Reader Comments

Post a new comment on this article

Clarification of citation context

Posted by tsbaguley on 11 Dec 2015 at 15:58 GMT

The authors present a useful and welcome set of research tools.

In addition, I agree with the authors that "many introductory statistics textbooks [1–5] do not even mention significance tests for correlations." However, at least one of the cited works not only covers the topic but also includes a worked example (by hand) and R code to compute a difference in correlations (as well as discussing whether it is sensible to compare correlations - as in many cases it would be better to compare unstandardised regression coefficients):

Baguley T. Serious Stats: A Guide to Advanced Statistics for the Behavioral Sciences. Basingstoke, UK: Palgrave Macmillan, 2012.

See pp.224-225. Possibly the authors pointed to these five texts as exceptions to the general pattern.

A related blog post included additional code and links to robust alternatives:

https://seriousstats.word...

Competing interests declared: I am the author of one of the cited books.

RE: Clarification of citation context

birkdiedenhofen replied to tsbaguley on 14 Dec 2015 at 11:22 GMT

I agree. Thomas Baguley's book is indeed a notable exception, and
unlike many other statistics textbooks it provides a thorough discussion
of significance tests for independent correlations. In the weblog
accompanying the textbook, significance tests for dependent correlations
are discussed as well. I can highly recommend his treatment of both
topics.

Competing interests declared: I am an author of the article.