The largest class I’ve ever taken was Introduction to Anthropology. Over 1500 students were seated in the convocation hall, facing a large-screen projector and a professor on the podium. The course rotated through four areas of anthropology (biological, archaeological, linguistic and social), and lectures were provided by four different professors according to their area of research.
What were the assessment methods?
The two exams – midterm and final – consisted of multiple choice questions, as many would have guessed. This seemed like the most efficient method of assessment for a large number of students.
The other deliverables consisted of four essays, one for each area of anthropological study. These were graded by teaching assistants, who were responsible for about 150 papers each.
Were these assessment methods appropriate?
Multiple choice questions get bad rep, and I think for the right reasons. They certainly served as efficient tools of assessment, since the instructor (or more likely, TAs) only had to insert answer cards into a machine; but they were not necessarily appropriate because they could only determine whether the student was proficient in memorizing facts. Ironic, since the field of anthropology is riddled with uncertainties, in terms of facts and perspectives.
The essays, on the other hand, were appropriate for learning the subject as they required greater critical thinking, but they were not necessarily appropriately evaluated. There was already inconsistency and unfairness by having multiple evaluators. Throw in the difficult task of assigning a single number – a percent grade – for 150 students, you can imagine the variability per evaluator as a result of fatigue, mood…
To be fair, attempts were made at demystifying the grading process. I remember there being a rubric that outlined the different facets of assessment, and the grades were assigned in increments of 5% (to avoid ambiguity between 85% and 86%, for example). The result still seemed to leave a number of students unsatisfied, possibly for the reason that the rubric was simply another scale in disguise, a complement and justification for the grade.
What would be an appropriate method of assessment?
Not sure. How about start by not having large classes? Idealistic, I know, especially when massive open online courses are on the rise. A large class is convenient for mass information transfer, but is also an obstacle on its own in terms of providing students with meaningful feedback. When instructors and TAs are saddled with the burden of grading so many assignments, they, too, are likely to succumb to the “get it done” mentality familiar to many students when completing assignments. Let’s not even mention attempting to track the development of individual students.
Thoughts on evaluation for large classes? MOOCs?