Improvements on UTD Grades

[WillieCubed]

Project Nebula is going to re-conceptualize UTD Grades from the ground up as a project codenamed Athena. The question is why?

UTD Grades vs. Higher Education

In my view, UTD Grades is contributing to the decline of American education. The university should be a place where students come to learn and embetter themselves, not merely to earn a credential. Obviously, this isn't to say that students should only come to a university to spend four years and thousands of dollars studying something that may not be "useful" in the workplace. People have to put food on the table, and a bachelor's degree is a ticket to a better life for many people. However, that doesn't mean obtaining a college education should be "easy." Education should be rigorous (if for no other reason than to not have planes fall out of the sky or bridges collapse), but it should be rewarding. UTD Grades plays a harmful role in commodifying higher education by reducing a degree down to the sum of its coursework and grades.

Why UTD Grades Fails

If we're being really critical here, the service reinforces class-based divisions between students who know how to game the system of higher education and students who come from backgrounds that simply have to take the experience they're given with the instructors available to them. For example, I see that many students with priority registration (usually honors students with high AES awards or receiving National Merit Scholarship, often coming from socioeconomically privileged or otherwise well-off backgrounds) use UTD Grades to choose instructors with more favorable grade distributions, leaving students with lower priority registration times to deal with instructors who may be less effective or less prefered instructors.

Of course, we can't entirely blame students for merely using the tools available to them; if given the choice between an instructor who teaches classes that produce high grades and one who teaches classes that produce lower grades, most students would always choose the former, even if the latter ultimately taught more. Furthermore, it's not uncommon that instructors with more favorable grade distributions simply teach in a way that students find easier to learn from.

Flaws in the Service

However, the current design of UTD Grades prioritizes quantity (number of high grades) over quality (instruction quality, pedagogy of instruction). For example, the user interface implicitly encourages students to focus on higher grades in class grade distributions and to avoid lower ones through a very deliberate color palette. (Green: good, red: bad.) Professors are reduced to names and grade distributions, and that's it. On a user experience note, the current service leaves much to be desired. The dashboard lacks useful filtering tools and could have a more intuitive layout that allows for comparison between instructors. The search functionality is limited to almost exact matching, and (dare I say) there isn't even a theme switcher for dark mode. (The horror!)

Even on a functional level, the current service is flawed: the conclusions people draw often aren't valid or are misleading because the tool doesn't properly contextualize them. The grade distributions, like any dataset, suffer from response biases like sampling bias (since only certain students, not an "average student" take specific coursework which influences the grade distribution of a course) and survivorship bias (students who make it to upper-level coursework are the ones who haven't dropped out of college). Speaking anecdotally, the judgements many students make implicitly make many assumptions about the data, namely that if they were to take a course taught by a specific instructor, their probability of obtaining a certain grade would look like this. However, this mindset fails to consider the process that causes students to obtain certain grades in the first place. A "C" student may try to use UTD Grades to earnestly find a course that tends to produce many "A" grades only to find out that a course was more rigorous than they anticipated and that students who took the course and got A's were already prepared for the work.

It then becomes clear that some students don't use UTD Grades to further their education but simply as a means to an end: they want to get their degrees as quickly and effortlessly as possible, no matter the cost.

A Future for UTD Grades

Personally, I want to dispel the notion that a "good" professor is the one with the "best" grade distribution (namely, one who teaches classes that produce higher grades like A's and B's). However, it's obvious that just because some instructors have better grade distributions, that doesn't mean they teach better. Within the Computer Science Department, certain faculty have a reputation of teaching simplified course content only for their students to be unprepared to tackle upper-level coursework. This is dangerous to the integrity of the university as an institution.

However, sometimes, there are course instructors who don't teach well - at least according to the university's own course evaluations. Whatever replaces UTD Grades should not demonize faculty or reduce instruction down to a simple letter grade but should allow someone to seek a balanced perspective based on data. For example, an instructor may receive low evaluations from students, but perhaps it's because the instructor was more comfortable teaching upper-level coursework or that the department as a whole was not preparing students for a particular course, measured by other the same course taught by other instructors. Furthermore, this should be used as an opportunity to personalize the university experience and remind students that the university is made of knowledgeable faculty who share knowledge, not just made of homework and exams. This could connect students with faculty who have teaching styles best for them to enhance the learning environment.

Our biggest challenge will be creating a tool that is familiar to the experience students currently have with the service but ultimately building something that works better for the community as a whole. If users find that the new experience offered by Athena makes planning out coursework simpler, especially connected to Planner and Guide, the replacement will be a success. Athena doesn't have to be some weapon in a nuclear arms race between faculty and students. If implemented correctly, it may be able to correct harmful perceptions of a university education among students and encourage a higher-quality classroom experience from instructors.

I invite other contributors to share their thoughts in this discussion.

(Cross-posted from https://github.com/UTDNebula/utd-grades/discussions/24)