LESSON 67 - RESEARCH METHODOLOGY || SECTION 3.8: DATA ANALYSIS TECHNIQUES || QUANTITATIVE DATA

Quantitative data analysis in a research proposal should be presented as a clear, step-by-step plan: reduce raw responses into manageable summaries, then use statistics to test hypotheses and generalize from a sample to a population. The core idea is that once questionnaires or other numerical instruments generate data, the researcher must specify how that data will be analyzed, what statistical tools will be used, and how results will be organized and presented—typically starting with descriptive statistics and moving to inferential statistics.

Data analysis is defined as the process of reducing collected data into meaningful summaries. In most social science dissertations, this work is largely carried out in Chapter 4, where the collected responses are summarized according to variables. The choice of analysis method depends on the type of data and the scale of measurement. Numerical data calls for statistical analysis, while narrative data requires different approaches (handled in later lessons). For quantitative work, manual analysis is no longer the norm; common tools include IBM SPSS for quantitative/numerical data and NVivo for narrative/qualitative data.

The lesson then separates quantitative analysis into two major categories. Descriptive statistics summarize a sample by using one number to represent a group of numbers. Inferential statistics go further: they support predictions and generalizations about the population based on sample data, relying on probability and hypothesis testing. Hypothesis testing splits into parametric tests—built on assumptions such as the population being normally distributed—and non-parametric tests, often described as “assumption-free” because they make fewer assumptions and frequently use ranking (analyzing ranks rather than raw values).

For presenting descriptive results, three organizing approaches are emphasized: graphical, tabular, and numerical summaries. Graphical methods include histograms, cutter diagrams, and frequency polygons, with specific guidance that some visuals suit categorical data while others suit continuous data. Tabular presentation includes frequency distribution tables and crosstabs, particularly for categorical variables. Numerical summaries include measures of central tendency and variability—mean, median, variance, and standard deviation—typically for continuous data.

Inferential statistics are also grouped into three families. Correlation examines relationships: Spearman rank-order correlation for ordinal data and chi-square for nominal categorical data, while Pearson product-moment correlation coefficient is used for continuous data. Regression is used for continuous outcomes, including simple and multiple regression, with binary logistic regression mentioned for earlier lessons tied to the type of data. Tests of comparison include t-tests (noted as having three types) and analysis of variance (ANOVA), both applied to continuous data.

Finally, the lesson lays out what researchers should write in Section 3.8: first identify the statistical tools for quantitative data, then explain how those tools answer the research questions and test hypotheses, and then describe the order of presentation—descriptives first (tabular/graphical/numerical), followed by inferential results. The next session shifts from quantitative to qualitative data analysis techniques.