ABSTRACT
As outlined in the first chapter, one of the main uses of statistics in the social sciences is to determine the probability of three different kinds of events or outcomes happening. The first kind is estimating whether a sample drawn from a population is representative of that population in terms of a specified characteristic. Suppose, for example, we know that the population of interest consists of an equal number of women and men and we draw a sample of five women and seven men. How likely are we to obtain this particular number of women and men by chance? The second kind is deciding whether one sample differs from another in terms of a particular variable. Assume, for instance, that ten people are randomly assigned to either having tutorials or not having tutorials. Three of the five having tutorials receive higher coursework marks than three of the five not having tutorials, while there is no difference between the other two. What is the likelihood of finding this difference by chance? The third kind is estimating the probability of two events being associated by chance. Imagine, for example, that three of four people having tutorials do well in their coursework compared with three of six people who do not have tutorials. How likely are we to come across such a relationship by chance?
