In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (dennis roberts) wrote:
> At 01:12 PM 1/12/01 +0000, Gene Gallagher wrote:
>
> >I do believe that regression to the mean is involved here.
>
> i just reiterate that regression in this case ... involves a
correlation
> between two columns of MEANS ... means for schools OR means for
districts
> ... and means do NOT change that much .... from year to year ... and
> certainly ... schools with low or high means one year just CANNOT
change
> their position much
>
The MCAS evaluation is based on means, but means based on a very small
sample size. Each one of the 1500+ 4th grades, 8th grades, and 10th
grades in each school is evaluated. In my daughter's 4th grade class,
there were 3 classrooms of about 25 students. The mean MCAS scores for
this group of 3 classes in one school in 1998 is compared to the mean of
the 1999 and 2000 classes in that school.
The school was expected to show a 2 point increase, and when it didn't
this 4th grade class in this school was graded a failure, as was the 8th
grade class in the junior high across town and the 10th grade class in
the high school.
This case is very much like Galton's regression to mediocrity. Even
with increasing mean test scores (not documented to date, by the way),
the tendency will be for the top performing schools on the 1st test to
fall back closer to the mean and the poorest performers on the 1st
test to increase.
I would like to get all of the 1998 1999 and 2000 MCAS scores so that
the correlation coefficients and sources of error can be calculated. If
the correlation is very high, then regression to mediocrity won't be
much of a factor.
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/
=================================================================
