[RE][Rasch] JAM Reviewers
MarkM at eddata.com
Thu Oct 12 02:08:00 EST 2006
I think there is a lot to recommend this design. It deftly makes security
concerns something of a non-issue, provides clear direction on how to
prepare for the test, and can be set up all at once through matrix sampling.
Items can be added to the bank at any time. It permits adaptive and
conventional form administration and allows you to equate forms across
different difficulty levels and target specific populations. The use of OE
items seems very sensible in this context.
On the down side, public exposure of the items is likely to degrade item
difficulty more rapidly over time, and there may be differential item
effects between examinees who figure out how to study them and examinees who
come to the test cold. They are also likely to increase "teaching to the
test", though the use of OE items makes this less of a bad thing.
I would track item drift by holding some items in reserve, used on rare
occasions to measure the degree to which the public items are drifting
relative to the reserved items.
I have actually heard of your design being implemented in one of the state
assessments -- one of the southern states, Kentucky maybe, ... I can't
remember. I do seem to recall that they were pleased with the results.
Educational Data Systems
From: rasch-bounces at acer.edu.au [mailto:rasch-bounces at acer.edu.au] On Behalf
Of iasonas lambrianou
Sent: Tuesday, October 10, 2006 11:05 PM
To: rsmith.arm at att.net
Cc: rasch at acer.edu.au
Subject: [RE][Rasch] JAM Reviewers
I posted this question again, but only got one response. I hope this time
I'll be lucky
I have a question and I hope that you can help me.
Here in my country, people discuss the issue of building a large item bank
with open-ended (not multiple choice, we never use them) calibrated items
(say 1000 items), publish this item bank on the web, and then use this item
bank to draw items to build tests for high-stakes tests. The concept is that
anyone who can memorize 1000 open-ended items deserves to pass these
high-stakes university entrance exams anyway. The idea is that this would
minimize the errors during test construction, and we would be able to keep
the difficulty of tests constant fro year to year. Has this ever been
discussed there? How do you personally see this? Are you aware of any
similar initiatives anywhere in the world?
---------[ Received Mail Content ]----------
Subject : [Rasch] JAM Reviewers
Date : Wed, 13 Sep 2006 00:42:00 +0000
>From : rsmith.arm at att.net
To : rasch at acer.edu.au (Rasch Listserve)
Dear List Serve Members,
Every three years we send out a call for new reviewers for the Journal of
Applied Measurement. Our belief is that we need to continually refresh our
pool of reviewers to reflect current trends in measurement and scholarship.
JAM is a peered reviewed journal and the success of the journal depends on
the timely and constructive reviews provided by our reviewers. Without our
reviewers and the support that they provide to the authors seeking to
publish in JAM, the entire process would come to a stand still. Many of the
authors who publish in JAM comment on the helpful advice provided by our
reviewers and the supportive nature of the reviews. If you would be willing
to review one to two manuscripts each year, please send your name, e-mail
address, and postal address to me at the e-mail address listed below. A
brief paragraph about your areas of interest would be helpful, as we try to
match the manuscripts for review with the reviewer's interests. As always
your sup! port of t
rnal is greatly appreciated.
If you want to learn more about the journal you can visit our website listed
Please do not reply to this e-mail, as it will be distributed to the entire
Richard M. Smith, Editor
Journal of Applied Measurement
P.O. Box 1283
Maple Grove, MN 55311 USA
(JAM web site)
voice: 763-268-2282 (w)
Dr. Iasonas Lamprianou
CFAS, School of Education
The University of Manchester
Oxford Road, Manchester M13 9PL, UK
Tel. 0044 161 275 3485
iasonas.lamprianou at man.ac.uk
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Rasch