Observation

• Sociologists often study group behavior by watching and recording what people do. A researcher may sit in the park and record when people choose to sit on an occupied bench. Though the sociologists is careful to be specific in recording choices, gender and age group of the people being observed, she does not do anything to change or influence what happens. Observation is often used to formulate a hypothesis that can later be tested using the experimental method. One hypothesis may be that people tend to choose an empty bench if one is available, and sit as far as possible from a person on an occupied bench.
  
  

Survey and Interview

• When trying to investigate attitudes, observation is not very accurate. A researcher cannot see inside a person's head to determine their attitudes or beliefs about sharing a park bench. The survey method is often used in this type of investigation, administers with a paper or computer programmed with set of multiple choice and open-ended short answer or essay questions. After the surveys are completed, the sociologist codes the answers and analyzes them with a statistical program. This analysis can determine whether there is a significant difference between the attitudes of one group and another. For example, men may be more accepting toward sharing a park bench with a stranger than women are, or vice versa.
  
  

Experiment

• The experimental method of investigating sociological issues is generally accepted as the most reliable. In this model, one or several hypotheses are formulated and tested by manipulating variables to see their effects on the study participants. If a researcher hypothesizes that people will sit closer or farther apart depending on how well they know each other, he may set up a study in which participants are randomly selected and asked to meet at the park. Unseen, the researcher records the distance between the participants while sitting on a bench together.
  
  

Secondary Analysis

• One method of studying sociological issues is to perform a secondary analysis on studies that have already been conducted. This approach can be controversial because different researchers use different populations, methodologies and statistical tests. However, secondary analysis can spot trends in group behavior that might be missed otherwise, since conducting multiple studies can be costly and time-consuming.