Descriptive analysis describes conditions, populations, and phenomena as they are. Contrast this with experimental analysis in which researchers place controls on conditions and administer different treatments or interventions to experimental and control groups.
In statistics, descriptive analysis takes the form of descriptive statistics. The purpose of descriptive statistics is to summarize or describe a set of quantitative data. Researchers use these statistics to describe or characterize the population or sample being studied.
Common descriptive techniques used in statistics include measures of central tendency, such as the mean (or arithmetic average) and median. Other descriptive statistics are measures of dispersion, such as the variance and standard deviation. In addition, researchers may use visual tools of descriptive analysis, such as bar graphs, pie charts, and line charts.
Not all descriptive analyses involve the use of statistics. A wide range of qualitative studies are descriptive in nature. Qualitative research involves subjects and data that are difficult to quantify. Researchers in anthropology and a variety of humanities fields, use qualitative research, in which the data come from field observations, interviews, documents, and similar sources. Qualitative data come in the form of words or images, rather than numbers, as with statistical data. A variety of descriptive analytical techniques exist for analyzing qualitative data, but in general, they require the researcher to sift through the material to identify distinct patterns, relationships, and themes that can describe the subjects being studied.
Descriptive analysis of qualitative data is sometimes perceived as less "scientific" than studies that use statistical methods. However, descriptive analysis of qualitative material is meticulous and complex, requiring a great deal of concentration of observation by the researchers. In some cases, depending on the data, it may be possible to use descriptive statistics to augment qualitative analysis.
For most researchers who use statistical techniques, descriptive statistics are only the starting point of an analysis. Descriptive statistics provide convenient summaries of the group or phenomena being studied, but most analysts are interested in studying the impact of a treatment, condition, or intervention (an independent variable) on an outcome of interest (dependent variable). For example, researchers will ask how education and income affect political participation, or how socioeconomic background affects students' academic achievement. These questions cannot be answered by descriptive analysis and require instead the use of inferential statistics (also called inductive statistics). Inferential statistics enable researchers to draw conclusions about a population based on the analysis conducted.