Assessment of Minorities' Ability Using Low-Stakes Tests - Evidence from PISA and a Field Experiment

Internally funded project


Start date : 01.09.2017


Project details

Scientific Abstract

The minority-majority educational performance gap is typically measured using low-stakes standardized assessment tests. However, these measures will be distorted if there is a difference in intrinsic motivation between students from different groups. I conduct a field experiment and use data from PISA 2015 to evaluate the differences in minority-majority motivation and performance gaps when using high and low-stakes tests. I find that performance gaps substantially differ depending on the stakes of the test. Using high-stakes tests reduces the gap by 0.3 standard deviations. The experimental results suggest that 60\% of the Arab-Jewish math performance gap measured on 8th grade Israeli National Assessment tests, could be due to differences in motivation rather than cognitive skills.

Involved:

Contributing FAU Organisations: