UNIVARIATE ANALYSIS
    Distributions
    Central Tendency
    Dispersion
    Continuous and Discrete Variables
    Detail versus Manageability
 

We'll begin our examination of quantitative analysis simply: univariate analysis, meaning one variable at a time.  There are several aspects of this, however.

Let's say we are interested in analyzing the ages of people participating in a large, national survey.  We might be look first at the central tendency or average.  This section will present three different measures: the mode is the most common attribute on the variable; the mean is what most people think of as the "average" (e.g., add up the ages and divide by the number of people); and the median is the attribute of the middle case if they were lined up from one extreme (e.g., youngest) to the other (e.g., oldest).

While the central tendency offers a summary description of the population under study, we don't know how well it represents all the individuals.  If we say the average age is 35, for example, are most of the people surveyed close to that age or are they widely spread out--from much younger to much older?  Measures of dispersion answer this question.

The way in which variables can be analyzed depends in large part on the nature of the variables themselves.  As we'll see, the attributes of discrete variables consist of relatively few categories, such as male/female or Democrat/Republican/etc.  The attributes of continuous variables, on the other hand, form a steady progression from one extreme to the other, such as the ages 18, 19, 20, 21, etc.

Often, the quantitative analyst can choose the degree of detail to be captured for analysis.  For example, we could ask people to report their ages in terms of years and days (23 years and 142 days old), but it is unlikely that such fine distinctions would be useful in the analysis.  On the other hand, we could simple ask if they are below or above 30 years of age, but that would not be sufficiently detailed for most uses.