Annual Report 2005‚Äď06

Parenting and families in Australia

(Taken from the unpublished report commissioned by FaCSIA titled Parenting and families in Australia by Steve Zubrick, Grant Smith, Jan Nicholson, Ann Sanson and Tanyana Jackiewicz.)

Parenting is one of the 'engines' of child development. Aspects of parenting include not only the provision of the necessary means to create safe, sustaining environments but entail parental expectations about the capacities of children and the provision of opportunities that prompt and facilitate their development. Parenting is a process that results in change for both the child and the parent and highlights the active participation of the developing child in the process.

Evidence from the first wave of data from Growing Up in Australia shows that the vast majority of parents are doing well in the task of caring for their families and parenting their children. It also reveals a clear link between parenting quality, parenting self efficacy and the overall functioning of infants and children. Family and parenting characteristics 'matter' for children's development, with parenting practices having a particularly prominent role. Even when adjusted for parental income, education, employment, family structure and parental wellbeing, the independent association of poorer parenting quality with poorer developmental outcomes remains for both the children aged 0-1 (B or infant cohort) and the children aged 4-5 (K or child cohort). Thus, parenting styles show clear associations with child development from very early in life and these associations are sizeable and, in all likelihood, persistent over time.


The outcome measure used in these analyses (LSAC Outcome Index) is designed to provide a general indication of how children are developing. It is a composite of three domains which are proposed to be the major components of current wellbeing and the future capability to be a successful member of society: health and physical development; social and emotional functioning; and learning and academic competency. Summary scores for each of these domains are calculated, and they are combined into the overall Index.

The analyses that follow show the significance of a number of independent variables in predicting the likelihood of a child falling below the negative cut-off (that is, the bottom 15 per cent) on the Outcome Index. This information is conveyed through odds ratios. The odds ratios presented below indicate how membership of a category within an independent variable predicts the likelihood of a child falling below the negative cut-off on the Outcome Index. Each category within the independent variable (using family size as an example: 2 children, 3 children, 4 or more children) is compared to the reference category for that variable (in the case of family size: 1 child). An odds ratio with a value above 1 indicates an increase in the odds of falling below the negative cut-off (compared with the reference category of the independent variable) while an odds ratio below 1 suggests a decrease in the odds. If the 95 per cent confidence interval does not include a value of 1, this is considered statistically significant. It is important to note that these are adjusted odds ratios, meaning they indicate the unique predicting power of each independent variables after all others have been taken into account.

A number of caveats need to be borne in mind in interpreting these results: causal inferences cannot be made from the cross-sectional data available in Wave 1 of Growing Up in Australia; the measure of outcome used here is a global one, and assesses only the 'problem' or 'negative' end of the spectrum of outcomes; and the same source (a parent) provided most of the data on both the predictors (for example, parenting practices) and the outcome, resulting in the possibility of respondent bias impacting upon findings.

For the infant cohort, ten of 15 predictor variables made significant contributions. Broadly speaking, infants were more likely to fall below the negative Outcome Index cut-off if they were male; if their primary parent was born outside Australia, reported lower community connectedness, and reported less support from sources outside their immediate family; and if there were more children in the family. In addition, the three parenting measures of global self-efficacy, warmth and hostility each made strong unique contributions to the prediction of Outcome Index scores, with lower selfefficacy, lower warmth and higher hostility being associated with a greater likelihood of having a negative outcome (see Figure 7).

The analyses also revealed some counter-intuitive results that indicate that lower levels of parental education and living in a step-family are associated with the child being less likely to fall below the negative cut-off.

The primary parent's age, employment status, and psychological distress, the family's income and the community rating of 'disadvantage' (SEIFA Disadvantage quintiles) were not significantly related to the likelihood of negative outcomes for infants in the multivariable analysis.

Figure 7 - Significant predictors of infants falling below the negative cut-off on the Overall Outcome Index

For the child cohort, eight of 16 variables made significant contributions to the model (see Figure 8). Broadly speaking, children were more likely to fall below the negative Outcome Index cut-off if they were male; their primary parent was younger, had less education, and was psychologically distressed; the family received less income (although trends here were non-linear); and the primary parent reported more hostility, less consistency and lower self-efficacy in their parenting. Some trends were close to significant: there was a trend towards more negative outcomes for children whose primary parents were not in the workforce (in comparison to those in full-time employment), for children in single-parent families (in comparison to married intact families), for children whose primary parent reported not getting enough support, and for children receiving less parental warmth.

The primary parent's country of birth, employment status, community connectedness, the family's structure, household size and the SEIFA Disadvantage quintile for the neighbourhood, were not significantly related to the likelihood of negative outcomes for children in the multivariable analysis.

Figure 8 - Significant predictors of child aged 4-5 years falling below the negative cut-off on the Overall Outcome Index


In the infant cohort, all three parenting measures were strongly related to negative outcomes, with the associated odds ratios often being higher than for any other variable. The findings for the child cohort suggest that parental hostility is a particularly potent predictor of poor developmental outcomes for 4-5 year olds. Consistency in parenting practices, particularly around discipline, also emerged as an important predictor of outcome in the child cohort. Compared to the infant cohort, parental warmth and parenting self efficacy were less salient.

In terms of parent and family characteristics, for the older cohort the primary parent's educational status made a consistent contribution to prediction of child outcomes, indicating that this is an important aspect of the psychological capital contributed by the parent. This was not the case for the infant cohort. Older parents tended to be beneficial for the child cohort, while community connectedness had some role for infants. However, parental work status, country of birth, family type, family income, and neighbourhood disadvantage had little role in prediction for either cohort.

While other aspects of the child's family and community context played some part in predicting children's negative outcomes, it appears that parenting, as the most proximal influence on the child, played the stronger part. Especially in early childhood, many aspects of the family's functioning can be expected to be mediated through the way that they impact on the parents' ability to be warm, responsive, and consistent and to use nonpunitive disciplinary techniques.

An important feature of these parenting data is that there is no apparent "threshold" for distinguishing 'good parenting' from 'poor parenting'. In this study, the measures of parental warmth, hostility, and consistency were all positively skewed, and their predictive importance was evident despite this. That is, almost all primary and secondary carers scored in a positive direction on the parenting measures. Among those who were classified for the current analyses as being 'higher' in hostility or 'lower' in warmth, consistency or self-efficacy, only a small proportion reported parenting behaviours that would be regarded as being clearly worrisome or in the clinical range of 'abnormal' or 'abusive'. Thus, the current findings suggest that even somewhat subtle variations occurring within the 'normal' range of parenting behaviours are potent predictors of children's outcomes. This gives general support to universal initiatives that assist all parents in their parenting skills.

Contents  Next