What is field experimentation?
As an alternative to studying behavior in the sometimes restricted and sterile confines of the laboratory, scientists can turn to field experimentation as a method of finding out how people or other organisms interact with their natural environment. As the term field experimentation implies, genuine science is being conducted; however, the research takes place in the context of the places where the subjects normally live, work, and play. Instead of removing subjects from their normal surroundings and placing them in artificial situations, a field researcher attempts to study behaviors as they occur spontaneously in the real world.
Royce Singleton Jr., Bruce Straits, Margaret Straits, and Ronald McAllister, in their book Approaches to Social Research (1988), make the point that field experimentation procedures were used long before the techniques were recognized by the scientific community. The authors also state there is a consensus that anthropologists—followed shortly thereafter by sociologists—first developed and then legitimized this approach to research. Anthropologist Franz Boas and sociologist Robert Park were among the early pioneers of field research during the late nineteenth century and the beginning of the twentieth century. Boas was noted for his research in cultural anthropology. He emphasized the importance of circumventing one’s Western cultural biases by living in another culture for an extended time and acquiring that culture’s perspective. On the other hand, Park, who taught for a number of years at the University of Chicago, was influential in encouraging students to use the city as an alternative laboratory—studying people where they lived.
Field experimentation grew out of a need to seek answers to questions that could not be brought into a laboratory setting. Foreign cultures, complex social relationships, and secretive sects are examples of the phenomena that lend themselves to this method. Laboratory research—research in which phenomena are studied in an artificial setting with rigorous procedures in place to control for outside influences—might be seen as a hindrance to understanding dynamic human behavior. An alternative method needed to be found, and field research filled this vacuum.
Early in its development, field experimentation used data-collection procedures that almost entirely consisted of informal notes. A long narrative describing a sequence of behaviors would not have been uncommon. There has been a gradual move toward the use of more “objective” techniques such as standardized rating scales, behavioral checklists, and structured surveys. These methods were created to quantify better the observations being made. Once the behaviors could be quantified (that is, once specific behaviors could be assigned numbers), they could be subjected to the same statistical analyses used by laboratory experimenters. This improved approach to data collection helped field experimentation methods play a significant role in the social and behavioral sciences.
Studying people in their natural environments can yield a number of advantages over more traditional laboratory research methods. For example, it has been found that when subjects are aware that they are being studied, their actions sometimes differ from their actions when they are unaware that they are being observed. This phenomenon is known as the Hawthorne effect. A field study can avoid the Hawthorne effect by enabling the researcher to go “undercover” and study the subjects without their being aware that a study is going on. A field study helps ensure that genuine, rather than contrived, behaviors will emerge.
Another advantage of the field experimentation method is that it lends itself to the study of complex behaviors, such as relationships among family members, that would be too difficult to simulate in a laboratory setting. Another important strength is that the researcher can maintain the interaction between the subject and the setting in which the subject lives or works. Under this set of circumstances, the field study is the method most preferred. In addition, there are some instances in which time does not allow the researcher to bring the phenomena under study into the laboratory. Such instances include those associated with natural disasters or national calamities. For example, a researcher might want to study the psychological reactions of people who have lost their homes in a hurricane. Since it would be imperative to begin collecting data immediately, taking the time to develop a comprehensive survey or to identify and eventually test the important variables in a more controlled setting would jeopardize the data collection of this dynamic, rapidly changing situation.
Conducting a field experiment does not come without its share of disadvantages. First, there are many topics worthy of study that are too difficult to stage outside the well-controlled confines of a laboratory. Studying memory loss or the processes involved in solving a complex algebra problem are examples of these kinds of topics. Second, some researchers argue that because so many uncontrolled outside influences are present in a field study, it is difficult, if not impossible, to understand causal relationships among the behaviors being studied. Third, field research is particularly susceptible to the biases of the researcher while the data are being collected. Since data collection is typically less standardized and formal than in other methodologies, it is possible that the researcher may be unaware that observations that support the researcher’s hypothesis may be recorded and given more attention than behaviors that go contrary to the researcher’s beliefs. Some of the research published by anthropologist Margaret Mead during the 1920s, for example, has been called into question for this reason by other researchers who have reached different conclusions.
Field experimentation usually entails going into a naturalistic setting to collect data that can be used to generate research questions. The researcher will take such information, begin to organize it, and try to draw some general conclusions from it. This process, referred to as inductive research, occurs when data are first collected and then used to formulate general principles. Thus, field research differs from many other kinds of research methodologies. Field research begins with a broad theory, then sets out to test specific aspects of the theory to see if the data support it.
Field experimentation represents a variety of strategies for studying behavior. One specific technique involves a researcher who goes into the field and chooses to identify herself or himself to the subjects; the researcher also becomes actively involved in the group’s activities. The researcher has become a participant observer. An example of this method would be a person who wanted to study a violent inner-city gang. The researcher might approach the gang’s leadership, then identify himself or herself and give reasons for studying the group. The researcher would also participate in the gang’s meetings and other activities. Perhaps a better approach, in this situation, would be to do everything described except participating in the group’s activities, especially if the gang’s activities were illegal or harmful to others. In that case, the researcher, who revealed his or her true identity to the group yet chose to play a passive, inactive role from a distance, would be considered a “nonparticipant observer.”
An equally important field study technique involves concealing the identity of the researcher from the group that is being studied. In a classic study by John Howard Griffin described in the book Black Like Me (1962), Griffin colored his skin to take on the appearance of a black man. He then traveled throughout the American South, documenting his experiences, especially those involving race discrimination. This kind of activity is called covert research.
Conducting research in the field does not prevent the researcher from manipulating or altering the environment. In fact, it is a rather common occurrence for the “field” to be contrived. For example, a study on altruism might be designed for field experimentation. A scenario would be designed to discover what kind of person would come to the assistance of someone in need. The “need” could be helping to fix a flat tire or helping a lost child find his or her mother. In either case, since both scenarios occur infrequently in the real world and would be difficult to study, the setting would need to be staged. The ability to stage events opens the possibility of studying a variety of phenomena in a convenient context.
Five steps need to be completed in field experimentation. First, an appropriate field must be selected. This is a crucial decision, because the quality of the research hinges on the vitality of the data collected. Second, specific methods and techniques (for example, nonparticipant observation) must be developed to ensure that the behaviors the experimenter wants to observe can occur. In addition, an attempt must be made to eliminate outside influences that might bias the research. Third, the data must be collected. Fourth, the data must be organized, analyzed, and interpreted. The fifth and final step is to report the study within an appropriate format, which might be either a journal article or a book. To show how these steps are implemented and how field experimentation can contribute to scientific knowledge, two examples will be explored in detail.
In his article “The Pace of Life,” published in American Scientist (1990), Robert Levine attempted to understand how different cultures perceived time. In his opinion, attitudes toward time could affect a society’s pace of life and ultimately might lead to detrimental health problems for its members. Levine chose to collect data from the largest city in six different countries: Japan, Taiwan, Indonesia, Italy, England, and the United States. To gauge the general pace of life, he chose to study three unique indicators: the accuracy of outdoor bank clocks, the average time it took pedestrians to walk a distance of 100 feet (about 30.5 meters), and the time needed for a postal clerk to complete a transaction that entailed selling stamps and returning some change. None of these measures relies on subjective evaluations of the pace of life by the person collecting the data. Levine preferred these particular “objective” measures over a survey approach, which might have required subjects to respond to how they “feel” about the pace of life. He was more interested in direct measures of behavior as indicators of pace.
Standardized techniques were employed while collecting the data to ensure that the pace-of-life indicators would be measured fairly. For example, walking speed would not be measured if it were raining outside. Levine chose a covert approach, since he did not want subjects to be aware that they were in a study, thus eliminating any Hawthorne effect. In addition, both participant and nonparticipant observations were made. Measuring walking speed some distance away from a subject would be an example of nonparticipant observation. On the other hand, the purchasing of stamps on the part of the experimenter was an example of the participant observer technique.
The data were collected primarily by Levine’s students, who visited the countries. The data were then analyzed via basic statistical procedures. The study revealed that Japan had the fastest pace of life of the six countries, scoring the highest on all three measures. The United States came in with the second-fastest pace, followed by England; Indonesia was last, having the slowest walkers and the most inaccurate clocks.
Levine extended this research by looking at associations between the pace of life and both psychological and physical health. He found that the tempo of a society is significantly related to the prevalence of heart disease. In fact, the time-related variables often turned out to be better predictors of heart disease than psychological measures that identify high-energy behaviors in individuals. He concluded that a person who chooses to live in a fast-paced city should take necessary precautions to keep from becoming a time-urgent person. Living in a busy and stressful city can lead to unhealthy behaviors such as smoking and poor eating habits.
In another field study, which came to be known as the Rosenhan experiment, David L. Rosenhan studied mental health professionals’ ability to distinguish the “sane” from the “insane.” Rosenhan later published the research in the article “On Being Sane in Insane Places” in 1973. He sent eight psychologically stable individuals to twelve different mental institutions to find out if they would be admitted as patients. Each “pseudopatient” went to an institution with an assumed name and a false occupation; this was necessary because three of the pseudopatients were psychologists and one was a psychiatrist, and they might be treated differently from other patients. The pseudopatients told the admitting staff that they had been hearing voices that appeared to say the words “hollow,” “empty,” and “thud.” All pseudopatients were admitted and diagnosed as schizophrenic or manic-depressive. From the moment the pseudopatients gained entrance into the institutions, they began to act in a completely normal manner.
Rosenhan’s study used both covert and participant observation techniques to collect the data. Field notes (the recorded behaviors and observations that make up the data of the field study) concerning the behavior of staff members were taken on a daily basis. Although Rosenhan was shocked that all of his assistants (as well as himself) were admitted, he was even more dismayed that the pseudopatients’ “insanity” was never questioned by the staff. When the pseudopatients were observed writing their field notes, the behavior was interpreted by many of the staff members as paranoid and secretive.
The pseudopatients were released from the hospital between seven and fifty-two days later. Field studies, as this example indicates, can be filled with risks. None of the pseudopatients truly expected to be admitted, let alone having to stay an average of nineteen days in the hospital before the mental health professionals declared them well enough to be released. Rosenhan’s study was significant because it underscored the problem of distinguishing the normal from the abnormal with conventional diagnostic procedures. Rosenhan applied the results of this study to the broader issue of psychological labels. He pointed out that categorizing an individual with a particular mental illness can be misleading and in many instances harmful. Rosenhan’s pseudopatients were discharged with the label “ schizophrenia in remission”—that is, according to the mental health workers, they had been relieved of their insanity, although perhaps only temporarily.
Baker, Therese L. Doing Social Research. 3rd ed. New York: McGraw-Hill, 1999. Print.
Berg, Bruce Lawrence. Qualitative Research Methods for the Social Sciences. 7th ed. Boston: Allyn, 2009. Print.
Griffin, John Howard. Black Like Me. New York: New Amer. Lib., 2003. Print.
Levine, Robert V. “The Pace of Life.” American Scientist 78 (1990): 450–59. Print.
Mehl, Matthias R., and Tamlin S. Conner, eds. Handbook of Research Methods for Studying Daily Life. New York: Guilford, 2012. Print.
Reason, Peter, and Hilary Bradbury. Handbook of Action Research: Participative Inquiry and Practice. Thousand Oaks: Sage, 2001. Print.
Reis, Harry T., and Charles M. Judd, eds. Handbook of Research Methods in Social and Personality Psychology. 2nd ed. New York: Cambridge UP, 2014. Print.
Rosenhan, David L. “On Being Sane in Insane Places.” Science 179 (1973): 250–58. Print.
Singleton, Royce, Jr., Bruce C. Straits, M. M. Straits, and Ronald J. McAllister. Approaches to Social Research. 4th ed. New York: Oxford UP, 2005. Print.
Swingle, Paul, ed. Social Psychology in Natural Settings: A Reader in Field Experimentation. New Brunswick: Aldine Transaction, 2007. Print.
Teele, Dawn Langan, ed. Field Experiments and Their Critics: Essays on the Uses and Abuses of Experimentation in the Social Sciences. New Haven: Yale UP, 2014. Print.