A research method is a systematic plan for conducting research. Sociologists draw on a variety of both qualitative and quantitative research methods, including experiments, survey research, participant observation, and secondary data. Quantitative methods aim to classify features, count them, and create statistical models to test hypotheses and explain observations. Qualitative methods aim for a complete, detailed description of observations, including the context of events and circumstances.
Now let's take a look at each of the different research methods in detail.
An experiment is a research method for investigating cause and effect under highly controlled conditions. When conducting an experiment, researchers will test a hypothesis. A hypothesis is a statement of how two or more variables are related. For example, let's say we wanted to examine whether studying affects a person's GPA. Our hypothesis might be: the more a person studies, the higher a person's GPA will be. In this example, studying would be the independent variable (the cause), while a person's GPA would be the dependent variable (the effect).
If we wanted to test this hypothesis, we would randomly assign subjects into two groups. The experimental group is a group of individuals that are exposed to the independent variable. The control group, on the other hand, is not exposed to the independent variable. We would require that the control group doesn't study at all, but that our experimental group has to study at least 10 hours a week. After one semester, we would then determine which group has the higher GPA. If the experimental group has a statistically higher GPA, we can assume our hypothesis is correct.
It is important to note that sometimes a change in the dependent variable could be the result of something entirely different than what was being studied (maybe, for example, those chosen for the experimental group were simply more intelligent than those in the control group). When two variables change together but neither one causes the other, we call this a spurious correlation.