Talk: Using IBL as an assessment strategy

January 15, 2013 — 5 Comments

Several weeks ago I was asked to take part in the Project NExT Alternative Assessment Techniques panel discussion at the 2013 Joint Mathematics Meetings, which recently took place in San Diego, CA. I was extremely honored to be considered for the panel, but at the time I was not planning on attending the JMM, so I declined the invitation. A couple weeks later, it turned out that I was going to make it to the JMM after all. At about 11PM the night before I was going to fly to San Diego, I received an email from the organizers of the panel discussion indicating that one of the panelists was unable to make it and that they heard was going to be there. They asked if I could fill in at the last minute and I accepted.

Here is the abstract for the panel.

Since classroom assessment is used to determine a student’s level of mastery, how can we vary our methods of assessment to accurately reflect the diversity of ways that students learn and understand the material? Traditional methods of assessment, such as exams, quizzes, and homework, may not accurately and robustly measure some students’ understanding. In this panel, we will propose alternative methods and discuss the following questions:
– What assessments exist besides the traditional ones and how can I use them for my course?
– How can I determine the validity of an alternative assessment?
– How can I develop my own alternative assessments?
– How can alternative assessments help me evaluate the effectiveness of a non-traditional classroom?

It is worth pointing out that I’m not an assessment expert by any stretch of the imagination. Also, given that I had less than 48 hours to prepare amidst a pretty full schedule, I didn’t have a lot of time to come up with something new and creative for my talk. Inquiry-based learning (IBL) is one of my passions and I’ve given quite a few IBL-related talks in the past few months, so I decided to “twist” the ideas from some of my recent talks into a talk about assessment. In my talk, I propose implementing IBL not only as a pedagogical approach but also as an assessment strategy. This isn’t really a stretch since in my view, an effective IBL class is all assessment, all the time.

My fellow panelists included Theron Hitchman (University of Northern Iowa), Bonnie Gold (Monmouth University), and Victor Odafe (Bowling Green State University). Theron gave a talk on using Standards Based Assessment (you can find his slides here), Bonnie spoke on a variety of summative assessment techniques, and Victor shared his experience with oral assessment. It turns out that the person that I was filling for is mathematics education superstar Jo Boaler. Me filling in for her is ridiculous.

Here are the slides for my portion of the panel.

Thanks to the organizers of the panel (Cassie Williams (James Madison University), Jane Butterfield (University of Minnesota), John Peter (Utica College), and Robert Campbell (College of Saint Benedict and Saint John’s University)) for providing me with the opportunity to speak on the panel.

Dana Ernst

Posts Twitter

Father of two boys, husband, mathematician, cyclist, trail runner, rock climber, and coffee drinker. Columnist for MAA blog Math Ed Matters.
  • Just an anecdote regarding IBL and assessment…

    In one IBL course I taught, we had a traditional midterm and final. The idea was that students would feel more in control of their grades if they had traditional exams where they could do all the traditional “learning” (= cramming, memorizing, etc.) they are used to for lecture courses. I had mixed feelings about it then and after a few years after the fact, I think this was a bad idea. First, I didn’t take the exams very seriously. I already had so much information about each student, the exams just gave me a weakly correlated random number to add to the count. Second, the students took exams very seriously and had the illusion that their exam grades had a very deep impact. Overall, students had less focus on what really mattered and the randomness had, as expected, random effects which were sometimes unpleasant.

  • I am teaching a general-education mathematics course this semester. I have taught similar courses before in a relatively traditional way. This semester I am trying something different: no exams. At all. I have several reasons for doing this (and for not doing it in my calculus class). In any case my gen-ed class will still have shorter quizzes every two weeks, along with homework, computer labs, and two longer prose assignments. I’m interested to see how it turns out, but one thing I am very confident about is that the students will have ample opportunities to show what they can do.

    At the same time, we should not take the “document” model too seriously. To quote Stanley Fish on the opposite position: “… higher education, properly understood, is distinguished by the absence of a direct and designed relationship between its activities and measurable effects in the world.” Every time I have seen a poll done in an IBL setting about what instructors would like their students to know five years after graduation, the dominant opinion was that the students should have a certain way of thinking, not that they should remember any particular material from the course. I think we can find a compromise that allows us to justify grades without losing sight of the larger picture.

  • François and Carl, thanks for the comments!

  • I was just reading this post on the Harvard cheating scandal. The post deals with collaboration on (take-home) exams and how this can make individual assessment difficult. I immediately thought about this post, though the relation is mostly tangential.

    One proposal is to combine collaborative exams with individual oral exam follow-ups, which is a reasonable idea in the traditional setting. IBL offers an even better solution since it opens up the collaboration process and integrates it into the evaluation. I’ve seen some students drag along in IBL courses but it usually only takes a nudge or two to reactivate them. Those who resist more quickly see how their inaction harms their performance as well as that of others and they eventually reactivate too. I have seen very few instances where that didn’t work (and none led to cheating).

    One difficulty that I experienced with IBL is that we often have to deal with students tendency to get by with as little work as possible through “gaming the system.” Once a few fall into this mode, it quickly becomes an epidemic that is very hard to contain. Though this kind of situation is very uncomfortable, I much prefer this form of open defiance to catching students cheat.

  • François, your comments are timely as I will be giving out my first take-home exam in my intro to proof course in the next couple weeks. My typical approach is to have an in-class portion that students complete individually followed by a take-home exam, where students are allowed to work together (with some guidelines), but outside resources are strictly forbidden. Perhaps surprisingly, I see far fewer “clone” exams since I’ve started allowing students to collaborate. Also, I provide students with choices about which problems they complete for the exam, and often students that work together don’t even do the same problems. Part of this isn’t my choice, however. For the in-class portion of the exam, most of the problems are exercises and a few require the students to write proofs for theorems they have not seen before. Again, I provide them with choices as to which problems they complete in class, but the remaining ones that a student did not choose are among the problems on the take-home exam. The upshot is that two students that work together may not have the same take-home exam. Your comment has me considering adding a twist to my usual approach. I may add an oral component to the exam, or perhaps leave it to my discretion as to whether it is necessary to have an oral follow-up to the exam.