ABSTRACT
Imagine that we have data on fifty-six students regarding which faculty they belong to at a university (see Table 5.1). The university has only four faculties: engineering, pure sciences, arts, and social sciences. Even though fifty-six is not a large number on which to have data, it is not particularly easy to see how students are distributed across the faculties. A first step that might be considered when summarising data relating to a nominal variable such as this (since each faculty constitutes a discrete category) is the construction of a frequency distribution or frequency table. The idea of a frequency distribution is to tell us the number of cases in each category. By ‘frequency’ is simply meant the number of times that something occurs. Very often we also need to compute percentages, which tell us the proportion of cases contained within each frequency, that is, relative frequency. In Table 5.2, the number 11 is the frequency relating to the arts category, that is, there are eleven arts students in the sample, which is 20 per cent of the total number of students.
