[Rasch] Scaling ranks

Mike Linacre (RMT) rmt at rasch.org
Mon Nov 10 10:25:25 EST 2008


Thank you for your question, Don.

Rasch-analyzing your rankings as ratings on a partial-credit rating-scale 
works well. JAM 7:1 p. 129-136. This also allows for  tied rankings and 
missing data. See also www.rasch.org/rmt/rmt82p.htm .

An example in which golf tournaments are used to rank players is shown at 
www.winsteps.com/golf2006.htm

Mike L.

At 11/9/2008, you wrote:
>   I have a survey wherein respondents rank the importance of 20 items (a 
> difficulty task, I know).  I'd like to aggregate the responses and scale 
> the results.  I could translate the ranks into paired comparisons and 
> then apply Thurstone, but it seems like there would be a much easier way 
> with Rasch.  Can someone point me in the right direction?

Mike Linacre
Editor, Rasch Measurement Transactions
rmt at rasch.org www.rasch.org/rmt/ Latest RMT:  22:1 Summer 2008
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20081109/63c352e5/attachment.html 


More information about the Rasch mailing list