Reliability data suggest that, although there are many problems, it is feasible to systematically evaluate a student's analytic and persuasive competence in free oral argumentation. The first part of the paper describes the contexts within which the evaluation project took place. Specifically discussed are the five areas of analysis on which the project concentrated: (1) problem identification and differentiation, (2) making explicit cross problem assumptions, (3) identifying and using appropriate strategies for dealing with different types of problems, (4) identifying common dialectical operations, and (5) identifying relevance problems. The second part of the paper describes the content analysis system used to quantify student behavior. The categories or units used to describe the interactions are examined (many examples are provided) and the importance of the frame of reference of the person who does the categorization is emphasized. Two evaluation studies were conducted. In the first, four trained scorers scored from 10 to 18 discussions between a student and adult interviewer, in which a student was challenged to defend a position on a controversial case. In the second study, two trained scorers scored 32 pupil-led discussions. On the average, there was a relatively high level of agreement among scorers in both studies. (RM)