In article <[EMAIL PROTECTED]>,
"Daniel P. B. Smith" <[EMAIL PROTECTED]> wrote:
> (Thanks, Rich Ulrich, for pointing out this thread)
>
> Thank goodness my kids are long out of school... Forgive me jumping in
> as a layperson with a post that may only be marginally on-topic...
>
> In my local community (Norwood), the same thing happened. The school
> with the best score--the Callahan--is the one that got slapped with
the
> "failing" grade. I have to say that "regression to the mean" was
> certainly the first thing that crossed MY mind.
>
> But what seems far more puzzling to me is that the stated figures are:
>
> School Score Change
>
> Callahan 245.3 -1.0
> Oldham 242.2 6.5
> Prescott 242.5 5.5
>
> These seem to me like minuscule differences.
>
> Needless to say, all public reporting and discussion of MCAS scores
> seems to assume that the scores are perfectly accurate, with no stated
> margin of error. I believe individual scores are being reported to
> students the same way.
>
> Can anybody possibly believe that a difference of one point in 245.3
can
> possibly be significant? We're talking about schools with a less than
a
> maybe sixty fourth-graders in them. This just runs against common
> sense...
>
> Worse yet, if such tiny differences are being taken seriously by
> officials, there would seem to be strong motives for all kinds of
> mischief and "gaming the system" in various ways.
>
> --
> Daniel P. B. Smith
> Preferred email address: [EMAIL PROTECTED]
> Alternate email address: [EMAIL PROTECTED]
> "Lifetime forwarding" address: [EMAIL PROTECTED]
> Visit alt.books.jack-london!
>
For each grade and school, the MA Dept. of Education had 3 scores: 1998,
1999 & 2000. The Mass Dept. of Ed. decided to make the evaluations
based on the mean of 1999 and 2000 vs. 1998. In Norwell, every school
failed. The School head pointed out that the 2000 scores were higher
than the 1998 scores but that a low 1999 brought down the mean. I think
much of the current evaluation needs to be redone, but at a start, there
can be little justification for using the 1998 datum as the baseline for
assessing change.
--
Eugene D. Gallagher
ECOS, UMASS/Boston
Sent via Deja.com
http://www.deja.com/
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================