Return-Path: <nifl-assessment@literacy.nifl.gov>
Received: from literacy (localhost [127.0.0.1]) by literacy.nifl.gov (8.10.2/8.10.2) with SMTP id g6PFkmX29426; Thu, 25 Jul 2002 11:46:48 -0400 (EDT)
Date: Thu, 25 Jul 2002 11:46:48 -0400 (EDT)
Message-Id: <NGBBJDHKPMGKJEBFMDKOAEKACDAA.dbaycich@literacy.kent.edu>
Errors-To: listowner@literacy.nifl.gov
Reply-To: nifl-assessment@literacy.nifl.gov
Originator: nifl-assessment@literacy.nifl.gov
Sender: nifl-assessment@literacy.nifl.gov
Precedence: bulk
From: "Dianna Baycich" <dbaycich@archon.educ.kent.edu>
To: Multiple recipients of list <nifl-assessment@literacy.nifl.gov>
Subject: [NIFL-ASSESSMENT:163] RE: norm vs criterion
X-Listprocessor-Version: 6.0c -- ListProcessor by Anastasios Kotsikonas
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0)
Content-Transfer-Encoding: 7bit
Content-Type: text/plain;
Status: O
Content-Length: 3216
Lines: 66
Hi,
First off, your posts are not overly wordy but rather good explanations that
are helping me gain a better understanding!
What you said makes sense. My interpretation is that portfolios are
formative and standardized tests summative - more or less anyway. Do you
know if there is any imformation about the predictive validity of
portfolios? Any studies? Or is it mostly anecdotal evidence?
Dianna B.
-----Original Message-----
From: nifl-assessment@nifl.gov [mailto:nifl-assessment@nifl.gov]On
Behalf Of John Sabatini
Sent: Wednesday, July 24, 2002 10:04 PM
To: Multiple recipients of list
Subject: [NIFL-ASSESSMENT:162] RE: norm vs criterion
Reply to: RE: [NIFL-ASSESSMENT:161] RE: norm vs criterion
Hi again,
I agree with the principle that a range or 'porfolio' of assessments is a
sound strategy to meet the various goals or purposes of assessments
(diagnosis, placement, formative instructional guide, selection,
achievement, evaluation). And that is the key, knowing the goal or purpose
before choosing an assessment type.
I'm all for portfolio assessment, but I wouldn't jump to the conclusion that
performance on any particular portfolio task generalizes to more than
performance on an almost identical real world task. A portfolio task may
only tell one how well an individual will do on other portfolio tasks. Is
writing a letter to the gas company the same as writing a letter to the
principal of your child's school, the same as writing a memo to one's boss?
And what is portfolio assessment of reading ability? What would make it more
or less authentic or real world? (I might add that not all tasks I've seen
called portfolio are authentic/real world outside of a classroom context.)
Predicting how any one will perform on a range of real world tasks is a
tricky business for all assessments including portfolios. Before one can
make the claim that a portfolio performance generalizes to other real world
tasks, one has to conduct validity studies, observing how well people who do
well on a portfolio assessment perform on similar tasks as they arise in
real world situations. Predictive or concurrant validity studies of
standardized tests typically check to see whether high scorers on a test are
also the ones who perform well in the settings that demand those abilities,
for example, that most successful college level students outperform
non-successful or high school or middle school students on a test predicting
college performance. Such validity studies don't guarantee that the test
results tell us real world ability, but they provide the kind of evidence we
should demand of any assessment that claims to predict performance in
settings outside the assessmen!
t situation.
Portfolio assessments can inform instruction better than most standardized
tests, since one often gets performance evidence that the learner and
instructor can reflect on and uses as a guide to improving performance on
the next try. That assumes that the portfolio is being used for the purpose
of aiding instructon, not for outcome assessment.
That was neither an endorsement of or against portfolio or standardized
tests, just another overly wordy response to a post.
Best,
John
This archive was generated by hypermail 2b30 : Fri Jan 17 2003 - 14:46:25 EST