Mapping Toronto school testing results

This project was originally intended to be my third major OpenFile mapping project (after the Second World War casualty maps and the Ontario baby name maps.) For various reasons that didn’t work out, and so I present it here.

It mashes up two data sets which are otherwise hard to come by: Toronto public school catchment areas and EQAO school testing results. I look at five-year averages of Grade 3 test results for reading, writing and arithmetic.

Boundaries
In early August, I filed an FOI request with the TDSB asking for a geographic file showing Toronto elementary school catchment areas. In September, the request was denied on the basis that the information was already public, which was true in a limited sense, in that catchment areas for individual schools (there are several hundred) could be viewed but not downloaded on their site. One helpful friend pointed out where the 400-odd boundary files were on the TDSB server – finding them was an ordeal in its own right – and another helpful friend wrote a Perl script to scrape them. I was then left with the job of manually associating them with the schools (since nothing in a .kml file name indicated what school it related to) and assembling them in a single map file. These maps are the only one-stop-shop map of Toronto catchment areas that I am aware of.

Data

The EQAO did not return calls asking for their data, and the research agreement on their site was designed for academics (data to be seen by a limited number of people, returned to them after a year, not hosted on a U.S. server, etc.). In the end, in late November, I filed an FOI request for the EQAO data for Toronto public schools with the Ministry of Education, and received it in early January.

The maps

EQAO scores for individual schools can vary from year to year (in some cases – some schools are fairly consistent). I felt it was more informative to look at the average scores over as long a time span as possible, so mapped the percentage of Grade 3 pupils with Level 3 or 4 marks on the test (ie. met or exceeded the provincial standard) over the five academic years from 2005-10 for reading, writing and math. All neighbourhood-based elementary schools with geographic catchment areas are depicted. Alternative or French-immersion schools are not. Schools with a small number of Grade 3 students don’t appear in the original data, which accounts for most of the grey on the map, though some grey areas have some data. Non-residential areas that don’t fall into any catchment area are in a very light grey. The Leslie St. Spit is excluded because its polygon was large and complex, and including it adds no information to the map, while adding noticeably to its size.

Readers will have their own impression of the value of the EQAO itself as a way of evaluating either students or schools. My own sense is that the broad patterns that the test results show are revealing.

Click to see:
Reading test results
Writing test results
Arithmetic test results


2 thoughts on “Mapping Toronto school testing results

  1. It would be interesting, someday, to correlate these results with socio-economic data to see which schools are over-performing (i.e. getting good results in more difficult catchment areas).

  2. Agreed, Dylan! I was thinking this could be done with average incomes from the CT profiles on statcan, but there’s no other way to map it (eg. by household) more specifically…