The list includes “hunting” but not “fishing,” so that’s cool. I wonder how they’d feel about a question involving different cuts of meat. In any case, I’m happy to see that “Bayes” is not on the banned list.
To ruin fun, I’ll say that I think it’s important that standardized tests be as culturally-neutral as possible. I remember teaching probability and learning that several of the (college) students didn’t know about card decks (much less the poker terminology used by the book).
Focusing on “upsetting” words, as the NYCDoE does, is a little silly to me. But banning words that students might be unfamiliar with seems like the right thing to do (within reason), at least for standardized tests. In the classroom, unfamiliar concepts can be explained, so not an issue there.
That’s awesome! We should have a contest for the most insensitive test question. My first entry is:
How many birthdays does a poor divorcee need to have before people dance on their grave like a cancer ridden dinosaur?
This is ridiculous. Is it for real?
Well, at least Motherfucker is still OK.
Actually, ETS has similar guidelines which govern its test construction. There are formal written guidelines, a mandatory training procedure, and all items are reviewed for “sensitive language”. I wasn’t ever an item writer and I managed to duck the mandatory training a couple of times, so I’m not 100% certain how much of the rules are written, and how much are an effect of the review procedure. But these are an important part of ETS’s policy on Fairness, and I believe information is available on the ets.org web site. I’m also sure that most of the other major testing companies have similar policies.
The reason they did it is related to validity, both true validity and face validity. The face validity case is obvious. Avoid sports items because it is widely believed that boys like sports more than girls. Even if it is not true, it avoids complaints.
There is also a true validity argument. If the student looks at content that is personally upsetting during the exam, the emotional response is unwanted measurement error. There is also a background knowledge effect, especially in Reading. You don’t want to use something that is common knowledge to one group but must be inferred from the passage for others. Thus, by avoiding potentially sensitive topics you generally get better measurement. Of course there are some exceptions: For example, it would be a gross distortion to never talk about slavery on a US History test. But unless there is a pressing need they should be avoided.
I agree that the result is bland tests. But the blandness cuts down on construct irrelevant variance. I’m sure that there are few silly examples in any given list, but in general policies like this make good psychometric sense.
When I was a student in the New York City public school system, back in the Pleistocene era, the kids used to carve “Bored of Education” on the rulers and other classroom objects. Now I see they might have been on to something. These banned words are so ridiculous that it would seem that this is something out of The Onion.
The motivation for the banning seems to be the desire to shield students from any and all unpleasantness no matter how trivial. How will these kids learn to cope with life outside of school when they grow up? The real world is full of bullies (especially in government) and other nasty people who will not spare your feelings. There was a lot wrong with the old English Public (really private) school system, but it did toughen the kids up. Places like Harrow and Sandhurst produced a resourceful generation in the 19th Century. So resourceful the British ran the whole Raj with fewer people than Stanford University has administrators. I doubt we would have had the Churchill we know without Harrow and Sandhurst.
The buffoons operating out of Livingston street (if the Board of Ed is still there today) have a lot to learn about life.
To ruin fun, I’ll say that I think it’s important that standardized tests be as culturally-neutral as possible. I remember teaching probability and learning that several of the (college) students didn’t know about card decks (much less the poker terminology used by the book).
Focusing on “upsetting” words, as the NYCDoE does, is a little silly to me. But banning words that students might be unfamiliar with seems like the right thing to do (within reason), at least for standardized tests. In the classroom, unfamiliar concepts can be explained, so not an issue there.
That’s awesome! We should have a contest for the most insensitive test question. My first entry is:
How many birthdays does a poor divorcee need to have before people dance on their grave like a cancer ridden dinosaur?
This is ridiculous. Is it for real?
Well, at least Motherfucker is still OK.
Actually, ETS has similar guidelines which govern its test construction. There are formal written guidelines, a mandatory training procedure, and all items are reviewed for “sensitive language”. I wasn’t ever an item writer and I managed to duck the mandatory training a couple of times, so I’m not 100% certain how much of the rules are written, and how much are an effect of the review procedure. But these are an important part of ETS’s policy on Fairness, and I believe information is available on the ets.org web site. I’m also sure that most of the other major testing companies have similar policies.
The reason they did it is related to validity, both true validity and face validity. The face validity case is obvious. Avoid sports items because it is widely believed that boys like sports more than girls. Even if it is not true, it avoids complaints.
There is also a true validity argument. If the student looks at content that is personally upsetting during the exam, the emotional response is unwanted measurement error. There is also a background knowledge effect, especially in Reading. You don’t want to use something that is common knowledge to one group but must be inferred from the passage for others. Thus, by avoiding potentially sensitive topics you generally get better measurement. Of course there are some exceptions: For example, it would be a gross distortion to never talk about slavery on a US History test. But unless there is a pressing need they should be avoided.
I agree that the result is bland tests. But the blandness cuts down on construct irrelevant variance. I’m sure that there are few silly examples in any given list, but in general policies like this make good psychometric sense.
When I was a student in the New York City public school system, back in the Pleistocene era, the kids used to carve “Bored of Education” on the rulers and other classroom objects. Now I see they might have been on to something. These banned words are so ridiculous that it would seem that this is something out of The Onion.
The motivation for the banning seems to be the desire to shield students from any and all unpleasantness no matter how trivial. How will these kids learn to cope with life outside of school when they grow up? The real world is full of bullies (especially in government) and other nasty people who will not spare your feelings. There was a lot wrong with the old English Public (really private) school system, but it did toughen the kids up. Places like Harrow and Sandhurst produced a resourceful generation in the 19th Century. So resourceful the British ran the whole Raj with fewer people than Stanford University has administrators. I doubt we would have had the Churchill we know without Harrow and Sandhurst.
The buffoons operating out of Livingston street (if the Board of Ed is still there today) have a lot to learn about life.
Pingback: Creating consistent test results in schools by banning words… | Ganesha's Scarf