next »

[scienceprof]

http://www.nytimes.com/2011/12/19/education/new-york-city-student-testing-over-the-past-decade.html?src=recg

Not from the Chronicle, but I found this piece from NYT on testing ( for NCLB ) interesting.  It seemed to confirm what I had already suspected about the state tests.

[polly_mer]

What greatly bothers me about the whole discussion of these NCLB evaluations is found in one of the links in the article

http://www.nytimes.com/schoolbook/2011/11/01/on-national-test-new-york-declines-in-math/

In the National Assessment of Educational Progress, scores correspond to achievement levels. Nationally, about a third of fourth and eighth graders met the cutoff to be assessed as “proficient” in math this year. New York’s students performed a little lower than that. But they did slightly better than the average in reading. About 35 percent of New York’s fourth and eighth graders performed at or above the proficient level.

Should we be diddling around with more tests and interpretations when the tests we have indicate two-thirds of children who have a reasonable amount of schooling are not proficient in reading? 

Even the numbers that so made the New York politicians happy indicated that a third or more of students were not proficient at grade level.  Those are not great numbers.

Yeah, let's put more time into testing instead of putting more people into classrooms, libraries, and what have you to teach people to read and do math.  That's a good use of effort.

[irhack]

What I don't get about NCLB is the expectation that districts must improve every year. I am in a really good school district--where at my kid's school say 95% of 3rd graders are proficient. Yet every year I get a letter from the district saying the school is under review or something because it's not making progress from year to year. What nonsense.

[mouseman]

NCLB is a typical case of trying to force a simplistic solution on a complex problem  It may cause some improvement at the beginning, but at the end it will not have solved the original problem, and will have created a whole new set of problems as well.  The only people who have really benefited from NCLB are the testing companies and the private schools.

[polly_mer]

What I don't get about NCLB is the expectation that districts must improve every year. I am in a really good school district--where at my kid's school say 95% of 3rd graders are proficient. Yet every year I get a letter from the district saying the school is under review or something because it's not making progress from year to year. What nonsense.

I remember the huge scandal the first year that failing schools in my state were announced.  The best schools in the state were on that list for that failure to make the required annual yearly progress.  The criteria for AYP were modified in short order so that schools that were already at 90+% proficiency did not have to improve at X% a year.  After all, the state looks terrible when nearly all the schools in the state are on the inadequate AYP list since most of the schools are terrible and won't improve for structural reasons and the fabulous schools fall on the list because some people are bad at statistics (graduates of the other schools in the state, no doubt).

[jackofallchem]

What I don't get about NCLB is the expectation that districts must improve every year. I am in a really good school district--where at my kid's school say 95% of 3rd graders are proficient. Yet every year I get a letter from the district saying the school is under review or something because it's not making progress from year to year. What nonsense.

Governments do this all the time.  They do it with air quality, water quality,and education.  The reasoning behind it is to enact an absolute standard would be disastrous to those with the worst problem.  If you simply took the 1975 California Achievement test and said every district needed to have 60% of the students reach a particular score, that could give you a realistic, absolute standard.  The problem is that the worst districts would find that only 10% of their students could achieve that score.  To meet the standard, they would have to do something drastic and no one in involved in education wants to do something drastic. 

Standardized testing won't work in the public schools because the people in charge of the testing don't want it to work.  They don't care about education or the children, they care about PR and preserving their jobs.  In my state the teachers all know what will be on the standardized tests and that is all they teach.  Algebra I didn't cover quadratics last year because the state test didn't cover it!  A month before the state tests, the students just take old copies of the state tests (that all say "Do Not Copy" and "Confidential") on them.  After the state tests, the school covers nothing for the rest of the year (a month).  Some days the teachers don't even show up!

Now, If "I" ran state testing, things would be different.  I would treat the school districts the same way I treat classes renowned for rampant cheating.

(1) I would base the tests on the textbooks.  I have looked at numerous textbooks, and the truth is, many if not most of them are pretty good.  If the teachers would just teach the information in the textbooks, the students would get a pretty good education.

(2)  I would have multiple exams covering different topics. Every topic in the books would be covered by at least one exam form.   These would be send out to the districts randomly, not being assigned until the last minute.  This would make them have to cover all the material.  Each district would also be sent more than one form and wouldn't be able to tell which test books were which forms until the seals were broken.

(3)  Test would be hand-delivered by my employees on testing days.  These employees would stay all day to monitor the testing.

(4)  My employees would cold-call school districts and offer to sell them advance copies of the exams.  I know this is entrapment, but I wouldn't be a police official and this wouldn't be a criminal case.  This would just let me know who is blatantly corrupt.  The tapes of the phone calls would also make good political ammunition.

(5) Tests would be graded on an absolute, not a percentile, grading scale.  If no student in the state met the standards, no one met the standards.  If every student did, great!

(6)  Schools would be judged and funding allocated based not on how well the district did on its tests, but on how well the tests correlated to the grades given.  If the districts students did poorly, but they were failed, the district would receive funding.  If their students did well, and they were passed, they would receive funding.  If their students failed the state testing and they were passed on grades...no funding for you. 

Schools would have to start teaching the material in their textbooks and giving honest grades again.  Isn't that all we really need to do to 'fix' the system?  Of course, this fictional scenario wouldn't work.  Anyone running such a testing scenario would be fired within a year.  That really just shows that the problem with the schools is that most people don't actually want them fixed.

[fiona]

I agree that many/most people in charge really don't want the schools fixed (if they could be, which I doubt).

If the schools were actually fixed, it would be the same thing that would happen if weight loss diets worked. Thousands of people would lose their jobs.

Of course, better-educated students and women who worried about things other than their weight might try to change society in a serious way.

That scares a lot of small-minded people.

The Fiona

[hmaria1609]

While in parochial grade school, the only standardized test I had to deal with every year was the Iowa Test. I was strong in some areas, others not. When I was in public high school, I had to take county issued assessments and finals for most of my classes except Advanced Placement.
Too time consuming!

[msparticularity]

The original impetus behind NCLB was a very legitimate concern: the only strongly positive correlation in the U.S. for education is between socioeconomic status and academic achievement. IOW, the interpretation of that phenomenon went, wealthier schools do a better job of preparing students to perform well academically. NCLB was premised on the assumption that all that was needed was accountability to make the lower-performing schools do better. There was also a series of secondary assumptions, including the idea that the path out of poverty is better education. Alternative causation, including the possibility that the path to better education can only be accomplished if you do something to relieve poverty, received insufficient attention. 

Remarkably enough (given my ideological bent), I recommend Diane Ravitch for her series of very lucid explorations of all of this.

[polly_mer]

Alternative causation, including the possibility that the path to better education can only be accomplished if you do something to relieve poverty, received insufficient attention. 

Yep.  The crab pot mentality is often overlooked..

However, if I ever become in charge of standardized testing, I'm drafting as many people like Jackofallchem as I can find.  While I care that students are failing and have ideas about how to work with people to stop so many falling-through-the-cracks failures, I care a whole lot more about stopping the machine that is awarding passing grades to people who don't know what they are supposed to know.  I'd rather have a C that's a true competent than continue to be tripped up by A's that evidently meant "showed up all the time and has a good short-term memory".

[mouseman]

I would add to Jack's list the additional part of sending people in to a failing school or school district to figure out why it is failing, and have people who could look at strategies to help.  The whole idea of cutting funding to schools just because the students are failing is stupid.  It is like cutting treatment to any patient who is not getting better because we automatically assume that it's because they weren't taking their medication.  In some cases it is worse - it is like not giving a patient enough medication and then cutting that because they are not getting better. 

[msparticularity]

I would add to Jack's list the additional part of sending people in to a failing school or school district to figure out why it is failing, and have people who could look at strategies to help.  The whole idea of cutting funding to schools just because the students are failing is stupid.  It is like cutting treatment to any patient who is not getting better because we automatically assume that it's because they weren't taking their medication.  In some cases it is worse - it is like not giving a patient enough medication and then cutting that because they are not getting better. 

The really frightening part is that in the cases when additional funding is available for "school improvement," it is contingent upon use of one of a very narrow selection of improvement plans (aka: scripted curricula put out by major publishers and testmaking corporations).

[polly_mer]

I would add to Jack's list the additional part of sending people in to a failing school or school district to figure out why it is failing, and have people who could look at strategies to help.  The whole idea of cutting funding to schools just because the students are failing is stupid.  It is like cutting treatment to any patient who is not getting better because we automatically assume that it's because they weren't taking their medication.  In some cases it is worse - it is like not giving a patient enough medication and then cutting that because they are not getting better. 

The really frightening part is that in the cases when additional funding is available for "school improvement," it is contingent upon use of one of a very narrow selection of improvement plans (aka: scripted curricula put out by major publishers and testmaking corporations).

Yep.  Those scripted curricula are generally the worst possible solution since they don't try to connect the kids with the rest of their lives.  The most efficient way to reach some of those "unteachable" kids is to toss academics as a separate endeavor aside and teach something cool as a major project, while bringing in the math, science, reading, writing, and other endeavors as students find they need such things.  Scripted curricula in separate areas unrelated to anything else in the students' lives are less likely to be useful than simply having people from the community come in and explain how they use various skills in their lives.

[jackofallchem]

I would add to Jack's list the additional part of sending people in to a failing school or school district to figure out why it is failing, and have people who could look at strategies to help.  The whole idea of cutting funding to schools just because the students are failing is stupid.  It is like cutting treatment to any patient who is not getting better because we automatically assume that it's because they weren't taking their medication.  In some cases it is worse - it is like not giving a patient enough medication and then cutting that because they are not getting better. 

The hard part is to get them to ask for and accept the help (not have it forced on them).  If you look at the scheme above, what happens in a failing school?  To keep their funding, they have to fail the students.  If they do this, the parents will suddenly become interested in the school.  The school district won't be able to just give the students passing grades, though, because that would result in a loss of funding.  Parents accept the current system because their kids pass (usually with good grades) so nothing that bad can be the matter, right?  When their kids don't pass, suddenly the (good) parents will be concerned and want something done.  They will want help, and they will be desperate enough to accept change.