http://www.cnn.com/2003/EDUCATION/05/23/florida.test.protest.ap/index.html <i>Statewide, more than 43,000 third graders and 14,000 12th graders didn't pass, and protesters have said that an unfair percentage of those are minorities. </i> Perhaps the minorities should study more rather than blame the test!
Without seeing a copy of the test, I can't comment on the claim that it's more difficult for minorities, but I can definitely say I am against standardized testing. Many standardized tests of the past used cultural references that were completely foreign to certain minorities. Most of these students were very intelligent, but certain terms and concepts in the tests baffled them. Whether that's the case in Florida or not, Standardized testing is bad for education. It encourages teachers to focus strictly on what will be on the test and nothing else. It's also been tracked by educational psychologists, and found to be an inacurate method of detemining a student's metacognitive ability.
At least with the SAT, the whole "cultural bias" charge has some weight, given how ETS chooses which questions to put into the SAT. Each prospective question will be given to a test group and ETS will see how the test subjects do with that question. Obviously, it's to verify that they're putting questions into the SAT which are of a comparable difficulty level to the ones that are already in there. Sounds pretty good, right? Well, the thing is that they also break the results down by race and gender and so on, and questions who have results outside "the norm" are thrown out. On some questions, for example, a high percentage of black test subjects got the questions right, a higher percentage than white test subjects. And there's no apparent reason WHY the black test subjects would be have so much more success with these particular questions, since they have nothing to do with race or culture or anything like that. It's a mystery, frankly, why black test subjects do so much better at these particular questions. But since the results for these questions fall outside "the norm," the questions are thrown out. So you've got a test where each ethnic group has varying degrees of success, and questions for that test are chosen in large part because they fit those profiles of success. It's a self-perpetuating system. I'm not sure if that's the case with the Florida test, but it's not exactly a baseless charge to say that some standardized tests are racially and culturally biased.
How the hell could a test possibly be unfair towards minorities. Do they ask black students questions about hockey and Michael Bolton?
Reread El Jefe's post. He's summarizing Jay Rosner's study for the Princeton Review Foundation. Rosner's study of every question considered and accepted on the SAT in the last 10 years found that ETS did not include a single question in which Blacks scored higher than Whites in pre-testing even though more than 30 existed, including some math questions. He concluded that the SAT has systematically disadvantaged minority testers through the question evaluation process. It's obviously not the Florida exam, but it's one fairly convincing example of how a pretty vital test is unfair to minorities, not because of open bigotry, but because of the test-makers' formulation of acceptable test questions. Since ETS is the leading company in designing standardized tests, it would not be surprising if the companies designing the state tests have copied their test formation methods.
This test just shows the disparity that still exists in our education system. I think the test should be given so all the schools are held up the the SAME standard. The people protesting should be protesting their school system not the test.
I understand what you're trying to say, but I'm still baffled as to what race/culture rather than quality of one's education has to do with this. What are these 'cultural references' people are talking about? I took an SAT test some 5 years ago. Normally, in Canada, there is no such a thing as an entrance exam for a post-secondary institution, but I wanted to go to school in the US, so I had to take it. Despite having never lived in the US, I didn't see a single question on there that I thought was unfairly biases towards people who were born and raised in the US. I saw no questions on Nascar, Carl's Jr., NFL or anything else we don't have up here. So, I'm wondering what kinds of questions could possibly be biased against blacks or Hispanics.
The thing about Rosner's study is that the questions that Blacks scored higher on were no different in appearance from other SAT questions. For some reason not readily apparant, Blacks performed better on certain questions. However, since the questions did not mirror the performance of high testers vs. low testers (which is correlated by race), it was omitted from the test. Similarly, Whites did disproportionately better on some questions (up to a 20% differential on some questions included on the test), but these remained in because they mirrored the expected racial results. I posted one of the questions about a month ago. I'll see if I can find it.
Here it is: The actor's bearing on stage seemed ________ ; her movements were natural and her technique ___________ . (A) unremitting ... blase (B) fluid ... tentative (C) unstudied ... uncontrived (D) eclectic ... uniform (E) grandiose ... controlled He writes quote: ------------------------------------------------------------------------ This looks like a typical SAT verbal question. Yet this question differs from others in one important respect: according to ETS, '8 percent more African-American than whites answered this question correctly.' I call this a 'black preference question.' I don't know why blacks did better here, but nearly all SAT questions capture something about race that can't be determined until pretesting. Because it favored blacks, who score lower on the test overall, this 'actor's bearing' question, which was pretested by ETS in 1998, did not favor high scorers and therefore was rejected for use on the SAT. I have identified several other examples, including a black preference SAT math question, that were rejected My considered hypothesis is that every question chosen to appear on every SAT in the past 10 years has favored whites over blacks. The same pattern holds true for the LSAT ... since they are developed similarly. ------------------------------------------------------------------------ BTW, the answer is C.
If there was nothing different in the appearance of these questions, isn't it very possible that the perofrmance on these questions could have been coincidental? I wouldn't buy this reasoning that some apparently identical questions are easier to handle for some ethnic groups than others until somebody shows specific questions and explains what specifically about them are sided towards "white people" and "black people."
For the purpose of his argument, "why" there is a racial difference is much less important than the tester-maker's reaction to questions that reveal racial differentials. "Black preference questions," as he terms them, are systematically excluded because they fail to mirror the demographics of the high-scorer/low-scorer continuum that the test-makers aim for. "White preference questions" mirror that continuum and are not excluded on this basis. The "why" question is fascinating and it would be great if some social psychologist took his study and tried to answer that question.
Just because the reason for it isn't clear to us, doesn't mean that the disparate performance of different racial/ethnic groups on particular questions is arbitrary or coincidental. Sometimes no reason is apparent. But there's a difference between not being apparent and not existing at all. I was a private S.A.T. and standardized test tutor, and I agree that these tests are a very poor measure of a student's skill. One of the many reasons minority students tend to score lower on the S.A.T. is that a large portion of the test is vocabulary. Vocabulary is learned mostly through environment. A student whose parents don't have a college education is immediately at a disadvantage on a vocabulary test. That's just one example. And culturally specific knowledge IS a concern. Farming terms, for example, are bizarrely common on the S.A.T., constituting a bias against urban students ("urban" not [necessarily] intended as a euphemism there). But the real problem with standardized testing is "teaching to the test." High school teachers become S.A.T. tutors, teaching kids only what they need to know to pass the test. It would be impossible to construct an exam which accurately tested a student's knowledge and ability, and was not biased in any way. Take five minutes to look at the reading comprehension section of any SAT or Iowa Test, and you'll see how ridiculous they are. They reward points to students with good test-taking skills, not necessarily good reading skills. The "correct" answer choice is rarely a good answer to the question posed. People like testing because it attempts to quantify what students have learned. It provides a simple, understandable schema in which to understand how well a student is doing. Unfortunately, depending on tests won't improve the education system, whose problems are more complex and difficult than most people are willing to face.
No, you racist bastard. It's all the fault of The Man®, the dirty swarmy evil devil who is responsible for all of the oppressed....
Re: Re: Opposed to High School Grad test in Florida So, actually suggesting that the real problem may be the minority children gets one labeled as a racist. Got to love free thinking in a supposedly free society.
Re: Re: Re: Opposed to High School Grad test in Florida Are you trying to argue with that logic? Bring it up with this guy:
Coincidental? Probably not. More likely is a misrepresentation or misinterpretation of the statistics. I've been searching for Rosner's actual research as opposed to his conclusions, and have thus far been unable to find it. I am interested in finding out whether he found a statistically significant difference in each ethnic groups answers when he compared only those test takers that scored similarly on the test as a whole. I'd be willing to bet that there is no bias when high scorers of one race are compared with high scorers of any other race.
Re: Re: Re: Opposed to High School Grad test in Florida No, ignoring years of psychological testing and studies that show the problem to be certain trends in test questioning in order to claim that the real problem is the minority children gets one labeled as a racist. There has been boatloads of work done on this concept. While said work does point to conclusions that these tests are slanted againts minorities, it also concludes that standardized testing in it's entirety is a poor measuring stick for ability and intelligence. That's what infuriates educators. They've known this for years and continue to lobby against many of the standardized tests (note: not all), but if anything, the state and federal departments of education continue to INCREASE the number of standardized tests in the schools. Not only is it a bad judge of ability, but it slants the curriculum in a negative way, forcing educators to narrow their lessons to a test specific line.
I'd love to see the results broken down for: -Poor white kids in rural districts, especially in the Panhandle. -Rich, Cuban kids in Miami. If the test is so culturally biased; the poor white kids should do better than the rich Cuban kids, right? Maybe they are. I'd just love to see a breakdown. If the SAT is so culturally biased, how come Asian kids whose parents were in Re-Education camps or making Barbie Dolls in Laogai camps are going Australia-American Samoa over white kids when the SAT scores come out?
Actually, economic background affects test performance, too. Race and ethnicity do not necessarily trump class in this case. But that's still a problem. If wealthy kids are outscoring poorer kids, it's still a problem.
Yeah...but this has absolutely NOTHING to do with the test, and everything to do with the quality of education received by the kid. Again, I'd be willing to bet that nobody can produce conclusive data showing questions with racial bias among candidates with similar test scores.