Research Methods in Sociology

Survey research collects data from subjects who answer questions about behaviors and opinions, often in the form of a questionnaire. Most surveys are done as quantitative research. However, qualitative research can also be done with surveys using open-ended questions. The survey is one of the most widely used research methods. It allows respondents a level of anonymity (not being identified) in which they feel freer to express personal and sensitive information. Sociologists conduct surveys to gather information on political ideologies, reported levels of religiosity, and a host of other issues.

Surveys target a specific population, or people who are the focus of a study. Because it is impossible to study an entire population, most researchers survey a sample, or a smaller number of subjects who adequately represent the larger population. In a random sample, every person in a population has an equal chance of being chosen for the survey.

Interviews are a type of qualitative research. In an in-depth interview, a researcher talks with an individual at length. Also called intensive interviews, these types of interviews are usually recorded so they can later be transcribed (written out) for analysis. There are advantages and disadvantages to interviews as a research method. In-depth interviewing allows the researcher into the world of the participant. However, information gained from interviews cannot always be generalized to a larger population. For example, a researcher could interview a mass murderer for hours and come to deeply understand why that person carried out multiple acts of violence. However, the information collected in the interview may not be useful in understanding the motivations of a different mass murderer.

A structured interview uses questions with no variation. Subjects are asked a series of close-ended questions—questions that only allow respondents to give certain answers. This can include yes/no questions or a set of possible answers that respondents can select. This allows researchers to collect information about preestablished categories and to have control of the interview script. Close-ended questions ensure that researchers get the data they want to collect and analyze. They are also useful because they do not allow for nonsense answers that are outliers, answers that are too different from the majority of answers from respondents and cannot be used as data. An unstructured interview uses open-ended questions that allow for more fluid interactions between the researcher and the respondent. Respondents are not forced to choose from predesigned answers but instead can elaborate on their statements. The goal in unstructured interviews is to understand respondents' thinking.

For example, a researcher studying drug use could use a structured or an unstructured interview. The structured interview has a set number of questions. No matter what type of answer a respondent gives, the researcher simply records the answer and moves on to the next question. However, if a 13-year-old respondent mentions using heroin or methamphetamine every day for three years, most researchers would want to know more about that. In a structured interview, the researcher does not probe further. In an unstructured interview, the researcher is able to ask questions such as, "Tell me why you started using these drugs at such a young age?" The unstructured interview allows the researcher to develop new lines of inquiry based on respondents' answers. The structured interview focuses on gathering data related to a specific question or questions, excluding issues that fall outside those questions.

In participant observation, researchers become part of the group they are studying. To do so, the researcher spends time with a group, such as a class, a group of neighbors, or a religious group. A type of qualitative research, participant observation can yield a great deal of complex information. Researchers using participant observation must guard against the Hawthorne effect, which occurs when research study participants’ awareness of being studied influences their behavior. In participant observation, the researcher's presence can heighten this awareness. Participants may alter their behavior when the researcher is observing or interacting with them. One way researchers can counteract the Hawthorne effect is to develop rapport with the study subjects. When the people being studied gain a sense of familiarity with the researcher, they are more likely to follow their usual patterns of behavior rather than adjusting them because of the presence of an observer.

After a day of conducting participant observation research, it is common for researchers to type out field notes of what happened. Note-taking during participant observation is an important part of data collection. Not only does it record the researcher's observations; it provides a framework to help the researcher make connections. For example, note-taking helps researchers frame responses based on respondents' membership in larger social institutions, or links them with political party, socioeconomic status, or religious affiliation. Participant observation research can be beneficial because it can provide sociologists with a lot of detailed information about a group, as well as how the group changes over time. For example, a researcher who spends time with a biker group over a period of months or years can get a rich, complex understanding of this group, such as a complex understanding of its characteristics, dynamics, and practices. A researcher who spends time with families in their homes, observing or taking part in daily activities, conversations, arguments, and celebrations, can gain a nuanced understanding of the behaviors, norms, and values of those families.

Content analysis is used to analyze the content of written and visual texts, such as books, film, magazines, television, and advertisements. The main goal of content analysis is to understand how particular issues are presented over time. Content analysis can be conducted using both qualitative and quantitative methods. As a qualitative method, not only is it unobtrusive and easy to access, but it also can show how images change over time. For example, this method can be used to study the representations and roles of women in advertisements. As a quantitative method, images can be coded numerically to show how often a behavior appears. For example, a researcher might look at occurrences of people of color in positive roles on TV shows. A researcher might analyze content on Twitter, looking at issues that are trending, in order to gain a sense of the mood of a society at a particular point.

Comparative-historical analysis is used by researchers to examine data that have already been gathered by someone else. This method can be used in quantitative and qualitative research. Examples of comparative-historical data are census data; old letters, diaries, or journals; and interviews previously conducted. Using data that is already available not only saves time and money, but it also adds depth to a study. This is because sociologists often interpret findings in a new way that was not part of an author's original purpose. For example, a researcher might use comparative-historical analysis to examine data that were collected by Western researchers in Asian societies in the early 20th century. Informed by a 21st-century perspective that values cultural relativism and considers issues of intersectionality, a researcher might look at the original data to draw different conclusions about the societies under study or to draw conclusions about the Western societies that produced the original research.