During my college years, I have seen a lot of grade curves. Most of the time, they are classic bell curves.
So, I would assume that profs knew what they were doing with their exams. Yes, the median usually centered around 55%-65% (at my large public u’s engineering program, that is C/B) - I think that is pretty good to allow differentiations between students.
Also, I do not believe that your understanding of a subject matter is binary - either you get it or you do not get it.
It is very “analog”. I am pretty sure you guys have sat through some technical interviews. I am sure that most of you can answer those questions given sufficient time - but the employers will invariably limit your time to answer these questions. For example, for you CS geeks, Google asks the question about coming up with an algorithm to detect whether a given link list data structure has a cycle. I am sure you CS guys can do that at home given enough time. But under the time pressure of 5-10 mins, that is quite different.
“For example, for you CS geeks, Google asks the question about coming up with an algorithm to detect whether a given link list data structure has a cycle. I am sure you CS guys can do that at home given enough time. But under the time pressure of 5-10 mins, that is quite different”
That’s a question Google asks freshman 
That’s akin to a pop quiz, and it’s not really something that really measures any reasonable idea of understanding of the subject (does it really make you a better software programmer if you answer a quiz question under artificial conditions?). Truth is that Google probably doesn’t even bring anyone in for an interview who wouldn’t really belong there for a job, and they differentiate arbitrarily. Google is actually well known for false negatives and for taking people after some manager gave them the correct interview questions that happened to resonate well with them.
Here is a decent article related to it: http://www.unlimitednovelty.com/2011/12/can-you-solve-this-problem-for-me-on.html
How does this relate to schools? Google, like top schools, has an overabundance of talent inside and want to differentiate between them. This is difficult to do properly because at some point there are more talented people than there is good means of differentiating them. Sometimes arbitrary criteria are used to do this, and that simply isn’t a fair or useful system. Arbitrary criteria are more political than meritocratic by nature and no one is immune to this.
University of Maryland grade distributions can be found at http://www.ourumd.com/grades/ .
It looks like D and F grades are few.
Those are final grades, we are talking about individual test grades. No way to tell what the test grades were or what the curves were that ended up at those grades.
@gettingschooled, part of the problem with this is that same experience that keeps your son from being overconfident is often interpreted completely differently by young women who get those grades, and discourages them so much that they drop out of STEM altogether. They typically aren’t overconfident to start with, and take the fact that they are getting low percentile scores to mean that they must not be competent enough to continue in a subject area.
@intparent
“that same experience that keeps your son from being overconfident is often interpreted completely differently by young women who get those grades, and discourages them so much that they drop out of STEM altogether. They typically aren’t overconfident to start with, and take the fact that they are getting low percentile scores to mean that they must not be competent enough to continue in a subject area.”
In parallel with that - boys can sometimes view as a “challenge” or “not a big deal” some grades that really indicate that they ought to drop the course or switch levels, etc.
In our professional development, one of the things we do is talk about how to work against those stereotypical fears and what safeguards to put in place, what kinds of specific conversations to have, etc. If I do my job right in HS, those girls should theoretically be immunized when they get to college, to the extent possible.
Mmm… I have no idea if that is happening in your classroom or not. This is pretty deeply ingrained in women, to blame themselves and assume they aren’t good at something. The bias in teachers honestly runs pretty deep, too.
In my D’s HS, the female Calc teacher would assure you she had no bias in favor of the male students and was very supportive of female students. But when they decided to add multivariable Calc to the curriculum and quietly invited some students to do pre-Calc over the summer to accelerate in preparation for this, guess what? She invited 7 boys and no girls to do this. D found out in the fall when the 7 boys were jumped to Calc (and was she pissed!). My D is a Physics major now at a top STEM school – but certainly does not see that Calc teacher as having helped her build confidence. She also had perfect math SAT and Math II SAT scores (not a huge accomplishment by CC standards, but only a couple of those boys managed it).
But getting back to the original discussion, I unfortunately think it takes more than one HS teacher to immunize young women against a grading system that systemically grinds down their self worth. I don’t think this conversation is taking place at nearly enough colleges, or if it is, the profs are making no or only token efforts to address it. Especially the older male profs who don’t even really see it as a problem. Of course there are exceptions, but not enough if them.
@intparent - I can’t reassure you adequately anonymously on the internet, and for all I know I have unconscious biases like everyone else. However, it is an active part of my teaching craft, to specifically fight against stereotype threat and other insidious sexist stuff in STEM.
@intparent I agree with what you are saying. That was what I was saying in my last sentence. I would have dropped a class where I got a 48 on a test in high school and never looked back. I am female. And I worry about other kids in his class who don’t have the self confidence to deal with this. He ended the semester with an A despite a few of these doozies in the grade book. Just by the hair of his chin. This semester he is carrying a much higher grade but it is early. Part of it is learning curve on her grading style and part of it is they have gotten to a more modern era in history that interests him- but now I think there may be another factor at play which is preparing them for the AP test and college where a relatively low percentage correct is somewhat normal.
Not sure it matters but the class my son is having this experience in is a history class. I don’t see this approach being taken with STEM in his school. Or if it is, the curve is being applied in a way that is hidden to me. His school has a large number and percentage of girls who do well in STEM and at least leave here intending to major in those areas. I will give them credit for that along with parents who must be encouraging girls that they are quite capable. I don’t have a lot of info on whether they stick with it in college.
I used to work in at a junior college and I was at a department meeting once and the department chair said she was going to start checking all the teachers grades to make sure no one was giving out too many A’s. Another teacher was bragging about how her class starts with 35 students and only 10 finish, and she was trying to get it lower. I started thinking, "Why not just give the test in another language? That way no one will pass. How can someone call themselves a teacher when everyone fails?
In all my classes, a curve showing all scores was posted after each significant exam. Kids could see how they scored in relation to the entire class. We were also told what the median would be set to, usually C+/B- or B-, and sometimes score cuts were indicated for each grade. So I think kids had a pretty good idea where they stood and I don’t think anyone was stressing over the exact numerical scores. That didn’t matter, just where they sat on the curve. I don’t think it’s ok to let kids stress about low numerical grades with no context of what they mean.
On another note, I was on a tour once at a top 20 school and the tour guide told us that the number one major of students was Psychology. Then when students I know started attending I realized why, they get flunked out of their major they entered with and switch to the easier major.
I think another thing that surprises students when they get the college is the attitude of the professors. As a high school teacher I am always reminded of how many failing (or D) students I have and what am I going to do to raise their grades. Society also holds k-12 teachers responsible for dropout rates. College professors pride themselves on how many kids they can get to drop. It is a big culture change once in college.
I think society looks at graduation rates and freshmen retention rates for colleges too. Key factors in many rankings. Why that doesn’t translate to individual classes and professors, I am not sure. Maybe the thought is you can graduate with a degree but not necessarily the degree you thought you would get.
I think it’s really wrong to say that college professors pride themselves on how many students they get to drop out. They are disappointed when students do poorly, unless maybe it’s abundantly clear that the student isn’t trying (eg. doesn’t hand in assignments, doesn’t attend lecture). They would love to have a class composed entirely of high-achieving students.
The bad ones will then proceed to make the course more difficult so some of the high-achievers will now be failures.
I have had professors who were proud of the fact that the program they helped to create had “the lowest average GPA and the lowest professor ratings of any program in the school.” Something like that is stupid arrogance and the kind of thing that gets a lot of professors a bad rap.
I read a book recently where one of the situations described was giving tests in a college course. The author mentions that students complained a lot about the test being too hard when they got low percentage scores, even though the low percentage scores corresponded to good letter grades. After changing the test so that number of points was a number greater than 100 and not as easy to do mental math on, the scores were higher, the percentage scores were similar, but there was less complaining about getting (for example) 90 points out of 137 compared to getting 66 out of 100.
The different grading scale changed the illusion of “doing well” versus “doing poorly” on the test. On a rational basis, it made no real difference. But people are not necessarily rational.
^^ I can understand how that works with your Average Joe, but if those are STEM students that’s pretty scary!
Of course, there are other tricks that can be done with test score scaling.
For example, if the instructor writes a difficult test where the typical distribution of student scores has a median of around 50%, then the instructor can rescale the scoring to a 30 point scale, but then add 70 points of very easy questions that passing students should take only 5 minutes to complete. Then every passing student will get at least 70% correct, just like in high school, even though the actual contribution to the course grade is really in the 30 points of more difficult questions. I.e. the easy 70 points that take 5 minutes is really just a placeholder to make students feel good about their scores on the test, even though it makes no contribution to their course grades.