Rasch Analysis

[Johnny Amora]

Hi everyone,

Rasch Analysis is one of the approaches of Item
Response Theory which is used to estimate the
probability of a correct response to a given item as a
function of item difficulty and person ability, where
item difficulty and person ability are parameters.
Please describe in detail how SPSS can perform this
analysis.

Thank you!
Johnny Amora


      ____________________________________________________________________________________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[SR Millis-3]

John,

Alternatively, I'd strongly recommend the Winsteps
software which focuses exclusively on Rasch analysis.
Extensively detailed output and excellent technical
support.  You can download (for free) the student
version (Ministep):

http://winsteps.com/ministep.htm

Scott Millis


--- John Amora <[hidden email]> wrote:

> Hi everyone,
>
> Rasch Analysis is one of the approaches of Item
> Response Theory which is used to estimate the
> probability of a correct response to a given item as
> a
> function of item difficulty and person ability,
> where
> item difficulty and person ability are parameters.
> Please describe in detail how SPSS can perform this
> analysis.
>
> Thank you!
> Johnny Amora

Scott R Millis, PhD, MEd, ABPP (CN,CL,RP), CStat
Professor & Director of Research
Dept of Physical Medicine & Rehabilitation
Wayne State University School of Medicine
261 Mack Blvd
Detroit, MI 48201
Email:  [hidden email]
Tel: [phone]
Fax: [phone]

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Johnny Amora]

Dear Scott Millis,

Thanks for your reply. Can we not use SPSS instead?

John

> Alternatively, I'd strongly recommend the Winsteps
> software which focuses exclusively on Rasch
> analysis. Extensively detailed output and excellent
>technical support.  You can download (for free) the
student version (Ministep):
http://winsteps.com/ministep.htm

> --- John Amora <[hidden email]> wrote:
>
> > Hi everyone,
> >
> > Rasch Analysis is one of the approaches of Item
> > Response Theory which is used to estimate the
> > probability of a correct response to a given item
> as
> > a
> > function of item difficulty and person ability,
> > where
> > item difficulty and person ability are parameters.
> > Please describe in detail how SPSS can perform
> this
> > analysis.
> >
> > Thank you!
> > Johnny Amora
>
>
> Scott R Millis, PhD, MEd, ABPP (CN,CL,RP), CStat
> Professor & Director of Research
> Dept of Physical Medicine & Rehabilitation
> Wayne State University School of Medicine
> 261 Mack Blvd
> Detroit, MI 48201
> Email:  [hidden email]
> Tel: [phone]
> Fax: [phone]
>
> =====================
> To manage your subscription to SPSSX-L, send a
> message to
> [hidden email] (not to SPSSX-L), with no
> body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send
> the command
> INFO REFCARD
>

      ____________________________________________________________________________________
Looking for last minute shopping deals?
Find them fast with Yahoo! Search.  http://tools.search.yahoo.com/newsearch/category.php?category=shopping

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Daniel Robertson]

SPSS does not have a procedure to perform Rasch analysis. Winsteps and
Bilog are among the specialty packages that can perform a Rasch analysis
directly. In a general sense, a Rasch analysis could be performed using
any tool that can handle multi-level data with dichotomous outcomes and
random effects, for example HGLM in the HLM package. But this excludes
SPSS.
Good luck.

> Hi everyone,
>
> Rasch Analysis is one of the approaches of Item
> Response Theory which is used to estimate the
> probability of a correct response to a given item as a
> function of item difficulty and person ability, where
> item difficulty and person ability are parameters.
> Please describe in detail how SPSS can perform this
> analysis.
>
> Thank you!
> Johnny Amora
>
>
>       ____________________________________________________________________________________
> Be a better friend, newshound, and
> know-it-all with Yahoo! Mobile.  Try it now.
> http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>

--
Daniel Robertson
Senior Research and Planning Associate
Institutional Research and Planning
Cornell University / irp.cornell.edu

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Johnny Amora]

 I am aware that there are a lot of packages that
would handle Rasch analysis, including Winsteps, SAS,
STATA, among others.  These packages can handle not
only dichotomous outcomes but also polytomous
outcomes.  But I dont have chances to use such
packages.

Yes, Rasch analysis cannot be found as a procedure
within SPSS, but I think it is possible perhaps
through syntax because somebody already did it.  I did
some searching in the internet and I found that Gary
Phillips presented a paper on “Rasch Analysis using
SPSS” during the Annual Meeting of the National
Council on Measurement in Education in Montreal,
Canada, in April 1983.  Unfortunately, only abstract
of his paper is available.  The abstract is as
follows:

"Ways in which the Statistical Package for the Social
Sciences (SPSS) can be used to perform some Rasch
analyses are described in detail. It is shown how SPSS
and a set of item calibrations can be used to estimate
person abilities, standard errors of measurement, test
characteristic curve, test information curve,
classification consistency on a mastery test, and item
and person fit statistics. Additional uses of SPSS are
suggested. Figures, tables, and sample printouts of
Rasch analyses using SPSC are appended. (GDC)"


If anyone has copy of Gary Phillips's full paper,
please share it.  Rest assured it would be handled
professionally.

Thanks!
Johnny

--- Daniel Robertson <[hidden email]> wrote:

> SPSS does not have a procedure to perform Rasch
> analysis. Winsteps and
> Bilog are among the specialty packages that can
> perform a Rasch analysis
> directly. In a general sense, a Rasch analysis could
> be performed using
> any tool that can handle multi-level data with
> dichotomous outcomes and
> random effects, for example HGLM in the HLM package.
> But this excludes
> SPSS.
> Good luck.
>
> > Hi everyone,
> >
> > Rasch Analysis is one of the approaches of Item
> > Response Theory which is used to estimate the
> > probability of a correct response to a given item
> as a
> > function of item difficulty and person ability,
> where
> > item difficulty and person ability are parameters.
> > Please describe in detail how SPSS can perform
> this
> > analysis.
> >
> > Thank you!
> > Johnny Amora
> >
> >
> >
>

____________________________________________________________________________________
> > Be a better friend, newshound, and
> > know-it-all with Yahoo! Mobile.  Try it now.
> >
>
http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

> >
> > =====================
> > To manage your subscription to SPSSX-L, send a
> message to
> > [hidden email] (not to SPSSX-L), with
> no body text except the
> > command. To leave the list, send the command
> > SIGNOFF SPSSX-L
> > For a list of commands to manage subscriptions,
> send the command
> > INFO REFCARD
> >
>
>
> --
> Daniel Robertson
> Senior Research and Planning Associate
> Institutional Research and Planning
> Cornell University / irp.cornell.edu
>
> =====================
> To manage your subscription to SPSSX-L, send a
> message to
> [hidden email] (not to SPSSX-L), with no
> body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send
> the command
> INFO REFCARD
>

      ____________________________________________________________________________________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Daniel Robertson]

Have you tried contacting the author?

But without seeing the article, I'm skeptical that it could really be
done using Logistic Regression/Log-linear Models only, which afaik would
have been the only reasonable option in SPSS in 1983 (I was in
middle-school then). IRT models generally are hierarchical models with
items nested within examinees. The only saving grace that allows Rasch
models to be estimated using non-specialized software is the fact that
when only the difficulty parameter is estimated, the summed item score
is sufficient to estimate the latent ability score without resorting to
a factor analysis procedure. But in estimating parameters for each item,
you would still need to be able to parse out items within examinees and
treat examinees as random. I'm not aware of any way to do all that in SPSS.

Here's a much more recent paper showing how it can be done using HLM:
Kamata, A. (2002). /Procedures to Perform Item Response Data Analysis by
HLM/. Paper presented at the annual meeting of American Educational
Research Association, New Orleans, April 2002.
http://garnet.acns.fsu.edu/~akamata/papers/AERA_2002.pdf

If cost is an issue, you could always learn to use R -- the ltm package
apparently handles latent trait models.
<http://garnet.acns.fsu.edu/%7Eakamata/papers/AERA_2002.pdf>

John Amora wrote:

>  I am aware that there are a lot of packages that
> would handle Rasch analysis, including Winsteps, SAS,
> STATA, among others.  These packages can handle not
> only dichotomous outcomes but also polytomous
> outcomes.  But I dont have chances to use such
> packages.
>
> Yes, Rasch analysis cannot be found as a procedure
> within SPSS, but I think it is possible perhaps
> through syntax because somebody already did it.  I did
> some searching in the internet and I found that Gary
> Phillips presented a paper on “Rasch Analysis using
> SPSS” during the Annual Meeting of the National
> Council on Measurement in Education in Montreal,
> Canada, in April 1983.  Unfortunately, only abstract
> of his paper is available.  The abstract is as
> follows:
>
> "Ways in which the Statistical Package for the Social
> Sciences (SPSS) can be used to perform some Rasch
> analyses are described in detail. It is shown how SPSS
> and a set of item calibrations can be used to estimate
> person abilities, standard errors of measurement, test
> characteristic curve, test information curve,
> classification consistency on a mastery test, and item
> and person fit statistics. Additional uses of SPSS are
> suggested. Figures, tables, and sample printouts of
> Rasch analyses using SPSC are appended. (GDC)"
>
>
> If anyone has copy of Gary Phillips's full paper,
> please share it.  Rest assured it would be handled
> professionally.
>
> Thanks!
> Johnny
>
> --- Daniel Robertson <[hidden email]> wrote:
>
>
>> SPSS does not have a procedure to perform Rasch
>> analysis. Winsteps and
>> Bilog are among the specialty packages that can
>> perform a Rasch analysis
>> directly. In a general sense, a Rasch analysis could
>> be performed using
>> any tool that can handle multi-level data with
>> dichotomous outcomes and
>> random effects, for example HGLM in the HLM package.
>> But this excludes
>> SPSS.
>> Good luck.
>>
>>
>>> Hi everyone,
>>>
>>> Rasch Analysis is one of the approaches of Item
>>> Response Theory which is used to estimate the
>>> probability of a correct response to a given item
>>>
>> as a
>>
>>> function of item difficulty and person ability,
>>>
>> where
>>
>>> item difficulty and person ability are parameters.
>>> Please describe in detail how SPSS can perform
>>>
>> this
>>
>>> analysis.
>>>
>>> Thank you!
>>> Johnny Amora
>>>
>>>
>>>
>>>
> ____________________________________________________________________________________
>
>>> Be a better friend, newshound, and
>>> know-it-all with Yahoo! Mobile.  Try it now.
>>>
>>>
> http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
>
>>> =====================
>>> To manage your subscription to SPSSX-L, send a
>>>
>> message to
>>
>>> [hidden email] (not to SPSSX-L), with
>>>
>> no body text except the
>>
>>> command. To leave the list, send the command
>>> SIGNOFF SPSSX-L
>>> For a list of commands to manage subscriptions,
>>>
>> send the command
>>
>>> INFO REFCARD
>>>
>>>
>> --
>> Daniel Robertson
>> Senior Research and Planning Associate
>> Institutional Research and Planning
>> Cornell University / irp.cornell.edu
>>
>> =====================
>> To manage your subscription to SPSSX-L, send a
>> message to
>> [hidden email] (not to SPSSX-L), with no
>> body text except the
>> command. To leave the list, send the command
>> SIGNOFF SPSSX-L
>> For a list of commands to manage subscriptions, send
>> the command
>> INFO REFCARD
>>
>>
>
>
>
>       ____________________________________________________________________________________
> Be a better friend, newshound, and
> know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>
>

--
Daniel Robertson
Senior Research and Planning Associate
Institutional Research and Planning
Cornell University / irp.cornell.edu

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Daniel Robertson]

Actually, Jon's probably right -- I'd forgotten about the GEE aspect of
GENLIN, so the capacity is probably there. I don't have time to work out
the model specification right now, but here's an example of doing the
same sort of thing in STATA.
http://www.stata.com/support/faqs/stat/rasch.html

Peck, Jon wrote:

> Could this be done with the GENLIN  procedure?  It seems to have most of the prerequisites.
>
> And surely it could be done with the R plugin to SPSS 16.
>
> Regards,
> Jon Peck
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Daniel Robertson
> Sent: Monday, January 21, 2008 7:43 AM
> To: [hidden email]
> Subject: Re: [SPSSX-L] Rasch Analysis
>
> Have you tried contacting the author?
>
> But without seeing the article, I'm skeptical that it could really be
> done using Logistic Regression/Log-linear Models only, which afaik would
> have been the only reasonable option in SPSS in 1983 (I was in
> middle-school then). IRT models generally are hierarchical models with
> items nested within examinees. The only saving grace that allows Rasch
> models to be estimated using non-specialized software is the fact that
> when only the difficulty parameter is estimated, the summed item score
> is sufficient to estimate the latent ability score without resorting to
> a factor analysis procedure. But in estimating parameters for each item,
> you would still need to be able to parse out items within examinees and
> treat examinees as random. I'm not aware of any way to do all that in SPSS.
>
> Here's a much more recent paper showing how it can be done using HLM:
> Kamata, A. (2002). /Procedures to Perform Item Response Data Analysis by
> HLM/. Paper presented at the annual meeting of American Educational
> Research Association, New Orleans, April 2002.
> http://garnet.acns.fsu.edu/~akamata/papers/AERA_2002.pdf
>
> If cost is an issue, you could always learn to use R -- the ltm package
> apparently handles latent trait models.
> <http://garnet.acns.fsu.edu/%7Eakamata/papers/AERA_2002.pdf>
>
> John Amora wrote:
>
>>  I am aware that there are a lot of packages that
>> would handle Rasch analysis, including Winsteps, SAS,
>> STATA, among others.  These packages can handle not
>> only dichotomous outcomes but also polytomous
>> outcomes.  But I dont have chances to use such
>> packages.
>>
>> Yes, Rasch analysis cannot be found as a procedure
>> within SPSS, but I think it is possible perhaps
>> through syntax because somebody already did it.  I did
>> some searching in the internet and I found that Gary
>> Phillips presented a paper on “Rasch Analysis using
>> SPSS” during the Annual Meeting of the National
>> Council on Measurement in Education in Montreal,
>> Canada, in April 1983.  Unfortunately, only abstract
>> of his paper is available.  The abstract is as
>> follows:
>>
>> "Ways in which the Statistical Package for the Social
>> Sciences (SPSS) can be used to perform some Rasch
>> analyses are described in detail. It is shown how SPSS
>> and a set of item calibrations can be used to estimate
>> person abilities, standard errors of measurement, test
>> characteristic curve, test information curve,
>> classification consistency on a mastery test, and item
>> and person fit statistics. Additional uses of SPSS are
>> suggested. Figures, tables, and sample printouts of
>> Rasch analyses using SPSC are appended. (GDC)"
>>
>>
>> If anyone has copy of Gary Phillips's full paper,
>> please share it.  Rest assured it would be handled
>> professionally.
>>
>> Thanks!
>> Johnny
>>
>> --- Daniel Robertson <[hidden email]> wrote:
>>
>>
>>
>>> SPSS does not have a procedure to perform Rasch
>>> analysis. Winsteps and
>>> Bilog are among the specialty packages that can
>>> perform a Rasch analysis
>>> directly. In a general sense, a Rasch analysis could
>>> be performed using
>>> any tool that can handle multi-level data with
>>> dichotomous outcomes and
>>> random effects, for example HGLM in the HLM package.
>>> But this excludes
>>> SPSS.
>>> Good luck.
>>>
>>>
>>>
>>>> Hi everyone,
>>>>
>>>> Rasch Analysis is one of the approaches of Item
>>>> Response Theory which is used to estimate the
>>>> probability of a correct response to a given item
>>>>
>>>>
>>> as a
>>>
>>>
>>>> function of item difficulty and person ability,
>>>>
>>>>
>>> where
>>>
>>>
>>>> item difficulty and person ability are parameters.
>>>> Please describe in detail how SPSS can perform
>>>>
>>>>
>>> this
>>>
>>>
>>>> analysis.
>>>>
>>>> Thank you!
>>>> Johnny Amora
>>>>
>>>>
>>>>
>>>>
>>>>
>> ____________________________________________________________________________________
>>
>

--
Daniel Robertson
Senior Research and Planning Associate
Institutional Research and Planning
Cornell University / irp.cornell.edu

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

["Marc Halbrügge"]

> If cost is an issue, you could always learn to use R -- the ltm package
> apparently handles latent trait models.
> <http://garnet.acns.fsu.edu/%7Eakamata/papers/AERA_2002.pdf>
You should also take a look at the MCMC package for R. Markov chain
monte carlo methods are a more robust alternative to maximum likelihood
estimation.
http://mcmcpack.wustl.edu/wiki/index.php/Main_Page

If I remember it right, you can find a worked example of a Rasch
analysis using the MCMC package somewhere on that page or in the package
docu.

Greetings
Marc

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD