With the recognition in recent years amongst many university teachers that the ability to communicate in group situations is an important skill required by graduates, there has been increasing interest and use of the assessment of student participation in class discussion. Despite the increase in its use there has been very little study of this mode of assessment. There has, however, been much discussion in many departments on the value of its use and, in particular, concern expressed as to the problems of subjectivity and reliability and its adverse effect on class discussion. This paper explores some of the issues involved in assessing students' participation in class discussion, the reasons for its use, its limitations and ways it can be used more effectively.