Letter to the editor,

It is now a commonplace observation that jointly-written papers are cited more than single-authored ones. Observations to this effect have been reported by many authors.

However, in designing these studies it is important to control for all the other relevant variables whilst manipulating the one in question.

My contention is that this has not been done in this research, and thus that the conclusions reached so far may be invalid.

I would have thought that one of the key variables that should be controlled for would be where the articles were published. To make a fair comparison, the citation rates of articles written by single authors should be compared with those written by the same authors with colleagues in the same journals.

In an earlier work I assessed the citation rates of single authors compared with joint ones, keeping the journals constant (Hartley, attached). In point of fact I found this very difficult to do. Few authors in fact appeared to write many single and jointly authored articles in the same journals.

Nonetheless, I was able to do it for 8 authors. But these studies were limited because, as just noted, these authors did not publish many articles in the same journal. And the paper I wrote on these issues was roundly rejected—twice—by eminent journals because of the limited data.

One referee missed the point entirely and addressed my attention to 20 or so more papers on joint authorship that I had not cited. But that is not the issue. I was trying to compare the citation rates of authors who published studies alone or with their colleagues in the same journals.

Of course I may not have been looking in the right disciplines. But I challenge the readers of this letter to find me any study of joint and single authorship where the papers appear in the same journal!