Re: [R] a questtion about Cronbach's alpha more than 1

From: Ted Harding <Ted.Harding_at_manchester.ac.uk>
Date: Wed, 20 Feb 2008 19:55:05 +0000 (GMT)


On 20-Feb-08 17:34:33, Smita Pakhale wrote:
> Dear All,
> I am a novice in R and am working on my thesis for
> master's in Epidemiology & Biostatistics. While
> calculating Cronbach's alpha for the SF-36
> questionnaire, I got all 8 of them more than 1. Is it
> possible to get alpha more than 1. If yes, please help
> me understand it. I read about the negative alpha
> value on your web postings. It does not answer my
> question though.
> Thank you.
> Sincerely,
> Dr. Smita Pakhale, MD
> McGill University, Canada

You should not get alpha > 1 if you calculate it by the standard definition. You can of course get negative values which are arbitrarily large.

I wonder, therefore, whether
a) You have calculated it by an incorrect method or a

   non-standard definition
or

b) You have somehow been presented with negative alpha < 1

   as positive values > 1.

Can you provide more detail about how you obtained the values you are writng about?

Best wishes,
Ted.



E-Mail: (Ted Harding) <Ted.Harding_at_manchester.ac.uk> Fax-to-email: +44 (0)870 094 0861
Date: 20-Feb-08                                       Time: 19:55:00
------------------------------ XFMail ------------------------------

______________________________________________
R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Wed 20 Feb 2008 - 19:58:42 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Wed 20 Feb 2008 - 22:30:15 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive