A Meta-Analysis Is Only as Good as What Went Into It
Pooling studies sits at the top of the evidence pyramid, which is why the word carries so much weight in an argument. Combining…
Reading the Evidence
Sample sizes, funding, preprints and how to read a study cited at you.
21 articles · updated August 11, 2026 · page 2 of 2
The Healthy User Effect and Who Buys Supplements
People who take vitamins also exercise more, smoke less and see doctors sooner, and studies keep finding out about the vitamins.…
Blinding Is Harder Than It Sounds and Often Fails
A double-blind trial is the standard everyone cites, and it is easier to declare than to achieve. Whether participants could guess…
Twenty Outcomes and the One That Reached Significance
Measure enough things and something will look impressive by chance alone. The defence against this is declaring what you will…
Predatory Journals and the Citation That Looks Legitimate
A reference list is meant to be reassuring, which is why the appearance of publication has become purchasable. Some journals will…
Running an Experiment on Yourself, and Where It Stops Working
Testing something on yourself is a real method with real rules, and it beats guessing. It also cannot separate you from the…
If More Does Not Do More, Something Is Wrong
Real effects usually get bigger with a bigger dose until they plateau or turn harmful. A treatment that works identically at any…
No Evidence of Effect Is Not Evidence of No Effect
The two statements sound almost identical and mean opposite things. Both sides of a wellness argument routinely use whichever one…
Number Needed to Treat and the Headline It Deflates
A treatment that halves your risk sounds transformative until you ask how many people have to take it for one to benefit. That…