Evans, Dolan & Toriola (1997)
Detection of intra- and cross-cultural non-equivalence
by simple methods in cross-cultural research:
evidence from a study of eating attitudes
in Nigeria and Britain

Page started by Chris Evans in 2017, last updated 11.xi.17

The history of this paper is a little odd. It's one my earlier papers (20 years old: scary) and one of my favourite papers because I think the method is simple and very broadly applicable and the findings are very clear and strong.

It's not very clearly written and these days I would prefer to use bootstrapped confidence intervals for Cronbah's alpha rather than the analytic ones based on the assumptions of Gaussian distributions that we used in this paper. However, I have yet to see major differences between the two methods with real questionnaire data. Equally, though I'm sure it's still a great program, I haven't used SAS for about 18 years as, not long after this paper came out, I left St. George's and no longer had access to SAS. Pretty much since then I've tried to use the wonderful R (https://www.r-project.org/) for all my statistical work. R is more flexible than SAS, has a better syntax (in my view) and far more extensions, above all it's free, open source and runs on most systems, certainly Linux, Macs and Windoze. I used R to create a web page that will give you the analytic CI around any observed alpha (https://www.psyctc.org/stats/R/Feldt1.html) and you're welcome to use that.

The "odd history" is that the then publisher of the journal (Kurtis) apparently went bust AND lost their copy of our paper so if you go to the online site of the journal (now with Springer) you see a hole in their contents page where this paper should be (https://link.springer.com/journal/40519/2/2/page/1).

I found this out a year or so ago when trying to answer a request for the paper. I then started searching for my own paper copy and finally established that (a) I had lost all paper copies I had had and (b) it was nearly impossible to convert the electronic copy I had, written when I used Nota Bene as my word processor, to something that really reflected the printed version. Fortunately, after some lost Emails, the current editor of the journal, Dr. Massimo Cuzzolaro, sent me a photocopy of his own paper copy of that journal and I scanned that.

Unfortunately, at the moment, because it's a scan of a copy, the PDF is huge and I can't seem to reduce the file size successfully. Even more unfortunately, I don't think Springer are in any hurry to put a copy of the paper up on their web site so I am uploading it here and at that page about the CI around Cronbach's alpha

If the paper is still of interest or some use to you, do please get in touch with me using the webmail form below.