Hard on the heels of my dissection of the American Journal of Clinical Nutrition paper in my previous post, I came upon an article in PLoS, a free access journal, that should be required reading for every publisher and every reader of scientific articles.
Although this paper can be downloaded in full by anyone, because I think this paper is so important, I’m going to include the entire Abstract in this post.
There is increasing concern that most current published research findings are false. The probability that a research claim is true may depend on study power and bias, the number of other studies on the same question, and, importantly, the ratio of true to no relationships among the relationships probed in each scientific field. In this framework, a research finding is less likely to be true when the studies conducted in a field are smaller; when effect sizes are smaller; when there is a greater number and lesser preselection of tested relationships; where there is greater flexibility in designs, definitions, outcomes, and analytical modes; when there is greater financial and other interest and prejudice; and when more teams are involved in a scientific field in chase of statistical significance. Simulations show that for most study designs and settings, it is more likely for a research claim to be false than true. Moreover, for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing bias. [my italics] In this essay, I discuss the implications of these problems for the conduct and interpretation of research.
Anyone truly interested should read the entire paper. It has some statistical math sprinkled throughout that might seen a little foreboding to the non-mathematically incline, but skip over them and just read the text because the math doesn’t detract from the message of the author.
Remember this paper the next time someone tries to prove something to you by the findings in a single (or even several) study. As I pointed out in my previous post, scientific studies should be read thoroughly and critically; this paper shows why.
I was doing some perusing of your old blogs and found that there is a bad link in this [Most scientific articles are false – November 11, 2005] one. The link “previous post” in the first line returns an error page. Thought you might like to know and get someone to fix it.
Thanks for the heads up. It should be fixed now.