|
The precise ways in which the data collected were to be analysed were not specified when the original protocols for the studies were prepared. In hindsight this could possibly have been done. When such situations arise it is important to prepare an addendum to the main protocol setting out the data to be used, the calculations to be done, the hypotheses to be tested and the methods to be used for statistical analysis. Such a framework allows the researcher to follow a step-by step approach to the analysis. We have decided under Study design to use time units of six months to coincide with the main wet and dry seasons.
|
