|
Typically, science journalism covers new studies that have not yet been replicated with very little to indicate that this is really a fairly preliminary result. Sometimes these studies have very small sample sizes. Sometimes they're sponsored by organization that have a vested interest in a particular outcome. Still, the findings get repeated as fact, sometimes to be contradicted later. This can lead to a general distrust in science, as well as a confused public. So here's my idea: I'd like to create a labeling system, somewhat similar to the warning labels on video games and so forth, that could be used to provide some context for writings about scientific studies. -Small sample size -Medium sample size -Large sample size -Potential conflict of interest -Unreplicated study -Study replicated: *** Times Once a system was worked out, I'd pay someone to design icons for the labels. The *s could be replaced with some sort of graphic. I'd use the system at Technoccult, obviously, but release it to the public so that other bloggers and journalists could use it as well. Publications could put them at the beginning of articles about studies, or incorporate it somewhere into the design to tip readers of easily and prominently as to the status of the study. |
From http://technoccult.net/archives/2011/06/17/labelingrating-system-for-scientific-studies/