@much2learn My 11th grade dd is just as good at math as her older brother. Her math scores are higher. She detests math, even though it is her easiest subject. She loves languages and literature, especially epic poetry.
Re PG’s post #257: I wasn’t suggesting that Amazon’s practices are good, just that they exist.
A number of law school professors tend to grade on curves that hold students’ grades down, and force a certain number of C grades, or below–from what I’ve heard. It’s not just STEM profs. In fact, the New York Times also ran an expose of law schools offering scholarships with a minimum GPA requirement that turned out to be surprisingly hard to meet.
My brother-in-law, who is very successful as an attorney, once remarked to me that he had received every possible (different) passing grade in law school.
Again, I don’t particularly like it, I’m just saying the practice is more common outside STEM than has been generally acknowledged.
@Mom2aphysicsgeek “@much2learn My 11th grade dd is just as good at math as her older brother. Her math scores are higher. She detests math, even though it is her easiest subject. She loves languages and literature, especially epic poetry.”
It certainly happens, and for students with the talent and interest, it makes sense. However, in an economy where families are increasingly being stretched to just tread water, and college costs are through the ceiling, students are choosing STEM, healthcare, and business majors more frequently. No to get rich, but just to keep what they have.
Many years ago, it was expected that most of the truly talented, capable students would get a liberal art degree, but that balance has shifted. Today you rarely hear top students planning to major in liberal arts, with a backup plan to shift to science or engineering if the liberal arts program is too difficult.
"What’s lazy is not, “handing out A’s vs not handing out A’s,” but creating a forced normal distribution instead of establishing an absolute, rather than relative, standard for what passes for any given grade. That doesn’t evaluate people based on how well they actually learned the material, but on how they rank relative to their classmates.
In a proper system, if they all meet the standard for an A then they should all get A’s and if none of them do then none of them should get A’"
Exactly. What’s so hard about this? Maybe this semester only 5 students get A’s. Maybe next semester 20 students get A’s. So? That’s how life goes.
Universities are generally evaluated on how many successful people come out of their program (usually the absolute number, not percentage), which is more strongly dependent on how many talented people enter into their program than anything else. Were they to be evaluated on how good they are at nurturing talent into success, it would very likely be found that the normal distribution ranking is as harmful in a university setting as it is in the business world.
Here’s a fun experiment: click the Wikipedia link I posted earlier (#244) and go down to the list of companies that use a forced distribution. See how many of those just so happen to be known for being in terminal decline. IBM is probably the one most worth reading.
“In many STEM majors today you will find a large number of students with reading and writing scores that are higher that the students actually majoring in English. I think you will find many fewer liberal arts majors with higher math and science scores than the STEM maj”
Ok, you win! STEM students are the very smartest and very bestest of them all! It must be really, REALLY important that everyone gives you your propers.
“Ok, you win! STEM students are the very smartest and very bestest of them all! It must be really, REALLY important that everyone gives you your propers.”
You are implying that I raised the subject to feel good about my STEM degree, but there are two problems with that.
- I didn't raise the subject, you did.
- I am a liberal arts grad with a BA, not a STEM grad at all.
Reality is what is happening in the world. It does not care what we like or don’t like. Not everyone just argues that what they wish to be true, is in fact true. Some of us actually try to understand the reality of the situation. Currently, the reality is that STEM majors are more difficult. Perhaps the pendulum will swing back at some point, but it isn’t right now.
Pretty sure my STEM kid would have higher grades in a non-STEM major, just looking at her grades in her non-STEM courses. By her own estimation, she puts in less than half the time on her non-STEM classes and gets higher grades . And that GRE test I mentioned from a few weeks ago that gave her self worth a boost – a point higher on the verbal than the quant! But what that says to me is that she is pretty equally talented in both areas, and her non-STEM classes are easier. She is taking those classes across the 5C campuses, so it isn’t just humanities classes at her tech-focused school, too.
intparent and Much2learn the pc police are coming for you for sure. 
NeoDymium, I get it. You want learning outcomes that are fully quantifiable. Over the years, I’ve worked with profs from English to Engineering, or if you prefer Anthropology to Zoology, to set up rubrics to clarify as best as can be done for purposes such as allowing TAs in different sections to grade fairly, making clear what outcomes are most valuable, and standardizing curriculum across the university.
Faculty do this all the time; even lazy faculty do it. Developing outcomes and rubrics isn’t rocket science. It is a basic assumption of teaching. The problem is that complicated things, like learning, don’t measure so cleanly and simply. Surprise. The more complex the material, the greater number of graders, and the more complex the outcomes, then the more diverse the range of interpretation. So even if I am committed to giving as many A’s as I fairly can and if I have the best rubric or set of quantifiable learning outcomes possible, I still am going to be making interpretations that some students (and parents) will view as edgy.
Why grade inflation is evil: Some years ago I had a student, from a different major, who had handed in nothing. I had the hardest time discussing this with her because she didn’t respond to email and came to class late and left promptly. I finally got her to wait after class, sat her down in my office, and recommended that she drop the course. She said, “I don’t care if you give me a C.” I said, “I can’t give you a C; you haven’t handed in any work.” She never asked a chance to do the work. She simply dropped, as she was seemingly able to find professors who would pass her through.
My university is reasonably selective (55% acceptable rate) and has a relatively defined range of students, but I have never taught a class where everyone did A work (though there was one course where the students year after year were so motivated that I gave about 50% As). The only way that all As would happen is to make what constitutes an A trivial. I have never taught a class where everyone did C (or less) work, but I have come closer there. In a particular case, I tried to make some spread in the class as opposed to hold the students to the same standard I used in prior years. It wasn’t “fair,” but it was kind.
I think there is confusion about the purposes of grades. They are not just measures for the students, the faculty, and outsiders. I have taught pass/fail courses, and I thought I would like them as they are suppose to encourage exploration, but they are a waste of time for the majority of students who simply slack off. I prefer grading contracts where I promise at least a B to students who perform certain, set activities; I use them in courses that involve taking risks, doing public activities (service learning), or have a really diverse cross-section of activities. Students like the lack of stress in knowing finite acts will protect their GPA, and they still attend to the “material” because B- to A is still in play. Grades are motivators as well as measures. As motivators, like tests, they help students learn, apply, and retain.
@mamalion
Well let me put this very simply: What quality about this general variation in student performance justifies the imposition of a forced normal distribution for grades?
You don’t really seem to talk about that point very much, which makes me wonder what exactly it is you are arguing for or against. I agree that it’s seldom practical to expect that everyone will work to an A level but if it somehow happens that everyone does, why would they not all get A’s?
If all of the students do the same, one can’t force a distribution, like, duh. BUT they don’t. I don’t know what Platoric ideal you are imaging, but it doesn’t exist. People vary in their performances, and in some situations, like tests, the variation is noted.
The variation can be noted without being forced.
In an ideal world where that happened, it might be a matter of some people being able to think more quickly than others. In the real world, as I have said, many students copy their assignments from each other or from the internet. There are always students (not a small percentage) who will do their level best to thwart your efforts to teach them something. You can lead a horse to water, as they say.
ETA: When I give exams I often question whether they are reasonable/fair. But I will invariably have students who score into the 90’s, which is an indication to me that the exam is not beyond the ability of those who put in the effort and were prepared for the course to begin with.
Sure. I don’t think anyone is saying grade inflation is a good thing.
That summarizes my view very concisely.
I’ve seen many forms of distributions in performance of students on exams. Normal, bimodal, uniform, fatter-tailed bell curves, just plain random, etc. Learning/performance isn’t always normally distributed and its mean/variance isn’t generally fixed on a year to year basis (though sometimes it can be).
In a large STEM class, the performance of the students generally is a normal distribution. It’s not being “forced” in some artificial way. That is how it comes out.
Then why curve?
- For large classes, it usually is but there are reasons to believe it isn’t always the case. Realistically I think I’d say 60% of the classes I’ve ever taken/taught had student understanding follow some form of normal distribution.
- You are also fixing the mean and the variance in student understanding of the material. This is a horrible assumption.
- Being better or worse than your classmates doesn’t mean anything. You could all be terrible or you could all be fantastic.
@OhMom2 - the curve is used to assign letter grades to test scores. Unless the natural curve of the test matches the “standard” 65-90 D-A bell curve, the scores are normalized.