Track Wheel question

[David L. Johnson]

On Sun, 11 Apr 2004 18:09:29 -0700, Mark Hickey wrote:

>>Well, Mark, actually Jobst was talking about the administration's own
>>misuse of the language -- by the head of that administration.
>
> The "illiterati" term was what I keyed in on. If he considers a
> Bachelor's degree in history from Yale and an MBA from Harvard "a lack
> of edjumacation", Mr. Jobst must have some truly outstanding
> educational credentials!

Still, Mr. Bush does not show much of that Ivy-league education in his
speeches. Every time he says the word "nucular", it's like fingernails on
the blackboard to me.

>>It is also highly questionable whether it will improve education at all.
>>Taking more tests, which is the usual effect of the program, is not more
>>education, but less. Teaching to the test, which is the local school
>>districts' usual response, is not better education, but worse.
>
> I wish I knew a better way to hold the schools accountable for results -
> I'd support it in a heartbeat.

So would I. But that does not, in my mind, justify the current program.

> But the testing that goes on isn't
> something that SHOULD take "special classes" to pass. If the schools
> produce kids with even remotely adequate reading, writing, math and
> logic skills they'll do fine. If they're turning out students that
> require special classes to pass the test, the problem isn't the test,
> IMHO.

Agreed, but neither is the test the solution. Problem is, that is
essentially all that is offered under the "no child" program. Maybe my
old-fashioned New-Deal liberalism is showing through here, but somehow, if
a school shows poor student performance, cutting the funding to that
school would not seem to be the way to improve that performance.

I have seen the difference in performance of schools in Philadelphia,
compared to the outlying suburbs. There are buckets of money going into
the Philadelphia school system, so it would seem, but I never have been
able to find out the difference in $/student between the systems. I do
see that the Philly schools look like ****, and they use hand-me-down
books and ancient computers. Suburban schools have all the latest.
Either Philly schools don't get the funding the suburban schools do, or it
is siphoned off before it gets to the kids.

Sure, some of it is discipline and all the other problems in big-city
schools. But I suspect a lot is also plain old graft. To think that
cutting funding will help is naive. The "privatization" that has happened
here has also been merely a way to re-channel the flow of money, but not
to the students.

To test the teachers and fire those who can't pass seems like a good
solution, but only if there are other, better-qualified teachers ready to
step in. There is no long line of prospective teachers trying to get into
the Philadelphia school system.

No, I do not know the solution to the problems. But that does not suggest
to me that I should support a clearly shortsighted program aimed more at
garnering votes than improving schools.

--

David L. Johnson

__o | A mathematician is a machine for turning coffee into theorems.
_`\(,_ | -- Paul Erdos
(_)/ (_) |

 

[Tim McNamara]

"David L. Johnson" <[email protected]> writes:

> On Sun, 11 Apr 2004 18:09:29 -0700, Mark Hickey wrote:
>
>>>Well, Mark, actually Jobst was talking about the administration's
>>>own misuse of the language -- by the head of that administration.
>>
>> The "illiterati" term was what I keyed in on. If he considers a
>> Bachelor's degree in history from Yale and an MBA from Harvard "a
>> lack of edjumacation", Mr. Jobst must have some truly outstanding
>> educational credentials!
>
> Still, Mr. Bush does not show much of that Ivy-league education in
> his speeches. Every time he says the word "nucular", it's like
> fingernails on the blackboard to me.

Most of what he says- delivered in that mug smarmy style of his- is
like fingernails on a chalkboard to me. Although I'm hard to please,
I suppose- I didn't like Clinton's, Gore's or any of the most recent
crop of Democrats' speaking skills.

>>>It is also highly questionable whether it will improve education at
>>>all. Taking more tests, which is the usual effect of the program,
>>>is not more education, but less. Teaching to the test, which is
>>>the local school districts' usual response, is not better
>>>education, but worse.
>>
>> I wish I knew a better way to hold the schools accountable for
>> results - I'd support it in a heartbeat.
>
> So would I. But that does not, in my mind, justify the current
> program.

"Holding schools accountable" the way it's being done now is simply
the wrong approach. Hold students and parents accountable, first and
foremost. But that's not going to be done in this century, since the
American Way seems to have become "hold someone else accountable for
my screwups."

>> But the testing that goes on isn't something that SHOULD take
>> "special classes" to pass. If the schools produce kids with even
>> remotely adequate reading, writing, math and logic skills they'll
>> do fine. If they're turning out students that require special
>> classes to pass the test, the problem isn't the test, IMHO.
>
> Agreed, but neither is the test the solution. Problem is, that is
> essentially all that is offered under the "no child" program.

The problem with "no child left behind" is that it neglects a simple
reality: half of the children in any school are below average.
America is not Lake Wobegon where "all the children are above
average." The only way not to leave some children behind is to dumb
the standards down drastically. Bill Clinton started this "no child
left behind" malarkey in one of his State of the Union addresses and
Bush & Co. just sucked it up 'cause it sounds great to the public
(half of whom are below average themselves).

> Maybe my old-fashioned New-Deal liberalism is showing through here,
> but somehow, if a school shows poor student performance, cutting the
> funding to that school would not seem to be the way to improve that
> performance.

It's not. This sort of thinking just shows the bankruptcy of creative
analytical thought in the Legislature and the administration- not that
this is a new thing or confined to the Republican side of the aisle.

> I have seen the difference in performance of schools in
> Philadelphia, compared to the outlying suburbs. There are buckets
> of money going into the Philadelphia school system, so it would
> seem, but I never have been able to find out the difference in
> $/student between the systems. I do see that the Philly schools
> look like ****, and they use hand-me-down books and ancient
> computers. Suburban schools have all the latest.

They have the latest, and these kids started out with social
advantages as well: they've grown up steeped in the myth that you can
make your own future in any way you want. Kids from poor families,
which are disproportionately concentrated in urban districts, do not
grow up with the myth that they can make their own future. They grow
up with the myth that the games is rigged against them.

> Either Philly schools don't get the funding the suburban schools do,
> or it is siphoned off before it gets to the kids.
>
> Sure, some of it is discipline and all the other problems in
> big-city schools. But I suspect a lot is also plain old graft. To
> think that cutting funding will help is naive. The "privatization"
> that has happened here has also been merely a way to re-channel the
> flow of money, but not to the students.

Yup, charter schools were all the rage here but several of them have
been found to be run fraudulently and this has given the privatization
of schools a black eye. IMHO that's not unexpected, as the whole
premise of saving money by privatizing schools is specious.

> To test the teachers and fire those who can't pass seems like a good
> solution, but only if there are other, better-qualified teachers
> ready to step in. There is no long line of prospective teachers
> trying to get into the Philadelphia school system.
>
> No, I do not know the solution to the problems. But that does not
> suggest to me that I should support a clearly shortsighted program
> aimed more at garnering votes than improving schools.

Exactly.

 

[Tim McNamara]

Mark Hickey <[email protected]> writes:

> The Washington DC schools have one of the highest (maybe it is the
> highest) $ per student ratio in the country - and among the worst
> results.

Ray Suarez, who lives in DC and sends his kids to DC schools, was
recently in Minnesota talking about this very issue. I found it quite
enlightening. Unfortunately there doesn't seem to be a viewable copy
of this on the Twin Cities Public Television Web site
(www.tpt.org). Anyway, Suarez pointed out that the very figure you
cite is misleading in that some schools in the DC area are drastically
better funded than others, and not surprisingly the students in those
schools perform better.

 

[Tom Sherman]

Tim McNamara wrote:

> ...
> The problem with "no child left behind" is that it neglects a simple
> reality: half of the children in any school are below average.
> America is not Lake Wobegon where "all the children are above
> average." The only way not to leave some children behind is to dumb
> the standards down drastically. Bill Clinton started this "no child
> left behind" malarkey in one of his State of the Union addresses and
> Bush & Co. just sucked it up 'cause it sounds great to the public
> (half of whom are below average themselves)....

I though that half the children were below median. If you take 9
squirrel monkeys and one gorilla, 90% of the primates are below average
weight.

--
Tom Sherman – Quad Cities

 

[David L. Johnson]

On Mon, 12 Apr 2004 18:48:43 -0500, Tim McNamara wrote:

> The problem with "no child left behind" is that it neglects a simple
> reality: half of the children in any school are below average.
> America is not Lake Wobegon where "all the children are above
> average." The only way not to leave some children behind is to dumb
> the standards down drastically. Bill Clinton started this "no child
> left behind" malarkey in one of his State of the Union addresses and
> Bush & Co. just sucked it up 'cause it sounds great to the public
> (half of whom are below average themselves).

It's not Bush, nor is it Clinton, who started that. Several years ago I
was amazed to hear the debates about school curriculum; how it could not
be more rigorous since "not every child would be able to succeed". So,
they dumbed it down so that every child could "succeed". Sorry, but that
is not my idea of success, nor of education.

I have occasionally been involved with pedagogical projects. One in
particular got me. It was an experimental program, and what they wanted
was "indicators of success" of the program, _not_ the results of the
experiment, which were somewhat negative. Lake Woebegone is alive and
well in colleges of education throughout the nation.

--

David L. Johnson

__o | Do not worry about your difficulties in mathematics, I can assure
_`\(,_ | you that mine are all greater. -- A. Einstein
(_)/ (_) |

 

[Tim McNamara]

"David L. Johnson" <[email protected]> writes:

> On Mon, 12 Apr 2004 18:48:43 -0500, Tim McNamara wrote:
>
>> The problem with "no child left behind" is that it neglects a
>> simple reality: half of the children in any school are below
>> average. America is not Lake Wobegon where "all the children are
>> above average." The only way not to leave some children behind is
>> to dumb the standards down drastically. Bill Clinton started this
>> "no child left behind" malarkey in one of his State of the Union
>> addresses and Bush & Co. just sucked it up 'cause it sounds great
>> to the public (half of whom are below average themselves).
>
> It's not Bush, nor is it Clinton, who started that.

Well, he started the popular use of "no child left behind" as national
public policy objective. It's a great sound bite even though it is
barely meaningful and is in practice impossible.

> Several years ago I was amazed to hear the debates about school
> curriculum; how it could not be more rigorous since "not every child
> would be able to succeed". So, they dumbed it down so that every
> child could "succeed". Sorry, but that is not my idea of success,
> nor of education.

I agree with you on that. Education should be about mastery of the
material in absolute terms, not relative terms. We should also as a
culture face up to the fact that not every student is college material
or executive material, and that there are other ways of making a
valuable contribution and (there should be) other ways of earning a
good living to support yourself and your family. Unfortunately in our
culture, prestige, power and privilege are intimately linked to income
and not to the value that the person provides society.

> I have occasionally been involved with pedagogical projects. One in
> particular got me. It was an experimental program, and what they
> wanted was "indicators of success" of the program, _not_ the results
> of the experiment, which were somewhat negative. Lake Woebegone is
> alive and well in colleges of education throughout the nation.

Indeed it is.

 

[David L. Johnson]

On Mon, 12 Apr 2004 23:29:01 -0500, Tim McNamara wrote:

>>> to the public (half of whom are below average themselves)....
>>
>> I though that half the children were below median.
>
> No, half the children are below the mean, not the median nor the
> mode. Confusing these terms renders the discussion useless.

Sorry, but he's right, Tim. Median is the score for which as many
data points are below as above. So, if one kid has an IQ of 200, but 3
others have IQs of 103, 100, 99, and 98, then the median IQ is 100,
whereas the mean is 120. 4 are below the mean, but only 2 below the
median, and 2 above.

--

David L. Johnson

__o | A mathematician is a machine for turning coffee into theorems.
_`\(,_ | -- Paul Erdos
(_)/ (_) |

 

[Mark Hickey]

"David L. Johnson" <[email protected]> wrote:

>,,,but 3 >others have IQs of 103, 100, 99, and 98,

Sorry, but I find the irony (in regards to a discussion on averages)
too funny not to point that out.

Mark Hickey
Habanero Cycles
http://www.habcycles.com
Home of the $695 ti frame

 

[Tim McNamara]

"David L. Johnson" <[email protected]> writes:

> On Mon, 12 Apr 2004 23:29:01 -0500, Tim McNamara wrote:
>
>>>> to the public (half of whom are below average themselves)....
>>>
>>> I though that half the children were below median.
>>
>> No, half the children are below the mean, not the median nor the
>> mode. Confusing these terms renders the discussion useless.
>
> Sorry, but he's right, Tim. Median is the score for which as many
> data points are below as above.

He's still using the wrong measurement for this discussion. When
talking about variance and central tendency in performance on a
standardized test, it is preferable to use the mean rather than mode
or median. Standardization tends to limit skew, which would affect
the mean to a greater extent than the median or the mode, and hence
there is not much to be gained and quite a bit to be lost by using the
median or mode as the measurement. Median and mode can be used for
purely ordinal data, of course, which is not the case with the mean.
If we were talking about ordinal ranking, then the use of the median
would make sense.

However, my reference to "half of all children are below average"
refers exactly to the distribution of scores on standardized IQ tests.
That is of course slightly oversimplified, since as I pointed out
earlier some 68% of children score within one standard deviation of
the mean; this is a significant difference, however, as a child who
scores 115 will likely perform much better academically than a child
who scores 85.

> So, if one kid has an IQ of 200, but 3 others have IQs of 103, 100,
> 99, and 98, then the median IQ is 100, whereas the mean is 120. 4
> are below the mean, but only 2 below the median, and 2 above.

Yes, the mean is sensitive to being skewed by extreme scores which is
why, for example, it is not used preferentially in describing the
distribution of incomes or residential property values across the
population of the United States. The median income is a more useful
figure in these instances, although the media more often than not
confuses the median with the mathematical "average.".

For standardized tests of performance, however, the mean is used,
rather than median or mode, because it is more mathematically useful
in analyzing the results. In the specific case of IQ testing, the
mean is used rather than median or mode in describing performance on
IQ tests. IQ tests (e.g., the Wechsler series, the Stanford-Binet,
etc) are standardized so that the mean is 100 and the standard
deviation is (or was) 15 for the Wechsler tests and 16 for the S-B.
This allows easy comparison of individual performance against the
standardization sample.

In the case of the Wechsler tests, ISTR that the curve is slightly
leptokurtic. There is (or was) also a tendency for the S-B to produce
a slightly lower z-score than the Wechsler tests so it is (or was)
more likely, for example to rate a child as mildly mentally retarded
than as having borderline intellectual functioning. I haven't done IQ
testing in years and the tests have since been revised, so I don't
know what the relationships between scores on the Wechsler vs. S-B
might be now.

 

[Mark Hickey]

Tim McNamara <[email protected]> wrote:

>... a child who
>scores 115 will likely perform much better academically than a child
>who scores 85.

Is it still PC to say that? ;-)

Mark Hickey
Habanero Cycles
http://www.habcycles.com
Home of the $695 ti frame

 

[Benjamin Weiner]

Mark Hickey <[email protected]> wrote:
> Tim McNamara <[email protected]> wrote:

> >... a child who
> >scores 115 will likely perform much better academically than a child
> >who scores 85.

> Is it still PC to say that? ;-)

Well, increasingly, academic performance is defined as
scoring well on standardized tests, which often bear some
resemblance to IQ tests (OK, not all of them do). So it's
not just PC, it's tautological.

Silly remarks aside, test-taking is a skill in itself,
which has various weird correlations with social/cultural
background, etc., and is increasingly important, damagingly
so I think. I'm not ready to discuss this at length, but
Nicolas Lemann's book "The Big Test" is interesting and
scary reading. Burn the SAT!

 

[Tim McNamara]

Mark Hickey <[email protected]> writes:

> Tim McNamara <[email protected]> wrote:
>
>>... a child who scores 115 will likely perform much better
>>academically than a child who scores 85.
>
> Is it still PC to say that? ;-)

Hah! Absolutely not. But then I've never quite managed to be PC very
often. I'm too much of a loudmouth.

What is PC is to pretend that everyone can do anything they want with
their lives. That's been the myth of America for a long, long time.
Your odds of self-direction and doing what you want might be better in
the US than in many places in the world, and perhaps no better or not
even quite as good as in some other places- depending on what you want
to do.

 

[Tim McNamara]

Benjamin Weiner <[email protected]> writes:

> Mark Hickey <[email protected]> wrote:
>> Tim McNamara <[email protected]> wrote:
>
>> >... a child who scores 115 will likely perform much better
>> >academically than a child who scores 85.
>
>> Is it still PC to say that? ;-)
>
> Well, increasingly, academic performance is defined as scoring well
> on standardized tests, which often bear some resemblance to IQ tests
> (OK, not all of them do). So it's not just PC, it's tautological.

I haven't looked at any of the stadardized tests used in the "Profiles
of Learning" model used currently in Minnesota (and soon to be
abandoned). The ACT and SAT do have some relation to IQ tests, but
those are college entrance tests and that's appropriate. Grad school
tests such as the MCAT, LSAT, Miller's Analogies, ete. are even more
like IQ tests. Their purpose is to predict future academic
achievement rather than to measure current achievement.

> Silly remarks aside, test-taking is a skill in itself, which has
> various weird correlations with social/cultural background, etc.,
> and is increasingly important, damagingly so I think.

Well, there are test taking skills that can be learned and can have a
measureable affect on test scores. However, tests of ability (such
as IQ tests) are different than tests of achievement (which is what
standardized tests for academic achievement ought to be).

> I'm not ready to discuss this at length, but Nicolas Lemann's book
> "The Big Test" is interesting and scary reading. Burn the SAT!

The SAT has been rendered relatively useless with the advent of
training courses for the children of overly competitive parents. It
is less and less a valid predictor of performance and more and more a
measure of preparation. Back in my day (1977), my 670 math/740 verbal
wasn't too bad. It doesn't look as good by comparison today. The
average math score in 2002 was 516 (534 for males, 500 for females), a
35 year high; the average verbal score was only 504, however- 507 for
males and 502 for females. The latter is actually surprising, as
typically girls show better language skills development early but boys
must close that gap by the end of high school.

 

[Mark Hickey]

Tim McNamara <[email protected]> wrote:

>The SAT has been rendered relatively useless with the advent of
>training courses for the children of overly competitive parents. It
>is less and less a valid predictor of performance and more and more a
>measure of preparation. Back in my day (1977), my 670 math/740 verbal
>wasn't too bad. It doesn't look as good by comparison today. The
>average math score in 2002 was 516 (534 for males, 500 for females), a
>35 year high; the average verbal score was only 504, however- 507 for
>males and 502 for females. The latter is actually surprising, as
>typically girls show better language skills development early but boys
>must close that gap by the end of high school.

Do I remember that they "adjusted" the SAT tests some years ago
(resulting in an artificial "increase" in the test numbers)?

Mark Hickey
Habanero Cycles
http://www.habcycles.com
Home of the $695 ti frame

 

[Dane Jackson]

Mark Hickey <[email protected]> wrote:
>
> Do I remember that they "adjusted" the SAT tests some years ago
> (resulting in an artificial "increase" in the test numbers)?

You do remember correctly. From various sources I've seen the jump was
about 100 points.

From: http://www.myscschools.com/reports/sat00/

"The scoring scale was recentered in spring 1995. The data in this summary
report reflect the new, recentered scale. Data for the prior years for all
students also are reported using the recentered scale."

--
Dane Jackson - z u v e m b i @ u n i x b i g o t s . o r g
Whenever I hear anyone arguing for slavery, I feel a strong impulse to
see it tried on him personally.
-- Abraham Lincoln

 

[David L. Johnson]

On Wed, 14 Apr 2004 22:14:09 -0500, Tim McNamara wrote:

> The SAT has been rendered relatively useless with the advent of
> training courses for the children of overly competitive parents. It
> is less and less a valid predictor of performance and more and more a
> measure of preparation. Back in my day (1977), my 670 math/740 verbal
> wasn't too bad. It doesn't look as good by comparison today. The
> average math score in 2002 was 516 (534 for males, 500 for females), a
> 35 year high; the average verbal score was only 504, however- 507 for
> males and 502 for females. The latter is actually surprising, as
> typically girls show better language skills development early but boys
> must close that gap by the end of high school.

I suspect that the increase of SAT scores has more to do with the "Lake
Woebegone" syndrome than anything else. Yeah, people are pushing SAT prep
courses, but most of them are of questionable value. I had a roommate in
college (1973-77) who had, along with some classmates, been coached
effectively by a dedicated teacher. They all got 800/800 scores. Most of
these courses don't come near that kind of performance.

But the scores, and the exams, are rigged. ETS, who run the tests, alters
the exam every year, the point being to better differentiate between
students. But they determine what the average will be, and there has been
a tendency to increase the average over the past 10-20 years.

Basically, the test, and the scoring, are not what they were in the 1970s.
Any increase in average scores is engineered into the test, not measuring
any increase in student performance. They are trying to fend off
(justified) criticism of cultural bias, by raising everyone's
score. Students like that, parents like it, and so do colleges, since
they can brag about the increase in average SATs at their school as if it
were a measure of the increase of the school's prestige.


--

David L. Johnson

__o | We have a record of conquest, colonization and expansion
_`\(,_ | unequaled by any people in the Nineteenth Century. We are not to
(_)/ (_) | be curbed now. --Henry Cabot Lodge, 1895

 

[Mark Hickey]

Dane Jackson <[email protected]> wrote:

>Mark Hickey <[email protected]> wrote:
>>
>> Do I remember that they "adjusted" the SAT tests some years ago
>> (resulting in an artificial "increase" in the test numbers)?
>
>You do remember correctly. From various sources I've seen the jump was
>about 100 points.
>
>From: http://www.myscschools.com/reports/sat00/
>
>"The scoring scale was recentered in spring 1995. The data in this summary
>report reflect the new, recentered scale. Data for the prior years for all
>students also are reported using the recentered scale."

Exactly - so Tim needn't lament over the fact that the current average
scores seem to be creeping up near his (unless updating the data for
prior years included someone crawling through his bedroom window and
slipping a recalibrated set of test results into his dresser drawer).

Mark Hickey
Habanero Cycles
http://www.habcycles.com
Home of the $695 ti frame

 

[Tim McNamara]

Mark Hickey <[email protected]> writes:

> Tim McNamara <[email protected]> wrote:
>
>>The SAT has been rendered relatively useless with the advent of
>>training courses for the children of overly competitive parents. It
>>is less and less a valid predictor of performance and more and more
>>a measure of preparation. Back in my day (1977), my 670 math/740
>>verbal wasn't too bad. It doesn't look as good by comparison today.
>>The average math score in 2002 was 516 (534 for males, 500 for
>>females), a 35 year high; the average verbal score was only 504,
>>however- 507 for males and 502 for females. The latter is actually
>>surprising, as typically girls show better language skills
>>development early but boys must close that gap by the end of high
>>school.
>
> Do I remember that they "adjusted" the SAT tests some years ago
> (resulting in an artificial "increase" in the test numbers)?

Hmm. I don't know. I did read that the ACT's scoring method has been
slightly adjusted compared to when I took it in 1977, but I didn't see
anything like that about the SAT (and I may just have missed it).
However, I'd bet there's educators in this newsgroup who would know.

 

[Tim McNamara]

Mark Hickey <[email protected]> writes:

> Dane Jackson <[email protected]> wrote:
>
>>Mark Hickey <[email protected]> wrote:
>>>
>>> Do I remember that they "adjusted" the SAT tests some years ago
>>> (resulting in an artificial "increase" in the test numbers)?
>>
>>You do remember correctly. From various sources I've seen the jump
>>was about 100 points.
>>
>>From: http://www.myscschools.com/reports/sat00/
>>
>>"The scoring scale was recentered in spring 1995. The data in this
>>summary report reflect the new, recentered scale. Data for the prior
>>years for all students also are reported using the recentered
>>scale."
>
> Exactly - so Tim needn't lament over the fact that the current
> average scores seem to be creeping up near his (unless updating the
> data for prior years included someone crawling through his bedroom
> window and slipping a recalibrated set of test results into his
> dresser drawer).

My Mom's attic actually. I'll have to ask if there's been any
evidence of intruders. ;-)

 

[David L. Johnson]

On Thu, 15 Apr 2004 15:06:16 -0500, Tim McNamara wrote:

>
>> Do I remember that they "adjusted" the SAT tests some years ago
>> (resulting in an artificial "increase" in the test numbers)?
>
> Hmm. I don't know. I did read that the ACT's scoring method has been
> slightly adjusted compared to when I took it in 1977, but I didn't see
> anything like that about the SAT (and I may just have missed it).
> However, I'd bet there's educators in this newsgroup who would know.

Mark is right in the particulars here. They did bump up the scores by a
fair notch recently. But the scores have been creeping up since the
1980s, and since they control the determination of the score ranges, that
can only be by design.

--

David L. Johnson

__o | If all economists were laid end to end, they would not reach a
_`\(,_ | conclusion. -- George Bernard Shaw
(_)/ (_) |