In contrast to testing the null hypothesis ofno difference between treatments, we might ask the question 'how large is the difference between the experimental and control treatment?' The best estimate of this difference is given by the observed difference from the trial, and is called a point estimate. Alongside this, some indication of the precision of the estimate is required. This is provided by the confidence interval. It is conventional to quote the 95 per cent confidence interval, which gives a range of values around the estimate that we can be 95 per cent confident contains the true underlying value of the difference between treatments.

0 0

Post a comment