12 June 2014

An update on #losadagate

Since I've been asked a few times what's happening on "the Losada matter", I thought I'd write it up here so I can point people to it.

The story so far: On July 15, 2013, American Psychologist published our article and Dr. Fredrickson's response [PDF] online.  We had expected to have the last word, in a standard A-B-A sequence, but this was initially denied.  American Psychologist's logic was that the original 2005 article by Fredrickson and Losada [PDF] was A, our article was B, and Dr. Fredrickson's reply to us closed the matter.

We were not very happy about this, to say the least.  Our article addressed two other papers published in other journals (Losada, 1999; Losada & Heaphy, 2004 [PDF]) as well as Fredrickson and Losada's.  Furthermore, we initially had our article summarily rejected because (according to AP) it was a comment, and comments are only accepted for 3 months after the appearance of an article.  (Quite how science is meant to self-correct if any mistakes which are not identified within 90 days become officially true, is an interesting but separate issue.)  Only after we wrote directly to the CEO of the American Psychological Association did we get an offer for our article to be reviewed (which was done in a very thorough and professional manner, I should add).  We expected that Dr. Fredrickson would be invited to reply --- indeed, we encouraged this, on the basis that we would have the last word, as is customary in these discussions.  So when we were later told that we were not going to be offered a final response, we were not very happy.

We considered writing our reply and sending it to a different journal, combined perhaps with some complaining via blogs or the news media.  However, in the end we decided to grit our teeth and embark upon the APA's publications appeal process.  It took a lot of effort (almost all by Harris Friedman), but in the end, we were allowed a 1,000-word reply --- and with some further negotiating, this became 2,000 words.  (I note, in passing, that when AP's position was that Dr. Fredrickson's reply was the end of the A-B-A sequence, she was allowed 5,800 words and 60 references, almost none of which addressed the points made in our article.)

After review, our response to Dr. Fredrickson's reply to our article is now in press at American Psychologist.  You can see the final draft version here [PDF].

But wait --- there's more!

After our article, together with Dr. Fredrickson's reply, appeared in the December 2013 print edition of American Psychologist (which, as part of the benefits of APA membership, has a print run of about 100,000, meaning that my photo is now in the bathroom of most of the psychologists in North America), several people wrote to the editor with comments.  Five of these were selected for publication.  After some further discussion with the editorial team, we obtained the right to respond to these comments, which one would have thought would be automatic... never mind.  We replied to each of the comments, concentrating in particular on the three which critically addressed our article.  Our reply is now also in press, but I don't want to share the full draft in this case because to put it in context requires reading the readers' comments, and I'm not in a position to share them.  (But for those who enjoyed the first article: there are a few zingers in this one to look forward to.)

I'll end on one of the things that didn't make the cut of our reply to those readers' comments.  On p. 821 of the December 2013 issue of American Psychologist, Fredrickson and Losada jointly authored the "withdrawal" of Losada's mathematical model "as invalid".  Yet, if you visit the web site of Losada's consulting company, you will find the model everywhere, right down to the stylised "Lorenz butterfly" which forms the favicon of the site.  Quite how the model can simultaneously be withdrawn as invalid while simultaneously forming the backbone of a training programme that has apparently been delivered to companies such as Apple, Boeing, GE, and AT&T (cf. the "Clients" tab of the Losada Line Consulting site) is, I suppose, just another miracle of mathematical logic to which we mere mortals are not to be allowed access.

PS: Hot off the press: American Behavioral Scientist, which published the Losada and Heaphy (2004) article, has recently issued an Expression of Concern about it.  The last paragraph is particularly intriguing; have you ever heard of the editors of a journal, whether permanent or a guest team for a special issue, taking an incoming article and asking a grad student to polish it up a bit for their target audience?  (I note that one of the guest editors of the special issue in question, Kim Cameron, is a senior professor at the Ross School of Business at the University of Michigan, where Emily Heaphy was studying at the time.)