Some studies include more than one unit of analysis. In these instances, the researcher must anticipate what conclusions he or she wishes to make with regard to each unit of analysis. For example, if a researcher is examining what kinds of college students are most successful in their careers, but also wants to examine what kinds of colleges produce the most successful graduate students, he or she is working with two separate units of analysis: individuals (college students) and organizations (colleges).
Common Units Of Analysis In Social Science ResearchIn social science research, there are several units of analysis that are commonly used, including: individuals, groups, organizations, social artifacts, and social interactions.
Individuals. Individual human beings are perhaps the most commonly used units of analysis in social science research. Researchers tend to describe and explain social groups and behaviors by analyzing and aggregating the behaviors of individuals. They can note the characteristics of individuals (gender, age, religion, attitudes, etc.) and can then combine these descriptions to provide a composite picture of the group the individuals represent.
Any type of individual can be the unit of analysis in social science research. Some examples of classes of individuals that might be studied include: college students, single parents, Catholic churchgoers, factory workers, gang members, etc. Notice that each of these implies some population of individual persons. As the units of analysis, individuals are commonly characterized in terms of their membership in social groups. Researchers typically study the individuals and then aggregate these individuals to make generalizations about the population they belong to.
Groups. Another unit of analysis commonly studied in social science research is the social group. A researcher may be interested in characteristics that belong to one group, considered as a single entity. For instance, if a researcher is studying criminals by looking at the members of a criminal gang, the unit of analysis is the individual (the criminal). However, if the researcher was studying all gangs in a city to learn the differences between them (big gangs versus small gangs, “west side” gangs versus “east side” gangs, etc.), the unit of analysis is the social gang as a group because the researcher is interested in gangs rather than their individual members.
Other examples of units of analysis at the group level include: friendship cliques, married couples, families, fraternities, etc. As with individuals, each of these terms implies some population. Researchers typically describe a population by generalizing from their findings about individual groups that make up that population.
Organizations. Another unit of analysis that is used in social science research is the formal social organization. For example, if a researcher is studying corporations, the unit of analysis is the organization (corporation). The researcher might characterize the individual corporations in terms of the number of employees, net annual profits, gross assets, the percentage of employees who are racial/ethnic minorities, etc. From here the researcher could look at things such as whether large corporations hire a larger or smaller number of minority employees than small corporations.
Other examples of units of analysis at the organization level include: church congregations, colleges, army divisions, academic departments, and supermarkets.
Social Artifacts. Another unit of analysis used in social science research is the social artifact. A social artifact is any product of social beings or their behavior. Examples include: books, newspapers, paintings, poems, automobiles, pottery, jokes, buildings, songs, photos, etc.
In the same way that people and groups imply populations, each social object also implies a set of all objects of the same class. For example, if a researcher is using newspapers as the unit of analysis, an individual newspaper could be characterized by it’s size, average article length, number of pictures, or number sold. Then the population of all newspapers could be analyzed for the purpose of description or explanation, such as which newspapers sell the best and why.
Social Interactions. Social interactions are another unit of analysis that social science researchers use in studies. For example, a researcher might study weddings and characterize them as racially or religiously mixed or not, having a religious or secular ceremony, resulting in divorce or not, or by descriptions of the marriage partners. When a researcher reports that weddings between partners of different religions are more likely to result in divorce compared to weddings between partners of the same religion, weddings are the unit of analysis, not the individuals involved.
Other social interactions that might be units of analysis in social science research include: court cases, traffic accidents, fistfights, friendship choices, divorces, race riots, final exams, and congressional hearings.
No matter what your unit of analysis is in a research project, the important thing is to be clear about what your unit of analysis is. For example, when you start a research project you must decide whether you are studying crimes or criminals, marriages or marriage partners, corporations or corporate executives, and so on. Otherwise, you run the risk of drawing invalid conclusions because your statements about one unit of analysis are actually based on the analysis of another unit of analysis.
Babbie, E. (2001). The Practice of Social Research: 9th Edition. Belmont, CA: Wadsworth Thomson.