Saturday, February 23, 2013

The value of formative assessment

Students on a first year key skills module have the opportunity to take a non-assessed practice quiz which is similar in format and content to the summative assessment on the module. For the students who submitted the assessment on time, I entered the data into R and had a quick look at it:

The marking scheme constrains the marks it is possible for students to score (O, 20, 40, 60, 80, 100%), thus this is not continuous data and a non-parametric test has to be used for comparison (Wilcoxon rank sum test = Mann Whitney U test). In this sample (n=262), 30% of students chose not to do the practice assessment. In general, this group scored lower marks (median = 60%) on the summative quiz than those who had engaged with formative assessment (median = 80%). As you can see, this difference is statistically significant (p < 0.001).

This analysis does not take into account confounding factors, such as the fact that students who bother with formative assessment are more engaged, and quite possibly more talented, than those who do not, so it does not prove the simple hypothesis that it was the practice quiz wot dun it.

No comments:

Post a Comment