Negative results are to be expected in experimental science. Most researchers have lab books and files full of negative data, blurry gel photos and graphs showing no significant differences. When I was still working at the bench, the normal procedure was to shelve this data and keep working until you had "publishable results." You would tell your lab colleagues, collaborators and anyone who asked at a conference about it but this information was not made available to the wider scientific community.
If I was to hazard a guess at how much negative data I produced during my time in research, I would say it was approximately 60% of my work. This includes data from my student projects, PhD, postdoc and numerous pet projects. I also have several files containing work from unfinished projects that had to be abandoned because contracts ended or because I couldn't dedicate the required time to finish them. This prompted me to question how much global science data is shelved or forgotten each year. A recent letter published in The Lancet by Michael Galsworthy and colleagues reported that more than 50% of research funded by two of the European Union's Framework Programmes (FP5 and FP6 - ended in 2006) produced no traceable publications. Could this be partly due to a high proportion of negative data?
Researchers are often loathe to write up their negative results and journals are rarely keen to publish negative data because it is not thought to be important enough. Yet, the publication of this data is an important contribution to the community - it would save time, resources and energy. As long as the scientific method and reasoning behind the results is sound, the information is valuable and it is an opportunity for the researcher to add another publication to their belt. Ideally, these results should not be called "negative" at all, as the name itself suggests that the data is less interesting. Perhaps they should be called "unexpected contributions?"
An initiative by the open access journal F1000 Research is currently encouraging researchers to share this sort of data. They have waived their article processing fee for papers built around negative results until the end of September 2013, so if you have results burning a hole in your lab book now is the time to write them up. Another open access publisher, BioMed Central, is also supporting the publication of negative results through its Journal of Negative Results in Biomedicine, which accepts papers "that promote discussion of unexpected, controversial, provocative and/or negative results." Hopefully this indicates that the landscape is changing and that more journals will start to accept papers based on negative results in future.
**This post was originally posted on the Journal Club 3.0 blog on 11.08.13.**