Download files
Abstract
I conduct a survey of the newly available academic performance and social context data relating to Australian schools. I set this in the context of available school data sets internationally, highlighting the education economics literature that has been made possible by those data sets. I set out the new research opportunities that have become available to researchers using the Australian data, before giving a practical example of the use to which the new data can be put, by constructing and estimating an empirical panel data model using a six-year panel of the Australian data.
The model I construct estimates school competition effects through an index measure of competition developed for each school. The index is derived from schools’ own standardised test scores and those of nearby schools. Since the index is a proxy signal for relative local market quality, I test the hypothesis that an increase in index values will be associated with an increase in test scores in the following year, as schools respond to competitive pressure.
In parallel, and to dismiss the possibility that positive and significant findings may be due to regression to the mean, I develop a synthetic index for each school. This is created by substituting for each school’s actual index score one derived from the scores of nearby schools for a school randomly drawn from a subset of schools with similar baseline scores. This gener¬¬ates a robustness check, whereby the validity of parameter estimates for the actual index would be supported by estimates for the synthetic index that are not significantly different from zero.
I find significant effects for the competition index, in particular for primary schools and schools in metropolitan areas. Findings are validated by the absence of significance for synthetic index parameter estimates.