Your choice of research design depends on your research question. “The function of a research design is to ensure that the evidence obtained enables us to answer the initial question as unambiguously as possible.”
A helpful analogy is to consider the process of building a house. Before you plan the logistics for the building project, you must first design the building. Before you decide how much material you need and how much the building will cost, you have to decide how many rooms it will have. Broadly, research designs fall into the following categories:
- Descriptive: one group
- Experimental: people randomly assigned to groups and watched to see the outcome
- Observational: natural groups observed to see links or “correlations” between outcomes and exposures/interventions
- Longitudinal: groups observed over time (see jpg file explaining differences)
- Case-control: looks at the exposures of a group that has the outcome compared with a group that does not have the outcome
- Cohort: looks at the outcomes in a group that is exposed compared with a group that is not exposed
- Cross-sectional: groups observed at one point in time
Note, that the decision about research design is not the same as the decisions about the methods. Various methods can be used in any of the above research designs. For example, a house design does not depend on the choice of materials or the colour of the paint.
While research design should be driven by the research question, in practice the choice of method is also usually determined by your available resources, expertise, data and tools.