You are here

Post date: Tuesday, 12 March 2013

STAR and the science of assessment development

Let’s be clear –the revised STAR test does not deliver inflated results. Norms for the revised STAR test were generated on the basis of a carefully drawn national sample involving thousands of students. These norms have since been validated by subsequent collection of data from tens of thousands of students who have used the test in the classroom. ( If you want to know more about our test development process, check out pages 8-9  of our annual report.)

Warwick Elley was the author of the original STAR test and a key member of the team who revised the test. Here's what he has to say:

“It is not true that the new tests are easier than the old editions. Comparisons based on hundreds of students’ scores show that the results are very close to expectation. One test is a little easier, and two are slightly harder, while all the others are virtually the same in difficulty as before. Correctly administered and interpreted, most teachers will find  little difference in the performance levels of their students on the new versions.”

So why have you read concerns about STAR results being inflated? It's to do with how schools are reporting and analysing the results at different times of the year.  At NZCER we need to do a much better job at supporting schools to understand the use of scale scores and the limitations of relying only on stanines. This new blog is one way we’re trying to do that. In future posting we will explore how to use stanines with STAR (and why we think you should use scale scores too), explain why we only provide norms at the beginning of the year for STAR and PATs and we’ll tell you a bit about the new PAT on the horizon.

Add new comment

Community guidelines

Please refer to the NZCER blogs guidelines for participation on NZCER blog posts.