[Rasch] Differences in alpha for Winsteps and R

Mike Linacre mike at winsteps.com
Wed Jul 12 07:40:21 AEST 2017


Thanks for your question, Stuart.

The answer is "(approximate due to missing data)" !!

Cronbach Alpha is defined for complete data, so when data are missing, 
assumptions must be made about those data. Winsteps computes Alpha as 
though the data are complete, and ignores the fact there are missing 
data, except for adding the "approximate" message.

Stuart, what happens when you analyze a single form in Winsteps? Do the 
Alphas of Winsteps and psych::alpha() agree?

OK?

Mike L.

On 7/12/2017 4:00 AM, Stuart Luppescu wrote:
> Hello fellow Raschies, I'm writing reports on a number of assessments,
> including Rasch and non-Rasch statistics. It's easier to calculate the
> Cronbach's alpha using the psych package in R than to read it from the
> Winsteps output so that's what I'm using for that. But I've noticed
> that what psych::alpha() calculates is rather different from what
> appears in Table 3 in Winsteps. For example, for 1 assessment, Winsteps
> reports (for all forms together):
> CRONBACH ALPHA (KR-20) PERSON RAW SCORE RELIABILITY = .86 (approximate
> due to missing data)
> but psych::alpha() gives 0.55 or 0.68 (depending on the form)
>
> Anyone know why there should be such a difference?

-- 
Mike Linacre, mike at winsteps.com or winsteps1234 at gmail.com
Winsteps: www.winsteps.com/winsteps.htm - current version 4.0.0 - June 2017
Facets: www.winsteps.com/facets.htm - current version 3.80.0 - March 2017
Winsteps and Facets Tutorial PDFs at www.winsteps.com/tutorials.htm
Rasch Forum: http://raschforum.boards.net
Sydney, Australia, Winsteps Workshop, August 10, 2017.


More information about the Rasch mailing list