Religion studies fail to end bias

This article presents opportunities for older students to consider a designed before-after survey with the treatment in between being a religious studies course at the senior secondary level. Although there is no control group in the study it is possible to ask questions about whether the changes in percentages reported appear to be significant on the surface and whether they are indeed statistically significant.

If we assume that n=2000 for the survey before and after the course then for the question of having Asians as neighbours, p1-hat = .153, p2-hat = .210, p-hat = .182, sp(D) = .012 and z= [(p1-hat)-(p2-hat)]/sp(D) = -4.75
which indicates a statistically significant increase in aversion to Asians as neighbours. Other tests are carried out in a similar fashion.

Where to next?

Student Questions for this article
Newspaper article
Index - Related articles
Index - Data Representation
Index - Inference
Main Index - Numeracy in the News