ABSTRACT
There are a number of reasons why a data set needs to be checked prior to any statistical analysis being conducted. There could be values which are not part of the legitimate range for the measures used. There could be missing data. There could be participants whose data have a disproportionate influence on the results. The statistical test which seems appropriate to test a hypothesis may have specific assumptions about the data. A number of solutions are offered. However, often there is no perfect solution for dealing with such data and it is important to conduct sensitivity analysis to check whether the choice of solution has affected the results of the analysis.
