[senco-forum] Regression to the mean and other maths info

Phillipa and others out there,
Thanks so much for the response, that many of us pour over.  I suspect that
many of you out there understand this completely, but could someone explain
to a "maths phobic" what "regression to the mean" means in simplistic terms,
ideally with examples.   AND I'm prepared to admit that although I accept
that %iles are less helpful than Standardized scores, I do not completely
understand the bunching issue.  Although I am aware of it.  Again simple
terms and examples would really help.
Could someone, in words of one syllable explain both of these?
Appreciated
Sharon F.

-----Original Message-----
From: senco-forum-bounces at lists.becta.org.uk
[mailto:senco-forum-bounces at lists.becta.org.uk] On Behalf Of Paul and
Philippa Bodien
Sent: 17 December 2007 16:05
To: senco forum
Subject: Re: [senco-forum] Expected progress in reading for 'strugglers' at
KS3

Hi Maggie,

Some time back I asked a similar question of Martin Turner.  I needed to
know when enough was enough for our kids in the Dyslexia Unit.  He described
a simple to understand pattern that has served our unit well over the
years...

Take the IQ (this means you need an IQ measure so not always available) then
take the mid point between that and 100.  So an IQ of 120 gives an expected
attainment score of 110 and 80 gives 90... this because of regression to the
mean, Martin advised.  That gives you the expected attainment score... so
standard scores obtained on all tests should be in line with that - within a
confidence band.

Standard scores are one measure you can use for comparisons of performance
over time... percentiles bunch up around the mean and so the intervals are
not equal.  They are great for understanding the position within a group but
not for comparison of perf over time.  Any discrepancy more than about 12
standard points from the expected attainment score is significant.

It was because of this expected attainment score calculation that I was able
to deduce that a Y5 boy had eye movement problems... his word reading and
word attack had come within expected levels but his passage comprehension
remained stubbornly below expectation when all else was in place (he was
compliant in lessons, enjoyed lessons, was trying hard, did all his homework
etc).  Once his eye movements were brought into function through vision
therapy prescribed by Keith Holland, his passage comprehension scores also
came with expectations.  Without expected attainment scores to point the way
he may still have been struggling...

Another lad had had his IQ tested at around 85.   Then with Occupational
Therapy, Omega 3 (Eye-Q) and dyslexia tuition, his scores shot beyond
expectations.  I asked for a reassessment by the same EP and his IQ had
risen so that his attainment scores were in line with the figure... what
that pointed out then is that the IQ was masked by the blocks he was
experiencing and when we worked on those blocks his IQ was measured at a
level that more accurately reflected his ability.

Philippa

On Dec 17, 2007 12:20 AM, Maggie Downie <maizie2004 at yahoo.co.uk> wrote:

>
> I wonder if anyone would be able to tell me what would be their, realistic
> (as opposed to govt. driven targets), 'expected' rate of progress for
> children receiving some sort of reading intervention at KS3, measured
either
> by the results of standardised tests or in English NC levels.
>
> The children I work with are all at L3 or below at the end of KS2. Reading
>  & Comprehension Ages on entry to Y7 are about  9.00 or below.  I think
> that their progress is generally good, but I don't have any comparable
data
> to measure it against.  I am hoping that some of you good people might
have
> some !
>
> I'm enjoying all the Christmas funnies!
>
> Maggie
>
>
>
>
>
>
>
> ---------------------------------
>  Support the World Aids Awareness campaign this month with Yahoo! for Good
>