[Rasch] WINSTEP or FACETS for Speaking Rating scales

Stone, Gregory Gregory.Stone at UToledo.Edu
Tue Jun 26 04:23:16 EST 2012


That is a very common question.  The simplest answer is that if all you were interested in was investigating a traditional test or survey, wherein a person (student) interacted directly with an item, you would engage with the basic Rasch model.  WINSTEPS works primarily with the common Rasch model - rating scale, dichotomous, partial credit, etc.  The rule of thumb being that the measurement is direct - the person interacts with the item directly.

In your case, students are engaged in a speaking test, being rated by a judge.  In order to best capture this relationship, you will need to use the Multi-faceted Rasch Model (also known as the Many-Faceted Rasch Model), which will include (account for) the judges being used to assess your students.  WINSTEPS does not allow for the capturing of three facets - FACETS, does.  The difference does not relate to your use of a survey or a rating scale, but rather because of your use of a judge.  Both WINSTEPS and FACETS are excellent programs that I believe you will find invaluable, so I doubt you'll find either a problem purchase.  Hardly a week goes by that I don't use one or the other.  There are options in WINSTEPS not available in FACETS as well that you may find useful for general surveys and tests.

I'll be happy to assist you with FACETS should need help.

On Jun 25, 2012, at 1:10 PM, Mohamad Imanian wrote:

Dear All,

As I have communicated with the group before, rating scales can be validated by WINSTEP and I have bought the software. Today I encountered a study which used FACETS for validating a speaking rating scale. To give you a preview of the problem
I have a rating scale with seven rating criteria. 5 Judges assess the performnce of the students in the speaking test.
My big question is why some people use FACETS not WINSTEPS although the Winsteps.com<http://Winsteps.com> (homepage)has suggested WINSTEP for rating scales? It seems I have to sell my WINSTEP and buy a FACET. Right?
Warm Regards
Mohamad Imanian
Kashmar Payameh Noor University

Rasch mailing list
Rasch at acer.edu.au<mailto:Rasch at acer.edu.au>
Unsubscribe: https://mailinglist.acer.edu.au/mailman/options/rasch/gregory.stone%40utoledo.edu

Gregory Stone
Research and Measurement
The University of Toledo
Toledo, OH 43606

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20120625/d985d2fe/attachment.html 

More information about the Rasch mailing list