Redesigning Microeconomics – Reflections

~600 words, ~3 min reading time

This semester I redesigned my Principles of Microeconomics course. I want to give a brief run-down of what happened and thoughts for going forward.

(1) Engagement/Mastery division – I still like this idea, though since I’m switching to specs grading (more detail in a later blog post), the “weighting” of each will go away. But, I think it is helpful to be clear with yourself what the point of an assessment is. Is it simply to get students to engage with the material, or is it to test students’ mastery of the material? It is helpful to separate the two.

(2) Short Paper process – This semester, I required rough draft, peer reviews, and final copy (including a response to peer reviews). I’m scrapping the process in the future. Or, rather, I’m making it optional. The justification for the process was twofold: (1) students write better for each other than for professors, and (2) students understand comments from other students better than from professors. Students that I talked to were mixed about the value of the peer review process’s value for them. And, from the reviews I read, a lot of them were pretty useless. As such, they were mostly busywork rather than meaningful.

(3) Engagement activity options – Students were a little confused by the fact that they had choices, but this went okay. One tweak: I’m going to scatter due dates through the semester rather than have everything due at the end, just for the sake of grading sanity.

(4) Grade proposal – Scrapping this. Most of them were written by students who had earned an A anyway. A few were by students who did worse – but they were rarely very convincing. Also, they didn’t prevent grade-grubbing. On the whole, I think was pretty useless.

(5) Comprehensive Final & Midterm Diagnostics (not for a grade) – I liked this. Building in a buffer worked, and scores turned out okay – no worse than when I used non-comprehensive tests.

(6) Class attendance – not going to be part of the grade going forward. Including it in the grade led to some students coming who were WAY disengaged in their time in the classroom, and others grubbing their attendance. Strong correlation between attendance and other grade elements suggest this is not necessary.

(7) Class preparation questions – went pretty well, though I need to refine them. Multiple choice need to be sure to reflect the final exam to some degree, and short answer should be more closely linked to Bloom’s taxonomy levels. I also asked students for “curiosity questions” that could be used to inform the class if time permitted. Scrapping these. Students often used these not for curiosity/discussion questions, but for “here’s a topic I don’t understand” questions.

(8) Flipping the classroom – worked pretty well. Class time was more focused on covering what students didn’t understand – so less wasted time. Classroom response system (“Plickers”) worked well, though I need to assign them at the beginning of the semester rather than have students pick them up and turn them in each time. I still need to work some on becoming a less-awful discussion facilitator.

Going Forward

Apart from changes listed above, I think that switching to Specs grading will be good. I’ve also debated switching to a more problem-based learning format, but don’t currently feel confident in making that switch for Principles of Micro. However, it might make sense to design my Environmental Economics course around this format…

Hacking Backward Design

~1000 Words, ~5 min reading time

The “state of the art” in course design is called “backward design”. This design philosophy isn’t new, by any means. But, let’s walk through what it is for those who might be uninitiated. First, though, let’s consider the opposite of backward design. I’ve seen a few names for this – I don’t like any of them. But, here’s how the course design runs…

Step 1: Pick the topics for the class. (Perhaps based on textbook.)

Step 2: Present the topics to the class using various methods (lecture, demonstrations, practice problems, etc.)

Step 3: Come up with a test or other assessments (papers, problem sets, etc.) based on what you did in class and assignments outside of class.

This is often the way that college professors will design a course, ESPECIALLY if we’re teaching it for the first time. The old joke runs that you just have to stay one chapter ahead of the students. And, often, that is basically the way that we approach course design – flying by the seat of our pants, so to speak.

However, there’s a serious weakness to this approach: the question of WHY you’re teaching what you’re teaching is often secondary. “It’s in the book” is hardly a good answer, seeing as so many things are in the book that we don’t teach. Now, that doesn’t mean that we don’t think about these things – but rather that the “why” question tends to be asked piecemeal, with little attention paid to the overall story of the course. A shame, since we know that one of the best ways that we learn is by making connections between what we know and what we’re trying to learn. A well-integrated course design that tells a clear “story”, then, can be profoundly beneficial.

So, what does “backward design” look like?

Step 1: Determine learning outcomes for the course and the various units within the course.

Step 2: Determine assessment instruments based on those learning outcomes.

Step 3: Determine teaching methods – lecture, demonstrations, etc.

Unlike the previous method, backward design is filled with intentionality. The “Why” question should be obvious to the professor, at the very least, even if it isn’t obvious to the students.

The point of this post, though, was not to convince you to use backward design. It was to help you “hack” it. Why hack it? Because most college professors have almost no pedagogical training. As a result, Step #1 above is extremely difficult. We may have learning outcomes handed to us by our departments (the Higher Learning Commission certainly wants us to!), but we generally don’t know how to write them ourselves. The fact that most learning outcomes that are given by departments are at the course level doesn’t help – as they tend to be so broad as to be basically useless. So, let me allow a small hint.

If you don’t know how to write learning objectives – or are even just a bit hesitant about doing that – skip step 1 above. Instead do this.

Step 0: Write a traditional test (essay, short answer, problems, multiple choice, whatever you’re comfortable with) based on what you anticipate teaching.

Step 1: Determine what learning outcomes that test would assess. A well-written learning outcome is basically a test question, but stated in terms of what the student has to do to answer it. (Notably, very few course level learning outcomes you find in syllabi are written this way.) Eliminate those that you realize aren’t actually important. Add new ones as you’re inspired.

Step 2: Determine how to assess the learning outcomes. (Note: it might be that the test you write is the best way to do it. But, it doesn’t have to be so.)

Step 3: Determine learning activities.

My own experience is that I’m not very good at writing learning outcomes in a vacuum. However, I love writing test questions. What this method does is let me start with something I can do, and then forces me to back up and ask the “why?” question.

Consider this example. Test question: “Suppose that the economy is booming, and incomes are rising. What happens to the demand for filet mignon, given that filet is a normal good?” This is a pretty standard Principles of Microeconomics multiple choice question. What learning outcome does it imply? Something like “Students will be able to identify the effects of changes in income on demand for normal goods.” Notably, “identify” is the verb I used. “Predict” would be appropriate as well. “explain” would NOT be. Why? Because the question doesn’t ask the student to explain the answer. So, as a professor, I can start asking myself – “Do I want students to identify or explain or both?” I can also use this outcome to imply other outcomes. Do I want students to be able to identify the effects on demand for inferior goods as well? What about other demand shifters?

Through this process, you can develop a refined list of learning outcomes based on the test you’ve written. Then, start working backward. If I decided, for example, that I want for students to “be able to explain the effects of changes in income on the demand for normal and inferior goods”, then I know that the original test question alone will NOT be sufficient. I may need to break it apart into a couple. If “explain” is the verb I want I’ll definitely need to change the responses, at the very least – I might even need to change from a multiple choice format to something else.

It’s a simple hack, really – but one that can be very time-consuming if done properly and completely. But, that’s what backward design IS. It’s going to be somewhat time-consuming precisely because it’s not slapdash. However, at the same time, it ends up saving you a lot of wasted time preparing to present material that doesn’t really matter, and lets you spend more time emphasizing and re-emphasizing what does matter.

Redesigning Microeconomics – Part 1

~700 words, ~4 min reading time

Because I, apparently, don’t actually believe in breaks, I’ve decided to do a major redesign of the main course I teach: Principles of Microeconomics. This has been inspired by a few things – some of it being the research which I’ve linked and summarized here.

Thus far, I have redesigned my syllabi for both my online and face to face courses – including the “Honors addendum” for my face-to-face honors students. I’ll summarize a bit of what I’m doing here:

Assessment

Both face-to-face and online sections are now basically 40% engagement and 60% mastery.

In my online course, I’m using a point system. Students accumulate points in 3 ways: Knewton Homework assignments (100 points, based on successful completion), Short Papers (300 points total), and Engagement Activities (100 points).

Knewton is an adaptive online homework system designed to help students achieve mastery of the course material. The system tries to teach through trial and error. So, if a student answers questions in a topic correct consistently, they won’t see as many questions on that topic. If a student answers them incorrectly, then Knewton tries to adjust the difficulty level to where students can start seeing what a correct answer should look like, and builds them up to the point of mastery. I’m counting this as “engagement” because their grade, in the end, comes from their willingness to keep participating until they “get it” rather than from a summative assessment.

The Short Papers are three papers, each of which is designed to evaluate one of the main course learning objectives in an “authentic” setting. For example: the first paper asks students to choose a good and to make a forecast for the price of that good 1 year from now, and explain that forecast in terms of supply and demand elasticities and supply and demand shifts. In the Spring, I am modifying this assignment a bit by requiring a rough draft and self and peer assessments before the final copy is submitted.

The Engagement Activities give students an ability to customize the course to their own interests. These are generally completion based assignments. Right now, I offer three different options: Excel projects (which teach some basic Excel skills), Economics in the News, and Book Reviews. Students are also free to make their own, if I approve them. These are all completion based, and really just have the goal of convincing students to think a little bit about economics as a field.

My face to face course uses a weighted system, which explicitly separates “Mastery” and “Engagement”. “Mastery” is evaluated based on their performance on the multiple-choice final exam, which can be modified by an optional Grade Proposal – in which they provide evidence that they should get a specific grade for their mastery of course material. This idea was taken from “Rethinking Exams and Letter Grades…” by Kitchen et al. So students aren’t in the dark about the final, I will also have them participate in a “Midterm Diagnostic” which will look a lot like the final, but won’t count toward the course grade.

The Engagement portion is evaluated based on 4 things: (1) Class attendance, (2) Class Preparation Questions, (3) Short Papers (like I use in my online course – but, here graded more on participating in the process than for mastery), and (4) Engagement Choice Activities (which mirror the Engagement Activities from the online course).

For my Honors students, half of their Engagement Choice Activities points come from their Honors Project.

Weekly Rhythm

I’ve established a “Weekly Rhythm”.

For the online courses, the Weekly Rhythm is: Reading, Lecture Videos, Knewton Assignments, Short Paper Step.

For the face-to-face courses, the Weekly Rhythm is: Reading, Lecture Videos, Class Preparation Questions, Short Paper Step, Class Activities.

Next Step

Now that I have my overarching design set up (sequence of course topics, etc), and an assessment plan in place, the next step is to set up my assessments – so I need to write the final and the rubrics/checklists for the papers for my online class – this should help me align everything during the semester with how students will ultimately be evaluated. (YAY for backward design!)