ABSTRACT
This chapter describes a range of different problems which can exist within a data set, how they can be identified and what can be done to solve them. They include values which are not sensible ones for the variable being considered, missing data, values which might be affecting the results and factorial designs with unequal sample sizes. The chapter introduces the notions of intention to treat and bootstrapping and suggests an order in which data checks should be conducted. It concludes with ways to deal with issues raised in factorial ANOVA designs are unbalanced (sample sizes are unequal).
