Some time ago I outlined the goals for GE 103, our liberal arts math class. In this post, I want to follow up and discuss assessment — what I am having the students do, how I am grading it, and how that stuff relates to the goals for the class. (I waited to post this until I had actually given and graded a substantive item so that I could post it; wouldn’t want students reading this blog to get the assignment before it’s actually assigned, right?)
My philosophy about assessment has changed considerably since I started my first professorial job in 1997. Back then, my courses were like Rube Goldberg machines, with all kinds of funky assessment tools — weekly journals, 15-page research papers, graded group exercises, etc. in addition to the usual stuff like quizzes and tests. Some of my early courses had 7–8 different kinds of assessments. These were included basically for one purpose: To make me look like a pedagogical genius to my colleagues. (It’s just one of the many ways my early career was corrupted by Project NExT and the MAA; but that’s another post.) In recent years though, I’ve tried to keep to this basic idea about assessment: Design a minimal number of assessment tools that gather maximally useful amounts of information about student learning. In other words, maximize the efficiency or bang-for-the-buck quality of your assessment tools, using a small number of highly effective items.
This change of thought stemmed from something I heard in a talk — that if you trace the word “assess” back to its etymological roots, it comes from the Latin meaning “to sit down with”. So in its purest form, an assessment — test, quiz, paper, project, lab, whatever — is a means for me to “sit down with” my students, either figuratively or literally, and help them get from where they are to where they should be. Assessments should convey information, in other words, and not just be fodder for statistical analysis, or a means for obtaining something like points on a scoreboard.
And the only way to really “assess” in a supportive way is to have a small enough number of assessments so that I really can address the needs of each student. If I have upwards of three dozen assessed items in a 14-week course, there won’t be enough time left for the human element of assessment, which is the whole point of assessment. And in order to be confident that the information you’re getting is actionable, the assessments have to be intimately tied to the course goals — in fact should be extensions of the course goals.
Anyway, here’s what students are going to do in GE 103:
- Four 50-minute tests, one for each main unit of the course (logic, statistics, social choice, personal finance).
- One final exam.
- Weekly feedback journals.
- Weekly lab assignments.
The tests and final exams are self-explanatory; they are designed around the ideas of being able to produce quantitative information (e.g., computing statistical data, making a truth table, etc.) and evaluate quantitative information (e.g., interpret the meaning of this boxplot, analyze using a truth table whether this argument is valid, etc.). I’ll include some essay-type questions to hit at the thinking effectively/making informed choices goal.
Every Tuesday we meet in a computer classroom, and to make use of that environment as well as to train students on useful technological tools, I have set up labs. The labs themselves consist of three components:
- A 5-minute multiple choice quiz over the main concepts of the previous week (5 points);
- An in-class activity done in groups of 2 or 3 in the lab session itself (20 points);
- A homework assignment that extends the in-class activity (15 points).
Those of you who have followed my POGIL-blogging before will notice a similarity to POGIL activities in my Methods of Problem Solving course and the structure of these labs. In fact I adapted the lab structure loosely from the POGIL activity structure. Students have to do stuff outside of class to prepare for the lab (I assign exercises from the book), do the activity in a group, and then do something after the activity. The assignments I give throughout the week aren’t taken up, but students who blow them off will have a difficult time getting through the lab material in the allotted time; notice that the in-class activity is the most heavily weighted of the three components.
The labs address both main course goals, but they are more oriented toward the “think effectively and make informed choices” goal than the “produce/evaluate quantitative info” goals. (At least, this will become more and more of the case the further into the course we get. At this stage students are still learning to do truth tables; we’ll use that a lot in next week’s lab on evaluating the validity of an argument.)
Finally, the feedback journals are just weekly submissions that address three main questions:
- What was the most interesting, cool, or compelling thing you saw this week in the class?
- What connections did you discover this week between what we did in class and (a) other ideas we’d seen previously in the course, (b) ideas from your other courses you are taking, and/or (c) your life?
- What was the most unclear or difficult idea from the class this week?
I grade each question on a scale of 0 (didn’t answer it, or it was completely unclear and/or showed no real thought) to 3 (thoughtful, clear, and grammatically correct). I think the feedback journals address the goal of “think effectively/make informed choices” but really they are intended as a catalyst for real assessment, and — more importantly — making personal connections with my students. I’ve already had two very nice outside conversations with students based on their responses to last week’s journal entries.
Students can submit these journal entries on paper, or they may email them to me; but they may also “submit” them by joining me for lunch or coffee and discussing them in person, or posting them to their Xanga or LiveJournal blogs and sending me the permalink. So I’m hoping students will take an interest in those last two options for submission to either build a good relationship with me or to elevate the level of discourse that I typically see on student blogs.
So that’s my plan — keep a small footprint with a small number of (hopefully) effective assessments that are tied explicitly to the course goals.
For what it’s worth, the first week’s journals were OK; about half the students need to write in more detail about their experiences, but that’s the first week and I expected this. The class average on the lab quiz was 2.57 out of 5, with most of the misses coming on the same two questions — which conveys information to me, namely that we need to take some time in class to discuss those two questions. The average on the in-class portion of the lab was 15.95 out of 20, which was pretty good; again lots of repetition of the same mistake, which tells me something. The homework is due tomorrow.