Percentile/SFO Comparison Charts

In 1996, ‘Like’ school groups were introduced to provide fairer comparisons between schools in relation to student achievement outcomes. ‘Like’ school groups were based on two factors: the proportion of students in receipt of EMA/Youth Allowance and the proportion of students from a Language Background other than English. These two factors were used because they were known to correlate highly with student achievement outcomes. So schools with similar cohorts of students could be compared as ‘like’ schools.

Why introduce the charts?

While the Like School Group methodology was widely accepted, there were a number of inherent weaknesses in the model. School performance was compared against a fixed group of ‘Like’ schools. This did not always offer fair comparisons, particularly for those schools located near the boundaries of each ‘Like’ group. Furthermore, ‘Like’ school groups consist of 9 school groupings of unequal size. For example, LSG 4 consists of almost ten times the number of schools as LSG 3.

The Percentile/SFO Comparison Charts methodology has replaced the ‘Like’ school Group model. The new methodology was the subject of extensive consultation. In term 4 2006, regions presented the methodology to schools, asking for feedback. In term 3 2007, schools were provided with new draft reports that used this methodology, again requesting feedback. As a result of the consultation, some enhancements to the methodology were made, but overall the feedback was very positive. Hence the methodology is replacing the “Like” school group model in the 2007 School Level Reports.

Importantly, this method of presenting the performance data of your school is in addition to the absolute scores which show actual change over time.

What are the benefits?

Schools can compare their performance against all schools, while at the same time comparing their performance to what would have been predicted given the school’s Student Family Occupation density.
The proposed method makes comparisons between percentiles, rather than comparing average scores with those of a fixed group of schools.
The approach is particularly advantageous for those schools that were located near the boundaries of each ‘Like’ group.
It provides an additional way of seeing the ‘big picture’ of your school’s performance on a range of student achievement measures.