[Rasch] Not a Fan of Lexiles?

Lang, William Steve WSLang at tempest.coedu.usf.edu
Wed Oct 31 22:09:27 EST 2007


I can't agree with Rense more.  Judy Wilkerson and I have a chapter on "why we chose the Rasch model" in a recent application.  We agree that value-added models, data mining, etc. might eventually come to the same conclusions about teacher performance measures that we do, but it is the unique insights of misfit and rater errors that have changed our interpretation and still been understandable to the users.

We like to say that "Rasch puts the people back in testings." The revolution is not in the use of the model or a particular piece of software, but comes with the new insights that were originally revealed because the model and software encouraged a different way of thinking.  

The Hubble telescope is still the same device that Galileo used and is still revealing new revolutionary insights.  Only a few astronomers probably know what the latest revolutionary discoveries mean, but they are still there because of the telescope.  We don't stop using optical telescopes simply because someone now has created a radio telescope.

If the tool has merit, then continuing use of the tool for revolutionary discoveries is justified. Because the Rasch applications continue to develop, the model continues to produce what I consider to be revolutionary results.  

Steve Lang
University of South Florida St. Petersburg


-----Original Message-----
From: rasch-bounces at acer.edu.au on behalf of Rense
Sent: Tue 10/30/2007 9:13 PM
To: 'Paul Barrett'; rasch at acer.edu.au
Subject: RE: [Rasch] Not a Fan of Lexiles?
 
Paul,
 
I was actually making the opposite case than you now do, namely that there
is more to measurement than adding raw scores however highly items are
correlated. Rather than doing away with quality control, equating, DIF
tests, working with incomplete data sets, CAT, item and person fit, etc. I
intended to say that addressing those issues may be more important than
obtaining a "score.". 
 
This is not only because in many cases (i.e., with complete data) sums
correlate highly with Rasch person estimates. But, mainly because knowing
such "scores" may not be the most important outcome of testing. I am
currently involved in a number of projects where respondents' misfit is
identified and then used to give feedback beyond (and sometimes instead of)
the "score" proper. Unless one is interested only in scores on fixed content
tests, Rasch scaling is actually eminently practical even in applications
other than Lexiles etc.
 
Rense Lange

  _____  

From: rasch-bounces at acer.edu.au [mailto:rasch-bounces at acer.edu.au] On Behalf
Of Paul Barrett
Sent: Tuesday, October 30, 2007 6:22 PM
To: rasch at acer.edu.au
Subject: RE: [Rasch] Not a Fan of Lexiles?





Hello Mike

 
I thought they were. Which is why there are no "stand out" achievements or
results using Rasch/IRT (which had not already been achieved using other
methods - however approximately or clumsily) which come to mind. 
 
Yes, there are many applications using Rasch/IRT which show "differences"
between the "old" and "new" - but these differences don't seem to amount to
anything which has fundamentally changed the way we think about human
behavior or cognitions, or predictions we might have made using our
"measures".
 
And we've had 50 years to wait for something. 
 
Frankly, it is only the Metametrics work which seems to qualify for the
"stand-out" epithet .. and look how much of that was based upon empirical
investigations of the phenomena of interest. 
 
I'm not seeking to convince anyone about anything - but just commenting on
why some (like me) might just look at the eulogies for Rasch or IRT, and end
up walking away from it as a "productive" methodology. Largely because what
may be defined as productive in science is not relevant to how  "productive"
is defined in educational or psychometrics circles?
 
Regards ... Paul
 
 



 


Paul Barrett, Ph.D.

2622 East 21st Street | Tulsa, OK  74114


Chief Research Scientist

Office | 918.749.0632  Fax | 918.749.0635


pbarrett at hoganassessments.com

      


 <http://www.hoganassessments.com/> hoganassessments.com

 

 

 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20071031/37499d6e/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: attc850d.bmp
Type: image/bmp
Size: 29886 bytes
Desc: attc850d.bmp
Url : https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20071031/37499d6e/attachment.bmp 


More information about the Rasch mailing list