Thin thighs don't come in a jar
This humorous example is nice because it is overwhelmingly conclusive. The small sample size would be criticised by the manufacturers of the cream used (and even others!); note that half of 17 is quite a small number. These are the types of studies, however, which are needed to invalidate claims of some cosmetic manufacturers.
It would be interesting to know whether the 17th woman had used the treatment or the placebo! Students might like to consider ways they would redesign the experiment if approached by the company to do another study (one obvious idea would be to continue the treatment for longer than eight weeks).
Where to next?
Student Questions for this article
Index - Related articles
Index - Inference
Numeracy in the News - Main Index