Formative assessment during the course of a module may help learners understand the subject better and perform better in the summative assessment. However, the exact role that formative assessment plays in enhancing students' learning is not clear. Students may choose to focus on the preparation to the only formal assessment, the final exam, rather than to participate in the informal formative assessment during the module. In this project, the relationship between students' participation in formative assessment during the module and their performance in the formal examination at the end of the module is investigated.

During the module Introduction to Industrial Economics, students were given the possibility to participate in several types of formative assessment. The types of assessment included automated tests developed in the VLE Blackboard, a role-playing exercise (also developed in Blackboard) in which students were playing the role of firm managers and were making decisions that led to (hypothetical) profit of their firms, and evaluation of tutorial assignments. Students' participation is measured by the number of tests completed and the score in them, by the profit earned in the role-playing exercise, and by the number of tutorial assignments handed in. The conjecture is that a more active and successful participation in formative assessment would mean a better performance in the final examination as measured by the exam mark.

It is found that there exists a positive correlation between the participation in all three types of formative assessment and the performance in the exam. Further analysis indicates that the performance in the automated tests in Blackboard has the largest effect. The performance in tutorials and in the role-playing exercise has a positive but not significant effect. Regarding the role-playing exercise, it is the participation in it rather than the profit earned that has significant effect on the exam mark. To separate possible effect of students' ability on performance, further data analysis is done using students' marks from a different module. The effect of the performance in the tests remains significant.

One interpretation of the findings is that formative assessment helps learners to prepare for the final exam. Separating the effect of students' ability indicates that there may be a causal relationship between performance in formative and summative assessment. Some formative assessment methods are more effective than others. The automated tests turn out to be most effective in the present study. Although exam questions included both conceptual and problem-based parts, the larger effect of tests may be because of the emphasis in exam questions on concepts learned during the module rather than on the practical application of the concepts that was useful in the role-playing exercise. Formative assessment seems to help those students who choose to participate in it, and thus could be used more in higher education. The study at present is limited to only one module but it is hoped that its conclusions can extend to other modules.