Descriptive research is a quantitative research design that describes phenomena as they exist, taking care not to influence subjects or events in any way. Descriptive and survey research designs are used by social scientists to describe human behaviors, by market analysts to look at the habits of customers, and by businesses seeking insight about company practices. Descriptive research can be used to indicate what variables might be worth testing quantitatively.
An experimental research design is considered the most accurate method of research. Usually applied to the physical sciences, it attempts to prove or disprove a hypothesis numerically, using statistical analysis. Because of the exacting structure of this design, results can be statistically analyzed, replicated and validated by other researches, greatly diminishing arguments about the accuracy of the results.
Used frequently in the social sciences and psychology, quasi-experiments lack a control group, making firm statistical analysis difficult. However, these experiments can be useful in generating data to indicate general trends. They are effective in obtaining a general overview that can be followed up with a quantitative or case study to focus on the underlying reasons for the results generated.
A historical research design can be applied to all fields of study, because it includes factors such as origins, growth, theories and significant persons. It can collect both quantitative and qualitative historical data. Primary sources -- i.e., first-hand accounts -- are the most highly valued in historical research, and include oral history interviews and recordings, eyewitness accounts and personal diaries. In conducting historical research, consider any possible prejudices or biases your sources may have.
Observational research compares subjects in situations in which the researcher has little control over the experiment, mainly for ethical reasons. For instance, in comparing the life expectancy of smokers with that of non-smokers, it would be unethical to ask a group of otherwise healthy people to smoke for 20 years to compare them with the control group of non-smokers. The weakness of this design is that results can be skewed by factors such as pollution, second-hand smoke and environmental working conditions.
A case study is a qualitative research design that performs in-depth investigation of a narrow situation rather than a broad statistical survey. It narrows a broad area of research into one case within that field. It is useful for testing whether scientific models or theories work in the real world. Case studies include data from field notes, interviews and archival information.