Re: Definition of Standardized Tests

["Deanna M. De'Liberto" <Ddeliberto@AOL.COM>]

In a message dated 10/4/98 10:48:37 PM Eastern Daylight Time, jhd@WNCNET.NET
writes:

>
>  -----Original Message-----
>  From: Nancy Patterson <patter@VOYAGER.NET>
>  To: ARN-L@LISTS.CUA.EDU <ARN-L@LISTS.CUA.EDU>
>  Date: Sunday, October 04, 1998 8:18 PM
>  Subject: Re: Definition of Standardized Tests
>
>  Nancy's original post:
>  >J. Davis,
>  >
>  >You are right that I did not mention math or science.  My area of
>  >"expertise" is literacy and hypertext theory.  I fall into a category
>  called
>  >"constructivists," meaning that I and others like me believe that human
>  >beings construct meaning for themselves.  Along with this notion, which is
>  >supported by a large body of academic theory and research, is that
>  >individual readers bring to each text and each reading their own meaning
>  >based on their own experience with the world and with other text.  The
fact
>  >that many readers find similar meaning in any given text is more about
>  their
>  >sharing of some common culture.  But we can all point to instances when
>  >someone we know watched the same movie we watched at saw something
>  different
>  >in that movie than we did.  That's an over-simplification, but it's the
>  best
>  >I can do in this limited format.  If you want to know more about
>  >constructivist theory or reader response theory, let me know and I will
>  >direct you to some texts.
>  >

When I taught mathematics at the high school level, I combined a number of
different approaches including but not limited to constructivism.  So, I too
understand Nancy's point albeit much of the material related to literacy and
acting is out of my area of expertise.

Nancy's definition of standardized testing:
>  >To address Deana's post regarding the definition of standardized testing,
I
>  >am defining it as any form of assessment in which more than one person is
>  >tested, but in which all those tested are measured in a standard way.  I'm
>  >not saying this well because I don't have the vocabulary.

J Davis's reply:
>  I can certainly accept this as a fundamental if not sufficient definition
of
>  standardized tests.  If anything we may need to add some other qualifiers,
>  but I agree that this is the core of the subject.

Nancy's definition would classify ALL tests -- even constructed response AND
performance assessments -- as standardized.  Generally I do not disagree with
this definition as this one is really not much different from the one I
proposed in an earlier email.  In fact, I even made the case that standardized
tests are not limited to multiple-choice tests.  I am glad that Nancy and I
agree.  I am a bit concerned in that Nancy earlier wrote that she "hates
standardized tests".  So, if standardized tests include constructed reopens
tests and performance tasks, it would seem that Nancy is opposed to ALL
methods of assessment.

She did make the point that we need to find creative ways to assess students
and again I agree as I have always supported and encouraged multiple forms of
assessment.  Thus, I must ask Nancy--what creative assessment methods are you
proposing that would not fall under your definition?  Or does the difference
between multiple-choice tests and constructed response, performance tasks,
etc. lie in the fact that these would also include qualitative analysis of the
student work--that is that a judgment is made about the quality of the work?
But here again, if we apply rubrics to the scoring of these types of items
does this now fall under the category of "standardized testing"?  And here
again, if this qualitative information is reduced to quantitative form, would
this now be labeled "standardized testing"?  So once again, we get back to
method by which results are interpreted and  what the user does with the
results.

This issue reminds me of my experience and training in administering and
interpreting psychological tests such as the WISC, and WJ-R.  Qualitative
analysis is very important in order for the to be valid--so much so that only
a qualified and trained examiner may administer these tests and the
qualifications for purchase are much more extensive--yet these tests are
standardized in the same that examiners must administer the tests as directed
in the Examiner Manual and any deviation must be noted in the report.  Such
deviation is always discouraged as it may affect the validity of the results.
My point has always been the the quantitative information did not really
provide answers to my questions even with the training I had.  I needed to
know more than that person A scored X on Y subtest of the WISC and that person
A has a W score of Z on the Arithmetic test of the WJ-R or Wide Range
Achievement Test, etc.  I needed to understand HOW the person arrived at that
answer and WHY the person got question C wrong.  The most detailed report did
not address those areas and so unless I administered the test, much of the
important information was forever lost.  Yet, I think that these assessments
serve a useful purpose if the purpose of the testing is clearly defined and if
the test is administered properly.

More recently, I have in a few cases (there has been only one such case I have
had involvement with), suggested that we deviate from the manual so that we
could gather qualitative information necessary to help the client who in this
case is 33 years old and has taken some of these tests twice already.  I
insisted that we would not use the normed scores and that such deviation would
be noted in the report.  The test publishers agreed with my recommendations
(given my intended purpose for administering the tests) but there was one
person who worked in training standardization examiners for one of the test
publishers who disagreed with this approach because he was not sure if someone
might obtain the scores without seeing the actual report.  This issue really
raised my awareness about testing causing me to reflect upon my own training
in administering and interpreting tests and perhaps I can hear what some of
you think about deviating from the prescribed administration methods.

Deanna
===================================================
Deanna M. De'Liberto, President/Director of Assessment
D Squared Assessments, Inc.
(Specialists in Test Development/Validation and Test Administration)
9 Bedle Road, Suite 250
Hazlet, NJ 07730-1209
Phone:  (732) 888-9339
Email:    Ddeliberto@aol.com
Web:     http://www.quikpage.com/D/dsquared

Member of the Association of Test Publishers
===================================================

--------------------------------------------------------------------------
To unsubscribe from the ARN-L list, send command SIGNOFF ARN-L
to LISTSERV@LISTS.CUA.EDU.