On Instagram @AndreaHardyRD, I talk all about how important it is to understand the biases in research, the ways to find unpublished data and the reasons why unpublished data is still important.
Are null findings dull findings? (#scicommunity Thursday)⠀
⠀
Null findings are research that do not find anything significant, novel, or exciting in a study. And, because of that, null findings are less likely to be published – leading to bias in the research!⠀
⠀
Publication bias is the bias towards only publishing data with significant outcomes. Because who wants to read a research paper that found the intervention had no benefit? ⠀
Well – we should ALL want to read those papers, because they’re crucial in developing our cumulative knowledge on a subject. For example, if a research trial found that omega-3 supplementation lowers the risk of heart disease, and then it’s published in a prestigious journal, but the 3 similar trials found no significant results and weren’t published – how could you possibly weigh and interpret the data?⠀
⠀
You can’t. Because you don’t have all the information. So – how do you find it? ⠀
⠀
Ensuring you identify and include unpublished outcomes by reaching out to those with registered trials, but no reported outcomes, seeing what PhD theses may be available for review but aren’t published, and contacting companies that perform research for their unpublished data to name a few.⠀
In the interest of ensuring we have as full of a picture as possible, all publications should be made available so that researchers can properly weigh the evidence – I know mandatory publication is controversial and the feasibility is hard, but research shouldn’t ‘go to die’ if the findings aren’t sexy enough to meet a publications standards. Null findings are important too! ⠀
⠀
Tools I use as someone not actively involved in research include:⠀
– Government trial registration websites (clinicaltrials.gov)⠀
– ResearchGate – where researchers can share their papers and findings, regardless of outcome, and a great place to find PhD Theses ⠀
⠀
Any other favourite tools you use to find unpublished data? Any other tips?⠀
⠀
DOI: 10.2147/OAJCT.S34419
Recent Comments