. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Nap Time!!!

Thursday, September 28, 2006
The surveyors are coming!!!

Some folks have been talking about this study which found that some seniors answered fewer questions right than some freshmen on some test about pointless factual details of American history.

Too bad The Daily Cal is statistically illiterate, or they might have pointed out the contradiction between this:
The study, which was conducted by researchers from the University of Connecticut, was based on questionnaires that were distributed to random samples of students at 50 college campuses across the country, Ratliff said.
and this:
[Danise Sugita], who was given the survey on campus last year and decided not to participate, said the questions were about specific details in American history that might not be relevant for students who are not specifically history majors.
But that's probably asking too much.
The study's implication that faculty are doing an inadequate job and that there should be a systematic focus on American civics threatens the autonomy of university faculty, [History department chair David Hollinger] said.

"I think it's part of a national move to put universities under pressure," he said.
Oh, noes! People are doing surveys! Academia is doomed!

posted by Beetle Aurora Drake 9/28/2006 10:32:00 AM #
Comments (8)
. . .
i don't really see the contradiction. a randomly selected student could easily have chosen not to participate. personally i think the headline is more worth making fun of. students lost their civics knowledge? when, two weeks ago? where can they find it?
if they choose whether to participate, they are self-selected.
makes sense, though the initial process of narrowing down the students could still have been random. i'm not a stats major, however.
Based on the knowledge that hasn't leaked from my brain since Stat 20, a refusal to decline is only significant if the people who are declining will score outside the random sample. If I were to take a guess, I would say that those who would score poorly would be more likely to decline to participate, which would actually push reported scores UP.
That's an enormous stretch. Supergeniuses, such as myself, would not lower themselves to answering questions from one of the inferior lifeforms that surveyors are.

While you're correct that if refusals occurred randomly, it wouldn't matter, but it's the burden of a person doing a study to show that it's the case. One doesn't do statistical research by saying "assume my sample is random." They have to show it.
Even at our lowest test level (54.8%) Berkeley students scored higher then 7 of the top 10 schools in the study.
The point of the article is that we're not improving. I'm wondering along with Beetle when civics became a responsibility of college rather than a, I dunno, responsibility of high school?
I dun mind seeing civics as a responsibility of college. We're pretty stupid in high school and probably can't understand such things as well as we need to. What I do mind is the definition of "civics" that includes pointless historical details. I would expect such things to fade in time.
Post a Comment

. . .