[Rasch] Rating Scale or Partial Credit?

Mike Linacre (RMT) rmt at rasch.org
Wed May 17 13:16:18 EST 2006

Dear Tom:

Thank you for your question. The literature is contradictory, so how about 
this ...

The threshold distances can exaggerate differences between the item 
structures. Rasch estimates are fairly robust against threshold variation. 
So, instead, compare the model ICCs (expected score ogives). See the first 
Figure on www.rasch.org/rmt/rmt151j.htm

If the differences are large enough to cause you to come to different 
conclusions about the responses, then the rating scales are different and 
the partial credit model is indicated. But check to make sure that there 
are at least 10 responses in each category per item. Small category 
frequencies can lead to idiosyncratic results.

Mike L.

At 5/16/2006, you wrote:
>I am analyzing data from a personality instrument that employs 5-option 
>Likert-style items.  Inspection of the results shows that the items vary 
>greatly in terms of how well-spaced and well-ordered their thresholds 
>are.  Should I be using a partial credit model, because empirically I know 
>the thresholds are different for each item?  Or should I use a rating 
>scale model, because the items were created with the intention of 
>producing consistent thresholds?

Mike Linacre
Editor, Rasch Measurement Transactions
rmt at rasch.org www.rasch.org/rmt/ Latest RMT: 19:4 Spring

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20060516/93f7845c/attachment.html 

More information about the Rasch mailing list