Assessing Critical Thinking Using Selected Response Items
Sorry for missing my self-imposed deadline yesterday! I was moving my youngest daughter down to Kendall College of Art and Design and it took much more time and effort than I expected. It seems like just last year that I was the freshman on the curb watching my parents’ reflection in the car mirror recede as they dropped me off at school. It seems very strange to be on the other side of the mirror now.
Well now to the task at hand. One of the most pointed criticisms of multiple-choice (selected-response) assessments is that they cannot be used to measure higher levels of cognitive ability. I will readily concede the point that most multiple-choice exams do not assess much beyond the level of understanding. The vast majority of the questions that ship out in the test banks from college textbooks do not require much more than simple memorization (remembering level stuff) with a few categorization/classification (understanding items) tossed in. This need NOT be the case. There is a fairly compelling argument that mid-level cognitive abilities (applying and analyzing) can be handled very well in a MCQ format. A relatively recent article addresses this specifically for the field of Biology (my personal interest) and I am providing a link here. I believe that it essential to include a substantial number of mid-level questions in our assessments regardless of whether the class is 100-, 200-, 300-, or 400-level. You can refer to my earlier post on leveling for a little more info on my assessment philosophy.
Some people think that MCQ assessments can also be used for upper level cognitive abilities (evaluating). I’m not so convinced about this. I know that problem-based learning can be evaluated with selected-response items. However, selecting the most appropriate choice isn’t really the same thing as independently evaluating a problem and its possible solutions. I occasionally try a MCQ for evaluation, but I’ve never been too happy with using those data. Creating just flatly cannot be assessed using these tools. Selected response exams are not a panacea for assessment.
I also use my selected response assessments to get at metacognition in my courses (students thinking about their thinking…). I have created a handout for my students that explains the different types of thinking required in my course and how I plan to measure it. Throwing problem-based learning at students without preparing them for it is unfair and leads to poor results (and upset students). I am attaching a copy of my document here for your comments. Next time, I want to discuss how selected response items can be implemented for course assessment. See your then.