Chapter 14 (Analyzing Quantitative Data) of The Practice of Social Research by Earl R. Babbie explores the methods for converting data to numerical forms and presenting them in the form of statistical analyses in terms of quantitative analysis and special features of univariate, bivariate, and multivariate analysis.
The first part of the chapter examines the specifics of the process of quantification of the data. The collected data should be quantified to make such computer programs as SPSS and MicroCase handle the qualitative analysis. The researchers should code data that appear to be not appropriate for being processed numerically. To be able to compare the research results with those of other studies, the researcher should use an established occupational coding scheme. The chosen scheme should also correspond to the theoretical concepts examined during the research. The codebook, a document describing the locations of variables, serves as a guide used in the coding process. Each variable should be identified by an abbreviated variable name. Besides, the codebook should present the definition of each variable and indicate the attributes composing it. To make the computer able to read and manipulate the data, the researcher should choose the most appropriate technique for converting the data into the machine-readable format.
The next part of the chapter examines univariate analysis, which has a descriptive purpose. Listing the attributes to each case under study in terms of the variable in question is considered the basic format for presenting univariate data. The provided example demonstrates the specifics of this process. Central tendency is indicated as another form of presenting the data. Three ways of measuring central tendency include the mean, the mode, and the median. The measure of central tendency should be chosen carefully, paying attention to choosing one that is the most appropriate for the situation. Such measures of dispersion as the range, the standard deviation, and the interquartile range are discussed. The understanding of the nature of continuous and discrete variables is vital for choosing an appropriate technique for their analysis.
Subgroup comparisons are explored in the example from the General Social Survey. Besides, this section investigates the ways to handle “collapsing” response categories, “don’t know” answers data, numerical descriptions of qualitative research illustrating each of them with relevant examples of dealing with the same aspects of existing studies.
The purpose of the bivariate analysis or the analysis of two variables simultaneously is indicated as focusing on the relationship between variables. It has an explanatory purpose. Numerous examples of tables demonstrate special features of presenting the data in terms of bivariate analysis. Contingency tables are defined as the most common form of presenting results in this type of analysis. Such tables show the effects of the independent variable on the dependent variable and should be easy to read and interpret.
The multivariate analysis is aimed at analyzing more than two variables simultaneously. While bivariate analysis focuses on one independent variable, multivariate analysis is used when there is more than one independent variable. The specifics of multivariate analysis are demonstrated through the extended example of religious service attendance used to illustrate the bivariate analysis. Multivariate analysis is described as a powerful tool for sociological diagnostics, which plays an influential role in serving our society.
As quantitative analyses are considered susceptible to subjective biases, the researchers should strive to provide protection from subjectivity. The researchers involved in the quantitative analysis should also pay attention to the protection of subject privacy.
This chapter appears to be very informative and explores both theoretical and practical aspects of the discussed issues by illustrating them with numerous relevant examples.