My latest developmental review statements (#1317)

Topics/tags: Triennial reviews, long

In which we take a break from reflections on Michelle and my state of sadness, as well as from shock at the election, to return to a traditional form of musing.

As I noted two-and-a-half months ago, Grinnell has instituted a new developmental review procedure. I started on my review work early. But life got in the way. A few days ago, I realized that my stuff was due on November 3 (a few days ago). Life has been busy, so I wasn’t able to sit down and write the material until the evening of the 3rd. I’d prefer to have more time to reflect and write and rewrite. However, one does what one must. And let me channel Michelle for a moment.

It doesn’t affect your pay. You already reflect carefully on your teaching, research, and service in your musings. Why are you spending any effort on this?

That’s not an exact quote, but it’s close. She watched me starting to think through these issues in mid-July, between the summer hospital visit and the fall hospital visit. It’s also something she’d asked in past years, when we had a merit system.

The raise pool is small. Your merit score barely affects what you’re paid. Why bother?

The Why bother? question is a difficult one. Back when the Dean read our triennial review statements, it seemed like it worth doing them for more than reflection; it’s good for your boss to know what you do. But the Dean doesn’t read them. Given the rate at which we’re going through Deans, it may not matter whether or not the Dean reads them; it’s unlikely that it would affect long-term insitutional understanding. Still, I wish our top administrators understood more what the faculty do, not just what our scholarship is (which they also don’t seem to know), but also what our work stressors are. Oh well. I said as much at the last faculty meeting in the spring. I’m not sure anyone heard me.

You may recall from the past musing that there are three fields to fill out. At a high level, the three fields are titled Reflect on the past 3 years, My plans for the next 3 years, and Is there anything else you think is relevant that was not covered above? We are limited to 6000 characters (about 1000 words) per field.

---

Reflect on the past 3 years. Some questions you may consider include, but are not limited to: Which activities were most meaningful? Which activities were especially difficult? What am I most proud of? Where do I see room for growth? As appropriate, with these general questions in mind, please consider how your activities over the past three years relate to broader efforts at the College regarding diversity, equity, and inclusion.

This section is the primary one; it asks us to reflect on what we’ve done. Back in the days of salary reviews, it’s where you’d argue for a high merit score. The new format should permit us to more readily admit failure.

I have trouble writing concisely. My first (partial) draft reached over 6000 characters before I even finished the teaching section. The solution? Excise the deeper discussion of mastery grading. (Bleh.) Okay, I didn’t really excise it; I moved it to the third part of the review form. Is that cheating?

After that, I wrote the rest. Some notes on teaching. Some notes on service. Some notes on scholarship. Some notes on personal issues. (Hey, they are listed on the preparation grid.) Yeah, I realize that’s backwards. But I’m at the stage of my career in which disciplinary service is central, including leadership positions of two major conferences. Unfortunately, my first complete draft (or my second draft) was over 7000 characters. Damn.

I moved the notes on personal issues to the third section. I did some minor edits. That got me down to 6400 characters. I found myself asking, Should I ask ChatGPT (or CoPilot) to help me cut more? But I decided against it. I should be able to focus my writing. Editing is thinking! So I spent time cutting those 80 or so words. It was harder than it should have been, even though I’m a fairly verbose writer.

Eventually, I got down to 5944 characters or 901 words. If I were more responsible, I would have run them through Grammarly or something similar. But it was Sunday night, and Michelle’s advice was in the back of my mind. Here’s what I ended up with. (Unfortunately, I immediately noticed two typos when I loaded the document into this musing. Maybe I should have used Grammarly.)

This period represents my first three years of post-pandemic work. Like many faculty, I was adjusting to being back in the classroom and to post-pandemic students.

Mastery Grading

I am particularly proud of incorporating mastery grading throughout my courses, including CSC-151, Functional Problem Solving, CSC-207, Object-Oriented Problem Solving, Data Structures, and Algorithms, and CSC-301, Algorithm Analysis. This equitable grading technique bases student grades on how well they demonstrate mastery of the various course learning outcomes without considering when they demonstrate mastery. If a student struggles with a topic early in the semester, they can still show mastery by the end of the semester. In CS, we’ve generally emphasized two primary measures: small learning assessments (LAs), similar to those from specifications grading, and larger mini-projects, in which students apply basic ideas to more complex problems.

Last year, I implemented a variant of mastery grading in CSC-207 that I call portfolio mastery grading. Like mastery grading, it relies on learning assessments. However, those learning assessments take the form of portfolio elements. That is, each presents a learning outcome and asks students to provide evidence that they’ve achieved that outcome, usually a combination of code they’ve written for the class and narrative text that demonstrates they understand the underlying concept. Students are clearly challenged by this approach: Many struggle to find appropriate examples or to narrate their examples; about half the submissions need to be resubmitted. But I see clear growth in understanding when students resubmit. As importantly, I’m helping students develop an important skill: How to concisely demonstrate what you know to someone else.

Other Teaching Issues

I added assignments and projects related to disabilities and computing to both CSC-207 and CSC-151.

I developed a new version of CSC-151 based on image making and manipulation, which included writing about a dozen new lab/reading combinations and six or so new assignments, and a new final project in which students take a computational approach to studying the works of Hilma Af Klint. I also made significant revisions to about two-dozen lab/reading combinations. Somewhere along the way, I wrote a slew of library code to support all of this (somewhere around 10,000 lines of code). For context, a new assignment takes about five to eight hours to write (including a sample solution) and then another five to eight hours for the autograder; a new lab plus reading is about ten hours of work.

The new version seemed worth the time. Students were engaged in the image making and the images helped students better understanding the underlying computational thinking concepts.

I advised a larger-than-normal number of students. I count 21 advisees in 2021-22, 14 in 2022-23 (while on leave), and 20 in 2023-24. These contrast to a College-wide average of 12 advisees.

Service

During this period, I looked for ways to better connect the department to national organizations associated with broadening participation in computing. I successfully applied to have us join the STARS Computing Corps, and STARS has twice funded Grinnell students to attend the CMD-IT/ACM Richard Tapia Celebration of Diversity in Computing Conference. I helped organize the 2023 Missouri-Iowa-Nebraska-Kansas Celebration of Women in Computing (MINK-WIC). I also had the department join AccessComputing, a multi-institutional collaboration that looks for ways to better support students with disabilities in technological fields and to encourage all students to consider people with disabilities in the programs they develop. I serve as representative to all these organizations, which usually involves bi-weekly meetings.

I also participated in normal departmental service during this period, including chairing a complete review and participating in two tenure reviews while on leave.

Beyond Grinnell, I served two major conference roles: I was Engagement Chair for the CMD-IT/ACM Richard Tapia Celebration of Diversity in Computing Conference and I was one of three Program Chairs for the 2024 SIGCSE Technical Symposium in Computer Science Education, the flagship conference in Computer Science education.

As Engagement Chair, I served as a member of the organizing committee, meeting weekly to plan the conference. I also was in charge of our conference communication platforms and served as point of contact for attendees during the conference. I received an Exceptional Service award from the Tapia conference in the fall of 2022.

The SIGCSE TS Program Chairs are responsible for managing the submissions process, selecting the papers for the conference, working with track chairs to select other kinds of work (e.g., panels, posters), and putting together the schedule. We had more than 1000 submissions and 1000 reviewers. For just the papers track, we had nearly 700 submissions, each of which received three or four reviews and one summary review. Of those 700 submissions, the Program Chairs selected about 210 papers for presentation at the conference and publication in the symposium proceedings. Starting in December of 2022 and running through March of 2024, I spent at least five hours per week on SIGCSE TS business; some weeks I spent more than forty hours.

My other disciplinary work during this period included serving as a scholarship reviewer for some conferences, a paper reviewer for others, and a grant reviewer for NSF.

Scholarship

My primary scholarship during this period is the proceedings of the SIGCSE Technical Symposium, an edited volume of 200+ papers. While this is not a typical scholarly work, managing reviewers, reading reviews and papers, and selecting among a large number of papers is much like the work involved in any edited volume.

See, 6000 characters (or slightly fewer) is not all that much writing.

What about the report itself? It’s not great. There’s not a lot of reflection, just reporting. But I feel like you have to report before you reflect, right? What are my reflections? Let’s see.

• I took on too much work, even with a year of leave in the middle. That showed up in a need to cut back significantly when personal issues came to play.
• Mastery grading is taking too much of my time and energy. I need to find a way to make it work better. Perhaps I shouldn’t have 48 learning outcomes for CSC-207. (I didn’t choose them; I just broke up the primary learning outcomes up into chunks that were of a size I found appropriate to assess.)
• I did not succeed in putting together a student group to work on Broadening Participation in Computing activities in the department, which I should do as part of STARS. I hope to make that a priority in the future, perhaps when I return from leave. Perhaps before then.
• Students clearly had mixed reactions to my teaching in 2023–24. Some found me the kindest (and some said best) teacher they’d had. Others were concerned that my distractions at personal issues interfered too much with the quality of my teaching.
• I didn’t publish much research during this time. I need to get back to publishing SIGCSE (or similar) papers, not least because it’s a good experience for my research students. I suppose I should also consider taking on research students again; the past few years have been a break. I also couldn’t have had students in summer 2024 or summer 2023; I needed to reserve time to be with Michelle.

My plans for the next 3 years. What are my goals/intentions during the next review period? Given these goals, what will be the ideal balance between service, research, and teaching? What resources would help me achieve my goals and commitments? As appropriate, with these general questions in mind, please consider how your activities over the past three years relate to broader efforts at the College regarding diversity, equity, and inclusion.

I found this section complicated, too. Unfortunately, I neglected to take notes about the steps I conducted to get down to 6000 characters.

I struggle a bit to think about the next 3 years, not least because I’m already through nearly half of the first of those three years. A number of other factors add to the struggle: I am nearing the end of my Grinnell career, I’m facing life without my life’s partner, and I have a sabbatical next year. I’m also still not sure that I’ve fully recovered from my heart attack. And I wonder what will happen with ITS and broader College policies.

Teaching

I expect to use my sabbatical to turn my materials for CSC-151 (Functional Problem Solving) and CSC-207, (Object-Oriented-Problem Solving, Data Structures, and Algorithms) into open interactive online textbooks. Along the way, I’ll be exploring the question of what Parsons Problems (a popular teaching/learning approach for interactive computer science) will look like for Scheme programs as well as how best to incorporate Scamper, Grinnell’s implementation of Scheme, into my courses, which were designed with a different version of Scheme. I expect that I will need to make many changes to Scamper. I also hope to involve students in this work, which includes scholarly, technical, and pedagogical issues.

I hope to further refine and document the portfolio mastery grading approach that I’ve been using in CSC-207. I don’t think the approach is appropriate for an introductory course like CSC-151, but I expect to be able to adapt it to upper-division courses like CSC-301.

A few years ago, I attended a workshop on having students from other disciplines (e.g., Education or Sociology) observe classes and discuss what they see about the techniques and culture of the class. I’d like to find funding for such an activity.

Given that I will be working on materials for these courses, I would prefer to return to CSC-151 and CSC-207 after my sabbatical. However, given staffing changes in the department, I understand that I may be asked to teach other classes. I will, of course, do what seems to best serve the department and our students. I would like to return to teaching CSC-281 (Learning from CS Alumni) and CSC-282 (Thinking in C and Unix). However, teaching those classes often seems to require more work than the one-credit provides. I’m also not sure that we have space in the curriculum for them.

I’m not sure if I’ll be asked to teach Tutorial again. Nonetheless, I’ve been toying with a Tutorial about four political (or political adjacent) comic strips: Walt Kelly’s Pogo, Gus Ariola’s Gordo, Dan O’Neill’s Odd Bodkins, and Keith Knight’s K Chronicles and (Th)Ink. However, I’m not sure whether that would succeed with students who likely know none of the strips.

I hope to participate actively in department discussions of rethinking the curriculum in light of newly released curricular standards. In my participation, I’ve tried (and will continue to try) to balance how I give the younger members of the department ownership of the curriculum while still contributing from my knowledge and experience. From my conversations with some early-career faculty, they seem very comfortable contributing their own opinions, so I may not need to worry all that much.

The sample review notes grid that was provided for this review included goal for reasonable number of advisees. I’d love to return to the average number of advisees. But I don’t see how that’s possible. Even with Sarah Dahlby Albright (about 18), Henry Walker (about 15), and Liz Rodrigues (about 12) taking CS advisees, it seems like each post-tenure CS faculty member needs to take about 20. The prompt for this section asks us about resources. I’m not sure if this is a resource question, but it seems like one.

Service

Particularly given my activities with STARS, I’d like to help students jumpstart a Broadening Participation in CS group in the department.

I don’t have strong feelings about institutional service. I would not mind returning to a role in either ISC or CSFS when I return from sabbatical. I will also leave my name on the ballot for the various elected positions.

This year, I’m serving as Program Co-Chair for SIGCSE TS 2025. I hope to use next year to gather up some of the materials and knowledge we’ve developed over the past few years and put them into a more usable format. Among other things, I’d like to work with SIGCSE colleagues to develop better training materials for our reviewers and Associate Program Chairs. I also expect to take on another role for SIGCSE TS, perhaps as a track chair or returning to my earlier role of Assessments Chair.

I also expect to take on an appropriate role for the 2025 Missouri-Iowa-Nebraska-Kansas Celebration of Women in Computing.

I also expect to continue normal professional activities, such as reviewing papers and doing external reviews of other department. (Whoops, I may not have mentioned those in the prior section.)

Scholarship

I hope to get my work on portfolio mastery grading into a form that would be applicable for submission to a conference or journal.

Grinnell’s CS department is perhaps unique in how many classes and how many faculty incorporate mastery grading techniques. We’ve even had one-semester visitors use mastery grading. I think the department could produce an interesting experience report on the use of mastery grading across the CS curriculum and hope to take a leadership role in getting such a paper written.

As I noted, I will be exploring a new model of Parsons Problems as part of my textbook development work. I hope that we will learn enough new things that the new model will be publishable.

As a consequence of the other work mentioned earlier, I’ll also be co-editor of the 2025 SIGCSE TS proceedings and author of two open textbooks.

And Beyond

I hope to continue to work on my health and to find an appropriate work-life balance.

Looking back on that, I find myself vaguely remembering that I wrote and deleted a paragraph about a potential willingness to serve as department chair, even though two years ago I would have said that my health couldn’t take it. I’m calmer these days. Really I am.

I’m also intrigued that I maintained the teaching, service, scholarship ordering. Perhaps I don’t prioritize scholarship enough. As I may have mentioned, these days, I submit work primarily to give my research students an opportunity to publish and present. Whoops! That’s something I didn’t mention in my plans for the future. I need to reflect on what role research students and summer MAPs play in my plans.

I think I’d like to take on research students again; some of my happiest teaching moments have been working with summer research students. And some of my greatest impacts as a teacher have been with my research students. But should I take them in summer 2025, as I head into a sabbatical? I’m not sure. It could be a good jumpstart to the sabbatical. Alternately, I may need some time away from everything. I’ll discuss it with my advisory panel over winter break.

That first paragraph is also a bit strained, isn’t it? But it’s true. This month is not the best time for me to think about such issues. I really have no idea what life will be like without Michelle.

Finally, I see that I use the term However a lot. Perhaps I should have run this through Grammarly, too. Or had ChatGPT write it. Or something.

Is there anything else you think is relevant that was not covered above?

Crap. Even the first draft of this was too long. Nearly 7000 characters. In part, it’s because this section was my opportunity to rant a bit. I had a section in which I ranted about being limited to 6000 characters. I had a section in which I ranted about the lack of quality of data at the College; we shouldn’t have to recheck every number we get from the Administration, but we do. (Well, I do; I’m not sure that other people do.) Our directory system should work correctly.

As I say in this section, I’m not sure I’ve ever felt less respected as a faculty member than I have in these past few years. I had more paragraphs about the topic, but cut those even before I was at 7000 characters. These notes are disjoint. I suppose that’s okay for relevant but not covered above.

These comments are not organized in any particular way.

Mastery Grading

Since not all readers may be familiar with mastery grading, and mastery grading plays a major role in my current and future teaching. I’m including some information here.

As noted earlier, mastery grading, at least as we use it in CS at Grinnell, emphasizes small learning assessments (LAs) that check individual learning outcomes and larger mini-projects that bring together multiple concepts into a larger project.

In CSC-151, where Peter-Michael Osera first employed mastery grading and where I first adapted the technique, we have about 24 learning outcomes which get broken up into four sets of six learning assessments (SoLAs). Each LA should take a student about ten minutes to solve. If a student can’t solve it on one SoLA, they have another opportunity for a similar problem on the next, and the next, and the next. We give one extra SoLA during finals week for students who aren’t able to demonstrate mastery during the semester. When I grade LAs, I try to give students enough feedback that they can address whatever they missed and do better the next time.

I also give weekly mini-projects (except in weeks in which students are doing SoLAs). I spend a lot of time writing autograders that provide students with preliminary feedback on how well students have done. Since mini-projects in CSC-151 primarily involve writing programs, the autograders often take the form of given this input, this part of the program should compute the following output, although it’s rarely quite that simple. In addition to the feedback from the autograders, which students can check before they submit a mini-project for a grade, students also get some feedback on the quality of their code. I also write a detailed rubric for each mini-project. Mini-projects are graded on an Incomplete / Redo / Meets Expectations / Exemplary scale (IRME). Meets Expectations suggests that the solution is mostly correct, but may have a few small errors. Exemplary suggests that it has gone beyond the basic expectations to reach a quality that we could comfortably share it with other students and say This is what a good solution looks like. Redo suggests that a student has made a good start, but has misssed some key ideas. Incomplete suggests that the student has not even made a sufficient start.

Grades in the class are based on how many LAs students successfully complete (e.g., they need to successfully complete 22/24 for an A and 20/24 for a B) as well as how many R’s and E’s they receive (e.g., at least 4 E’s and 4 M’s for an A, 3 E’s and 4 M’s for a B).

Once they get used to the somewhat puzzling grading scale, students react quite positively to mastery grading. They say that they feel less stress throughout the semester and that the ability to catch up on something they’ve missed means that the grade better reflects what they understand at the end of the semester. I’ve noted that mastery grading also helps students deal with the occasional times that external forces (illness, family issues, work in other classes) lead them to submit work that doesn’t reflect their capabilities.

My musings

As you may know, I write a ’blog (SamR’s assorted musings and rants) about computer science, teaching, Grinnell, myself, and more. I don’t classify it as teaching, scholarship, or service, but it serves a bit of each. Certainly, I’ve heard from new faculty, staff, and students at the College that they learned important things about the College from my postings. I know that colleagues at other institutions sometimes get teaching ideas from my postings. And I hope that the ’blog helps demonstrate the value of writing, as I often explicitly write to learn. Plus, there’s the important liberal arts model that even a CS prof can write.

Personal issues

This was not a particularly good period for me or my family. I had a heart attack in Spring 2021, so 2021-22 was a year of recovery. Michelle was diagnosed with multiple myeloma in late 2022 and then with pancreatic cancer in summer 2023. My post-heart attack health clearly limited (and continues to limit) how much work I can do, and taking care of Michelle also occupied much of my physical and mental energy.

Professional issues

In spite of many successes (innovative classes that students appreciated, external accolades), I found these past three years incredibly frustrating on a professional level. In particular, I have felt less respected and less supported by the College than I have ever felt in my career. Many things contribute to this experience. Here are two.

There are the direct markers that Grinnell does not value its faculty as much, such as a significant switch in raise amounts (formerly more than cost of living, now less than cost of living) and a failure to meeting the Trustees’ promise to keep our salaries at the median of our peers. In 2023-24, full professor salaries were a full $10K less than the median of our peers.

The rise of ITS Governance (governance that often explicitly excluded faculty) also significantly interfered with my job as a CS professor. Like all faculty, I found myself hamstrung by policies. But I also found my expertise undermined, both in proposed policies (e.g., having ITS review any software we use in the classroom or release to the public) and in the selection of software. Part of our expertise as CS faculty is knowing what software students will be expected to use when they move off into the profession; being told that we cannot teach students about such software is an insult. I’m glad to see that things may be getting better.

There are also many other examples of the de-prioritization of the academic program at Grinnell; I’m sure you know most.

Data

The enrollment information for CSC-281 on my FAR is incorrect; it has a cap of 24, not 29. Data inaccuracies are a regular College problem.

Yeah, I’m clearly facing issues of frustration. I was worse in a few of the sections I deleted. Here are the main ones.

Data

Institutional data incompetence has also been a major frustration. Our online directory doesn’t work correctly; no one thinks it’s worth addressing. We can’t get accurate advisee counts for this review. The Dean’s office sent out incorrect numbers of majors on data sheets for majors at the start of the semester. The list goes on and on. It’s frustrating to feel like I have to recheck each piece of data I get from the institution. And it’s not clear whether all the errors are incompetence or attempts to mislead.

The new developmental review process

I found the 6000-character limit for section 1 frustrating. 6000 characters did not seem to be enough to describe my key activities, let alone to reflect carefully on them. I had to move a narrative about mastery grading to this section, even though it fit better in section 1.

The new process still does not achieve one of my key goals: I’d like our President and Dean to know what faculty work lives are like and to understand (broadly) our scholarship. A process that doesn’t provide some of this information to them seems incomplete.

And there you have it. The developmental review, from my side. What happens next? I think one of my departmental colleagues has to read it. Let’s see …

The chair of the department will review the faculty member’s materials, indicating in Interfolio that they have reviewed them. They will also be able to provide additional comments.

Whoops! Sorry department chair (or designee).

Members of the faculty review committee, which consists of the elected division chairs and the chair of the faculty, will review the faculty member’s materials and offer developmental feedback.

Whoops! Sorry division chairs and chair of the faculty.

But I’m confused. Wasn’t the new process supposed to make things easier for you? It seems like you’ll have a similar amount of reading. At least you won’t have the frustration of having to award scores.

The final step in the developmental review process will be a meeting with the Associate Dean for Faculty Development, who will share feedback from the review process, including comments from the review chair and the faculty review committee, and provide an opportunity to consider how to work toward the faculty member’s goals.

Meeting with an Associate Dean is not the same as meeting with the Dean. But I look forward to considering how to work toward my goals. Some of those issues are things to discuss with my Department Chair, such as what courses I’ll teach. But perhaps the Associate Dean can help solve our advising problem or brainstorm with me about ways to find time and energy for the BPC group. We’ll see.

All in all, I spent four hours on the developmental review and perhaps another hour or two on the wrapper text in this musing. Were the first four hours a good use of my time? Doing that work meant that I didn’t get grading done. I suppose I’ll see after I get feedback.

---

Version 1.0 released 2024-11-07.

Version 1.0.1 of 2024-11-07.