The title pretty much says it all.
Teachers Give Girls Better Grades on Math Tests When They Don’t Know They Are Girls
Things have changed a little. Dr. Mrs. Dr. F. studied engineering in another time and place, where teachers felt perfectly comfortable saying things like ‘I didn’t think a woman should have the highest score in the class, so I lowered it.’ That guy was an outlier even for his time but the idea of consequences for saying something like that was inconceivable. Still things have not changed enough. Not by a long shot.
In reality women do the same as men in fairly scored math tests (this study is pretty authoritative). So what’s the problem? First, as shown above fairly scored math tests don’t happen as often as they should. But we can become our own worst enemy as well. Expectations have more weight than people realize. Believe that you can do something stupidly hard and your odds go up by a non-trivial margin. But the flip side has power as well. Psychologists call the phenomenon stereotype threat: girls taking high school tests, for example, do significantly better if they ‘prime’ themselves with positive examples of women in science and mathematics beforehand. Boys already get that priming; we call it science class.
So the good news is we no longer live in the dark ages where people can make a general argument against women in the hard sciences. That means the will to fix the problem already exists, we just need to do it. Yes the problem has two faces – teachers’ prejudice on the one hand and self-destructive ‘stereotype threat’ on the other. But we know how to fix both of them. The sting survey works extremely well at finding institutional bias. Make those part of the how you evaluating a school or instructor and the problem will fix itself in a hurry. Stereotype threat is even easier. Most of the time you just need to know it exists. Marie Curie won two Nobel Prizes before most women could vote. You personally might or might not suck at math but it has absolutely nothing at all to do with being a girl (or a guy).
Another Holocene Human
What about the study that just came out of Boston and New York public schools showing Black girls were over 10 times as likely in New York to be suspended as white girls, if you can make that estimate given how many times the words “no white females were identified as being expelled” appear and so on.
Combine that with tracking and ka-boom! There’s your achievement gap, not, pace Bush, not spending all your school board taxes on third party standardized tests.
Tim F.
@Another Holocene Human: Just saw that. It fits with this episode of This American Life, which almost gave me an aneurysm.
SatanicPanic
Yeah, how does this work? Are they lowering grades based on how they arrived at the answer? This is weird.
gussie
Doesn’t this actually indicate that, barring institutional sexism, females actually have _more_ aptitude for math? That is:
A) Men and women do equally well in fairly scored tests, and
B) Women’s performance is negatively affected by stereotype threat and undermining messages.
So women are performing as well as men in specific tests despite a generalized handicap.
wormtown
Happened to me. Well, actually, gave the guys some extra points :). I am old enough that just shrugged.
trollhattan
@SatanicPanic:
since my kid is now common-coring I can say that yes, they must ‘splain how they got to their answer. Since the kid is a wordy thing as well as good in the maths, this has not been a problem for her to date. I’m guessing it’s a bigger hurdle to boys, the majority of whom still concern themselves with hitting other boys in the ‘nads and such.
Belafon
I don’t have any girls (three boys), but it kind of sounds like a way to push an attitude in favor of your child is to tell the teachers at the beginning of the year that your child is good at math/science.
Tim F.
@gussie: It occurred to me. Someone should test that formally.
Tim F.
@Belafon: You joke (?), but that works a lot better than it should.
Gene108
Scary your wife had to put up with those comments as you both seem to still be a few years shy of 40, from your posts, ie it was not all that long ago, as compAred to my mom getting her PhD 40 years ago.
SatanicPanic
@trollhattan: huh. I mean, I guess that’s good in some sense, but is there a right way to get to an answer? Just matters if you get there without cheating, right? I don’t get how you’d look at someone’s process and go “Nope, you got there the wrong way” weird
p.a.
@SatanicPanic: Yes. That occured to me also. Except maybe for high school geometry ‘proofs’ (if it’s still even taught that way), where is the wiggle room in math scoring before college level stuff?
trollhattan
@SatanicPanic:
I’m no expert but I think it’s a way to take kids out of just rote-learning the mechanics of getting to the answer (“Teach to the test!”) and digging deeper into other ways to get there and (hopefully) why it actually matters in the real world. I’m likewise hoping they have the opportunity to teach others who are having problems, since that’s the best way to really master something just learned.
prufrock
I feel fortunate that my mother was both a math teacher and a computer programmer. As a result, I never for a second bought into the idea that boys are better at math.
As a matter of fact, she is kind of the epitome of a math eccentric. At age 65, she can still do calculus better than I can, but she needs me to turn on the TV.
schrodinger's cat
@prufrock: My mother too was high school a math teacher and better than me at arithmetic. She can multiply fractions in her head, also too percentages. So I never brought into the idea that women are not good at numbers.
NonyNony
@SatanicPanic:
I think you’re looking at it a bit wrong – if the answer is right and the process is right then it’s going to get marked right.
Where you are likely to find grade differentials is in questions where the answers are wrong and partial credit is assigned for using the right process but making small mistakes.
In that kind of situation – which is mostly how math gets assessed – the bias in the grader can be as bad as the bias in someone grading an English paper. You’re going to be more likely to forgive minor mistakes from a student you perceive as “getting it” than from a student you perceive as “lost”.
(This is why exams in all fields should be graded as with students as anonymous as possible. No matter how unbiased the teacher thinks they are, unconscious bias can potentially impact everyone.)
skerry
@Tim F.: @Belafon: I did that with my 3 girls. I also told the girls that they were good at math and science. It seemed to help until they were in high school. Other pressures took over at that age. It is a hard fight. I wonder sometimes if single-sex schooling would help – but don’t know.
FWIW, I am a 50-something woman with advanced degrees in engineering earned in the 70s and 80s. When I did well in math and science in high school and college, I was told that “I think like a man.” Like that was a compliment. I was also told by college professors that “I was taking a man’s seat” in their classroom.
srv
liberal education is misogynist and racist. Who would have thought.
Oh, Victor David Hanson would.
SatanicPanic
@NonyNony: OK, that makes a bit more sense
Roger Moore
@SatanicPanic:
If it’s just arithmetic where you’re trying to give a correct answer, it would be hard to see how this could happen. But higher-level math involves things like proofs and multi-step problem solving. So teachers might take away credit for skipping steps in a proof, or give partial credit for getting partway through a problem correctly even if the final answer is wrong. Once you introduce that kind of subjectivity, there’s room for bias to creep into the process.
CurveFitter
I read the actual paper and it does not say that teachers are biased more towards boys (or girls).
The authors come up with a metric to measure bias (boy-girl gap in scoring for blind vs. non-blind). This measure for bias (for math) is actually very symmetric. In other words, there are as many math teachers biased towards girls as boys.
The bias distribution is plotted on the last page of the paper and it is centered at zero and has basically a gaussian shape.
What the authors do show is that response to bias is asymmetric. Girls with boy-biased teachers do worse in STEM than boys with girl-biased teachers.
Shantanu Saha
I guess my school is an outlier. I teach computer programming to 7th and 8th graders, and if you were to break down achievement by gender the female members of my classes perform better than the males. There are boys in class who do very well, but the group of marginally engaged (and thus low achieving) students are primarily male.
mai naem mobile
I have a female extended family.member who went to dental school in the 70s. There were only 2 women in the class. The other woman was a pretty white woman who according to family member flirted with the instructor and got by. Said family member put up with crap till the second semester when she went to the dean who assigned her a different instructor. Shes still pissed off because that one semester was the difference between graduating magna cum laude and summa cum laude. I feel similar stuff is happening with my hs niece but not to that extent. Part of me feels though that you might as well get used to it because thats.how it is in real life. Not saying its right, just reality.
Tim F.
@Shantanu Saha: I think that it’s easy to have outliers because the bias is so damn fragile. All you need is a teacher without BS expectations and students who believe in themselves.
NonyNony
@Shantanu Saha: This isn’t necessarily the school being the outlier, just an indication that your biases aren’t skewed towards thinking that boys are going to be better at computer programming because they have a Y chromosome.
Bitter Scribe
Wow.
It’s been well documented in classical music that when totally blind auditions became the norm–i.e., the auditioner was hidden behind a screen and the judges couldn’t tell his or her gender–the number of women in orchestras suddenly shot up.
I thought that was the clearest case of gender bias being exposed in a discipline that is supposed to be gender-neutral, but it looks like we have a winner and new champeen.
Punchy
Of course chicks are good at math….it takes some good algebra skills to know how to adjust that cake recipe for 4 guests and how to correctly dilute the floor cleaning solution.
Cermet
@wormtown: Also happened to me but in a strange way; first off, I am a guy; so the story is that in my E&M physics class final at the university, I tied for the second highest grade on the final exam and was just two points below the highest class grade. The person who did get the highest grade and the person I tied with were both females. Now both females got an ‘A’ for the course and I got a ‘B’; the justification was that the final was 80% of your total grade or solely your final grade (if your HmW sucked the final could be your highest grade during the course), If you had an ‘A’ on HmW these could be averaged with the final if the final was a poor grade (by the way, I had an ‘A’ average on HmW.) So he (the Prof) made just the final my total grade (and since he made the A/B cutoff just above my final grade I was screwed) but he used averaging(final + HmW) to give the girl tied with me an ‘A’. Obviously the girl that got two points above me got an ‘A’ since the Prof used the “highest grade” rule for them. The Prof openly admitted that they would never give me an ‘A’ and found a way to do it. To know Cermet is to hate Cermet … just ask my ex-wife …LOL .
FlyingToaster
@Tim F.: It does work. My parents always went in and ragged on my science and math teachers.
I did the same for my daughter, and they’re letting her and two other first-graders play with multiplication while the rest of the class masters addition.
HRA
I had 2 things that happened to me here and would not have happened in my birth country. A male student and I both scored 100 in our tests and overall class work in math. I was told we only have 1 medal to give at graduation and it will go to the male student.
In high school l chose Mechanical Drawing as an elective. I was called to the guidance counselor’s office where I was told girls are not allowed to take Mechanical Drawing.
Some things you just do not forget.
RSA
@SatanicPanic:
I’ve never taught a math class but I have covered some logic in undergrad computer science courses. Sometimes on a test I’ll ask students to prove something using a specific proof technique, and they will not receive credit if they don’t use that technique. Basically you want to ensure that students have a reasonable set of tools at their disposal, because a hammer isn’t always appropriate, and you can’t always know in advance what the best tool will be.
elmo
I’m 48 years old. When I was a senior in high school, our math teacher (who was also the head of the math department) would answer questions from girls with, “Come up here and sit on my lap and I’ll tell you, honey.”
He also refused to offer a calculus elective at all, even though a group of us students petitioned for it, because “None of you is ever going anywhere.” But that was a group of both boys and girls, so he was equal opportunity for that at least.
I started Harvard a year behind all of my peers in math, and never got over it.
David
The cited study was done in Israel. Are there any in other countries?
JCJ
@SatanicPanic:
Way back in high school I used to lose points in math class because I would do so many steps in my head and not write them down that the teachers would question how I got the answer. It wasn’t until I would do the problems in front of them that they believed that I wasn’t cheating. In college I became a complete stickler for writing down every step so this wouldn’t happen, even with relatively easy problems.
slag
@CurveFitter:
Give an example of how this finding might be operationalized, please.
PurpleGirl
@Punchy: Or how to adjust a knitting/crochet pattern for size and individual tension in the work.
NonyNony
@JCJ:
Well good – your high school teachers successfully taught you something! :)
(I tell my students that the answer is almost the least interesting part of their exams. I care about process, because I’m assessing their understanding of the problem and the methods used to solve it, not the answer itself. I already know the answer – I wrote the damn test! Getting the right answer is necessary to receive full credit on a problem, but it sure as hell isn’t sufficient.)
Cervantes
If you feel it is not inappropriate, could you please identify this other time and place a little more precisely? (Thanks.)
Lurking Canadian
@Roger Moore: They do it even with arithmetic. My son is learning it now and he has to give his “reasoning” behind arithmetic sums. Like he can’t just “know” 6+9=15. He has to say:
“Well, if I take one from the six and add it to the 9, I get 10, using the axiom of Counting To Ten. And then I add the remaining 5 to the 10, and I get 15 using the axiom of Adding Ten to Stuff”.
I guess they do it so the students can learn problem solving skills, instead of doing stuff by rote. I normally don’t like stuff done by rote, but this seems needlessly reductive to me.
Tim F.
@Cervantes: Nobody called that appropriate. I said there was no chance he would face consequences for it.
Cervantes
@Lurking Canadian:
Both ways help, I think, but the point of balance depends on where the child is. An observant teacher will get it right.
Cervantes
@Tim F.: When I said “If you feel it is not inappropriate,” I was referring to your providing further information re time and place.
The Gray Adder
One problem. It’s easy to tell the gender of the test taker by his or her handwriting. Unless they were required to use computers, well, don’t know how valid that study was.
Tim F.
@Cervantes: Oh right. It was in France and a while ago.
Emily68
My high school math story: either in the 10th or 11th grade (I had the same teacher both years) Miss Kubasik asked me if I wanted to go with a group of students she was taking to some place where there was a computer and we’d learn how to program the computer. This was in ~1966 in the San Francisco Bay Area. I thought for about 30 seconds and couldn’t think of anything I wanted to program, so I turned her down. Hmmm. Maybe I should have thought about it a little longer.
Also, I read Sonya Sotomayor’s autobiography and she said her geometry teacher in high school, a nun everyone called (behind her back) “rigor mortis” consistently marked her down because she didn’t do the proofs the way the teacher thought they should be done. In face, when Sonya aced New York’s Regents Exam (or whatever they call it) rigor mortis accused her of cheating. When Sonya produced the old tests, showing how she’d got the right answer using a different route, the teacher, to her credit, was convinced that Sonya was indeed a good student and not a cheater.
Stella B
I had it happen to me at Georgia Tech in the early 80s. There was no question about what was happening because the professor wanted us to work together and each turn in our own copy of the problem sets. I don’t think he realized what he was doing though. Ironically, two people were doing the bulk of the problems while the rest of the class coasted and copied and one of the two was me. The guys were getting better grades from my work than I was getting.
There were urinals still mounted on the walls of the women’s restrooms, too.
Cervantes
@Tim F.:
Thanks.
BruceFromOhio
I’ll admit I am biased when reading the post and the comments. Mom was a lab tech, then a comp sci master. Dad was an EE, and always said Mom was pretty good at what she did. Now the TeensFromOhio have both run the honors calculus gauntlet with the same high school teacher, an outstanding lady who pushes her students and exalts in helping them, all of them. They’ve done chem, physics with women who were awesome teachers. Role models abound. My parents were smart, my kids are smart, and their teachers were smart. This is a suitable outcome, all it takes is letting people do what they do best.
QED.
John M. Burt
@Stella B: The WACS used the urinals in their latrines as planters.
My father once observed that a woman actually could use a urinal, but that this required “supreme effort above and beyond the call of nature.”
As for the study, it’s strictly a matter of suspicions once again brutally confirmed.
Cervantes
@Tim F.: Thanks.