How to rank public schools. (Part 1)

One of the eternal question that get asked all the time is how does the school rank and how do they compare with each other. The primary source of the school information come from the relative newly minted myschool.gov.au that came out during the Julia Gillard’s labor government period. This site is absolutely monumental for someone like me, for the first time in ever I had access to just a wealth of information.

However now the question is how to interpret all these data and make sense out of them. We first need to understand ICSEA commonly being mentioned, official definition as following, in simple term higher the number more Socio-Educational Advantage a school has from the prospective of the Department of Education.

In using the Index of Community Socio-Educational Advantage (ICSEA) calculation to describe school populations on the website, the median value of all schools’ ICSEA values is set at 1000. This represents the ‘middle ground’ of educational advantage levels among Australian school students. The median is used because the ICSEA values of all schools are unevenly spread, reflecting the great diversity of student populations across Australia.

Most of stats on the site will compare with schools with similar ICSEA and another with all schools. One brief look at a single school such as Summer Hills Public School will show that it is only compared with 57 other school and ten of which is actually public school from NSW. Another thing with Summer Hills Public School is that it hosts OC classes for the local area as well, so the number for the year 5 are artificially inflated by this factor as well. What this means is that the raw result of year 5 of a school that host the OC classes can give a false impression of what general performance level.

There are five factor that NAPLAN measures

  • Reading
  • Persuasive Writing
  • Spelling
  • Grammar and Punctuation
  • Numeracy

It appears what myschool.edu.au gives a straight up measurement of number on above five which is possible to measure the schools against each other. However given the number of the schools in Sydney alone, let alone NSW and entire Australia, this is a mammoth task indeed.

There is no really a good way of mashing up the measurement from different year and also account for things like OC classes etc to extract a true general level of performance for a school. I will try to do a sample comparison on two schools to illustrate the difficulties in comparing the results.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.