David Colquhoun. 2014. An investigation of the false discovery rate and the misinterpretation of p-values. Roy. Soc. open sci. 1:140216
Summary of the summary
If you use p = 0.05 to suggest that you have made a discovery, you will be wrong at least 30% of the time. If, as is often the case, experiments are under-powered, you will be wrong most of the time. This conclusion is demonstrated from several points of view. It is concluded that if you wish to keep your false discovery rate below 5%, you need to use a three-sigma rule, or to insist on p ≤ 0.001. And never use the word ‘significant’.
Any more thoughts to add?
I think everyone knows the basic conditional probability examples mentioned at the article. The problem is that we, falsely, have accepted that a small p-value guarantees us for the hypothesis 1 (since we reject the null hypothesis) and this does not seem to change at the near future...
It is the orthodox view at every scientific discipline which makes our life difficult and does not leave us to try and use a different approach. Additionally the statistical softwares reproduce the p-values without checking the power of the tests...
So, you speak about "reproducible research'. Well this is a big story.
Here in RG there exists an option to check if a paper is reproducible, but the results are not so encouraging, since all reviews are positive!
So, it is a difficult case...
Just to increase your 'anger' ( :) ) I will mention that in most articles with econometrical arguments the R2 is remarkable low, less than 10%, so the 'articles' explain less than 10% of reality and are proud for it!
Dear Demetris,
No feeling of anger. Just remarks and observations. Nature and its understanding is complex.
Do humans determine the speed of discoveries, and if so, how? Just by increasing financial investments?
Dear Marcvel,
I don't think that dominant elites care about what we, the thinking scientists, care. There exists a simple moto: Continue until it will stop working.
So, if there aren't any kind of degrading cancellations, they will insist on their current view, whatever you, me or even all RG community write...
A small example is the 'colotoumba' (somersault) that Greek government did just one week after its election...
So, this is the tragedy of modern Science: To be another one servant of what financial and political elites want to implement by giving 'scientific reasons' for accepting the 'new' policies...
- Badges
- Science topic
- Similar topics
- Composition Studies
- Articles