Collecting data is the first step in completing any research study or paper. Data comes in two forms: primary and secondary. Primary data is obtained from studies, interviews and experiments conducted by the researchers personally. Secondary data is collected from published sources and research conducted by other people. Primary data is harder to obtain, but is tailored specifically to your study and therefore more important as a source than secondary data.
Generally used for studies in the social sciences and behavioral studies, researchers will create surveys with questions pertaining to their study and hand them out to participants, who will respond. Survey sample groups are usually larger than for other research methods because responses are generally lower. The researchers then tabulate the answers and results of the survey to determine if the research proves or disproves their hypothesis. Most research surveys are anonymous to prompt more honest answers from participants.
Observational data collection exists in many forms. In structured observation, the researchers create a specific set of tightly controlled circumstances and observe how study participants behave, while in unstructured observation, researchers simply study behavior in an uncontrolled environment. Study participants are usually aware they are being observed in this type of study, but under certain circumstances, disguised observation can occur in which participants do not know they are being watched. Observational data collection eliminates subjective bias, but it can be expensive to set up strictly controlled testing environments.
Similar to surveys, interviews help collect information from study participants. Unlike surveys, they are not anonymous, although you should obtain permission before you use participants' names in the final published study or paper. Interviews consist of the researcher asking questions to participants in person. The questions are designed to elicit responses related to the study at hand. This type of data collection is time-consuming, as interviews are usually conducted one-on-one and a large sample is needed, which means many interviews must be conducted.
Mainly used for scientific studies, experiments consist of a control group and a test group. The test group is administered whatever is being studied. The control group must be very tightly controlled to eliminate all variables that could affect the result and ensure that any differences between the two can be attributed to the substance or product being tested. Experiments need to be carefully designed and conducted multiple times to ensure the same results. An experiment is only considered scientifically sound when its results are proven to be repeatable.