Making sense of student evaluations

I'm at a bit of a crossroads as a teacher, and want to see what you all think. Here's the story in a nutshell. I've historically had really good student evaluations, well above my university averages and usually accompanied by very positive written comments. I also had a number of students take class after class with me, and was able to get these results in a way that I thought had a lot of integrity (e.g. my courses were not easy, but rather rated by students as significantly more challenging than other courses). Indeed, my methods played into my teaching philosophy, which has always been to bring out the best in as many students as possible, 'leaving no one behind.' One feature of this philosophy - which I still utilize - is to allow and encourage students to revise-and-resubmit at least one of their term-papers as many times as they wish (which I've found to be very important pedagogically). While my courses were rated as quite challenging (students often received very low grades on early paper drafts), the methods I used often resulted in fairly high average final grades (e.g. in the 'B' range) - something which I expect led most of my students to walk away from my courses pretty happy. 

However, several years ago during my pretenure review I was encouraged to change my practices in a couple of ways:

1. Expand my methods of evaluating student work beyond the primarily essay-focused exams, reading responses, and term-papers I was giving (e.g. to include things like multiple-choice quizzes, etc.).
2. Make it more difficult to get B-or-higher final grades (ostensibly, to increase the overall challenge of my courses).

While I was a bit hesitant to change practices that seemed to be working, I not only wanted to get tenure - I also wanted to trust my senior colleagues (the department and college pretenure committees). So I went out of my way to change my teaching methods in the ways suggested...and the consequences of the changes I have made are both interesting and challenging. So much so in fact that I feel torn about what kind of teacher to be moving forward. Allow me to explain.

The interesting thing I found was that, from my vantage point at least, the recommendations I was given by the pretenure committees (and my dean) were pedagogically good ones. First, the different kinds of assessment methods I'm now using (many of which are still essay-based, but which now include things like multiple-choice quizzes and more challenging weekly written homework) seem to me to measure important differences in student performance, doing a better job of separating out who has better learned and retained the relevant philosophical material (which seems important to me, as I want students to leave my courses actually remembering what they've learned!). Second, the new assessment methods do indeed appear make my courses even more challenging than before, making it much harder to get a grade of B or higher (and especially an 'A' grade) than before. This also seems good to me, as it makes higher grades in my courses more of an achievement. 

The challenging thing I've found, however, is that the changes appear to have led to significantly-less-positive student evaluations, and fewer students repeatedly retaking courses from me. The changes are by no means 'terrible.' On the contrary, my quantitative numbers are still pretty high given the level of difficulty my students report. Whereas my student ratings used to consistently be well-upward of 4.5 (on a 5-point scale, where 5 = 'outstanding'), this past semester they were closer to '4' (i.e. 'very good'). From my vantage-point, this is still pretty good, especially given that my students reported the overall difficulty of my courses approximately one full standard-deviation above my college averages (on assignment difficulty, exam difficulty, etc.). I've doctored an image from this Slate post on how 'hotness' and difficulty affects ratemyprofessor rating to illustrate roughly how my student evaluations have 'moved' quantitatively:

This still looks pretty good to me. Relative to the level of difficulty my students are reporting (i.e. super high), my overall evaluations are still quite 'good.' However, the change is significant. I'm also receiving more qualitative 'complaints' from students in their comments - the most common of which have recently been that my courses are too difficult, I move too quickly through the material, and teach the material at too high of a level. And I'm also getting fewer students who take my different courses from me over and over again. 

The problem I'm facing (viz. the 'crossroads' I mentioned at the outset of this post) is that I'm just not sure what to do with this. Now that I have tenure, I don't have some of the self-interested reasons to worry about this that I might have had pre-tenure (viz. a tenure-committee being worried about unhappy students). It's also not clear to me--despite the comments I've been receiving--that I'm doing anything wrong as a teacher. Yes, my courses may now be super-difficult; yes, I may move at a breakneck speed; and yes, I may teach material at a high-level - but I have some really excellent students who meet all of these challenges (viz. the few 'A' grades I've been assigning), and it's not clear to me that other students are worse off for it (despite their perception that the course may be 'too difficult' or 'too fast'). It's not clear to me in part because of recent studies indicating that student perceptions of learning may have a tenuous (at best) relationship to actual student learning. Since I'm not sure I'm doing anything wrong, the only 'clear' reason I seem to have for changing course (i.e. slowing down, lessening course difficulty somewhat) seems to me self-interested, viz. having more happy students, more 'repeat customers' (students taking multiple courses from me), attracting more majors to the department. These may actually be good reasons. At many institutions these days, the health and viability of small departments (e.g. humanities departments like philosophy) can very much depend on things like class enrollment, major numbers, etc. Still, these are self-interested considerations, and should (presumably) be weighed against other ones, such as pedagogical integrity. I've always wanted to be more than a good teacher. I've wanted to be a great teacher--and an inspiring one--in large part because I was inspired to become a philosophy by ones I had, and I've seen in the most personal way how much of a difference a great teacher can make in a life. On the flip-side, it has never been clear to me whether truly great teachers are always (or even usually) the most well-liked teachers. For instance, one great teacher who made an enormous difference in my life wasn't all that well-liked by a good number of students, in part because of how demanding they were as a teacher. 

Hence, the puzzle I'm facing: I'm just not sure how to proceed. My evaluations are 'worse' than they used to be, and I have more complaints than before - but they are still 'pretty good', and a fair number of students still give very positive feedback, despite how challenging my courses have recently become. When it comes to the question of whether to 'change course', the 'balance of reasons' is unclear to me. Which is why I share my predicament with you all. What do you all think? How do you respond to student evaluations? How do you think one should respond? Looking forward to hearing your thoughts!

Comments

[Amanda]

Hi Marcus,

Have you considered slowing things down while still incorporating the new methods? One problem I used to have, is that a large number of students would do horrible on my quizzes, even though they (at least claimed) to have done the reading. I realized I very much overestimated reading comprehension abilities. Now my quizzes are very easy. I try to make them so it would be very easy to pass, provided you did the reading. That is okay with me, because for me the point of the quizzes is to make sure the students actually do the reading. Given the level my students are at, this is enough of a win for me. My guess is you have better students, but you could still try to make the quizzes a bit easier.

The struggle on how to balance the brightest students with the average ones is hard. It seems to me there is no way to be completely fair to both. The best I have come up with is offering extra assignments for extra credit, as well as having assignment options. The assignment options has really helped. I give my students a list of possible assignments which includes, tests, essays, presentations, summaries of difficult (not mandatory) readings and even making videos. This way students can learn in the way that feels best to them, even if it is difficult they had a say in choosing the result.

Unlike you, I am actually not concerned with my students remembering the material. What I mean is I don't care if they remember what utilitarianism is or how Aristotle defines the golden mean. I think it is very, very, hard to teach in a way that students remember this. What is far more important to me is that students leave with better critical thinking, argumentative, and writing skills.

As to whether it is pedagogically a good or bad sign that one is getting slightly lower evaluations...I am conflicted on that.

[F. Contesi]

Thanks for this, Marcus. Besides the thing about you going “at a breakneck speed” and using multiple-choice tests (both of which I suspend my judgement on), it doesn’t seem like you’re being a worse teacher than before. You’re probably being a better teacher, in fact. I would stick with your current teaching and live with the worse student evaluations. If you really want to try something, perhaps try to be explicit with your students about why you are being tough etc.

These are two additional studies (weirdly not cited by the more recent study you link to) that cast serious doubts on the validity/usefulness of student feedback (at least as it is currently used):

Scott Carrell et al. (2010), “Does Professor Quality Matter? Evidence from Random Assignment of Students to Professors”, Journal of Political Economy 118, 409—432

Michela Braga et al. (2014), “Evaluating Students’ Evaluations of Professors”, Economics of Education Review 41, 71—88

[Marcus Arvan]

Amanda & F.: Thanks for sharing your thoughts. Oddly enough, your two views are more or less the ones I'm vacillating between!

Part of me is drawn to what Amanda says: that I should "slow down", since a number of students are calling for it. However, another part of me wonders whether students calling for it is a sufficient reason to do it - as a fair number of students *can* keep up (and even excel) at the pace I currently move.

As Amanda notes, "The struggle on how to balance the brightest students with the average ones is hard." Indeed. This is a big part of my struggle right now.

[G]

I strongly agree with Amanda. Unless I am missing something the change in teaching methods seems to have revealed something that was going on all along: the students weren't doing/comprehending all of the reading. Of course, the best students were, but I am mostly thinking here about the not-quire-so-best students. The same thing, more-or-less, happened to me. I started giving multiple choice quizzes at the beginning of every lecture. The quizzes had super easy questions (along the lines of "Searle imagines someone in a room who does know know what language? a) English b) Chinese c) French d) html") and other questions that really attempted to tell whether they understood the philosophical content. At first, the students did astoundingly poorly, and this was at a not-quite-Ivy-level-but-close R1. I was surprised. It turns out that many of my students had not been regularly doing the reading. Then they started doing the reading, because of the quizzes, and began to get all the easy questions, but were still failing the substantive questions. Then I made all of the readings shorter and easier, with less material that was irrelevant to what I actually wanted us to discuss in class. And that is when things really changed - the class discussions exploded with excitement, the papers improved, students stayed after class to talk more philosophy, and signed up for more philosophy courses. (All of this applies to lower-division courses. I would not want to so dramatically reduce the readings for upper-division courses.) It turns out that a philosophy course where all of the students, even and especially the least-keen students, do the reading is a wonderful thing. I think that as instructors we can get ourselves into a space of focusing on the best students somewhat more than we should, especially in intro-level courses. So my suggestion is the same as Amandas: keep the new teaching methods, but make the readings shorter and easier, or, to use the standard euphemism with which we are more comfortable, "slow down."

My sense is that disciplines in the humanities can be divided into two groups: those that take pride in how difficult their undergraduate courses are and think of those courses as weeding out the weaklings (e.g., engineering, math) and those that try to bring out how fun the material is, don't worry about how oppressively difficult it is, and try to increase the number of majors (e.g., psychology, many preprofessional disciplines). Philosophers are definitely in the former group, but we could sure increase enrollments if we moved a little toward the latter.

[G]

One more point regarding question being discussed of whether to focus on the best students or the more average ones: in intro-level courses, focus on the average students. The best/keenest students are already likely to take more philosophy courses. If enrollments are to increase, we are going to have to pick up other students. In my experience a very, very small minority of the best students will wish the course was faster, but that's good. They are the best students, after all.

[Marcus Arvan]

Hi G: Thanks for weighing in.

In many ways, I agree with you. As I mention in the OP, I've long had a "nobody gets left behind" pedagogy opposed to the "weeding out" approach you have concerns about (which I share!). I am also seriously considering "slowing down" for roughly the reasons you give. But I'm still a bit torn, for reasons I give in the post and alluded to by F. Contesi.

I actually already do quite a few of the things you describe. First, because I have small class-sizes, I have every student submit short daily (in some cases weekly) reading response assignments that require them to engage substantively with the reading. These assignments require demonstrable comprehension and critical thought - and while students usually begin the semester doing terribly on them, they tend to get the hang of it as we go along. I also have tried assigning fewer pages. As a result, my classes (much like yours) routinely "burst with excitement" and discussion.

The main issue, as far as I can tell, doesn't have to do with daily student excitement, getting students to do the reading, or engaging in class. My students appear to me to do all of the above as much as they used to. Further, as far as I can tell, I'm not "leaving more students behind" than before. Indeed, I don't even think I *am* teaching things any faster than I used to (I always taught at a quick pace, even when I routinely got ratings like 4.8 out of 5). Rather, the biggest difference--following the Slate article (and other studies)--seems to me to be my assessment methods and the student *grades* that result. Which is where I think my ambivalence stems from.

Because of the new assessments I've introduced, my average final grades have gone significantly down (like my pretenure and tenure committees suggested). Although I could be wrong, it seems to me like it may be this primarily that has resulted in my somewhat lower evaluations and higher "difficulty" ratings. For instance, this semester I taught three courses--and surprise, surprise: I had the highest evaluation scores and best comments in the course with the highest final grades, my second best set of evals and comments for the course with the second highest grades; and the lowest evals and worst comments for the class with the lowest grades. The issue here is: aside from changing assessments (and assigning lower grades), I don't think I'm doing *anything* differently than I used to.

This, I guess, is one of the main roots of my concern. It's not clear to me that I really AM doing anything "worse" than before, except for introducing assessments that have depressed average student grades a bit, separating out truly outstanding students from very good and average students. While, yes, while this does gesture in the direction of "separating the wheat from the chaff" (something antithetical to my "no one gets left behind" philosophy), I guess my concern is that there may be a legitimate point to it: some students truly *are* excellent, others very good, others average, etc.--and it may be important to teach in a way that reveals these distinctions, even if it's not the most popular way with students (bearing in mind, once again, that for the level of reported difficulty, my evaluations are still "pretty good", just not nearly as superlative as they were as before).

Does this make sense? I really do get everything you are saying - as your points correspond to the teaching philosophy I've always had@ I just have remaining uncertainties emerging (I think) from the sense that my "lower" evaluations may be (partially?, largely?) due more to demanding forms of assessment resulting in lower average grades than anything else (i.e. anything I might be doing "wrong").

[Marcus Arvan]

G.: Good point about lower-level v. upper-level courses. While I think I might have run afoul of this tip in one lower-level course this semester (which I taught for the first time and nevertheless got pretty good evals for), my issue here appears to be occurring in upper-division courses as well.

I used to have groups of upper-level students take courses from over and over and over again. Now that I've become more "difficult" (viz. more difficult assessments and lower average grades), there seems to be a bit less of that: I still have "repeat customers" (to use a terrible word!), just not quite as many as before, as well as somewhat more critical evals.

In one sense, the answer to all of this is easy: I could go back to what I was doing before (or at least closer to that) - if only for the sake of student enjoyment and enrollment. But that, again, is a source of ambivalence. Should one really determine one's teaching style on the basis of making as many students happy as possible, or should one be willing to hazard somewhat lower evals for the sake of being a *real* challenge where high final grades are really difficult to earn (perhaps for good pedagogical reason)?

[G]

Marcus, now I get it: the big change is that the grades got worse. The courses are only more difficult in the sense of *difficult to get an A in* as opposed to *difficult to understand the material, etc.*.

So then, put most bluntly, the question is: should you give higher grades to improve your evals, but more importantly to generate greater future enrollment in philosophy courses? (Of course, by "give higher grades" we really mean "give different kinds of assessments that tend to lead to higher grades".)

Well, I have an answer to this blunt question and I suspect I will be in the minority among professional philosophers. The answer is: yes, give higher grades. When I was a PhD student I was such a hard-ass when it came to grading my students' papers. I totally bought into the philosophers-are-so-much-smarter-than-everyone-else-so-let's-separate-the-wheat-from-the-chaff mentality. Then the professor who was the primary instructor for a course for which I was a teaching assistant basically told me to stop being such a hard-ass. Most other disciplines are giving inflated grades. Getting a C in a college course in 2018 means that you did poor work. Maybe it didn't mean that in 1960. Maybe it still doesn't meant that at Swarthmore (I have no connection with the school, but it has a certain reputation).

The fact of the matter is that students care about their grades. When I was an undergraduate, I did. If they know they are going to bust their butt and get a crummy grade, they are not going to take as many philosophy courses.

That being said, I still grade very harshly at the beginning of the semester on writing assignments because that is the only way to get students motivated to improve their writing. But, Marcus, if your old method of assessment still gets them to do the reading, and still teaches them how to write, and while doing all that it gives them slightly higher grades and gets more students to take more philosophy courses, then I say, go back to that.

One lingering worry is that grade compression will not sufficiently distinguish the excellent students from the average students. As you put it "it may be important to teach in a way that reveals these distinctions, even if it's not the most popular way with students." But I take it that here we really mean 'it may be important to *give assignments and grade* in a way that reveals these distinctions.' My only thought about this is that by giving lower grades to the average students you don't actually help the excellent students because you don't actually make those high grades appreciably more rare. To do that you would need to have control over all the grades in your university, or in all of higher education.

[Marcus Arvan]

G: Thanks for following up.

Funny, yours is broadly the same reaction *I* had when the pretenure committee, etc., suggested my grades were too high and that I needed new assessments. I thought to myself, "Why, when what I'm doing seems to be working well?" It's not like I was giving B's or A's out like candy. I was always considered a demanding instructor, even when my grades were higher. Indeed, like you, I always grade pretty brutally early on and then give students opportunities to improve as the semester goes on. Alas, I ended up doing what the committee asked (make new assessments, bring grades down, etc.), and it worked at least in the short run (viz. tenure, etc.). But yeah, I've struggled with for the reasons you give: I don't want to lose student interest, majors, etc.

So, maybe I'll transition back to what I was doing before. The funny thing is, although I didn't really like messing with a good thing--and I don't really like getting worse evaluations--I sort of "got it" pedagogically. Like I said, the new assessments that brought the grades down really do seem to be measuring things that matter (from my perspective). The reason I wrote this post - and the reason I've been struggling with this issue so much - is that I've been worrying whether the pedagogical advantages (which seem to me real) are worth the costs (in terms of unhappier students).

You seem to think the answer here is clear: go for happier students, as long as I can do so in a way that has integrity (i.e. my older ways). And I totally get it. Philosophy struggles to get students and majors as-is. Why make it harder on ourselves by being "hard-asses"? Indeed. So maybe I'll go back in the direction of how I was doing things before. I've given some thought to that today.

I guess the frustrating thing here is that, as faculty, we have inconsistent incentives. On the one hand, we have incentives pulling in one direction not to be too easy on our students (e.g. bring your grades down). On the other hand, we all know student evals matter (in terms of tenure, promotion, major numbers, and so on) and lower grades lead to unhappier students. Hence my quandary. Oh well, guess it's just the way the world is - but it's helpful to talk it out! :)

[Amanda]

think there is another option here that hasn't been mentioned. So the issue is that you are using new assessment measures which are pedagogically valuable, however, these methods result in lower grades, which in turn, results in lower evals and less repeat customers. Another problem with lower grades, is they might be unfair given grade inflation. For instance, a B or B- today reflects, I would argue, average work. Hence if a student is doing above average work and gets a B-, they have a legitimate complaint.

The choices that seem to be on the table are change evaluation methods which results in higher grades, or leave things as is with the lower grades. But If you do think the new assessment methods are valuable, why not keep them and change things in other ways to raise grades? You could do this by weighting things differently, dropping a certain number of low scores, offering extra credit, etc. And Marcus I am not sure the fact that some students can do well on your quiz means you aren't being too hard. Students get frustrated when they feel like no matter how hard they work they can't get high scores. That is why unlimited essay rewrites are great. With quizzes, it might be a good idea to have the main questions be easily and then have a few extra credit quizzes at the bottom.

[Grade]

There is a real problem with grade inflation that you need to acknowledge. When you give many student very high grades, then they get other expectations. For example, the many students who get A- and A grades will ask you for letters for graduate school, on the assumption that they are stellar students - after all they got an A- on your course. And you are THEN going to (perhaps in a letter they never see) say what you really think of their work. "It was okay". When you give high grades you are signaling to students that they are doing really well, and deserve the things we associate with academic excellence. STOP THE GRADE INFLATION.

[Amanda]

Grade, that ship has sailed. It just has.

I make a big difference between an A- and an A. I get maybe 4 or 5 A's per course. And grad school has managed to live with grade inflation by having gre scores, writing samples, etc.

[Marcus Arvan]

Grade: I am sympathetic with you, at least in principle. I've made my courses more difficult recently (viz. lower grades) in part because of the sense that it is important to better make these kinds of discriminations (between truly exemplary performance, very good performance, etc.). Wider grade distributions seem to me to do that, and grade inflation seems to me problematic for the reasons you mention.

The problem is, we face something of a prisoner's dilemma. It doesn't take much digging to see that humanities departments (and philosophy departments in particular) are increasingly being eliminated, largely it appears due to declining student/parent demand for these degrees.

Because philosophy already struggles to attract students and majors, if we lower our grades and other departments don't, we run a serious risk of losing students we very much need for our departments to remain viable. Personally, I would half like to see administrators require lower grade-distributions across colleges as a whole - as I do think grade inflation has really gotten out of hand. The problem is, few (if any) colleges will presumably be willing to take this kind of step, as *they* face a prisoner's dilemma of their own: if one university lowers their grades but others don't, students will likely flock to the universities that hands out higher grades, avoiding the ones that hand out lower grades.

I honestly don't know what to do about this. As this post expresses, I've taken on the task of lowering my grades - but the costs of doing so have been real. Hence this post!

[Grade]

I feel we really are not being true to ourselves. Our aim should be educating people. Here you have reason to believe that you are being a more effective teacher, teaching students a wider range of valuable skills and knowledge. But you are being tempted to go for the appearances - (who am I, Plato!). Resist. We are making people more shallow by pandering to their lower selves.

[Marcus Arvan]

Grade: I am totally sympathetic. It's why I wrote this post.

[Amanda]

All the evidence shows grad inflation has happened. It's done- we inflated grades and this is the spot we are in. So it is unfair to give a student a C today even if you would have given them a C 20 years ago. The letter "C" reflects something different today, so using the same standards is misleading. And while I think we need to find a solution to this mess, I don't think students should have to be the sacrificial lambs. If student "Jimmy" gets a C, and complains that makes him look like a poor student, a professor might reply that a C is for solid, middle of the road work. And then Jimmy points out that no, today a C means something worse than that. And then the professor explains he is fighting the system and the injustice of grade inflation. But poor Jimmy didn't sign up to fight this. All he knows is now he is going to lose a scholarship, or upset his family, or not get into grad school.

The point about discouraging students from taking philosophy is important too. At my old school students would avoid philosophy because of the lower grades. Given where we are with grade inflation, I think the my main purpose of my teaching is to teach...not to assess. Assessment has gone out of wack. But thank goodness teaching is different from that. So I am doing my very best job of being a teacher, and helping students learn. Assessment is just a small part of my job compared to what I think is the important part.

[Grade]

Amanda:
"All he knows is now he is going to lose a scholarship, or upset his family, or not get into grad school."
He never was going to get into grad school. But with grade inflation we just failed to fulfill a responsibility as his teacher. Why should the student work under the illusion that his work is appropriate for his goal of going to grad school? I would think that is part of our job to aid students with gaining this understanding. Perhaps many of the people who end up studying at low ranking (unranked) grad programs would have benefited from a more frank assessment.

[Morgan Thompson]

It's worth pointing out that student learning can also be a motivation to teach in a way that increases students' interest in taking further courses (from you or in philosophy). If students are learning more in one course but feeling discouraged due to their expected grades not matching their received grades, then they will opt out of further courses in philosophy and miss out on future learning opportunities. Students might learn some of the skills that they would have if they had taken more philosophy classes, but that depends on what courses they choose to take instead. So, I don't think the only motivation for Marcus's first teaching method is self- or philosophy-centric; it can also be motivated by student learning for those students who would have been otherwise motivated to take further philosophy courses and continue learning the skills taught in those courses.

It's also worth pointing out that a letter grade at the end of the semester shouldn't be the only way to communicate with students about their achievement and improvement according to the learning goals of the course. Students should also be receiving feedback about their skills and suggestions for how to improve throughout the course. Ideally, this type of feedback can signal to students whether they should, in your estimation, continue on and apply to graduate school much better than a simple letter grade.

[anon]

Grade, I think that in the context of grade inflation, "A/A-" no longer necessarily signals "I would definitely be happy to write you a letter for graduate school". And if a student misinterprets the signal, you just need to have a difficult conversation where you explain that to them.

(So how do we signal to the students with potential for gradate work that we are so willing? Write something to that effect in your comments on their final paper, mention it during a meeting, something like that.)

(I also think that aiding students understand how ready they are for graduate work is a relatively small part of the job, given how few of them are interested in it. This may vary by institution.)

[Marcus Arvan]

anon: one problem here is that students may *think* A/A- grades indicate suitability for grad school, and get pretty upset if they can get such high grades only to be told it's not good enough to pursue grad school.

[Amanda]

Grade, it's partly my fault for making a vague hypothetical, but I think whether or not he "should have" gotten into grad school depends on so many variables (how good he was, what grad school, the competition, etc.) that I can't really give a meaningful reply.

[Amanda]

Also, I wasn't referencing philosophy grad school. I am just thinking of students who care about grades because they want, say, a master's in social work and their low philosophy grade might prevent them from pursuing (or make them nervous about) a career in social work. That is just one of many possible situations. While I always make my students work for their grade, I don't see much of a benefit in being one of the few hard asses out there.

[Daniel Brunson]

An alternative viewpoint: https://www.jessestommel.com/why-i-dont-grade/

[NS]

Quizzes and really any *kind* of evaluation method doesn't seem to in themselves lead to lower grades. Am I missing something?

Regardless, any kind of evaluation seems to have ways to both give a good range of feedback while also being adaptable to grade inflation. Just dropping a certain number of low quiz scores seems to be one way. I've heard one recent intro class had three or four total exams, the first exceedingly challenging and the rest more reasonable. Then at the end of the term the instructor revealed that everyone's lowest exam score would not be included in their final grade. I've also seen a weekly paper/reflection grading scheme in which out of 20, the first 10 are for completion, and then the next 10 are graded rather harshly with only the best students (relative to the level of the course) getting all 10. So what might otherwise get a D or F gets a B or C.

These seem to be information-preserving for students to get good feedback while also adjusting for the upward grade compression prevalent everywhere else.