Letter to HERA's Editor: Response to HERA Editorial on 'Negative Findings'

摘要

A long overdue discussion of "negative findings" in research investigations appeared recently (Johnson and Domingo 2013). Their discussion focuses on the issue of whether or not to publish the results of studies that produce only or mainly negative results. "Negative" is defined as lack of statistical significance for a specific effect. I agree with Johnson and Domingo (2013) that the term "negative study" is ambiguous. In a health study, a statistical negative result does not necessarily mean that the biological effect observed is not important. A negative result does not necessarily indicate that no further action should be taken. A negative result, based solely on a lack of statistical significance, often may be the result of an inadequate sample size that lacks the power to achieve statistical significance for the biological result obtained. Ultimately, it is the size of the biological effect that is important. If the observed biological effect in a study is large enough to be important, then further study may be warranted.