoms. Early risk factors for persistent distress symptoms were:

  • Younger maternal age;
  • Being from a non-English speaking background;
  • Not completing high school;
  • Having a past history of depression;
  • Antidepressant use during pregnancy;
  • Child development problems;
  • Lower parental self-efficacy;
  • Poor relationship quality; and
  • More stressful life events.

This research sheds light on the early risk factors that may predispose women to enduring symptoms of psychological distress across the early parenting period, offering opportunity for early identification and targeted early intervention.

Back to top

Self-regulation and maternal mental health interactions from birth to age seven: Mother or child-driven effects?

Kate E. Williams1, Donna Berthelsen1, Jan M. Nicholson2 ,1, Sue Walker1
1School of Early Childhood, Queensland University of Technology, 2Parenting Research Centre, Melbourne

Children's self-regulation skills develop through dynamic processes between parent and child. There is substantial existing evidence for the influence of maternal mental health on children's general development, but limited studies on the longitudinal relationships between children's early self-regulatory capacity and maternal mental health. Further, although transactional models of child development hypothesise that child-driven effects as well as mother-driven effects are likely, there are limited empirical findings to support this view.

This study used data for 2880 children, participating in The Longitudinal Study of Australian Children (LSAC) Baby cohort. Sleep regulation and temperamental reactivity and persistence from birth to age five were used as self-regulation indicators in a series of longitudinal structural equation models (SEMs).

The mother-driven and child-driven models fitted the data well. Mothers with a history of depression were more likely to have infants with poor sleep and emotional regulation at Wave 1. Mothers with poorer mental health at Wave 1 were also more likely to have children with poorer sleep, emotional and cognitive regulation two years later. Evidence was also found to support child-driven effects. Even when mother's prior mental health was accounted for, children with poorer emotional regulation at Wave 2 were associated with poorer maternal mental health two years later, and poor sleep and emotional regulation in children at Wave 3 also contributed small but unique variance to maternal mental health at Wave 4. These findings contribute to a greater understanding of the mother-child system and highlight the importance of targeting maternal mental health in preventive strategies.

Back to top

The OECD Education and Social Progress Project (ESP), an international comparative study of the role of non-cognitive skills on the social progress – evidence from LSAC

Galina Daraganova, Ben Edwards
Australian Institute of Family Studies, Melbourne

The ESP project is an ambitious cross-country study that aims to identify learning contexts associated with the development of cognitive and non-cognitive skills and the role of skills in the development of social outcomes such as health, income and employment, crime and volunteering. While the ESP Project includes analyses of studies from over 10 OECD countries, in this paper we focus on the results for Australia only and will only refer to some key findings from other countries.

Using data from the K Cohort of LSAC and applying non-parametric dynamic latent variable models specifically developed for this project, the paper examines how family and learning contexts measured when children were 4-5 years, influences the development of non-cognitive and cognitive skills at age 6-8 years. We also examine the role of skills on obesity and conduct problems of children aged 10-11 years of age.

Our findings suggest that different contexts influence different skills. For cognitive skills the home education environment (e.g. visits to library, number of books and time spent with the computer) was an important influence while for non-cognitive skills parenting style was important.

Our analyses have also found that non-cognitive skills were more influential for the increase in conduct problems than cognitive skills while cognitive skills influenced obesity. The development of the non-parametric dynamic latent variables models are on-going, and as such these results are preliminary as future versions will incorporate more than one latent variable of non-cognitive skills.

Back to top

Parent-child engagement in home learning activities across early childhood: Socio-demographic predictors of change and associations with learning outcomes in the early years of school

Nicole Hayes1, Donna Berthelsen1, Sue Walker1, Jan M. Nicholson1 ,2
1Queensland University of Technology, 2Parenting Research Centre, Victoria

The importance of parent-child engagement in home learning activities in young children's early development and learning has been well established in international research. Less is known about the home learning environments of Australian children. Nationally and internationally, there is little evidence about the nature of changes across early childhood in the frequency of engagement in home learning activities and the effects on child outcomes as children begin school.

Drawing on data from Waves 2, 3 and 4 for the LSAC-B cohort, the current study used latent growth curve modelling to address the following three questions:

  1. What is the nature of change in engagement in home learning activities across the ages of 2-3, 4-5 and 6-7 years?
  2. What child, maternal and family characteristics are related to change in engagement in home learning activities across the ages of 2-3, 4-5 and 6-7 years?
  3. How does change in engagement in home learning activities across the ages of 2-3, 4-5, and 6-7 years relate to children's language and literacy, and numeracy outcomes on the Peabody Picture Vocabulary Test and Academic Rating Scale at age 6-7 years?

Results revealed that level of engagement in home learning activities decreased across the early childhood period. Child gender, maternal ethnicity, education, and family income were significant predictors of the rate of change in engagement. A faster rate of decrease in levels of engagement was also negatively related to children's language and literacy, and numeracy outcomes at age 6-7 years.

Back to top

Does school type affect academic achievement in young children? Evidence from Australia

Son Nghiem1, Ha Nguyen1, Rasheda Khanam2, Luke Connelly1
1University of Queensland, Brisbane, 2University of Southern Queensland, Toowoomba

Conventional wisdom and some existing research holds that private schooling enables children to achieve better academic results. If this is true, schools in the independent sector provide an educational model to which public schools should aspire and parents who are willing and able to pay for a private education may improve their child's educational attainment.

This paper investigates the effects of school choices on academic achievement in Australia using the Longitudinal Study of Australian Children (LSAC) linked with individual-level National Assessment Program - Literacy and Numeracy (NAPLAN) test scores for Year 3 and Year 5 as indicators of achievement. We mitigate the problems associated with individual unobserved heterogeneity by exploiting both the richness of LSAC data and contemporary econometric methods. We find that sending children to Catholic or other independent schools has no significant effect on their NAPLAN test scores. Moreover, using the value-added approach to measure school performance, we find that students in Catholic schools have significantly lower test scores on spelling, grammar and mathematics for grade 5. Support for the conventional wisdom is only found in naive estimates, where unobserved heterogeneity is not properly controlled.

Back to top

Sugar sweetened beverage and high fat food consumption are related to raised BMI z-scores among a cohort of Australian children from 4 to 10 years of age

Lynne Millar1, Bosco Rowland2, Melanie Nichols1, Boyd Swinburn1 ,3, Catherine Bennett2, Helen Skouteris2, Steven Allender1
1Deakin University, Geelong, 2Deakin University, Burwood, 3University of Auckland, Auckland, New Zealand

Introduction: There is little evidence from longitudinal studies on the determinants of obesity among Australian children. Much of the evidence is cross-sectional and most of the international data originates from the USA. It remains unknown if the findings from the USA where the culture of soft drink and fast food consumption is strong are generalisable to other countries where consumption is lower.

This paper describes relationships between consumption of sugar sweetened beverages (SSB), high fat foods (HFF) and longitudinal changes in standardized body mass index (BMI z-score) over 6 years among a prospective cohort of Australian children aged 4 to 5 years at baseline (2004).

Methods: Data from 4,164 children participating in the Longitudinal Study of Australian Children (LSAC) were analysed. A multi-level growth model was used to test relationships between BMI z-score and daily frequency of consuming SSB and HFF. BMI z-sore was calculated using WHO standards.

Results: The final model identified that higher levels of BMI z-score were associated with greater levels of HFF and SSB consumption, independent of BMI z-scores at wave 1. For every additional occurrence of daily SSB intake an increase of 0.015 in BMI z-score was observed. Similarly for every additional occurrence of HFF intake an increase of 0.014 in BMI z-score was observed.

Conclusion: While there has been a plateauing of childhood obesity at a population level, for additional gains in the prevention of obesity urgent action to address the impact of the consumption of SSBs and HFFs in childhood must be considered.

Back to top

Social determinants of Indigenous health: Factors associated with healthy childhood weight in the Longitudinal Study of Indigenous Children

Katherine Thurber1, Emily Banks1, Cathy Banwell1, Terry Neeman2
1National Centre for Epidemiology and Population Health, Australian National University, Canberra, 2Statistical Consulting Unit, Australian National University, Canberra

The increasing level of sedentary behaviour among Australian children has been implicated in the rise of overweight and obesity. Television viewing consumes, on average, a quarter of Australian children's waking hours; an estimated 80% of Australian children exceed the recommended daily maximum of two hours of screen time. Research has demonstrated a significant association between sedentary behaviour (commonly operationalized as television viewing) and Body Mass Index (BMI), independent of physical activity, among non-Indigenous children. The association has not, however, been widely examined within Indigenous children despite the high prevalence of overweight and obesity.

This paper will explore the association between sedentary behaviour and BMI in the Longitudinal Study of Indigenous Children (LSIC). Findings will be adjusted for physical activity, diet, family- and neighbourhood-level factors. The prevalence of sedentary behaviour and overweight are both high in LSIC: in Wave 3, around 90% of children over three years of age reported watching more than an hour of television daily, and nearly 25% of children aged 5-7 years were overweight or obese according to international BMI cut-offs.

LSIC presents a valuable resource for the longitudinal examination of the association between sedentary behaviour and BMI among Indigenous children, taking into consideration the social and cultural context. This research aims to uncover modifiable risk factors for the development of childhood overweight and obesity to inform interventions and policy. Interventions to date have focused on increasing physical activity and have met limited success; decreasing sedentary behaviour could present an alternative approach.

Back to top

The associations between child weight, dental problems and diet in young Australian children: A prospective study

Merrilyn Hooley, Helen Skouteris, Lynne Millar
Deakin University, Burwood

The association between dental problems and BMI was tested using data from 4149 children (51.5% male) participating in the Longitudinal Study of Australian Children (LSAC). Data from the first three waves were included with children aged 4-5 years, 6-7 years, and 8-9 years. Multivariate cross-sectional and prospective analyses were conducted to determine the relationships between child weight, diet and dental problems. Cross-sectionally, children aged 6-7 years had increased risk of dental problems if they were underweight and decreased risk if they were overweight. At 8-9 years, children had increased risk of dental problems if they were overweight or consumed high fat diet or high sugar diet. Prospectively, children who consumed high sugar diet at age 6-7 years had a higher risk of being overweight at age 8-9 years. Children with high consumption of high fat foods at age 4-5 years and 6-7 years, had higher risk of dental problems at the next wave, but children who consumed a high-fat diet at age 6-7 years had lower risk of overweight at age 8-9 years. Children who were under- and overweight at age 6-7 years had higher risk of dental problems at age 8-9 years.

The results suggest underweight and overweight at early ages predict future dental problems, but early dental problems do not predict underweight or overweight. Diets high in fat predict dental problems and high in sugar predict overweight. Combined and coordinated interventions designed to combat both obesity and dental caries may reduce incidence of both diseases.

Back to top

Using LSAC data: Bayesian network modelling for factors influencing Australian breastfeeding rates

Margaret Rolfe1 ,4, Lesley Barclay1 ,2, Jo Longman1 ,2, Sandra Johnson3, Kerrie Mengersen3
1University Centre for Rural Health North Coast, Lismore,2University of Sydney, 3Queensland University of Technology, 4Southern Cross University, Lismore

The aim of this study is to:

  1. Identify and estimate the relative strengths of the determinants of breastfeeding rates in Australia; and
  2. Understand the conditions under which the recommended rate of breastfeeding in Australia (80% when an infant is six months of age) may be achievable.

Waves 1 (2003/4) and 2 (2006) of the B cohort of the Longitudinal Study of Australian Children (LSAC) were used to determine breastfeeding rates at six months after birth. The national breastfeeding rate (defined as any breastfeeding) was estimated to be 51%, ranging from 46 to 56% between jurisdictions, thus falling well below the NHMRC 2003 guidelines of 80%.

To identify possible reasons for this low rate, we constructed a Bayesian network model using all known covariates (maternal, child and area specific) of breastfeeding rates that were available in the LSAC database.

Bayesian networks use a graphical representation of a decision problem by mapping the relationships among key variables. Probabilities are then assigned which represent the extent to which key variables relate to each other, and to include uncertainty. The advantages of Bayesian network models are the ability to view the problem graphically and for reasoning under uncertainty.

Using different probability scenarios (sensitivity analyses) of key breastfeeding determinants, for example by changing the maternal smoking rate, the conditional probabilities from the Bayesian network model have provided the basis for determining under what conditions the recommended rate of breastfeeding in Australia might be achievable.

Back to top

Protocols for the collection of sensitive data from adolescents in Wave 6 of The Longitudinal Study of Australian Children.

Amanda Christian1, Jennifer Renda2, Joanne Corey3
1Department of Social Services, 2Australian Institute of Family Studies, 3Australian Bureau of Statistics

Adolescence is a key developmental period, marked by substantial physical, social and emotional changes. Collecting information about these changes and associated behavioural changes as children become more independent is essential for a longitudinal study. As the K Cohort of the Longitudinal Study of Australian Children (LSAC) transition through adolescence, more potentially sensitive questions have been added to the study, including drug and alcohol use, antisocial and criminal behaviour, sexual feelings and experiences, contraception, pregnancy, STIs, suicide and self-harm. These topics can raise concerns about potential adverse impacts on the adolescent or their family, which may lead to withdrawal from the study. To help address these concerns, new protocols have been established for Wave 6 to ensure we maintain the on-going trust of our study families and respondents. Key changes to process include informing teenagers and parents about the topic areas covered in the interview, as well as allowing parents to refuse particular modules. These changes necessitated the redesign of the parental pre-interview letter and brochure, and the introduction of a child brochure. In-home protocols for the interviewer were developed, and training updated to give them the knowledge and confidence to manage the new procedures during the interview.

This paper examines past research on the impacts on adolescents of asking sensitive survey questions, explores methods used in other longitudinal studies to address the balance between moral responsibilities and informational needs, discusses the development process for these protocols, and outlines the results from the Wave 6 Dress Rehearsal.

Back to top

The trajectories of overweight/obesity among Indigenous, Non-English speaking and English-speaking Australian children

Jack Chen1 ,2, Lixin Ou ,2
1Simpson Centre of Health Services, Australian Institute of Health Innovation Research, University of New South Wales, Sydney, 2South-Wesern Clinical School, Faculty of Medicine, University of New South Wales, Sydney

Background: There is significant uncertainty in terms of the overweight/obese developmental trajectories among children from Non-English speaking background (NESB), Indigenous children (ATSI) in comparison to English-speaking (EB) children in Australia.

Methods: We explored the rates of overweight/obese among NESB, ATSI and EB Australian children using the data from the K-cohort of Longitudinal Study of Australian Children (for boys and girls, respectively). We estimated the rates through taking into account the complex survey designs. We also examined the association between the rate of overweight/obese and the ethnicity, mothers' marital status, mothers' age groups and children family's socio-economic position through survey-based Poisson regression.

Results: There was an absolute 7.7% increase of overweight/obese rate from 20.6% (4/5 yrs) to 28.3% (10/11 yrs) in overall sample. Such an increase was more pronounced in NESB children population (10.1%) in comparison to Indigenous children (7.8%) and EB children (7.3%). NESB children also had significantly higher overweight/obese rates in each of four waves in comparison to EB children. These differences were mostly due to the higher rates in NESB boys. In comparison to EB children, the overweight/obese rates of Indigenous children in Wave 2 (30.1% vs. 18.2%) and in Wave 3 (39.1 vs. 23.5%) were significantly higher. EB girls had higher rates of overweight/obese in comparison to EB boys between 4/5 to 8/9 but the gap disappeared at 10/11 years.

Conclusions: The grave challenges of combating overweight/obese and its life-course health impact remain. The sweeping policy initiatives and tailored interventions in targeting specific children groups should be considered.

Back to top

The effect of paid parental leave on child health

Barbara Hanel1, Guyonne Kalb2, Daniel Kuehnnle1, Miriam Maeder2
1Melbourne Institute of Applied Economic and Social Research, Melbourne 2The University of Nuremberg, Department of Economics, Nuremberg, Germany

We estimate the effect of eligibility for employer-provided paid parental leave on child health using LSAC. Eligibility for paid parental leave is strongly associated with the age when the child stops being breastfed. Care arrangements also differ across entitlement status, and children whose parents are eligible for paid leave spend more time with their father and mother at very young ages. Both could potentially impact on health outcomes later in life.

A large number of health outcomes are analysed, including the children's general health and weight, use of prescribed medicine and hospitalisation, prevalence of long-term health conditions, as well as specific health problems such as respiratory diseases, sensual disorders, infections, and developmental and mental health issues. We control for a wide range of personal characteristics and family characteristics, as well as health inputs during pregnancy and health status at birth. We follow children over all 4 waves of LSAC, and find that the immediate impact of paid parental leave on child health is relatively small, but positive health effects show up later in the child's life, particularly at age four to five years. If their parents are eligible for paid parental leave, the prevalence of digestive problems is lower in toddlers, and infections and respiratory diseases are less common among children of kindergarten age.

Back to top

A safe and supportive family environment for children

Killian Mullan, Daryl Higgins
Australian Institute of Family Studies, Melbourne

Families play a crucial role in protecting and supporting children. Previous research has shown that though the majority of children experience broadly cohesive family life, child wellbeing suffers in families where parents are overly disengaged from their children (referred to as disengaged families), or families where conflict between parents seeps into relationships between parents and children (referred to as enmeshed families). In Australia, governments have placed renewed effort towards protecting Australia's children through the promotion of, among other things, safe and supportive family environments as part of a broader public health approach to child protection. However, beyond research looking at the most 'at-risk families', we know relatively little about different family environments in the broader Australian population, and about the relationship between different family environments and child wellbeing.

With data from the Longitudinal Study of Australian Children (LSAC), this paper used latent class analysis to identify different family environments for children aged 2-3 years and 6-7 years (in the B cohort) and 4-5 years and 10-11 years (in the K cohort). In line with theoretical and empirical work, we identified three broad family environments (cohesive, disengaged, enmeshed). The paper considered links between these family environments and child outcomes at particular points in time, as well as links between changes in the family environment and child outcomes over time. The paper reinforces the importance of the family environment in promoting positive child outcomes, but draws attention to some areas of concern. The paper discusses some implications for policy.

Back to top

Fathers' leave, fathers' involvement and child development: Are they related? Evidence from four OECD countries

Jennifer Baxter
Australian Institute of Family Studies, Melbourne

Previous research has shown that fathers taking leave around childbirth are more likely than other fathers to be involved in childcare related activities, and children of more involved fathers perform better during the early years. This paper analyses data from Australia (using LSAC), Denmark, UK and US to describe how leave policies influence fathers' behaviours and how involvement translates into positive child cognitive and behavioural outcomes. Fathers who take leave, especially those taking two weeks or more, are more likely to do childcare related activities. Children with highly involved fathers performed somewhat better on cognitive scores but findings were weaker for behavioural outcomes. Results also suggested that the kind of involvement matters, indicating that what matters is the quality and not the quantity of father-child interactions.

Back to top

Symposium title: Parent mental health in the perinatal period and beyond: Trajectories, determinants and consequences
Presentation: Risk factors associated with fathers' persistent psychological distress across the early parenting period

Rebecca Giallo1 ,2, Daniel Christensen3, Fabrizio D'Esposito1 ,2, Amanda Cooklin1, Jan Nicholson1 ,2
1Parenting Research Centre, Melbourne, 2Murdoch Childrens Research Institute, Melbourne, 3Telethon Institute for Child Health Research, Perth

One in ten fathers report high levels of psychological distress in the first year after having a baby. However, little is known about the course of distress and associated risk factors beyond the postnatal period. Therefore, the aims of the study were to:

  1. report on the course of fathers' distress when their children were aged 3-12 months, 2-3 years, 4-5 years, and 6-7 years;
  2. identify distinct trajectories of distress over time; and
  3. identify risk factors for persistent distress.

Data from 2470 fathers participating in the Longitudinal Study of Australian Children were analysed. Latent growth modelling revealed that for the overall sample, distress was highest in first postnatal year and then decreased over time. Latent class growth modelling identified two distinct trajectories. The majority of fathers (92%) reported moderate distress in the first postnatal year which decreased over time, whilst 8% of fathers reported high distress during the first postnatal year which increased over time. Risk factors for persistent distress were poor relationship quality, low job quality, poor maternal mental health, and low parental self-efficacy.

The findings of this study highlight the existence of a group of fathers who experience persistent distress across the early parenting years. Timely interventions to address these difficulties are not only important for promoting the wellbeing of fathers, but also their children and families.

Back to top

Symposium title: Parent mental health in the perinatal period and beyond: Trajectories, determinants and consequences
Presentation: Postpartum maternal separation anxiety, over-protective parenting and children's social-emotional well-being

Amanda Cooklin1, Rebecca Giallo1 ,2, Fabrizio D'Esposito1 ,2, Sharinne Crawford1, Jan Nicholson1 ,2
1Parenting Research Centre, Melbourne, 2Murdoch Children's Research Institute, Melbourne

Moderate anxiety after separating from an infant in the postpartum is a normal maternal state believed to underpin bonding and protection of the infant. High levels of maternal separation anxiety (MSA) in the post partum may have deleterious effects on parenting and children's outcomes.
This study examined prospectively the relationships between postpartum MSA, over-protective parenting and children's social-emotional well-being at 2-3 years. Wave 1 and Wave 2 data from N=3103 mother-child dyads participating in the Longitudinal Study of Australian Children (Infant Cohort) were analysed using structural equation modelling. Postpartum MSA, psychological distress, over-protective parenting and children's socio-emotional functioning were included in the model, with socio-demographic characteristics included as potential moderator variables.

The overall model was a good fit to the data, explaining 24% of variance in children's social-emotional well-being at age 2-3. Higher maternal separation anxiety was a stronger predictor of over-protective parenting at 2-3 years (β = 0.49, p<0.001) than postpartum psychological distress (β = 0.07, p<0.01), and in turn associated with poorer child socio-emotional functioning at age 2-3 years. Findings suggest women with high postpartum maternal separation anxiety may sustain this heightened vigilance across the first years following birth, promoting over-protective behaviours, and resulting in increased behaviour problems in their children. Early postpartum support for women vulnerable to these early parenting difficulties may prevent the establishment of a repertoire of parenting behaviours that includes unnecessarily high monitoring and anxiety about separation.

Back to top

Symposium title: Parent mental health in the perinatal period and beyond: Trajectories, determinants and consequences
Presentation: Maternal mental health problems during pregnancy and associations with childhood asthma and food allergy

Rebecca Giallo1 ,2, Anita Kozyrskyj3, Amanda Cooklin1, Jan Nicholson1 ,2
1Parenting Research Centre, Melbourne, 2Murdoch Childrens Research Institute, Melbourne, 3University of Alberta, Edmonton, Alberta, Canada

Antenatal maternal distress has been associated with a range of health and wellbeing problems in children. This study specifically investigated the relationship between antenatal mental health difficulties, asthma and food allergy in a nationally representative sample of Australian children. Participants were 4226 children and their biological mothers from the infant cohort of the Longitudinal Study of Australian Children (LSAC). Mothers reported on antenatal mental health difficulties, and whether their child had experienced wheeze or whistling in the chest, on-going food or digestive allergies, or been diagnosed with asthma when aged 0-12 months, 2-3 years, 4-5 years and 6-7 years.

Approximately 19% of mothers reported antenatal mental health difficulties. After adjusting for known factors associated with asthma and allergy, antenatal mental health difficulties were associated with an increased risk of wheeze at age 0-12 months (OR, 1.34; 95%CI, 1.07-1.67) and 2-3 years (OR, 1.38; 95%CI, 1.13-1.69), as well as allergies at age 0-12 months (OR, 1.45 95%CI, 1.00-2.08). Asthma was 1.5 times more likely at age 4-5 and 6-7 years following antenatal mental health difficulties (OR, 1.45; 95%CI, 1.15-1.83 and OR, 1.45: 95%CI, 1.18-1.77, respectively). Finally, wheeze and asthma that was persistent across childhood was 1.4 times more likely for children of mothers who had antenatal mental health problems (OR, 1.37; 95%CI, 1.00 -1.89 and OR, 1.35: 95%CI, 1.04-1.75, respectively). These data implicate antenatal mental health difficulties in the development of childhood asthma and food allergy. Implications for policy and practice focussed on antenatal mental health screening and support will be discussed.

Back to top

Symposium title: Parent mental health in the perinatal period and beyond: Trajectories, determinants and consequences

Warren Cann, Rebecca Giallo
Parenting Research Centre, Melbourne

Parents are at heightened risk of mental health difficulties in the perinatal period, with approximately 16% of mothers and 10% of fathers reporting significant distress in the first year after having a baby. Such difficulties can have an adverse impact on daily functioning, couple relationships, parenting and children's wellbeing. The economic costs to the Australian community are also substantial, with the direct cost of maternal and paternal perinatal depression to the Australian health care system estimated to be $41 million in 2012. This symposium presents a collection of papers that build upon our knowledge of the extent to which parents experience mental health difficulties across the perinatal and early parenting period, the risk factors associated with such difficulties, and the consequences for child health and wellbeing. The implications of this body of research for both policy and practice focussed on perinatal mental health and parenting support will be discussed by Parenting Research Centre, CEO, Mr Warren Cann.

Back to top

Introducing Growing Up in Australia's Child Health CheckPoint: A physical health and biomarkers module for the Longitudinal Study of Australian Children

Melissa Wake1, Susan A. Clifford1, for the Child Health CheckPoint team1
1Murdoch Children's Research Institute, Melbourne, 2The Royal Children's Hospital, Melbourne

Background: Supplementing LSAC's rich dataset with objective health measures and bio specimens has been mooted since 2007. Made possible by one of the NHMRC's largest-ever project grants, LSAC's Child Health CheckPoint will take place in 2015, led by LSAC's senior health researchers and conducted via the Murdoch Childrens Research Institute working closely with the parent study.

Objective: To describe the Child Health CheckPoint protocol and anticipated output to LSAC researchers.

Methods: The CheckPoint is an additional data collection for the LSAC B cohort between main Waves 6 and 7. The 10-11 year old child and a parent attend a three-hour session at a 'pop-up' study assessment centre travelling through major Australian cities. Their comprehensive health assessment includes cutting-edge measures of cardiovascular (blood pressure, arterial ultrasound, pulse wave analysis, retinal photography) and respiratory (lung function) health, physical activity (accelerometry), time use (MARCA), fitness, body composition, vision, health-related quality of life, healthcare utilisation and others. Children may also provide a blood sample. De-identified data will be released to LSAC license holders with the Wave 7 data release in 2017. Access policies will be developed for further analyses, e.g of digital images and bio specimens.

Conclusions: LSAC's Child Health CheckPoint will integrate with lifetime trajectories of mental and physical health. The result will be a better understanding of how these trajectories interact with social circumstances to impact both on children's quality of life and the physiological and structural changes that precede disease. This evidence is essential to develop appropriate, early and effective prevention and intervention.

Back to top

General practice services for Australian children - results from linked data between the LSAC and the Medicare database, 2003-04 to 2010-11

Lixin Ou, Jack Chen, Ken Hillman
University of New South Wales, Sydney

Objective: To evaluate general practitioner (GP) visits for Australian children from 2003-04 to 2010-11, and to examine the relationship between socio-demographic characteristics and GP visits.

Methods: The data from the Longitudinal Study of Australian Children (LSAC) was linked to Medicare Australia claims data between 2003-04 and 2010-11. A total of 4779 infant cohort children and 4606 child cohort children were included in this study. GP consultations were identified by the item numbers 1-51, 597, 599, 601, 602, 5000-5067, 715 claimed from the Medicare Benefit Schedule (MBS).

Results: The adjusted incidence of GP visits decreased for the infant cohort children from 2007-08 to 2010-11 and for the child cohort children from 2004-05 to 2010-11, respectively. Boys visited less than girls in the infant cohort (P=0.04), and similar in the child cohort. Children from a non-English speaking background were more likely to visit GPs than Indigenous and English speaking background children (p<0.001). Children whose mothers were born in Australia, New Zealand, the UK, the US, Canada, South Asia and China were less likely to visit a GP compared with others (P<0.05). Poor socio-economic status associated with more GP visits. Children without private health insurance (P= 0.007) and children living in non-metropolitan areas (P<0.001) visited GPs less.

Conclusion: The average number of annual visits to GPs for Australian children declined and there were marked disparities in access to GP services among different ethnic and socio-economic groups. Policy intervention initiatives are needed in reducing the disparities in GP visits.

Back to top

How children's sleep patterns change from 0-9 years: Australian population longitudinal study

Anna Price1, Jude Brown2, Michael Bittman2, Melissa Wake1 ,3, Jon Quach1, Harriet Hiscock1 ,3
1Center for Community Child Health, Murdoch Children's Research Institute, The Royal Children's Hospital, 2School of Behavioural, Cognitive and Social Sciences, University of New England, 3Department of Paediatrics, The University of Melbourne, Melbourne

Background and Aims: To provide accurate population normative data documenting age-specific sleep patterns in Australian children aged 0-9 years old.

Design and Setting: The first three biennial waves of the nationally-representative Longitudinal Study of Australian Children, comprising two independent cohorts recruited in 2004 at ages 0-1 years (n=5107) and 4-5 years (n=4983).

Participants: Children with analysable sleep data for at least one wave.

Measures: At every wave, parents prospectively completed 24-hour time-use diaries for a randomly selected week or weekend day. 'Sleeping, napping' was one of 26 pre-coded activities recorded in 15 minute time intervals.

Results: From 0-9 years of age, 24-hour sleep duration fell from a mean peak of 14 (SD 2.2) hours at 4-6 months to 10 (SD 1.9) hours at 9 years, mainly due to progressively later mean bedtime with age from 8pm (SD 75 minutes) to 9pm (SD 60 minutes), and declining length of day sleep from 3.0 (SD 1.7) hours to 0.03 (SD 0.2) hours. Number and duration of night waking's also fell. By elementary school, wake- and bed-times were markedly later on weekend days. The most striking feature of the percentile charts is the huge variation at all ages in sleep duration, bed-time and, especially, wake-time in this normal population.

Conclusions: Parents and professionals can use these new percentile charts to judge normalcy of children's sleep. In future research, these population parameters will now be used to empirically determine optimal child sleep patterns for child and parent outcomes like mental and physical health.

Back to top

Healthcare costs associated with language difficulties up to 9 years of age: Australian population-based study

Emma Sciberras1 ,2, Elizabeth Westrupp3 ,1, Melissa Wake1 ,2, Jan Nicholson3 ,1, Nina Lucas4 ,3, Fiona Mensah1 ,5, Lisa Gold7 ,1, Sheena Reilly1 ,6
1Murdoch Children's Research Institute, Melbourne, 2Centre for Community Child Health, The Royal Children's Hospital, Melbourne, 3Parenting Research Centre, Melbourne, 4National Centre for Epidemiology and Population Health, Australian National University, Canberra, 5Clinical Epidemiology and Biostatistics Unit, The Royal Children's Hospital, Melbourne, 6Department of Paediatrics, University of Melbourne, 7Deakin Health Economics, Deakin University, Melbourne

Aim: Surprisingly little is known about the healthcare costs associated with language difficulties, precluding estimations of the economic implications of service improvements. This study aimed to quantify the non-hospital healthcare costs associated with language difficulties across the early to middle childhood years.

Method: Data from waves 1-3 of LSAC (B and K cohorts) were used to estimate the healthcare costs associated with language difficulties from birth to 9 years. Language difficulties were defined as scores ≤ 1.25 SDs below the mean on measures of communication (0-3 years) and vocabulary (4-9 years). Participant data were linked to administrative data (Medicare) on non-hospital healthcare attendances and prescription medications.

Results: Between 5 to 12% were defined as having a language difficulty at each wave. Two-year healthcare costs were higher for children with language difficulties at each age compared to those without language difficulties, most notably 36% higher (mean $AU206, 95% CI: $90 to $321; p=0.005) at 4-5 years (B cohort) and 29% higher (mean $AU141, 95% CI: $32 to $251; p=0.01) at 8-9 years (K cohort). The majority of costs were attributable to healthcare attendances rather than prescription medications. Two-year government costs ranged from $AU1.2 million at 6-7 years, to $AU12.1 million at 0-1 years when modelled to the Australian population. Six-year healthcare costs increased with the persistence of language difficulties (p=0.002).

Conclusions: Language difficulties are associated with substantial excess population healthcare costs in childhood. Healthcare costs increased with the persistence of language difficulties highlighting the importance of early preventative and remedial interventions.

Back to top

The intergenerational transmission of Indigenous languages within Aboriginal and Torres Strait Islander families

Walter Forrest
Australian Institute of Family Studies, Melbourne

More than half of the known languages that were spoken in Australia at the start of European colonisation are still used today, but three-quarters of them are severely or critically endangered. Failure to arrest these declines could mean that the vast majority of surviving Aboriginal and Torres Strait Islander languages will no longer be spoken within the next 30 years. The intergenerational transmission of languages (in which children learn languages from their parents and/or grandparents) is a key mechanism through which such declines may be reversed, but only about half the number of indigenous children whose parents speak an indigenous language also speak (or are learning) that language (Commonwealth of Australia, 2009).

In this paper, I analyse the results of the Longitudinal Study of Indigenous Children (LISIC) to identify the major protective and risk factors that influence the successful intergenerational transmission of indigenous languages within Aboriginal and Torres Strait Islander families. Particular attention is given to the community-level factors that influence successful language transmission independently of child and parent characteristics. Lessons and implications for the successful preservation of Australia's linguistic diversity are discussed.

Back to top

The role of early language impairments in the developmental pathways of emotional problems from ages 4 to 12: Evidence from the Longitudinal Study of Australian children

Richard O'Kearney, Shaun Goh
The Australian National University, Canberra

Not only do communication problems in children commonly co-occur with emotional difficulties but children with early language impairments also show higher levels of emotional problems such as depressed mood and anxiety and a higher risk for internalising disorders later in childhood and adolescence compared to young children with typically developing language.
This presentation has two aims. First, using latent growth curve modelling (LGM), it will document the trajectory of change in internalising problems from the beginning of the primary school years until the transition into high school and adolescence for LSAC, K- cohort children. It will then investigate whether there are clear, separate trajectories of change for fear and for self-focused distress across this period. The second aim is to provide evidence about how early language impairment contributes to the course of emotional difficulties during childhood either as a unique factor or in interaction with other child (temperament; gender), family (quality of parent-child interactions) and environmental (quality of peer interactions) influences.

The evidence coming from LSAC can be important contributors to improving interventions to reduce the psychological impact of early communication difficulties, and also to developing better targeted programs to enhance the modest benefits of current early intervention and prevention strategies for adolescent emotional disorders particularly depression.

Back to top

Delaying school entry- differential impacts for disadvantaged and relatively advantaged children?

Anna Zhu
Social Policy Research Centre, University of New South Wales, Sydney

Disadvantaged children have lower cognitive and non-cognitive developmental scores than their relatively advantaged counterparts. Narrowing these gaps when children are young can help to align their school learning trajectories, allowing for greater equality in later outcomes.

This paper explores how starting school at an older age affects the developmental score gaps between disadvantaged and relatively advantaged children. Using several waves of LSAC, it exploits the interaction of school eligibility rules and the child's birth month to exogenously randomise the school starting age. While previous findings suggest that delaying school entry may improve school readiness, little is known about whether it has differential effects for disadvantaged and advantaged children. For disadvantaged children, starting school early may be a better alternative to staying at home for longer as school provides a more stable, nurturing and educational environment than the family home, overcompensating for the penalties of starting school early. This trade-off may be less applicable to relatively advantaged children who generally have access to higher quality parenting, home environments and neighbourhood resources and who are more likely to utilise formal child care or pre-school services. However, delaying school entry may have differential effects on disadvantaged and relatively advantaged children for other reasons. For example, the latter children may adjust to the home-school transition differently from those in the former group as they have higher levels of school readiness at all ages. Alternatively, their parents may be more likely to delay their entering school until the year after they are first eligible.

Back to top

Reading and television viewing examined through the time use diaries of the first 3 waves of the Longitudinal Study of Australian Children (LSAC) K-Cohort

Jeanine Willson, Bernd Heubeck
Australian National University, Canberra

Since the advent of television there have been concerns about its impact on children's pursuit of developmentally important activities, because of the relatively large amounts of time children have devoted to viewing, even from its earliest days. Even with the advent of other screen based activities, television still dominates, at least for young children.

The impact of television on reading has received a particularly large share of research attention, because of the importance of reading for school achievement, future job prospects and pro-social life choices. Various ways in which television may impact on reading have been considered including language acquisition and vocabulary, emergent literary skills and leisure time reading, an important determinant of reading proficiency. While there has been research evidence that viewing educational content can have a positive impact on language and reading acquisition, research has generally focussed on the negative impact of viewing on reading.

Apart from studies examining content, researchers have in general explored the relationship between viewing and reading using total amounts of viewing, for reasons of ease and convenience. The time use diaries in LSAC provide a unique opportunity to examine patterns of time use and provide a more fine-grained analysis of the relationships and possible trade-offs between viewing and other activities, particularly reading activities. It is also possible to examine these patterns in relation to computer use. LSAC also provides the opportunity to examine how these patterns change in the K-cohort from 4 to 8 years.

Back to top

Symposium title: Fathers, mothers and jobs: How do they matter for child outcomes?
Presentation: Impact of income, job quality and time pressure on child outcomes in sole-parent families

Fabrizio D'Esposito, Nina Lucas, Amanda Cooklin, Jan Nicholson
Parenting Research Centre, Melbourne

Background: Since January 1st 2013, all sole-parent families receiving the Parenting Payment-Single (PPS) are being transferred to a lower income support payment, NewStart Allowance (NSA), when their youngest child turns eight. The implications of such policy reform for child development remain unclear.

Aim: To investigate the relationship between reduced income, parental employment factors and outcomes for Australian children aged 4 to 6 in sole-parent families led by mothers.

Methods: This study used data from 1,468 sole mother families who participated in the Longitudinal Study of Australian Children.

Results: Unadjusted analyses revealed that children of mothers whose income was below the PPS and/or below NSA performed more poorly across all ten child development outcomes assessed. When analyses were adjusted for maternal employment factors, associations between income and child outcomes were no longer significant. Feeling pressed for time, however, was found to be a risk factor for total developmental problems above income and other employment factors.

Conclusion: Policy reforms affecting income in sole-parent families should be considered in the context of the impact that they may have on maternal employment, competing demands and ultimately child outcomes.

Back to top

Symposium title: Fathers, mothers and jobs: How do they matter for child outcomes?
Presentation: Longitudinal effects of maternal work-family conflict and psychological distress at four stages of the family life-cycle (from child birth up to age 11)

Elizabeth Westrupp1 ,2, Angela Martin3, Lyndall Strazdins4, Amanda Cooklin1, Stephen Zubrick5, Jan Nicholson1 ,2
1The Parenting Research Centre, Melbourne,2The Murdoch Childrens Research Institute, Melbourne, 3University of Tasmania, Hobart, 4Australian National University, Canberra, 5University of Western Australia, Perth

The complexities associated with combining paid work and caring for children have increasingly gathered research interest in recent decades. Established literature has identified work-family conflict (WFC) as having negative associations with employee mental health; however, what remains unknown is whether WFC is a cause or a consequence of mental health.

This study aimed to more precisely document the nature of the relationship between work-family conflict and psychological distress for mothers across different stages of the family life-cycle (over 12 years of the early childhood period). Mothers of infants and young children require specific attention, given the unique pressures they face when combining work and family roles in this life stage. We used cross-lagged structural equation modelling with four waves of nationally representative survey data across two cohorts of mothers reporting sustained employment over the period their child was 0-5 (N>900) and 4-11 years (N>1300).

We found that work family conflict and psychological distress were distinctive, separate constructs and were relatively stable over the early to middle childhood period. Using cross-lagged models, we then showed a pattern of bidirectional influences between the two constructs, and also provided evidence that timing associated with the family-life cycle may shape the onward strength and nature of these relationships. Given the prevalence of work-family conflict, our results suggest that many employed mothers would benefit from policies and interventions targeting job conditions associated with work-family conflict, as well as interventions designed to directly enhance mental health may also be useful in reducing work-family conflict.

Back to top

Symposium title: Fathers, mothers and jobs: How do they matter for child outcomes?
Presentation: Fathers at work: Work family conflict, work-family enrichment and parenting

Amanda Cooklin1, Elizabeth Westrupp1 ,3, Lyndall Strazdins2, Angela Martin4, Jan Nicholson1 ,3
1Parenting Research Centre, Melbourne, 2Australian National University, Canberra, 3Murdoch Childrens Research Institute, Melbourne, 4University of Tasmania, Hobart

Contemporary fathering is characterised by the combined responsibilities of earning an income and being an involved, nurturing parent. The ensuing work-family interface is likely to affect fathering behaviours. However, the relationship between work-family conflict, work-family enrichment and specific fathering behaviours known to determine children's development has not been widely investigated.

Secondary data from fathers of 4-5 year old children (K-cohort, Wave 1) participating in the Longitudinal Study of Australian Children (LSAC) were analysed (N=2679). Results revealed that higher work-family conflict was associated with more irritable (β = 0.06, p<0.001), less warm (β = -0.04, p<0.01), and less consistent parenting (β = -0.07, p<0.001), when socio-demographic and child characteristics were controlled for. Protective associations were found between work-family enrichment and optimal parenting behaviours. Fathers' employed for long hours, and those who were the sole-earner in the household were most at risk of reporting high work-family conflict. Findings provide impetus for workplace and public policy to extend optimal, family-friendly employment conditions to all parents, including fathers, of young children.

Back to top

Symposium title: Fathers, mothers and jobs: How do they matter for child outcomes?
Presentation: Fathers' work-family conflict and mental health in the postnatal period – What matters most in the workplace?

Amanda Cooklin1, Rebecca Giallo1, Angela Martin2, Lyndall Strazdins4, Jan Nicholson1 ,3
1Parenting Research Centre, Melbourne, 2University of Tasmania, Hobart, 3Murdoch Childrens Research Institute, Melbourne, 4Australian National University, Canberra

Employment conditions are associated with work-family conflict (WFC), work-family enrichment (WFE) and mental health outcomes for all parents. However, the specific workplace characteristics associated with WFC, WFE and mental health for fathers in the postpartum period have not been identified in the Australian policy and workplace context.

This study analysed data from 1700 fathers participating in the LSAC B-cohort, with infants aged 0-12 months (wave 1). Path analysis was conducted to test a model of the relationship between a broad range of employment characteristics (e.g., work hours per week, occupation type, job quality) and fathers' psychological distress, where WFC and WFE mediate this relationship.

Results revealed that increased work hours per week, night shift, job insecurity, a lack of autonomy over workload, and inflexible work hours were associated with increased WFC, and this was in turn associated with increased distress. Job security, autonomy, and being in a more prestigious occupation were protectively associated with WFE and mental health. The strongest pathway from employment characteristics to fathers' distress was via WFC. These associations remained significant after adjusting for known risk factors for fathers' postnatal mental health difficulties (relationship quality; maternal mental health). Identifying key workplace factors that are associated with better outcomes for fathers is necessary to inform the provision of 'family-friendly' workplaces specific to this critical stage in the family lifecycle - the postnatal period.

Back to top

Who are tiger moms? An investigation of native and immigrant mothers' time investments in young children

Jen-Hao Chen
National Chengchi University, Taiwan

Children of immigrants often show more optimal developmental outcomes as compared to their native counterparts. The relative advantages of immigrant children are puzzling given their disadvantaged social positions. Conventional wisdom suggests that immigrant mothers' high level of supervision and time investments may explain such advantages. However, relatively few studies directly examine the immigrant-native differences in maternal time investments in details.

The present study used the time-use data from the Kindergarten Cohort of the Longitudinal Study of Australian Children (LSAC) to quantify and compare the levels of maternal time investments between native and immigrant children. Results show substantial native-immigrant differences in maternal time investments, after controlling for socioeconomic status and a wide range of demographic variables. Specifically, immigrant children spent substantially more time in educational activities, meals, and personal hygiene but they spent less time in free play and in sedentary leisure activities as compared to native children. Yet, such high level of time maternal investments was only observed among non-Anglophone immigrant mothers. There was no difference in maternal time investments between native and Anglophone immigrant mothers in Australia. Furthermore, immigrant mothers' patterns of time investments varied by maternal length of stay. Those who stayed in Australia for more than 10 years showed no difference in time investments as compared to native mothers.

Back to top

Separated parents' preferences regarding fathers' involvement in the lives of their children

Lixia Qu, Ruth Weston
Australian Institute of Family Studies, Melbourne

Consistent with an objective of the 2006 family law reforms, surveys undertaken around the time these reforms were being rolled out suggest that there is a general consensus among parents that the continuing involvement of each parent is beneficial for the children after parental separation. Nevertheless, separated mothers in these surveys were less likely than non-separated mothers to hold this view, while the reverse pattern applied to separated and non-separated fathers. The parent with minority care time (typically the father) may want increased time, while the other parent may resist this for various reasons. While many separated parents are able to maintain friendly and cooperative relationships with their former partners, this is not the case for some separated parents. Toxic relationships between separated parents can affect their preferences for the level of involvement of non-resident parents and their capacity to focus on children's needs. On the other hand, different views about the level of each parent's involvement in their children's lives may generate or intensify conflict.

This paper focuses on the views of mothers and fathers whose child spends most or all nights in the care of their mother (the most common arrangement). It examines the preferences of the mothers and fathers regarding the fathers' involvement in their children's lives and factors that are linked with such preferences.

Back to top

Exposure to inter-parental conflict during the early childhood period

Natalie Rose1, Elizabeth Westrupp1 ,2, Jan Nicholson1 ,2, Stephanie Brown2
1Parenting Research Centre, Melbourne, 2Murdoch Children's Research Institute

Background and aims: Rates of child exposure to inter-parental conflict (IPC) are high and appear to be increasing, with at least one million Australian children affected annually. To date, there are no established prevalence rates for more common but less severe forms of IPC for young families in the wider Australian community. The aims of the current study were to report on the prevalence and longitudinal course of IPC, and examine the co-occurrence and persistence of verbal and physical conflict over time.

Method: Analysis was conducted using all four waves of LSAC data from the infant (B) and kindergarten (K) cohorts. Verbal and physical IPC were defined from maternal report on the Argumentative Relationship Scale. Parent, family and child characteristics are described for parents with and without IPC at any wave.

Results: Findings indicate 19.8% (B-cohort) and 20.8% (K-cohort) of mothers reported IPC at any wave. Episodic rather than persistent IPC was more common: 6.7% (B-cohort) and 8% (K-cohort) of parents reported IPC at two waves, 3.8% (B-cohort) and 3.6% (K-cohort) at three waves and only 2% (B-cohort) and 1.5% (K-cohort) reported IPC at all four waves. Finally, parents experiencing verbal versus physical IPC represented distinct groups; co-occurrence of verbal and physical conflict ranged from 1.5-3.3% (B- cohort) and 1.9-3.2% (K-cohort) across waves.

Conclusions: Establishing accurate prevalence rates for the less severe but most common forms of IPC will allow family and child support services to more effectively allocate finite resources and develop targeted interventions to promote children's positive development.

Back to top

Parental joblessness, financial disadvantage and the wellbeing of parents and children

Jennifer Baxter1, Matthew Gray2, Kelly Hand1, Alan Hayes1
1Australian Institute of Family Studies, Melbourne, 2Australian National University, Canberra

Joblessness among families with children remains high in Australia relative to many other OECD countries. This report uses the first four waves of LSAC to analyse links between joblessness/short part-time hours of employment and the wellbeing of parents and their children. This is done through examination of the characteristics of single and couple parent families according to their employment status (jobless, short/part-time hours or longer hours). These categories are used to consider differences in income, indicators of financial wellbeing and social capital. Links between employment and financial measures were observed, but associations were less clear for social capital. Parents' and children's outcomes were also considered, to show how joblessness and lower levels of employment can flow through to poorer outcomes for families.

Back to top

Footprints in the workplace: Indigenous mothers' employment

Jennifer Thompson, Fiona Tetley
Department of Social Services (DSS), Canberra

One of the targets for Closing the Gap is to halve the difference in employment outcomes between Indigenous and non-Indigenous Australians within a decade. Mindful of the employment target, this paper explores the labour force participation of Indigenous mothers.

The scene is set with a picture of Indigenous and non-Indigenous mothers' labour force participation. ABS Census and administrative data are used to paint this picture, exploring characteristics such as age at birth of first child, the number of children, partnered status, level of education, housing tenure and geographic location.

The four waves of LSIC data provide further insights into the employment experiences of Indigenous mothers. The characteristics of mothers, including their experiences with paid employment prior to the birth of their children, are explored. The paper makes use of the LSIC data to examine the subsequent changes that take place with mothers' paid employment as their children grow up.

The findings provide a base from which policy development and employment service initiatives can address the circumstances and lived experiences of Indigenous mothers.

Back to top

The hidden side of reconciling work and family life: Young children's patterns of sleep and its consequences for parents' experience of mental health and work-family balance

Michael Bittman1, Lyndall Strazdins2
1University of New England, Armidale, 2Australian National University, Canberra

It has long been recognised that reconciling work and family life can cause difficulties for both the workplace and for family relationships. When workplace demands intrude into family life it is described as 'work-to-family-spill over' and, conversely, when the obligations of family life intrude upon the workplace this is described as 'family-to-work-spill over'. Generally more attention has been paid to work-to-family spill over. As a result there is more written on how characteristics of the workplace can be more compatible with family life - flexible start and finish times, part-time hours of work, family leave and so on. Less attention has been paid to family-to-work-spill over, and the extent some family demands may complicate parents' work performance. Interruption to sleep is often unavoidable for parents with babies and young children, and, when parents are employed, it is likely to undermine both productivity and parent wellbeing.

Using an analysis of the first wave of the Longitudinal Study of Australian Children (LSAC), which contains time-diary information about children's activities, we describe the patterns of young children's sleep behaviour that may be disruptive for parents. Our analysis shows that there are wide variations among children at different stages of maturation, and for a few families, sleep can be disrupted over several years. We then examine the extent children's sleeping patterns may be linked to parents' experience of being harried, psychologically distressed, and their perception of work-family balance.

Back to top

Symposium title: Educational contexts: Vulnerability, access and learning outcomes

Convened by Linda Harrison and Sue Walker

Presentation: Does speaking a language other than English at 4- to 5-years-old impact literacy, numeracy, and social-emotional outcomes at primary school?

Sharynne McLeod1, Sue Walker2, Chrystal Whiteford2, Linda Harrison1
1Charles Sturt University, Bathurst, 2Queensland University of Technology, Brisbane

This study examined the impact of a language background other than English (LOTE) when 4- to 5-years on teacher-rated literacy, numeracy, and social-emotional outcomes at 6- to 7-years and 8- to 9-years for children with and without speech and language concerns. Participants were from the LSAC - K cohort.

First, a matched-case control design enabled comparison between four groups of 109 children: a) English + typical language, b) LOTE + typical language, c) English + language concerns, and d) LOTE + language concerns. Next, two hierarchical multiple regression analyses were undertaken for the entire LSAC sample to test child outcomes at 6/7 years and 8/9 years. Variables were entered in three steps: child and family factors, language concerns and language status, and competencies at 4- to 5-years. Although 4- to 5-year-old children who spoke a LOTE performed more poorly than English only children on school readiness measures (PPVT and the Who am I?), at both age 6/7 years and 8/9 years, speaking a LOTE did not predict academic or social-emotional difficulties. Significant predictors for Language and Literacy included SEP, parent and teacher reported use of speech therapy, expressive language concerns, and 4/5 PPVT.

Significant predictors for Mathematical Thinking included child sex, SEP, parent and teacher reported use of speech therapy, and 4/5 PPVT. SDQ significant predictors included child sex, SEP, and receptive language concerns. These results indicate that speech and language concerns at 4- to 5-years was a greater predictor of school-age academic and social-emotional difficulties than speaking a LOTE.

Back to top

Symposium title: Educational contexts: Vulnerability, access and learning outcomes

Convened by Linda Harrison and Sue Walker

Presentation: Following the path of Australian children with disabilities as they progress through school

Sarah McDonagh1, Loraine Fordham1, Julie Dillon-Wallace2
1Charles Sturt University, Bathurst, 2Queensland University of Technology, Brisbane

The landscape of education in schools and classrooms continues to undergo rapid change for children with disabilities. Children whose educational needs previously resulted in specialised services and separate placements are now primarily receiving their education in mainstream regular class and school settings. While much discussion has centred on teacher preparation and professional development to support the academic progress of children with disabilities in regular school settings, there exists a paucity of research that examines how successfully the ideal of inclusion has translated into practice for students with disabilities in the Australian context.

Using the teacher questionnaire data collected for the K cohort over Waves 1-4 of the LSAC study, this presentation will examine both the representation of and outcomes for children with disabilities in Australian schools. Categories of disability identified in this analysis include: intellectual, sensory, physical, speech/language impairment, learning difficulties, emotional/behavioural problems, and autism. The presentation will examine longitudinally the contexts in which education is provided to LSAC students identified as having a disability, and determine how these children are performing in English language and literacy, mathematical thinking, social development, and approaches to learning as rated by their classroom teachers.

Findings will be discussed in relation to policy decisions in regard to programs to support the successful inclusion of children with disabilities in inclusive school settings, including how teacher preparation and professional development can best meet the needs of diverse student populations in a variety of educational settings.

Back to top

Symposium title: Educational contexts: Vulnerability, access and learning outcomes

Convened by Linda Harrison and Sue Walker

Presentation: Identifying the cumulative effects of disadvantage on children's utilisation of ECEC through the application of a Disadvantage Index

Sandie Wong1, Chrystal Whiteford2, Corine Rivarlland3, Linda Harrison1
1Charles Sturt University, Bathurst, 2Queensland University of Technology, Brisbane, 3Monash University, Melbourne

This paper draws on Wave 3 data from the Birth Cohort of Growing Up in Australia: The Longitudinal Study of Australian Children (LSAC)* to identify utilisation of early childhood education and care (ECEC) by disadvantaged children. Further, through the development of a Disadvantage Index it examines the cumulative effects of disadvantage on children's utilisation of ECEC. The Disadvantage Index was constructed as a continuous, summed measure of three child-level indicators of disadvantage (disability/health condition; Language background other than English; Aboriginal and/or Torres Strait Islander) and four family-level indicators (Socio-economic Position; Maternal and paternal health; Maternal and paternal mental health; Household drug and alcohol problems).

Results showed that children with multiple indicators of disadvantage (3 or more) were more likely to be in exclusive parental care, less likely to be using preschool, and to attend fewer hours of ECEC than their more advantaged peers. However, there was no effect of disadvantage on children's use of long day care. These findings suggest that there may be barriers to utilisation of preschool services for children and families for whom ECEC potentially has the most benefit.

*Restricted to the subsample of children who had not yet started school, n = 3,615.

Back to top

Disadvantaged families and disadvantaged areas: Relationships in LSAC waves 1-3

Bryan Rodgers1, Bina Gubhaju1, Lyndall Strazdins1, Peter Butterworth1, Tim Crosier1 ,2, Tanya Davidson1
1The Australian National University, Canberra , 2Department of the Prime Minister and Cabinet, Canberra

Current policies and the resourcing of services are influenced by the view that "disadvantage is highly concentrated" geographically. This presentation explores:

  1. the extent to which disadvantaged families in LSAC are located in disadvantaged areas defined by SEIFA scores;
  2. the degree to which families living in disadvantaged areas remain in disadvantaged areas over time; and
  3. whether living in a disadvantaged area affects the likelihood that disadvantaged families will remain disadvantaged over time.

These analyses use comprehensive measures of family disadvantage covering 12 constructs and two higher-order components of disadvantage (material and psychosocial). Findings are somewhat different for different measures of family adversity.

  1. For material adversity, disadvantaged families are more likely to live in disadvantaged areas. However, they are not "highly concentrated" in these areas. There is little association between psychosocial family disadvantage and living in disadvantaged areas.
  2. Over a four year period (e.g. wave 1 to wave 3 of LSAC) about half of the families originally living in disadvantaged areas, no longer live in a disadvantaged area and there is comparable movement in the opposite direction.
  3. Area of residence has little influence on whether families escape from disadvantage over time (either material or psychosocial).

The connections between area disadvantage and family disadvantage are not as strong as often assumed. Family disadvantage is not highly concentrated and living in a disadvantaged area is not linked to entrenchment of family disadvantage.

Back to top

How do housing conditions and housing instability affect the health of Indigenous and non-Indigenous children?

Anastasia Sartbayeva, Leo Bild, James Caffrey, Jason Brandrup
Department of Social Services (DSS), Canberra

There is limited research published on how housing conditions affect the health of Australian children. International studies have reported a number of connections between housing conditions and the physical health outcomes of children. This paper aims to examine whether these connections also hold for children in Australian families.

Several housing conditions have been identified as potentially affecting physical health pathways. For instance, lack of appropriate household equipment can influence the likelihood of facing infections and disease, which may lead to long-term health problems. Overcrowding may also contribute to poorer health by spread of infections and parasitic diseases.

Growing up in Australia and Footprints in Time bothprovide an opportunity to investigate these proposed pathways. Both surveys ask questions on the physical health of the study children and issues related to housing problems. The longitudinal nature of these surveys also enables us to observe how housing conditions and housing stability may associate with the children's health over time. Footprints in Time's focus on Indigenous children enables us to compare the effects of housing on our indigenous population and the general population represented within LSAC.

Previous published work on these datasets has shown that poor quality housing can influence children's social and emotional wellbeing and general health. Our analysis here intends to expand upon this work to examine the effect housing has on specific physical health conditions for children.

Back to top

Participation in community groups, maternal mental health and child outcomes

Hugh Webb, Helene Shin
Department of Social Services (DSS), Canberra

Extensive research demonstrates that an individual's participation in diverse social groups improves their mental and physical health. However, there is limited research examining the benefits of social group participation at a population level. Research suggests that benefits vary depending on the circumstances of individuals and the type of group involvement. Social policies that support participation in community groups could be particularly helpful for disadvantaged mothers with young children, given the relatively high rates of maternal mental health difficulties. In improving maternal mental health these policies may also strengthen mothers' attachment to the workforce and improve their children's outcomes.

Using data from the K cohort of the Longitudinal Study of Australian Children, the current research tests whether participation in social groups, such as volunteer and community organisations, is prospectively associated with improved maternal mental health. Whether such improvements in turn predict better socio-emotional and behavioural outcomes for children between 4-5 and 12-13 years of age is also examined, as well as the moderating effects of economic disadvantage, family circumstances, and culturally and linguistically diverse background. The findings provide insights into the potential benefits of social involvement for women and their children, and for which particular groups of women such benefits are most significant.

Back to top

Caring for a child with persistent chronic health problems and maternal participation in the labour force, job tenure, job security and short working hours (<10 hours)

Nick Spencer1 ,2, Lyndall Strazdins2
1University of Warwick, Coventry, UK, 2Australian National University, Canberra

Objective: To study the impact of having a child with persistent chronic health problems on maternal participation in the labour force, job tenure & security, and working <10 hours.

Methods: Using data from the 2nd, 3rd & 4th waves of the B cohort of the LSAC, we created a dummy variable for persistent chronic health problems among cohort children (chronic problems reported at >1 wave) which was entered into logistic regression models with maternal participation in the labour force, job tenure & security, and working <10 hours at the 4th wave as dependent variables and family hardship and lone parenthood at the 2nd wave as potential confounding variables.

Results: Complete data were available on 2283 mother/child dyads. 292 children were reported to have chronic health problems in more than one wave of the B cohort. In univariate analysis, only maternal job insecurity was significantly associated with caring for a child with persistent chronic health problems (OR 1.69 (95%CI 1.11,2.58) at the 5% level. Crude ORs for not in the labour force, causal job tenure and short hours were 1.28 (95%CI 0.99,1.65), 1.41(95%CI 0.94,2.18) and 0.91(95%CI0.58 1.41) respectively. Adjustment for family hardship and lone parenthood at the 2nd wave slightly attenuated the association with job insecurity (AOR 1.65 (95%CI 1.08,2.54).

Conclusion: Our findings suggest that caring for a child with persistent chronic health problems during the first 6 years of life is associated with increased odds of the mother experiencing job insecurity when her child is aged 6 years.

Back to top

Perinatal risk and physical health outcomes in Indigenous Australian children: Footprints in time: The Longitudinal Study of Indigenous Children

Elizabeth Westrupp1 ,2, Fabrizio D'Esposito1 ,2, Fiona Mensah2 ,4, Jane Freemantle3, Bree Heffernan3, Jan Nicholson1 ,2
1The Parenting Research Centre, Melbourne, 2The Murdoch Children's Research Institute, Melbourne, 3The University of Melbourne, 4The Royal Children's Hospital, Melbourne

Background and aims: Low birth weight, preterm birth and small for gestational age are perinatal risks experienced frequently amongst Aboriginal and Torres Strait Islander (Indigenous) children in Australia; 12.5% of Indigenous live births are less than 2500g compared to 6.2% in the non-Indigenous population. The long-term health and the potential pathways to improving outcomes for these children are unknown. We compare the health of Indigenous children born with perinatal risks to healthy term children and explore associated maternal and socio-economic factors.

Method: Participants were children from the Baby and Child cohorts of the Longitudinal Study of Indigenous Children (N=1440). Data from three waves of each cohort were examined (Baby cohort: 3 months-6.6 years; Child cohort: 1.1-8.0 years). Mild (N=285, 19.8%) and moderate-to-high (N=140, 9.7%) perinatal risk groups were identified using birth information provided by the primary carer. Carer-rated global health and disability, and interviewer-measured body mass index (BMI), were compared across perinatal risk groups using regression models, adjusting for maternal and socio-economic factors.

Results: We found that moderate-to-high perinatal risk predicted poor global health, disability, and lower BMI, while mild perinatal risk predicted lower BMI only. The effect of moderate-to-high perinatal risk on global health and mild risk on BMI appeared to be mediated by maternal and socio-economic factors, while the other effects were robust to adjustment.

Conclusions: Long-term outcomes of perinatal risk are a significant health problem for the Indigenous community. Health promotion focusing on maternal and socioeconomic factors may be a potential route to addressing these disparities.

Back to top

Is social disadvantage on the causal pathway to, or a consequence of, chronic disabling conditions in childhood?

Nick Spencer1 ,2, Lyndall Strazdins2
1University of Warwick, Coventry, UK, 2Australian National University, Canberra

Objective: To study the temporal relationship of social disadvantage with chronic disabling conditions (CDC) in childhood.

Method: Using data from waves 2 and 4 of the LSAC K cohort, the dependent variable was derived as follows: those reported to have no CDC at age 6/7 but had developed one at age 10/11 (index=1) and those without a CDC at either age (comparison=0). Logistic regression models were fitted on the dependent variable with household income quintiles at 6/7 years as the main independent variable adjusted for potential confounders. Change in mean household income and material hardship between waves 2 & 4 was compared for index and comparison groups.

Results: Of 4010 children with data for both years, 157 (3.9%) were included in the index group and 3526 (87.9%) in the comparison group. After adjustment for confounding, the children in the lowest household income quintile were almost 4 times as likely to develop a CDC as those in the highest income quintile. The difference in mean household income in the index and comparison groups between waves 2 & 4 did not change significantly; however, households with a child developing a CDC between waves 2 & 4 were significantly more likely to experience greater hardship in wave 4 (p<0.01).

Conclusion: The findings suggest social disadvantage at 6/7 is part of the causal pathway to CDC at 10/11. Having a child who develops a CDC between ages 6/7 and 10/11 does not significantly depress household income but appears to increase household hardship possibly by adding to household expenditure.

Back to top

PanelWhiz: A graphical user interface in stata for extracting data from the LSAC and LSIC

John Haisken-DeNew
University of Melbourne, Melbourne

It will be demonstrated how the user can extract data from the LSAC and LSIC in Stata format by using the graphical user interface created by PanelWhiz. Users are able to click on items, i.e. vectors of variables linked over time and select all variables automatically. The PanelWhiz system then generates Stata code on the fly to extract the data required. All matching logic is done by PanelWhiz. PanelWhiz allows extracting data from many Australian datasets such as: HILDA, LSAC, LSIC, MABEL, CASiE.

Back to top

Redressing child health disparities: What should be our outcome measure?

Fiona Mensah1 ,2, Jan Nicholson1 ,3, Melissa Wake1 ,2, Darren Wraith1 ,4, Donna Berthelsen5, C. Jane Freemantle6, John Carlin1 ,2
1Murdoch Childrens Research Institute and The Royal Children's Hospital, Melbourne, 2Department of Paediatrics, University of Melbourne, Melbourne, 3Parenting Research Centre, Melbourne, 4Centre for Molecular, Environmental, Genetic and Analytic Epidemiology, Melbourne School of Population and Global Health, University of Melbourne, 5Centre for Learning Innovation, Queensland University of Technology, 6Centre for Health & Society and Onemda VicHealth Koori Health Unit, The University of Melbourne

Health promotion is integral to child well-being and the potential benefits of such investment are evident. Most affluent countries are making substantial social and health investments in prevention research and programs to support parents and communities in raising healthy children. Approaches that address the strong, shared social determinants common across multiple aspects of health and development offer a comprehensive public health-oriented strategy for redressing health disparities.

In efforts to reduce health disparities in early life, the development of innovative multi-outcome approaches to redress the wide reaching vulnerability engendered by social disadvantage is encouraged. This raises the challenge of how we may best evaluate such approaches, particularly with regards to what the key child outcomes may be and how and when these may be evaluated most informatively. 'There is no single directly observable variable that can be said to capture health itself'.

Typically, the impacts of actions to promote health and prevent disease are evaluated in terms of the changed prevalence of a range of specific outcomes, usually considered separately. However joint tests that consider multiple outcomes have the potential to enable more powerful and comprehensive evaluations than univariate methods alone.

The LSAC cohorts give us a rich catalogue of health and developmental outcomes. Their measures extensively assess physical, cognitive and socio-emotional health and development from infancy through to early adolescence. In this paper we examine these data to make recommendation of how we may comprehensively evaluate health interventions and social policies for redressing health disparities.

Back to top

Fixed effects matter: The importance of panel data for understanding cognitive development in Australian Indigenous children

Richard Brown1, Bethany Davies1, Prabha Prayaga1, Simon Quinn2
1School of Economics, University of Queensland, Brisbane, 2Department of Economics and Centre for the Study of African Economies, University of Oxford, UK

We use the LSIC panel dataset to measure the effect of life shocks upon cognitive development of Indigenous children in Australia. We consider determinants of standardised test scores, and we focus attention on a range of potential shocks to child progress (including shocks to child health, guardian health and to family circumstances). We use the Mundlak (1978) method to test the importance of fixed effects while allowing clustered errors. We find significant correlation between child fixed effects and a wide range of shocks, implying that simple cross-sectional comparisons are likely to be highly misleading in diagnosing the challenges to Indigenous children's test scores. We show that this result is robust to controlling for a wide variety of observable baseline characteristics; such controls produce more accurate estimates, but do not adequately resolve the fundamental problem of time-invariant fixed effects. We conclude that panel data are critically important for understanding the causes and consequences of Indigenous poverty in Australia.

Back to top

Investigating the predictors of NAPLAN attendance- a measure of the impact of policy reform

Paula Cronin, Rebecca Reeve, Stephen Goodall
CHERE, University of Technology, Sydney

This study aims to investigate the predictors of NAPLAN attendance.

A sample from Wave 4 of the K-Cohort of the LSAC (N≈4048) was analysed using MNL models. The models examined the impact of child, parent and school related variables on NAPLAN attendance. Four models were developed. The initial model is a main effects model. Model 2-4 further examines the impact of school-type on NAPLAN attendance.

Results indicated that the teacher assessment of child's overall performance relative to their peers was a significant predictor of a child's non-attendance in NAPLAN testing (-0.276***). Children with higher overall performance ratings were less likely to be non-attenders in NAPLAN. Children in Catholic and Independent schools were less likely to be non-attenders in NAPLAN (-0.908***, -0.535** respectively). In contrast Indigenous students were more likely to be non-attenders (0.532***).

Further analysis of the impact of school-type showed that in the government schools model, the teacher assessment of performance and Indigenous status remained significant predictors of non-attendance, (-0.436***, 0.720* respectively). However, these effects dropped out of the Catholic and Independent schools models. Frequent absences from school were found to be important predictor of NAPLAN non-attendance in the Catholic schools sample (1.336*).

There were no significant effects found for low-SEP students or parental-assessment of overall progress.

This research suggests that the introduction of the government's education-policy-reforms is impacting the behaviour of government schools with regards to influencing NAPLAN attendance, specifically towards below average performing students. Further research is required to investigate the impact on other disadvantaged groups.

Back to top

Learning to read English: what can we learn from the Aboriginal and Torres Strait Islander children who are doing well?

Fiona Skelton, Deborah Kikkawa, Sophie Balch, Stephen Zubrick
Department of Social Services (DSS), Canberra

Governments across Australia are keen to close the education gaps between Indigenous and non-Indigenous children. Parents and carers of Indigenous children want their children to have a good education, as indicated by responses in waves 1 and 2 of Footprints in Time. Many of the Indigenous children in Footprints in Time are doing well, have strong vocabularies and are successfully learning to read. What can we learn from those who are doing well to help those who are experiencing difficulties? What helps children get a good education?
In wave 4 the older cohort, children in Footprints in Time (aged 7-9) completed a reading assessment achieving a wide range of scores. This presentation will examine the link between a variety of characteristics of the children and their families on their wave 4 reading scores. The analysis will build on existing findings such as the importance of reading to children and parental education and explore other ways that parents and carers of Indigenous children support their children's learning.

Of specific interest is how strongly outcomes and activities recorded in earlier waves help explain the wave 4 reading assessment scores, compared to the impact of contemporary activities and characteristics. There are important policy implications for the design of Indigenous early childhood education programs depending on whether early outcomes from a very young age appear to be persistent over time or whether children are shown to be more responsive to current activities and interventions.

Back to top

Early childhood assessment in the German National Educational Panel

Anja Sommer, Ina-Sophie Ristau
Otto-Friedrich University of Bamberg, Bamberg, Germany

Early childhood is nowadays acknowledged as an important and crucial phase of life, which lays the ground for further development and achievement. Although many study results highlight the importance of familial and institutional learning environments and educational processes in early life, there is still further research needed to explore the exact interplay of different aspects. For example, interaction between family and institutional settings and the influence of context, structural and process variables, such as quality of interactions, parental health or socio-economic status.

To close this gap, the German National Educational Panel Study has been set up to find out more about the acquisition of education in Germany to plot the consequences of education for individual biographies, and to describe central educational processes and trajectories across the entire life span. Two of the six representative starting cohorts within the NEPS focus on the early childhood. To picture a wide range of variables on early childhood, assessment is not only limited to interviews with parents, but also includes caregiver questionnaires and different competence testing with the children. Since 2010, 3,000 7 months old infants and 3,000 4-year-old kindergarten children are followed through different life stages, educational settings and trajectories.

The data of each wave gets released as a scientific use file, which allows further analysis on early childhood and the interplay of its influencing variables.

Back to top

Children's perceptions of their parents' jobs

Jennifer Baxter2, Lyndall Strazdins1
1Australian National University, Canberra, 2Australian Institute of Family Studies, Melbourne

This research uses LSAC K cohort data from Wave 4, when children aged 10-11 years were asked their perceptions of their parents' work. Children have a mix of views about parents' jobs, many of them positive. A minority of children wished their parents did not have to work at all, most children thought their parents worked about the right amount of hours and most children thought that their parents liked their job. Boys and girls tended to show similar views about their parents' jobs, and there were just a few distinctions based on parent gender. These analyses also examined how children's perceptions varied with the characteristics of parents' jobs, and with parents' own reports of work to family spill over.

Back to top

Parenting through domestic violence: Examining the effects of the mother's mental health and parenting techniques on child outcome in the context of partner violence using LSAC K Cohort data

Helene Shin, Helen Rogers
Department of Social Services (DSS), Canberra

Domestic violence awareness has increased across Australia over the past few decades. The Government's commitments to the 'National Plan to Reduce Violence against Women and their Children' and a range of national actions under the plan including establishing a national telephone and online counselling service including support for workers in other services and the expansion and reform of training programs for health workers, also reflects recognition of the importance of local support services for people experiencing or at risk of experiencing violence at home. Exposure of children to intimate partner violence (IPV), even without direct abuse, is recognised as a form of child maltreatment. Approximately one-third of women who experience violence from their partner report their children witness the violence (ABS, 2005). Previous research shows children exposed to IPV are more likely to develop depression; exhibit aggressive or anti-social behaviour; experience low self-esteem; and have difficulties at school including peer conflict. In the longer term, they have greater chance of teenage pregnancy, eating disorders, and intergenerational transmission of IPV (e.g. Humphreys, 2007). Using several indicators of partner violence in Wave 4 of the Longitudinal Study of Australian Children (LSAC), this preliminary study investigates the extent to which a mother's mental health and parenting techniques, along with other contextual variables, protects her children from being negatively affected by the partner violence. Findings identify key issues to stimulate policy discourse concerned with children vulnerable to domestic violence. Further analyses and policy implications will be discussed in detail during the presentation.

Back to top

Are single mothers bad mothers? Evidence from LSAC

Jan Nicholson1 ,2, Fabrizio D'Esposito1 , 2, Nina Lucas3 , 1, Elizabeth Westrupp1 , 2
1Parenting Research Centre, Melbourne, 2Murdoch Childrens Research Institute, 3Australian National University, Canberra

Background: Children from sole parent family's fare more poorly on multiple outcomes than those in two-parent families. Most explanations for these differences assume that compromised parenting and parent mental health play a central role.

Aims: To explore the contribution of a range of child, socio-demographics, employment, family and environmental factors to the parenting and mental health of single mothers in Australia.

Methods: This study used data from wave 3 of the Longitudinal Study of Australian Children. We compared data from 994 single mothers with data from 7562 mothers in a couple relationship when their children were aged 4-5 (Baby Cohort) or 8-9 years (Kindergarten Cohort).

Results and Conclusions: Our findings show that single mothers are more likely than couple mothers to experience parenting (OR: 1.50-1.63; p≤0.001) and mental health difficulties (OR: 2.7-3.0; p<0.001), however they also face heightened adversity in their home and extra-familial environments. Importantly, our comparison of single and couple mothers facing similar levels of adversity show no difference in poor parenting practices, although single mothers remain more vulnerable to psychological distress. These findings have policy implications as they challenge the prevailing view that single-parent families inherently provide suboptimal environments for raising children.

Data workshop

Data Workshops were held on the day following the conference. The focus of the training was to assist users of the data, those considering becoming users, or those who are interested in learning more about LSAC data, to gain confidence in understanding and navigating the LSAC datasets. The training covered a range of topics designed to give a comprehensive overview of the conduct of the study, its datasets and supporting documentation.

To register your interest or find out more, contact the LSAC Team at AIFS.

Back to top

LSAC Data Users newsletter

Newsletter thumbnail

Keep up-to-date with LSAC news for data users.

Sign up to our email alert service.