[Rasch] neutral categories in scales
david.andrich at uwa.edu.au
Sat Feb 23 15:50:49 EST 2013
Have a look at
Andrich, D., de Jong, J. H.A.L & Sheridan, B.E. (1997).Diagnostic opportunities with the Rasch model for ordered response categories. Chapter 4. In J. Rost and R. Langeheine (Eds.), Applications of Latent Trait and Latent Class Models in the Social Sciences. Waxmann Verlag GMBH: Münster and New York, pp. 58-68.
If you cannot locate it, I am happy to send you a pdf separately.
David Andrich, BSc MEd W.Aust., PhD Chic, FASSA
Chapple Professor david.andrich at uwa.edu.au<mailto:david.andrich at uwa.edu.au>
Graduate School of Education
The University of Western Australia
M428, 35 Stirling Highway, Crawley, WA, 6009
Telephone: +61 8 6488 1085; Fax: +61 8 6488 1052
From: rasch-bounces at acer.edu.au [rasch-bounces at acer.edu.au] On Behalf Of Kulas, John T. [jtkulas at stcloudstate.edu]
Sent: Saturday, 23 February 2013 9:26 AM
To: rasch at acer.edu.au
Cc: 'Alicia Stachowski'
Subject: Re: [Rasch] neutral categories in scales
Hi Ian –
Alicia Stachowski and I have some studies (from which you can back-reference additional support for “believing it is not implicating a continuum of the latent trait”):
1) The predominant endorsement intent is not consistent with a unidimensional continuum (and in fact tends to be conditional/context-specific):
Kulas, J. T., & Stachowski, A. A. (2009). Middle Category Endorsement in Likert-Type Response Scales: Associated Item Characteristics, Response Latency, and Intended Meaning. Journal of Research in Personality, 43, 489-493.
2) Further support for a conditional/non-unidimensional continuum predominant orientation:
Kulas, J. T., & Stachowski, A. A. (in press). Respondent Rationale for neither agreeing nor disagreeing: Person and Item Contributors to Personality Assessment Middle Category Endorsement. Journal of Research in Personality.
We have another that possibly provides some evidence toward your “reliability” question, but keep in mind our study focused on personality assessment with a strongly agree to strongly disagree 5-point scale (with neither agree nor disagree being the middle category) and that our sample was students. The test-retest reliability estimates of scores (yielded from this sample and context) were not much affected by including or excluding the middle category, but I probably wouldn’t advocate using our results as a basis for your decision:
Kulas, J. T., Stachowski, A. A., & Haynes, B. A. (2008). Middle response functioning in Likert responses to personality items. Journal of Business and Psychology, 22, 251-260.
Beyond these articles that I think address your specific questions, I would strongly advocate taking a look at your specific scale and how it’s functioning (we are, after all, on a Rasch listserv). The empirical support will likely be far more convincing and informative than anything I’ve listed above :):
Linacre, J. M. (2002). Optimizing rating scale category effectiveness. Journal of Applied Measurement, 3, 85-106.
Good luck and let me know (off list) if you want any of the reprints or reference lists from the stuff that Alicia and I have taken a look at…
From: rasch-bounces at acer.edu.au [mailto:rasch-bounces at acer.edu.au] On Behalf Of Ian Blackman
Sent: Friday, February 22, 2013 6:48 PM
To: rasch at acer.edu.au
Subject: [Rasch] neutral categories in scales
Could I please have some advice.... I have some data with 5 categories with a middle category being neutral (rated as 3). I wish to remove that neutral category as i do not believe it is measuring a continuum of the latent trait being estimated. Is there any literature that supports this belief and secondly, would the reliability of the items be influenced if the neutral category was recoded to zero. Many thanks
Dr Ian Blackman
phone: +61 8 82013477
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Rasch