LESSON 12 - QUANTITATIVE RESEARCH DESIGNS: SURVEY, EX POST FACTO & EXPERIMENTAL

Quantitative research design is the “master plan” that dictates how data will be collected, analyzed, and reported to answer research questions or test hypotheses—and choosing the right design depends on the problem, the study’s purpose, researcher expertise, the target population, available resources, and who will use the findings. Without a plan, data collection and analysis become aimless; with one, researchers can align sampling, instruments, and statistical tools to the kind of evidence needed.

A research design is defined as procedures and a blueprint for collecting, analyzing, interpreting, and reporting data. It also guides practical decisions such as sample selection methods, the instruments used to measure variables, and the statistical techniques applied during analysis. The lesson breaks design down into core elements: (1) the research approach (quantitative, qualitative, mixed), (2) data collection methods (including instrument development, administration, and timing), (3) the population and sample size with sampling techniques, and (4) data analysis strategy—statistical procedures for quantitative data, inductive analysis for qualitative data, or a combination for mixed methods.

Selection of a design starts with the research problem: whether it calls for quantitative data, qualitative data, or both. Next comes the purpose of the study—descriptive, explanatory, or exploratory—which narrows the design choice. Researcher knowledge and experience matter too, since comfort with quantitative or qualitative data collection and analysis affects feasibility. The population of interest also shapes the decision: accessible large populations often fit quantitative approaches, while smaller, information-rich groups are more common in qualitative work. Constraints like time and money influence what can realistically be conducted, and the intended audience’s preferences—narrative versus statistical reporting—can further steer the choice.

The lesson then lays out three main quantitative designs used in social science. First is the survey, a non-experimental design that questions people and records responses to describe population characteristics. Surveys typically rely on probability sampling for large populations, use structured questionnaires or schedules (including interview or observation schedules), and measure variables numerically. Survey analysis uses statistical methods to describe trends and test hypotheses. Three survey types are highlighted: cross-sectional surveys (one point in time), correlational surveys (quantifying the degree of relationship between variables without implying causality), and longitudinal surveys (follow-ups over time). Longitudinal work includes trend analysis (tracking changes in a population over time), cohort studies (following groups with shared characteristics using different samples across periods), and panel studies (following the same sample over time).

Second is the ex post facto (causal comparative) design, which studies variables after the fact—when the “exposure” has already occurred and the researcher looks back to identify factors behind changes in a dependent variable. Because researchers do not manipulate the independent variable, groups are pre-existing, and the design mimics experiments through comparisons of similar groups. Limitations are emphasized: lack of control over groups, risk of spurious or reversed cause-and-effect relationships, and the need for repeated measures to strengthen conclusions. Third is the experimental design, which determines cause-and-effect through manipulation; detailed coverage is reserved for the next lesson.