Science is not a string of successes. It has its share of errors and misconduct, and acknowledging them does no disservice to the value of scientific research that stands the test of time. So it was a pleasure to review a new book, Brilliant Blunders, by Marco Livio, for the New York Times Book Review. No one is perfect, Livio shows us, even some of the greatest scientists of the modern age. Check it out.
It’s been very gratifying to listen to the conversation that’s been triggered by my essay in this Sunday’s New York Times on scientific self-correction. Here, for example, is an essay on the nature of errors in science by physicist Marcelo Gleiser at National Public Radio. Cognitive scientist Jon Brock muses on how to get null results published.
I also got an email from Eliot Smith, the editor of the Journal of Personality and Social Psychology who accepted the controversial clairvoyance paper I described in my essay. I wrote that three teams of scientists failed to replicate the results and that all three studies were rejected by the journal because they don’t accept simple replication studies.
Your recent Times column stated the following:
Three teams of scientists promptly tried to replicate his [Bem’s] results. All three teams failed. All three teams wrote up their results and submitted them to The Journal of Personality and Social Psychology. And all three teams were rejected — but not because their results were flawed. As the journal’s editor, Eliot Smith, explained to The Psychologist, a British publication, the journal has a longstanding policy of not publishing replication studies. “This policy is not new and is not unique to this journal,” he said.
In fact, JPSP has received only one submission reporting failed replications of Bem’s studies. I did reject that paper based on the reason your column stated.
And to put that in context, I also rejected another submission to the journal that reported successful replications of some of Bem’s studies, on the same grounds.
I believe that a published correction is warranted; the difference between one and three papers is quite meaningful in this context.
I’ve passed on Smith’s message to my editor at the Times, and I’ll also take this opporunity here to apologize for the error.
I’m not sure how meaningful it is in the context of my essay, since my point was that policies against publishing replication studies get in the way of science’s self-correction. But a mistake is a mistake.