Empathy, Listening, and Hearing

I have more stories I want to share here than I have made time to share them, but here's one that I find my idle mind keep returning to. I want to capture it here to make sure I don't forget it.

We are reading Design of Everyday Things in my HCI class (CS345/545), and several meetings ago, the students read Norman's presentation of the UK Design Council's double diamond model for design. After some thinking and Googling, I developed a series of in-class exercises to help students understand the model. Particularly helpful was Heffernan's overview of activities associated with various stages of design. I decided that an exercise on empathy mapping might be just the right way to start.

Of course, to build empathy, you need a focus and a context. I decided to use, as a running example, students' experience dealing with the triage grading system that I use. This is a brilliant grading system that I learned from William Rapaport at University at Buffalo when I worked as a TA with him. It is coherent, philosophically-sound, and unfamiliar to almost everybody. I get the occasional question about it and the student whinging in course evaluations, but by and large, student experience with it is unrecorded: it happens in the shadows or in passing conversations. It's also something that all of my current HCI students are experiencing, and practically all of them have either experienced it before in one of my other classes or know someone who has.

I introduced the context in class and asked students to work in small groups to develop a map, writing them on the classroom whiteboards. We used the conventional map labels as shown in Heffernan's overview: what do students think & feel, hear, see, and say & do in their experience with triage grading.

It's helpful to have just a passing understanding of triage grading before we move on. Whereas conventional grading is based on percentage correct, triage grading is based on discretely measured quanta. For example, if you assume that 90% is an A, then as a test designer, you would design the exercise so that A-level work is attained by completing 90% of the prompts correctly. Notice that this is the tail wagging the dog: the fact that you are using conventional grading determines your test structure. With triage grading, any given item is scored out of three points: essentially correct (3 points), essentially incorrect (1 point), or somewhere in between (2 points). The letter grade is determined by weighted linear interpolation across scores, assuming that "A" means correct, "D" means incorrect, and "C" means middling.

Almost every group wrote down that they see percentages when they first encounter triage grading. That is, they see "1/3 points" as 33%, which they interpret as "Low Failing Grade" even though in triage grading it is a low passing grade. (You might consider 33% in triage grading as having a qualitative interpretation like 65% in conventional grading, right on the border of poor and failing.) There was broad consensus about this in the class.

I pointed out that the maps appeared to come from initial experiences with triage grading, but one of my bright students—who has taken my classes before—noted that his was more of a mid-semester view. He had recorded in his map that he became able to see the feedback as qualitative, as coarse-grained values that drove him to change his behaviors. I do not remember exactly the words he wrote on his empathy map, and indeed I didn't understand what he meant by the words he chose, but our conversation came back to the concept "seeing quanta" rather than "seeing percentages."

That was pretty interesting in itself, but here's where it kicks up a notch. A student in the back chimed in, saying essentially, "But it's still a percent." The first student acknowledged that mathematically it was, but that's not what he saw, and the one in the back insisted more strongly, essentially, "But it's some number of points out of a total, and so it's a percent, and so you're still seeing it as a percent."

Wow! What a teachable moment for empathy! I pointed out, treading carefully, that this was an example of the second student showing no empathy for the first student. The second student saw the world in his way, insisted that it was the right way, and that everyone else must also see it that way. I introduced the idea that whether or not there is an objective reality, perception drives a person's lived reality, and perception is subjective. Two people can look at the same thing and "see" two completely different things. The expression on the second student's face told me that he understood what I was saying, but he was still working on the implication; of course, maybe I observed this wrong.

We are continuing to work with this example, and I am finding it a rich context for discussion. I hope to share a few more stories here on the blog, but for now, it's time to head off to class. Thanks for reading!