Talk:Publication bias

"A study of published and unpublished clinical trials found that positive results in the sample were three times more likely to be published."

Ah, but if the positive results had been no more likely to be published than the negative ones, would that study have been published??? Wehpudicabok (talk) 08:23, 7 April 2013 (UTC)

Self-interest

OK, I understand why a researcher might feel awkward about publishing negative findings, but, in my limited academic promotion experience, a CV of negative findings wouldn't necessarily be bad. Am I naive or just in the one fair university? FWIW, I got my PhD based on a negative finding. MarmotHead (talk) 16:45, 21 July 2014 (UTC)

My experience is that our research group never published negative results over a period of many years; this may be because we were more data-oriented rather than experiment-oriented. An allied research group performed a 2-year study and did not published the negative results. There were some flaws in the study but this was more on the results side than on the methodology side. I don't know why they did not publish it. The lead researcher had some emotional and formerly financial interest in getting a positive result. An unrelated research group performed a similar experiment several years later and did published the negative result. Bongolian (talk) 04:38, 12 August 2014 (UTC)

My MSc. research paper was based on a negative finding as well, so I'm with MarmotHead. I imagine scientists employed outside of academia may have more of an aversion to publishing results that could make their employers unhappy (whether that be negative or positive results), but who knows? Among academics, among whom research is often done for the sake of research, I would be surprised if many avoided publishing negative results. - Grant (talk) 05:05, 12 August 2014 (UTC)

I guess I'll happily stay in the halls of academia! To be fair, I never submitted dissertation work to journals. I was tired of it, but, if it were positive, maybe I'd've been just excited enough to do the extra work. MarmotHead (talk) 15:20, 12 August 2014 (UTC)