Individuals who conduct research on literature, such as English or comparative literature professors, use books and stories as source material in analytical research. While this type of research is usually qualitative rather than quantitative in nature, the researcher still forms a hypothesis on a literary topic and has to collect data that he can analyze to support that hypothesis. In this type of research, data usually takes the form of literary passages or historical details that are relevant to an author's life.
Finding out the opinions held by the general public is something both businesses and politicians very much desire. Discerning these opinions is the job of the pollster. These individuals gather data by administering polls to a number of people, either in person, over the phone or over the Internet, in which preferences and opinions are coded into quantitative data. With this data, researchers can use a variety of statistical techniques to estimate the prevailing opinions in the general public.
For scientists that work in a laboratory, such as biologists, chemists or physicists, the material for their analytical research comes from experiments. The scientist in question first determines what concrete information she would need to support, or fail to support, a particular hypothesis. The scientist then designs an experiment that will produce the information, or data, she needs. These can be things such as substance presence, absence or magnitude following the course of an experiment. Each outcome is a single point of data, so researchers run the experiment several times to have a reliable body of data to analyze.
While most analytical research projects collect and then analyze a body of data, advances in statistical techniques and raw computing power have given rise to meta-analysis: a study of studies. In this type of research, the analysis researchers performed on data becomes the data for a new set of analysis. This kind of analytical research can answer questions about how a particular research method is affecting the data studies produce and examine the strength of a given finding by comparing it to several different results.