
Programs & Projects
The Institute is a catalyst for advancing a comprehensive national literacy agenda.
[LearningDisabilities 4226] Regression to the mean
Michael Gyori
tesolmichael at yahoo.comMon Nov 2 19:33:32 EST 2009
- Previous message: [LearningDisabilities 4219] Re: What does a grade level gain mean?
- Next message: [LearningDisabilities 4234] Re: Horse vs Cart
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Greetings Brant and all,
Thank you, Brant, for continuing to participate in the discussions!
Regression to the mean occurs due to the set of responses that are selected by guessing or chance, rather than knowing the correct answers, does it not? If so, I agree that this phenomenon would appear to have little or anything at all to do with test reliability.
My other question remains: if you have significant variations across measures, whether or not instruction occurs between them, I would assume that such variation could not only help inform instructional practices, but that the practice of maintaining longer intervals between assessments in order to (hopefully) demonstrate learning gains may not reveal much about the actual learning that has taken place. What about pre-testing repeatedly at or near the beginning of instruction?
I do a lot of formative assessment with my students. I believe that answer choices - whether right or wrong - are not nearly as important as the articulation of reasoning that has led to a particular answer choice.
Michael
Michael A. Gyori
Maui International Language School
www.mauilanguage.com
________________________________
From: Brant Hayenga <bhayenga at rrps.net>
To: The Learning Disabilities Discussion List <learningdisabilities at nifl.gov>
Sent: Mon, November 2, 2009 11:32:59 AM
Subject: [LearningDisabilities 4219] Re: What does a grade level gain mean?
Michael:
I might have misunderstood Tom’s post, but he said that they did nothing in between the tests, and the students’ scores changed (the number below 6th grade was smaller). I think his point was that the testing instruments are unreliable because the students’ scores were changing. My point is that the reason the students’ scores were changing was possibly not due to an unreliable standardized test, but to the phenomena of regression to the mean, which occurs even with reliable assessments.
I didn’t take his point to be about whether or not the instruction was working.
Thanks,
Brant Hayenga
Educational Diagnostician
Stapleton Elementary/Rio Rancho Middle School
(505) 896-0667 ext. 226 (District Office)
(505) 891-8473 ext. 519 (Stapleton Elementary)
bhayenga at rrps.net
Please note that RRPS email addresses have changed. Please update your
address books or distribution lists.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.nifl.gov/pipermail/learningdisabilities/attachments/20091102/2086e3e9/attachment.html
- Previous message: [LearningDisabilities 4219] Re: What does a grade level gain mean?
- Next message: [LearningDisabilities 4234] Re: Horse vs Cart
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the LearningDisabilities discussion list



