~2000 words, ~10 min reading time
Part 1, Part 2
So, in Part 2, we learned that my students will have lots of ‘splaining to do, as the verb “understand” in my course learning objectives more closely align with the facet of understanding called “explanation”. So, that should play some role in my assessment strategy.
Elements of Assessment
A few elements that I’ve decided need to be part of this, for one reason or another:
(1) “Chapter Assignments” – to convince students to read, they have to, in some way, respond to that reading. So, I’m going to have my students do Chapter Summaries. Based on the course learning objectives, I’ve decided to make these written assignments where students select some combination of economic models, problems, and solutions, and explain them. The main purpose is to get students started thinking about these, and convince them to at least skim the chapter with enough attention paid to write a summary of it. Graded for completion.
(2) “Article Responses” – this is an opportunity for students to go a bit deeper into topics they find interesting. They will read papers from academic journals (or similar sources). Totally their choice. The purpose is to get students to engage with the material a little beyond what we do in class. Graded for completion.
(3) “Case Studies” – the department declares that “problem sets” are part of this course. The professor who has taught this before has provided me with the details for 5 case studies that they assign. These are reasonably advanced assignments (he noted that his students needed help with these), and require some of the more technical skills that we’ll be talking about. This hits the “Applying” aspect of “understanding”. Here, I will want to grade for quality rather than just completion.
(4) “Term Paper” – the department declares that a “research paper” is part of the course. So, here I’m doing that. I’m forcing students to go through a 4 step process – proposal, annotated bibliography, rough draft, and final draft – with a response to comments and a reflection.
(5) Reflections – every assignment has a required “From this assignment, I learned that…” section. This is to encourage reflection. Also, the form of this statement is important. “I learned that” forces a sentence to follow. “I learned…” doesn’t. Example: “From this assignment, I learned about elasticity.” That’s just a word. Only tells me that you saw the title of the chapter. Not nearly as good as “I learned that elasticity is how much quantity responds to a change in price.” In addition, I have built in “Midterm Reflections” every few weeks, to encourage students to keep track of their progress in the course. These, however, are not required.
(6) Final Exam – I’m going to follow Linda Nilson’s advice and give them the final exam early in the course – ideally, on the first day, if I can get my act together. Final exams of some variety are required by the University, and I think they are a good way to see what students actually absorbed. Because this can’t be revised, I do grade this one with partial credit.
(7) Revisions – Students are allowed to revise any unsatisfactory assignment except the final exam, as long as they submit a “Revision Form”. (This is a Word document where the student has to answer four questions: which assignment they’re revising, why they didn’t meet specs originally, what they changed to meet specs, and what they plan to do to ensure they meet all specs in the future.)
Specs Grading – The Core Concept
The core concept of specs grading is grading assignments on a pass/fail, satisfactory/unsatisfactory basis on the basis of clear “specifications”.
What these specifications look like is up to your goals for the assignment. A good rule of thumb: look at the rubric you use now. Write your specs to be somewhere around the top 2 levels of your rubric. Naturally, you don’t want to be too mechanistic about doing it this way – take the opportunity to think about WHY things are in the rubric and what is really acceptable and what would lead you to hand the paper to the student and say “do it again”.
For my purposes, I write up two sets of specs, depending on the assignment: format specs, and content specs. Format specs lay out things like length, file type, work cited pages, numbers of sources. Content specs say what should be in the paper. If I’m grading for quality, the content specs will include descriptors like “correctly” or “reasonable”. Otherwise, they won’t.
Just as a couple examples: for my chapter assignments, students must meet these specs:
A satisfactory submission will:
(1) Be at least 300 words.
(2) Be submitted as a Word document.
(3) Contain a summary that deals with at least one of these elements: (a) economic models, (b) problems in the economics of the environment, or (c) solutions to these problems, and for each of these you must include:
(a) For economic models – describe the economic model, explain its assumptions, uses, and limitations.
(b) For problems – describe the problem, and the conditions that lead to it occurring. Examples are encouraged, as appropriate.
(c) For solutions – describe the problem being solved, the solution, why the solution could alleviate the problem, and any limitations to the solution.
(4) End with a paragraph starting “From reading this chapter I learned that…”
Remember, this is a “completion” grade. For these, I can use “power grading” – or what I like to think of as “grading at a glance”. Import the file to word, check the word count. Check the last paragraph. Skim the middle to make sure it’s relevant. Done. Doesn’t need to take more than a few seconds – but as a written assignment, I can get some “deep” data about what students think is worth including and what they think they are learning.
For the final draft of their term paper, my students will have this set of specs:
To be satisfactory, the final draft must be at least 1,000 words long, be “complete” in terms of its structure (with a clear introduction, body, and conclusion), have a bibliography or works cited page in MLA, APA, or a similar format, use at least 5 sources, be submitted on Blackboard in Word format, and the body must state what the problem is and evaluate two possible solutions for the problem, clearly stating the strengths and weaknesses of each approach. Finally, after the works cited page, the Final draft must have two additional sections. The first is a “response to comments” – in this section, you must respond to any suggestions that were made on the first draft, either stating how you incorporated those suggestions or stating why you decided not to. The second is a paragraph that starts “From writing this term paper I learned that…”
Here, there’s more concern about quality. “Clearly stating the strengths and weaknesses of each approach” implies that they can’t just write nonsense. If they say something is a strength that ISN’T, then the paper isn’t satisfactory. Also, notice the last parts: one of the most annoying things as a professor is when I give students comments on a rough draft telling them how to improve the paper, hand the paper back, and the student proceeds to turn EXACTLY the same paper in for the final copy. Requiring a “response to comments” section prevents that situation. Even if a student doesn’t incorporate my comments, they at least have to say why they didn’t. Both of these require that students do something they often don’t: read my comments and think about them on some level. Another thing this does: it speeds up grading of final drafts by letting me just check the revised parts.
Because pass/fail grading is somewhat high stakes, students should be given SOME chance to revise assignments. There are a number of ways to do this. Some only allow the first assignment to be revised. Some allow a fixed number of revisions for each student. Some employ a “token economy” in which students are given or awarded “tokens” (perhaps for answering questions in class, turning work in early, etc.), which can then be redeemed for various things (like a bump in the final grade, opportunity to turn an assignment in late, an additional “skip day”, etc.). Personally, I find limiting revisions to be too time consuming to track, and question the pedagogy of telling a student not to bother redoing work that wasn’t good enough. I’d rather just leave them unlimited opportunities, but force them to be a little reflective about the situation.
Putting it Together: Getting to Letter Grades
Under Specs Grading, there are two main approaches to turning a set of S&U grades into a course grade: a points-based approach and a “bundle” approach. Personally, I like the “bundle” approach, but I’ll talk a bit about each.
The points-based approach does as is typical in any other point-based system. There are X points in the course, and you need 90% of them for an A, 80% for a B, etc. (adjust as you like). The only “specs” part of it is the satisfactory/unsatisfactory grading with some revision opportunity.
The bundle-based approach ties each grade to a bundle of assignments that must be completed. For example, in my course, here are the C and A bundles:
To earn a C you must: complete 10 chapter assignments satisfactorily, complete 2 article responses satisfactorily, and complete 3 case studies satisfactorily.
To earn an A you must: complete 13 chapter assignments satisfactorily, complete 3 article responses satisfactorily, complete 4 case studies satisfactorily, and complete a term paper satisfactorily.
Notice: there is a combination of “more” and “harder” to move up the grade scale. A student who just wants a C (or a B, for that matter), can forego the term paper entirely. In my opinion, this ability to tailor the TYPES of assignments to specific grade levels is a strong reason to use the bundle-based system.
What about the final exam? I’ve decided to make the final exam determine +/- grading as follows:
>85% on the final adds a “+” to the grade.
50-69% on the final adds a “-” to the grade.
<50% on the final bumps the grade down to the “+” of the next letter down. (So, a B becomes a C+ if the final is really bad.)
For Principles of Microecon, I have the final exam as part of the bundle (so, an A in the course requires at least an 80% on the final exam, plus the other work). There are pluses and minuses to both methods. The Econ of the Environment method increases certainty for the students going into the final (“I’m going to end up with something between a B+ and a C+ in the course!”). But, it leads to interesting gaps. For example: I expect a D+ would be very rare, as would an A-, and if these DID happen under this system, I would suggest it’s a weakness of course design somewhere. (A student can do all the A work otherwise, but still end up with <50% on the final? Either your final is super-hard or your A work isn’t actually encouraging much learning – or something is poorly aligned.) On the other hand the Principles system makes the final more high-stakes, and led a few students to complain to me about how their final dragged down their course grade. (“I’m a bad test taker!”) Though, practically, the final by itself never seemed to make more than a letter grade of difference for any student. Still, high-stakes testing tends to unnecessarily increase student stress, even though I try to include a fairly significant buffer (requiring an 80% for an A, 65% for a B, 50% for a C).
So, how to figure out what I’m actually doing IN class? That’s a subject for another entry (or more!).