NYTimes: Tutors see stereotypes and gender bias in SAT, testers see none of the above

Parent Cafe
1

http://www.nytimes.com/2016/06/27/us/tutors-see-stereotypes-and-gender-bias-in-sat-testers-see-none-of-the-above.html

(Sorry to intrude in parents forum, just thought it was the article was interesting and of the type that I see here sometimes)

2

I’ve written exam items, and I’ve reviewed exam items - not for the SAT, but for other companies’ exams. Yes, there is a lot of dumb stuff that gets past the multiple layers of screening. Depending on the specific exam, it can seem (to the item writer at least) that sometimes dumb stuff is specifically added to the items.

It is good that the tutors are raising this issue. Both they, and the test producing companies, need to know if adults view the items differently from the way that the students who are being tested view them.

3

Can you provide an example?

4

The NYT article says there was 1 math & 1 verbal question the tutors thought reinforced negative stereotypes:

Go forbid if you even mention that negative stereotypes used to exist.

This is beyond ridiculous. Should historical passages about American sufferage be excluded because they mention that women were unable to vote?

5

just pseudo intellectual babble.
“The two items, one in the verbal portion and one in the math section, posed what some test-prep experts considered a textbook example of “stereotype threat.” When people are reminded during a test of a negative stereotype about their race or sex, psychologists say, it creates a kind of test anxiety that leads them to underperform.”

it is funny and sad at the same time.

6

You mean to say US kids are now stereotyped as wimps?

7

I don’t think a student would be concerned about gender stereotypes while working a math question during the SAT. . I just see these as examples.

Similarly, if the reading text is copied from a literary work, or scientific article, what is there to argue? You are to read it and answer relevant questions. Not analyze if there is gender bias.

8

No?

9

Whatever we may think of these examples, if good research shows certain kids do worse or better after reading certain things (and apparently that’s what the research says), then that is something a test should work to avoid. IMO.

10

I would put it differently: “if good research” on these questions show that bias exists on test outcomes, they should – and will be – tossed. For example, 90% of the boys get it correct, and 90% of the girls get it wrong. (of course, CB can easily assess this in multiple ways since they have millions of data points.)

11

But the issue isn’t who gets that question right or wrong, it’s how they do on the rest of the test after reading the question. The objection to the reading passage was that it was at the beginning of the test.

Some of the studies are linked here: http://reducingstereotypethreat.org/definition.html

12

if true, CB can easily test for that too. Did, all of a sudden, the female mean drop way below 500? Was this math test score was much lower than a previous test taken by the same student? Did the females all of a sudden bomb the next math sections, or did they do even better than the first?

CB has a gazillion data points, and can verify/refute the veracity of the claim.

btw: do we even know if the questions being challenged were on the ‘real’ test or were they experimental questions?

13

This was only the 2nd revised SAT. CB does not yet have a gazillion data points.

14

Manufactured outrage.