work, Fish showed how they supported each other’s work and scholarship and provided emotional encouragement and intellectual stimulation. Fish also considered women’s professional styles as compared with the styles of men. Whereas men often received their training and support from a mentor (an older, established, and respected man in the field), the women of Hull House operated within a network of egalitarian relationships and interactions.
According to Fish’s research, the data and documents collected by their leader, Jane Addams, and other Hull House women provide baseline information that has been used as a starting point or comparison for later studies—for social researchers in the fields of immigration, ethnic relations, poverty, health care, housing, unemployment, work and occupations, delinquency and crime, war, and social movements.
In experiments, all variables except the one being studied are controlled so researchers can study the effects on the variable under study. An experiment usually requires an experimental group, in which subjects in the group are exposed to the variable being studied; this process is to test the effects of that variable on human behavior, and a control group, in which the subjects are not exposed to the variable the researcher wants to test. The control group provides a baseline with which the experimental group can be compared, as shown in the example of Hector.
Experiments conducted in a lab can often provide the most accurate test of cause and effect. They make it possible to control most variables (eliminating irrelevant spurious variables) and determine the sequence in which variables affect each other. By separating the sample into experimental and control groups, the researcher can see if the study’s independent variable makes a difference in the behavior of people who are exposed to that variable compared with those who are not. Psychologists use lab experiments, but few sociologists use this method because many sociological questions cannot be studied in controlled settings. For example, Hector’s environment in the favela cannot be studied in a laboratory setting.
Control and experimental research projects outside of a lab setting are more common among sociologists. For example, researchers may want to determine whether a new teaching method using technology might help children from Hector’s favela. Researchers can do so by comparing a control group, exposed to the usual teaching method, and an experimental group, provided with the new method or experimental technology. We must ensure that the control and experimental groups of children are at the same academic level and that the teachers are equally motivated and prepared when teaching both classes. With this carefully designed research project, we can conclude that the new approach increases learning if the children in the experimental group score significantly higher on the final exam than those in the control group.
Triangulation refers to the use of two or more methods of data collection to enhance the amount and type of data for analysis and the accuracy of the findings. To study Hector’s situation, a research study could use macro-level quantitative data on poverty and on educational statistics in Brazil and micro-level interviews with Hector and his peers to determine their goals and attitudes toward education. If all findings point to the same conclusion, the researcher can feel much more confident about the study results.
▲ Social scientists are not the only professionals who use triangulation. Journalists also consult a variety of sources, including social scientists, to put together news broadcasts.
© Getty Images/Benjamin Lowy/Getty Images News
Selecting a Sample.
It would be impossible to interview or send a questionnaire to every school dropout in Brazil to determine why the teenage dropout rate is so high. It is possible to study a portion of that population, however. In Step 5, the research design process includes determining how to make sure the study includes people who are typical, that is, representative, of the total group (or population) you want to learn about. This involves careful selection of a sample, a group systematically chosen to represent a much larger group.
Researchers use many types of samples. A common one, the representative sample, attempts to accurately reflect the group being studied so that the sample results can be generalized or applied to the larger group or population. In the case of studying why so many boys from Hector’s favela drop out of school, a representative sample for a study could be drawn from all 13- to 16-year-olds in his region or city in Brazil.
The most common form of representative sample is the random sample. All people in the population being studied have an equal chance of being selected for the study. By observing or talking with this smaller group selected from the total population under study, the researcher can get an accurate picture of the total population and have confidence that the findings apply to the larger group. Developing an effective sampling technique is often a complex process. In the case of Brazil, people constantly move in and out of the favela. Those who have just arrived may not have the same characteristics as those who have been living there a long time, but it is important to have a sample that represents the whole group being studied. Samples also must be large enough to accurately represent a population and to use statistical programs to analyze the data. If you take a methods course, you will delve further into these details of sampling and data analysis discussed next.
Collecting the Data.
We have now made a research plan and selected our methods to use and our sample. In Step 6 we actually collect the data following our research plan. Once we have data, we have to determine what to do with the data to answer our research question.
Thinking Sociologically
Define, distinguish between, and provide an example of each of the following important research concepts:
cause-and-effect relationships
spurious relationships
evidence
correlation
dependent variable
independent variable
hypothesis
sample
triangulation
Now think of a question you would like to research. How would each research concept relate to your research question—or not?
Making Sense of the Data
Once you have collected your data from your sample, you have to analyze it.
Analyzing the Data.
In Step 7, the researcher uses statistical and other techniques to analyze what the data say in order to answer the research question. Imagine that you have 100 interviews from residents of Hector’s favela, plus a notebook full of field observation notes from “hanging out” with the youth there. What do you do with the data? Social researchers use multiple techniques to analyze data, but whatever techniques they use, they look for patterns in the data and then use theories and findings from past research on the topic to make sense of those patterns.
Presenting the Findings.
One
