Observation allows researchers to experience a specific aspect of social life and get a firsthand look at a trend, institution, or behavior. Participant observation involves the researcher joining a sample of individuals without interfering with that group's normal activities in order to document their routine behavior or observe them in a natural context. Often researchers in observational studies will try to blend in seamlessly with the sample group to avoid compromising the results of their observations.

Observational research is a type of descriptive research that differs from most other forms of data gathering in that the researcher's goal is not to manipulate the variables being observed. While participants may or may not be aware of the researchers' presence, the researchers do not try to control variables (as in an experiment), or ask participants to respond to direct questions (as in an interview or survey based study). Instead, the participants are simply observed in a natural setting, defined as a place in which behavior ordinarily occurs, rather than a place that has been arranged specifically for the purpose of observing the behavior. Unlike correlational and experimental research which use quantitative data, observational studies tend to use qualitative data.

For example, social psychologists Roger Barker and Herbert Wright studied how a sample of children interacted with their daily environments. They observed the children go to school, play with friends, and complete daily chores, and learned a great deal about how children interact with their environments and how their environments shape their character. Similarly, anthropologist Jane Goodall studied the behavior of chimpanzees, taking careful notes on their tool making, family relationships, hunting, and social behavior. Her early work served as the basis for future research on chimpanzees and animal behavior in general.

Advantages of Observational Studies

By observing events as they naturally occur, patterns in behavior will emerge and general questions will become more specific. The hypotheses that result from these observations will guide the researcher in shaping data into results.

One advantage of this type of research is the ability to make on-the-fly adjustments to the initial purpose of a study. These observations also capture behavior that is more natural than behavior occurring in the artificial setting of a lab and that is relatively free of some of the bias seen in survey responses. However, the researcher must be careful not to apply his or her own biases to the interpretation. Researchers may also use this type of data to verify external validity, allowing them to examine whether study findings generalize to real world scenarios.

Laboratory Observation

Laboratory observation can feel artificial to participants and influence their behavior. Observation in a natural setting allows researchers to document behavior without this influence.

There are some areas of study where observational studies are more advantageous than others. This type of research allows for the study of phenomena that may be unethical to control for in a lab, such as verbal abuse between romantic partners. Observation is also particularly advantageous as a cross-cultural reference. By observing people from different cultures in the same setting, it is possible to gain information on cultural differences.

Disadvantages of Observational Studies

While observational studies can generate rich qualitative data, they do not produce quantitative data, and thus mathematical analysis is limited. Researchers also cannot infer causal statements about the situations they observe, meaning that cause and effect cannot be determined. Behavior seen in these studies can only be described, not explained.

There are also ethical concerns related to observing individuals without their consent. One way to avoid this problem is to debrief participants after observing them and to ask for their consent at that time. Overt observation, where the participants are aware of the researcher's presence, is another option to overcome this problem. However, this tactic does have its drawbacks. When subjects know they are being watched, they may alter their behavior in an attempt to make themselves look more admirable.

This type of research can also be very time consuming. Some studies require dozens of observation sessions lasting for several hours and sometimes involving several researchers. Without the use of multiple researchers, the chances of observer bias increase; because behavior is perceived so subjectively, it is possible that two observers will notice different things or draw different conclusions from the same behavior.