Before I became a film major, I was very heavily into social science, I had done a lot of sociology, anthropology, and I was playing in what I call social psychology, which is sort of an offshoot of anthropology/sociology - looking at a culture as a living organism, why it does what it does.
The Tools of Sociology
(Part C)
by
Charles Lamson
Experiments
Although for both moral and practical reasons sociologists do not have many opportunities to perform experiments, a large body of literature in the social sciences is based on experiments. There are two experimental models that social scientists use frequently. The first and most rigorous is the controlled experiment conducted in a laboratory. The second is the field experiment, which is often used to test public policies that are applied to some groups but not to others.
 |
Controlled Experiments The controlled experiment allows the researcher to manipulate an independent variable in order to observe and measure changes in a dependent variable. The experimenter forms an experimental group that will experience a change in the independent variable (the "treatment") and a control group that will not experience the treatment but whose behavior will be compared with that of the experimental group. The control group is similar to (The experimental group in every other way.) This type of experiment is especially characteristic of studies at the micro level of sociological research.
Consider an example: Which line in Figure 1 (b) appears to match the line in Figure 1 (a) most closely? Could anything persuade you that a line other than the one you have selected is the correct choice? A simple diagram formed the basis of a famous series of experiments conducted by Solomon Asch in the early 1950s. They showed that the opinion of the majority can have an extremely powerful influence on that of an individual.
FIGURE 1 Lines Used in the Asch Experiment on Conformity
Cards like these were used in the Asch experiment. Subjects were asked to judge the lengths of various lines by comparing them with the three lines on the bottom card. The line on the top card quite obviously matches line B on the bottom card; all of the judgments were this simple.
Asch's control group consisted of subjects (The term subject refers to a person who participates in a controlled experiment.) who were seated together in a room but were allowed to make their judgments independently. When they looked at sets of lines like those in Figure 1, the subjects in this group invariably matched the correct lines, just as you no doubt have. But in the experimental group a different result was produced by the introduction of an independent variable: group pressure.
Asch's experimental group consisted of subjects who are asked to announce their decisions aloud in the group setting. Each subject was brought into a room with eight people who posed as other subjects but were actually confederates of the experimenter. When the lines were flashed on a screen, those "subjects" all chose a line that was not the matching one. When it was the real subject's turn to choose, he or she was faced with the unanimous opinion of a majority of "subjects" who had picked the wrong line. Thirty-two percent of the real subjects went along with the majority and chose the wrong line as well. And even among the subjects Asch called independent (the 68 percent of the real subjects who gave the correct response despite the pressure of the majority) there was a great deal of variation. Some gave the correct response at all times, whereas others gave it only part of the time. This confirming response was less likely when there was at least one other person in the group who also went against the majority. By varying the number of people who said that the shorter line was longer, Asch was varying the degree of group pressure experienced by the subject. Higher levels of the independent variable (group pressure) thus produced more "errors," or choices of the wrong line.
Field Experiments Field experiments are used extensively in evaluating public programs that address specific social problems. In these experiments there is usually a "treatment group" of people who participate in the program and a control group of people who do not. In one example of this type of experiment, Angelo Atondo, Mauro Chavez, and Richard Regua of California's Evergreen Valley College attempted to test the theory that if students at risk of failure are linked with adult mentors from their own communities, their chances of success will improve Machi. This popular Theory stems in part from an influential article in the Harvard Business Review titled "Everyone Who Makes It Has a Mentor" (Collins & Scott, 1978). Many successful people from all walks of life, especially people who have escaped from impoverished backgrounds, say that they benefited greatly from the guidance of a mentor who steered them toward constructive goals. Most mentors do not actually tutor the individuals they spend time with; instead, they teach through conversation and example.
Researchers have found little evidence to prove that mentors actually make a difference. Unless one can compare students who had mentors with similar students who did not, it is impossible to say within a scientific certainty---that is, beyond individual stories of success or failure whether the mentors made a difference. The Evergreen Valley College researchers therefore assigned 115 entering Latino students with low English proficiency to Latino mentors with excellent English skills. Their control group was composed of a comparable number of Latino students with similar characteristics; these students were not assigned to mentors. Both groups attended the same classes and took the same exams. At the end of the semester 89% of the students with mentors passed the Freshman English course, but only 46% of the students in the control group passed the same course. Later, more of the students with mentors went on to 4-year colleges. While much more controlled research of this nature needs to be done before we can fully understand how mentors help students overcome their educational problems, this example shows how a field experiment can advance our understanding of a social issue like success and failure in schools.
The Hawthorne Effect A common problem of experimental studies is that just by paying attention to people in an experimental group the researcher may be introducing additional variables. This problem was recognized for the first time in the late 1930s, when a team of researchers led by Elton Mayo conducted a famous series of experiments at Western Electric Hawthorne plant. The purpose of the study was to determine the effects of various environmental and social conditions on the worker's productivity. One hypothesis was that improvements in the physical features of the workplace (such as better lighting and improved social conditions (coffee breaks, different methods of payment, and so on) would result and greater productivity (the dependent variable). But it appeared that no matter what the experimenters did, the worker's productivity increased. When the experimenters improved the lighting, the worker's productivity increased over that of the control group. When they dimmed the lights, productivity went up again. This was true even when the workers were subjected to somewhat worse conditions or were returned to their original conditions.
At first the series of experiments was considered a failure, and the researchers concluded that the variables they were introducing had little to do with the changes in productivity that resulted. But on further reflection they realized that the real independent variable was not better working conditions but, simply attention. The workers liked the attention they were getting; it made them feel special, so they worked harder. This realization led the researchers to design experiments dealing with the effects of attention from supervisors and better communication between workers and their managers; those experiments led to a new philosophy of worker-management relations. Today the term Hawthorne effect is used to refer to any unintended effect that results from the attention given to subjects in an experiment.
The Hawthorne effect occurs in a variety of experimental situations. For example, in the Evergreen Valley College experiment described earlier, it is possible that the students with mentors were most or more successful than the students in the control group because they were getting extra attention from adults. Over time the effects of that attention may disappear and the experimenters may conclude that once the novelty of the extra attention wears off, students with mentors do not do appreciably better and others. Sophisticated evaluations of educational change try to anticipate the Hawthorne effect by making sure that the experimental group does not get more attention than the control group.
*SOURCE: SOCIOLOGY IN A CHANGING WORLD, 6TH ED., 2003, WILLIAM KORNBLUM, PP. 35-38*
end
|
No comments:
Post a Comment