These types of experiments are often used by anthropologists, psychologists and social scientists to observe natural behaviors without affecting them in any way. It is also used by market researchers to judge the habits of customers, or by companies wishing to judge the morale of staff.
The results from a descriptive research can in no way be used as a definitive answer or to disprove a hypothesis but, if the limitations are understood, they can still be a useful tool in many areas of scientific research.
The subject is being observed in a completely natural and unchanged natural environment. A good example of this would be an anthropologist who wanted to study a tribe without affecting their normal behavior in any way. True experiments, whilst giving analyzable data, often adversely influence the normal behavior of the subject.
Descriptive research is often used as a pre-cursor to quantitative research designs, the general overview giving some valuable pointers as to what variables are worth testing quantitatively. Quantitative experiments are often expensive and time-consuming so it is often good sense to get an idea of what hypotheses are worth testing.
Because there are no variables manipulated, there is no way to statistically analyze the results. Many scientists regard this type of study as very unreliable and ‘unscientific’.
In addition, the results of observational studies are not repeatable, and so there can be no replication of the experiment and reviewing of the results.
Descriptive research design is a valid method for researching specific subjects and as a precursor to more quantitative studies. Whilst there are some valid concerns about the statistical validity, as long as the limitations are understood by the researcher, this type of study is an invaluable scientific tool.
Whilst the results are always open to question and to different interpretations, there is no doubt that they are preferable to performing no research at all.