Most empirical research relies on using the scientific method to conduct large studies that use many participants in a control group and an experimental group. The purpose is to determine the effect of some experimental factor by introducing it to the experimental group, but not to the control group, and seeing what, if any, effect the experimental factor has. However, in some cases researchers will use an alternative method called single-subject research design.
Single Subject Research Designs (SSRDs) work by designing an experiment where, instead of a control group of subjects and an experimental group of subjects whose results are compared to one another, the control and experimental measurements come from a single subject. Researchers measure the metric of interest before introducing the experimental factor for a control measurement, and measure the metric of interest after introducing the experimental factor for the experimental measure. While studies may include more than one subject, each subject is treated as a unique experiment instead of one trial in a larger experiment.
An advantage of using an SSRD is that, instead of comparing the percentage of people that responded to an experimental factor to the percentage of people that did not, the study examines how an individual subject, with his own unique characteristics, responds to the experimental factor. This is particularly useful when studying specific subsets of a population, rather than the population as a whole.
In an experimental group versus control group study, the researcher has to find a large number of participants to act as subjects. This is necessary for the data from the experiment to yield statistically relevant results. This requires the time and resources to not only gather the participants, but to run trials of the experiment on all the subjects to gather all the data. An SSRD allows researchers to quickly design and run their study without having to find so many participants.
While the fact that the researcher does not use a large number of participants has its advantages, it also has a downside: Because the experimental trials are run on only one subject, it is difficult to empirically show with the experiment's data that the findings will generalize out to larger populations. All the trial can show is what happened with the individual subject, whereas traditional research designs that use large numbers of participants are specifically designed to show if a result is statistically valid for the general population.