After constructing a frequency distribution, a researcher involved in quantitative analysis will usually compute sample descriptive statistics. Often, research papers do not include a frequency distribution, just descriptive statistics. Descriptive statistics convey two basic aspects of a sample: central tendency and dispersion. The former describes the most common variate of the sample, and the latter how the sample is distributed around the most common variate.

A word should be said about what calculation method may be easier in what situation. In a situation in which a researcher does not have access to a computer and needs to compute descriptive statistics of a large sample, he would probably choose to group the data into a frequency distribution because it is easier to compute the statistics in this manner. However, if the data are grouped, the results will not be as precise because of rounding error. With the use of computers, descriptive statistics are rarely computed by hand and therefore frequency distributions are rarely used for computation purposes. The computation of descriptive statistics with SAS/ASSIST will be illustrated at the end of the chapter, since all descriptive statistics can be obtained with one operation.

Measures of central tendency

This section discusses three central tendency statistics: the mean, the median and the mode. The three are different kinds of ‘averages’, used in different situations. Their general purpose is the same, namely, to find the single most representative score in the sample.