ABSTRACT
This chapter discusses several important significance tests appropriate for differing kinds of problem frequently encountered in the social sciences. The underlying problem in research design is that one often has insufficient understanding of which variables or attributes to use for matching purposes or an inability to select appropriate cases even when one has this knowledge. There is a need for procedures directly comparable to tests involving differences of means and proportions which can be used with ordinal scaling. These methods are particularly useful in the social sciences, since one so often encounters ranked data, and they can also be usefully employed even with some interval level data – for instance, when the rather strict conditions needed for the application of the difference of means test fail to apply.
