The next three chapters deal with methods for analyzing quantitative data. We begin with descriptive and inferential univariate analysis. Then, in chapters 21 and 22, we move on to bivariate and multivariate analysis.
Descriptive analysis involves understanding data through graphic displays, through tables, and through summary statistics. Descriptive analysis is about the data you have in hand. Inferential analysis involves making inferences about the world beyond the data you have in hand.
When you say that the average age of people in a village is 44.6 years, that’s a descriptive analytic statement. When you say that there is a 95% probability that the true mean of the population from which you drew a sample of people is between 42.5 and 47.5 years, that’s an inferential statement: You are inferring something about a population from data in a sample.
In univariate analysis, we examine variables precisely and in detail and get to know the data intimately. Bivariate analysis involves looking at associations between pairs of variables and trying to understand how those associations work. Multivariate analysis involves, among other things, understanding the effects of more than one independent variable at a time on a dependent variable.
Suppose you’re interested in the causes of variation in the income of women. You measure income as the dependent variable and some independent variables like: age, marital status, employment history, number of children, ages of children, education, and so on. The first thing to do is examine carefully the properties of all the variables. That’s the univariate part of the analysis.
Next, you’d look at the association between each independent variable and the dependent variable. You’d also look at the association between pairs of independent variables. That’s the bivariate part. Finally, you’d look at the simultaneous effect of the independent variables on the dependent variable or variables. That’s the multivariate part.
Each part helps us answer questions about how things work.