Re: st: Creating Scale with Ordinal Data using IRT

[Sara Mottram <s.mottram@cphc.keele.ac.uk>]

I would be quite interested in this myself. My understanding is that 
Rasch analysis (a form of IRT) can be performed in Stata 
http://www.stata.com/support/faqs/stat/rasch.html, although I haven't 
actually tried to do it. I'm not sure that a Rasch model will fit your 
data though - I don't really know enough about this area (still working 
on that!) but I think it may be restricted to a binary response 
variable. I think some forms for IRT require specialist software (e.g. 
RUMM). I'm not sure this is entirely helpful to you, maybe someone else 
could clarify this further?
Sara

n j cox wrote:

> The abbreviation "IRT" which is here assumed familiar to
> all I understand to mean "item response theory".
>
> A -findit item response theory- finds some items. -findit IRT-
> finds some more (and some false positives).
>
> I have no idea whether any is the answer to your question.
>
> Nick
> n.j.cox@durham.ac.uk
>
> gz6931@albany.edu
>
> I have a two level structure, items within persons. There are five items,
> each measured on a five-point Likert Scale (responses are 1,2,3,4,5). I
> need to create a scale for each person based on the five items using IRT
> instead of the simple average of items in order to make use of all of the
> data. How would I go about this? I can first reshape the data, but where
> do I go from there?
>
> For example, my data may look like this originally:
>
> person  item1 item2 item3 item4 item5
> 1       3     2     3     4     5
> 2       5     4     2     4     3
> 3       3     2     1     2     2
>
> I can reshape the data into:
>
> person  item   scale
> 1       1      3
> 1       2      2
> 1       3      3
> 1       4      4
> 1       5      5
> 2       1      5
> 2       2      4
> 2       3      2
> 2       4      4
> 2       5      3
> etc.
>
> What commands might I use to create one scale for each person using IRT
> based on the responses to the 5 items?
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
--
Sara Mottram	
Research Assistant: Biostatistics
Primary Care Musculoskeletal Research Centre
Primary Care Sciences
Keele University
Staffordshire, ST5 5BG
Tel:  01782 584711
Fax:  01782 583911 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/