Annual Report 2005‚Äď06

Does it take a village?

(Taken from Family Matters, edition 72, "Does it take a village? An investigation of neighbourhood effects on Australian children's development" by Ben Edwards.)

There is a large body of international research that documents neighbourhood influences on children's developmental outcomes. However, there are few studies of neighbourhood effects on Australian children. This issue has particular relevance to policy given the federal and state governments' recent emphasis on community development policies that aim to foster positive childhood development.

This analysis examined whether a measure of neighbourhood socio-economic advantage and disadvantage (SEIFA Index of Advantage/Disadvantage) is associated with 4-5 year old Australian children's physical, social/emotional and learning outcomes (as measured by the Outcome Index, which was developed specifically for Growing Up in Australia), even when controlling for several child and family socio-demographic factors. Using these factors as controls limited the likelihood that neighbourhood influences reported in this study were the result of selection bias, as socio-demographic factors such as family income may be associated with parents' decisions to live in a particular neighbourhood.

Do neighbourhoods affect Australian children's developmental outcomes?

Analysis shows that the SEIFA Index of Advantage/Disadvantage had a statistically significant association with the Overall Outcome Index, and the Physical, Social/Emotional and Learning domains when it was the only variable included in the model. When the variables that were used to control for selection bias were also included (weekly family income, child age and gender, child is of Aboriginal and Torres Strait Islander origin, single parent family, at least one parent employed, mother's highest level of education), the Index of Advantage/Disadvantage was still significantly associated with children's Overall Outcome Index, and the Social/Emotional and Learning domains but not the Physical domain.

Figures 15 to 16 illustrate the mean scores of children on the Overall Outcome Index and Learning domains for children living in the five levels of neighbourhood advantage/ disadvantage after statistically adjusting for the control variables. Note that graphs are not centred around 100 because variables that were used to control for selection bias in the statistical analyses explain a portion of the variability in the Outcome Index.

Figure 15 - Neighbourhood effect on Continuous Outcome Index

Figure 15 suggests that children's scores on the Overall Outcome Index were higher in more advantaged neighbourhoods. Children in the most disadvantaged quintile had significantly lower scores on the Overall Outcome Index than children living in all other neighbourhoods except the second most disadvantaged neighbourhood quintile. Conversely, children in the most advantaged quintile had significantly higher scores than children living in all other neighbourhoods except those living in the second most advantaged quintile. Other significant differences are also evident from inspection of the confidence intervals.

The overall pattern and differences between neighbourhood quintiles for the Social/Emotional domain was the same as for the Overall Outcome Index.

Figure 16 - Neighbourhood effect on the Learning domain

The pattern of scores for the Learning domain was different across the neighbourhood quintiles (Figure 16) than for the Overall Outcome Index and the Social/Emotional domain. Children living in the middle and the two most disadvantaged neighbourhood quintiles had similar scores on the Learning domain while children from the two most advantaged neighbourhoods had significantly higher Learning domain scores than the other three neighbourhood types. Children living in the most advantaged neighbourhood quintile also had significantly higher scores on the Learning domain than children living in the second most advantaged neighbourhood quintile.

Thus, the preliminary evidence from this study suggests that neighbourhoods do matter to children's development and supports the community emphasis of many federal and state government policies. However, it should be noted that there is a very limited evidence base on neighbourhood effects on Australian children.

Further data about how neighbourhoods affect Australian children is needed and Growing Up in Australia will become an invaluable source of such information, especially as it will allow us to document effects over time. Further support for research within this area will enable policies that target community development to be tailored for the maximum benefit of Australian children and their families.

Contents  Next