New Zealand schools data

A very practical example of exploratory data analysis, descriptive statistics and visualisation with R by Luis A. Apiolaza (Quantum Forest).

It’s a series of posts on schools data in New Zealand, merging a sample of assessment data with existing data on schools socio-economic and demographic profiles.

The sample is self-selecting, schools consenting to early-release of their data ahead of the complete set, to be released in a contentious educational policy initiative.

Eric Crampton of Offsetting Behaviour adds some more exploratory regressions, using Stata.  He also notes the perspective from Emil’s Demense, a policy analyst, who cautions that since assessment is by overall teacher judgement, not standardized testing, it’s quite unsuitable for ranking.  It may be more useful in identifying best practice and enabling meaningful teacher appraisal in the context of national data.

Analysis of the data also relies on feedback, in interpretation of coding and the policy framework which defines the schools covered.

~ by mdev on September 28, 2012.

Leave a comment

 
OUseful.Info, the blog...

Trying to find useful things to do with emerging technologies in open education and data journalism. Snarky and sweary to anyone who emails to offer me content for the site.

Registrarism

Registrarial comment and observations