Materials referred to in the paper Damaging the Case for Improving Social Science Methodology Through Misrepresentation: Re-asserting Confidence in Hypothesis Testing as a Valid Scientific Process
by James Nicholson and Sean McCusker can be found using the links below.
Simulations: we have replicated Gorard’s simulation with a number of different sample sizes for each group. The paper explains the relevance of the 3 different sample sizes we have used
Click on the following links to download the Excel files
Explaining the mathematics in the paper: A document giving a fuller treatment of the mathematics behind Lehr’s approximation, the exact calculation of minimum sample size, and what the exact test should be in each case. Note that the first part of the first section about Lehr’s approximation is accessible (we think!) to most readers, but there is some more sophisticated mathematics later in the document. The brief summary at the end is worth reading even if you can not work through all the detailed mathematics.
Click to download the MS-Word document