Cracking the code of biology tests: let’s get metacognitive!

This is a guest post from Greg Crowther

In a previous post here on Scientist Sees Squirrel, Steve raised an important and hard question: aside from helping students learn the specific content of our courses, how can we help students get better at learning in general?

Although it’s a hard question, I think I have a pretty good one-word answer: metacognition.

Yes, it’s faddish educational jargon. But (like “Bloom’s taxonomy” and “growth mindset” and “backwards design”) it’s good educational jargon, with lots of logic and research behind it.

Metacognition is often defined as “thinking about how you think” (Tanner 2013). Some researchers have distinguished between two subtypes of metacognition: self-awareness and self-regulation (Schraw & Dennison 1994, Dang et al. 2018). Self-awareness would include, for example, a student’s recognition that she does not yet understand the genetic code, while self-regulation might mean seeking out a friend who can explain it well.  In general, acute self-awareness combined with skillful self-regulation helps people learn material that does not come easily to them.

Encouraging metacognition in one’s students is now fairly well established as an educational “best practice” (Hacker et al. 2009). I’m sure Steve, like most thoughtful instructors, does this in one form or another* (though he has yet to use the word “metacognition” in 5.5 years of blogging, thus leaving a niche for me to fill…). Among many options, one personal favorite is the “muddiest point,” in which students identify the most confusing aspect of each lecture (Mosteller 1988).

The muddiest-point exercise is quick, but it demands self-awareness (what am I struggling with the most?) and may lead to self-regulation (how can I fix this problem?). And while each iteration zeroes in on one particular bit of mud, broader patterns may eventually become evident. A biology student may realize, for example, that many of his muddiest points are math-related, and that reviewing the video lectures usually doesn’t help but that doing additional textbook problems usually does .  Thus, the student may reach broader lessons about how to learn more effectively — the goal noted in Steve’s previous post.

Of course, metacognitive exercises only work if students take them seriously. To achieve maximum compliance, many of us (A) bribe students with points and/or (B) sell metacognition as a way of helping them do better on test questions.

I’m OK with (A), but I like (B) better.  And that brings us to  a new paper, Testing in the Age of Active Learning: Test Question Templates Help to Align Activities and Assessments, which, among other things, could be seen as a guide to pulling (B) off.

Test Question Templates (TQTs) are a possible bridge between regular classwork and exam questions. In short, TQTs carefully define the formats of possible test questions without committing to any particular details.  Students can’t just memorize their way to a high grade, but they can use TQTs to think carefully and metacognitively about questions similar to actual test questions.

To see how this works in practice, let’s look at the sample TQT for an introductory biology course in Table 1 of the paper:

Template: “Given a table of the genetic code and a point mutation (reported in terms of the DNA coding strand OR the DNA template strand OR the mRNA), determine the likelihood that the mutation will affect the function of the corresponding protein. 

Example: In an exon of the coding region of the coding strand for a particular gene, most people have the codon 5’-AGG-3’, but Jesse has codon 5’-CGG-3’. Is the corresponding protein likely to function differently in Jesse than in most other people?”

The “Jesse” example isn’t special, but the framing TQT arguably is. The TQT is the instructor’s attempt to show students that this example is not simply a particular question with a particular answer, but a part of a broader pattern that can generate many questions. TQTs thus encourage broader metacognitive thinking about the content and the questions that can be asked about it.

As students try other versions of a problem (e.g., what if Jesse’s sequence was GCC?), metacognitive comparisons of different versions should reinforce core concepts. For the TQT above, changing the DNA sequence sometimes changes the amino acid sequence, and sometimes it doesn’t; some amino acid substitutions are more conservative than others; the amino acid sequence can be predicted from the DNA sequence, but not vice versa; and so on.

Content aside, TQT-fueled metacognition should help students grasp often-unwritten rules of test-writing and test-taking. In the TQT above, students are explicitly granted access to a table of the genetic code (i.e., which mRNA codons go with which amino acids).  Metacognitively inclined students will notice that they’re expected to know which nucleotide bases go with which, and how to use the genetic code, but that they’re not expected to memorize the code. The rules of the game have become clearer.

If we use TQTs or similar frameworks to promote metacognition, we’ll be aiding all students, but especially those who are still figuring out how biology classes work. Students who are afraid to ask questions. Students who are good at memorizing facts but who must now get better at applying those facts.  Students who, for any number of reasons, will not quite know what we expect of them unless we show them, using realistic examples, in advance of the actual test.

By this point, you may have noticed that the TQT paper I’m yammering on about is my paper (with wonderful coauthors Ben Wiggins and Kiki Jenkins).  It’s an “ideas paper,” with no real data on student learning, faculty attitudes, or anything else. And so, given that TQTs have yet to be deployed much beyond my classroom, I would love to hear your feedback. Can you imagine using TQTs in your own biology survey courses? Would they work better in some of your courses than others? Or do TQTs seem too rigid, too test-focused, too much, too little, too late? Your comments, please!

© Greg Crowther  June 1, 2020

You may remember Greg from a previous guest post as the singing Biology instructor – if not, head on over to “Becoming a Science Writer: A Musical in Three Acts”. You’ll enjoy it.

Image: RNA code, adapted from blogs.plos.org/dnascience/2015/01/22/ by Subin Hyun


*^Steve here – this comment made me think about the way I teach thinking about thinking – metametacognition, if you will.  And Greg is right, I do regularly work metacognition into my courses.  It’s often in very simple ways like adding “why do you think so” to a question, or in asking one student to explain their answer to another.  But I haven’t been very deliberate about it, and this post is prodding me to be more so.

1 thought on “Cracking the code of biology tests: let’s get metacognitive!

Comment on this post:

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.