Fifty-five sixth graders will be left behind, or at least part of their scores will be, thanks to three teachers that decided to prep them for the science portion of the mandatory ISTEP+ standards test by showing them the questions and answers ahead of time.
"In their minds, they thought they were doing what was fair for the students..." said the school superintendent, Dr. Tom Warmke, indicating that the guilty teachers were ot-nay oo-tay ight-bray, if you know what I mean.
At least one of the teachers involved is probably going to get fired, which should give them plenty of free time to hang out down at the statehouse and protest education reform.
Subscribe to:
Post Comments (Atom)
21 comments:
We have seen similar things here in Florida, and that is why I think tying teacher pay and raises to test scores is not the way to do it. All this does is result in teachers focusing on the test instead of the student.
Please don't be naive enough to think that any of these students will be left behind or that one of these teachers will be fired.
You think IMPD takes care of their own, you ain't seen nothin' sister.
Joseph,
"...one of these teachers will be fired."
Word is that one is getting thrown to the wolves to lighten the sleigh.
They've caught entire schools doing that down here in Texas. Note to administrators - when your school goes from 40% passing all parts of the state exam to 95% passing all parts, in one year, without any personnel changes, someone just might get a wee bit suspicious.
LittleRed1
Harumph......
In most public schools, mine included, some students will be given the test in advance and allowed to practice taking it. These are 'special needs' students (45% of student population).
What these teachers did was 'wrong' only in context of the state mandated test.
Okay... it's wrong period.... but today's society can't stand the idea of wittle Bobby flunking school, so the system is designed to shove his lazy butt along till he falls out the other end, all edumacated with a piece of paper to prove.
Me? I write many of my tests in such a way that my students can go right ahead and read them ahead of time. It won't help. They have to THINK to get a good grade.
It's the problem with teaching to a test rather than teaching the material (which most teachers are incapable of).
"Me? I write many of my tests in such a way that my students can go right ahead and read them ahead of time. It won't help. They have to THINK to get a good grade."
And that is what separates teachers from morons. You, Sir, are a teacher.
I remember in college, the non-engineering majors being scandalized that most of the engineering exams allowed cheat sheets, or were even open book, and that half 1/3 to 1/2 of the class failed out every year anyway.
on a sceince exam, knowing the the questions shouldn't make much of a difference, unless they have have science confused with trivia.
Owen,
"on a sceince exam, knowing the the questions shouldn't make much of a difference, unless they have have science confused with trivia."
Sixth grade, public school.
"Sixth grade, public school."
ZINGGGGGG!!!
You know you have a dumb kid when the teacher gives them the answers to standards test.
Gerry
I'm actually a fan of standardized testing. It's a basic skills test, so there should be no "teaching to the test" needed.
My oldest daughter is now in 3rd grade and taking the TAKS test for the first time. They have spent at least one day a week practicing for this test the whole year, and Bobbie is one of two students in her class not in need of extra tutoring.
This, combined with the fact that my first-grader is bored out of her mind while the teacher re-teaches everything she learned in kindergarten, has led me to believe kids are simply treading water and being passed along until time for the test, when there's a collective HOLY SHIT from the instructors and the cramming starts.
We had must-pass standardized testing when I was in school too (the TAAS test back then), but things were not handled like this. We went through the school year normally and just took the test when it was time. I don't know why things have changed.
Julius and Ethel were teachers? They did the same thing, sharing crib-notes with the Soviets. Teachers are really starting to piss me off, and my dad was a teacher.
I mind my first-grade teacher, who was white-headed in 1956. That means she was prolly a 19th-century gal.
I recall her praising one of the girls in the class for not whining and not crying when she was injured, and holding her up to the rest of us as a good example.
The Bene Gesserit had nothing in the way of an advantage on that teacher.
Take off and nuke the education system from orbit. It's the only way to be sure.
I'm reminded of a question from an actual British physics exam, which directed students to write a short story about what might happen if an ambulance had radio troubles. No mention of how radios work or what kind of troubles they might have; it was purely a "let's all pretend!" problem.
The guy who pointed it out, a physics teacher, was furious that they were killing his profession.
Carteach0: The tests were not for the students.
The feds are testing the schools. This school cheated.
Sabra, not to date myself, but I took the CATS when I was in TX public school. The best part was those of us who finished early were allowed to read at our desks.
LittleRed1
You can justify almost anything if you say that, "in your mind," you were doing the right thing . . . . and also start from the assumption that the world is entirely subjective.
The problem is not how to justify your actions to yourself, but how to get all those bothersome other people who live outside your mind to accept your rationalizations instead of firing you and giving your job to a gun nut from Illinois.
(For example.)
The problem is that you have failed lawyers enacting standardized testing for educatation majors.
And no one in the process with a QA engineering background -- becuase that's all these tests really are, QA tests.
As with the SAT or ASVAB, if you can "teach the test", then your test sucks. You can teach how to take the test in terms of tactics, but not "teach the test".
As with the SAT or ASVAB, your score may well not look anything like a academic grade.
Students aren't nuts and bolts (or jet turbine blades, or F22 airframes, or rocket nozzles -- all of which I have done QA process testing on. . . actually testing the testing of the production parts, so to speak), but we can apply additional factors to account for the human part of the equation, just as we do when trying to keep tester bias out of the picture on teh casting floor.
Real simple --
Create a question pool so large it is friggin' impossible for anyone without an eidetic memory to memorize it, or even a decent chunk of it.
Proctored testing not under the supervision of those who are directly answerable for the performance of the little golliwogs in the room at the time (you can proctor some other teacher's testing, but none of your own students).
Random assignment of sealed testing materials that cannot be determined which test version each is until the seal is broken.
NO answer sheets available at test site -- tests are sent in and graded offsite, preferrably with a blinding process in place.
Run the damned test, follow that cohort for several years, and compare test results (on an individual, but anonymous, basis) to see how raw score test result "X" correlates to actual performance criteria "Y". Scale pass/fail criteria based on the resulting raw scores. If 90% of students who get 40% or more of the questions end up as high-performing students three years later, then 40% is a pretty damned good score. Arbitrarily setting a "Pass" score so it looks "right" to someone used to the ABCDF/90/80/70/60 "ten point scale" is stupid.
Implement the SAME DAMNED TEST each year, instead of changing it every year. (If you change it, it has to re-validated. Which means a certain percentage of randomly selected golliwogs might well be taking TWO standardized tests in a particular year, so the answers on the proposed test can be validated, whilst only using the already validated version in stats.)
I love the way "No Child Left Behind" works. Poorly constructed, never truly validated, tests; badly administered; used without any understanding; while the minimum passing scores get arbitrarily higher every year; AND the minimum pass rate for the collective population being analyzed gets higher every year (they seriously have it set at "100% pass rate" by 2014).
More proof people are simply monkeys with car keys.
Geodkyt - your analysis is impressive, but the more basic problem is that standardized subject testing is generally a bad idea in the first place, usually because it's not testing the main thing we want to know in the first place - beyond taking a test, can the person actually do useful shit with what they have learned?
professional certification testing often does employ complicated psychometric techniques like you suggest (drawn from psychological testing techniques instead of QA work), yet I consistently find unless the tests require the most "hands-on" skill demonstration possible for the field, they are shitty and useless guides to practical competence.
I can count on one hand the number of IT certifications that are worthwhile in terms of actually demonstrating a basic level of skill to an employer. They all require a hands-on demonstration portion.
sadly, the biggest obstacle to this is the sheer cost. Having a state require their English students to produce a novella, well-written technical report, or informative paper to progress to the next grade level and giving it even basic levels of reasonably impartial evaluation would make most budget hawks weep at the money expended to do this for relatively inexpensive disciplines like English.
Post a Comment