The aim of this chapter was to give you an introduction to some of the statistical techniques that you can use when trying to interpret the data you have collected as part of your mobile HCI study. These forms of data analysis were based on data collection approaches that are discussed in Chapter 3. As stated at the start of the chapter, it is important that you choose appropriate forms of data collection before you start your study. If you don't do this, you will find it very difficult to carry outany worthwhile statistical analysis of your data; garbage in, garbage out, as the old maxim goes.

The chapter started off with a discussion on how to describe your data both numerically and in tables and diagrams. By presenting your findings in a clear and unambiguous way, interested individuals will be able to understand the results of the statistical analyses you have presented and also, hopefully, understand why you have chosen to interpret your results in a particular way.

The actual tests themselves covered aspects of parametric and non-parametric analysis ranging from testing for differences (e.g. t-tests, Mann-Whitney U and Wilcoxon), looking for relationships between scores (e.g. correlation analysis) and also how to analyse and interpret categorical data (e.g. content analysis, CHI-square). Although you were asked to go through and calculate the test statistics by hand, I am aware that some of you may be familiar with commercially available statistical software packages and use these to help you...