When we do research in Medicine it’s usually to test whether a new treatment works (by testing it against placebo) or better than the established treatment we’re already using.
At the beginning of our study we have to come up with a null hypothesis (denoted as H0).
The null hypothesis is a statement that assumes no measurable difference between whatever you’re studying.
The null hypothesis is therefore usually something along the lines of:
‘Drug A won’t be better than Drug B at treating this condition.’
We then set out to test this null hypothesis. If we find Drug A is better than B then we reject the null hypothesis and conclude Drug A is the superior treatment. If Drug A is found to be no better (i.e. the same or worse) than Drug B then we accept our null hypothesis and conclude that Drug A is non-superior (or inferior).
Error comes when we either wrongly reject or wrongly accept the null hypothesis.
Error means we come to the wrong conclusion. There are two types of error, the next blog will look at the first, Type I Error.