Are Laboratory Experiments the Best Way to Carry out Research - Essay Example

Are laboratory experiments the best way to carry out research in social psychology In recent years the field of social psychology has advanced sufficiently so that it is now possible to do strictly controlled laboratory experiments in many areas. What is important is that it is now possible to do these experiments so that they effectively add to our understanding of the dynamics of the behavior of groups in real-life situations. The social psychologist who attempts to do such experimentation, however, is faced with a number of difficult decisions which he must make. He must ask, first: do I know enough about the area I am investigating so that I may be confident that I will be dealing with important factors Second, do I know enough about these important factors so that I can adequately measure, control, and manipulate them in the laboratory And third, is my theoretical understanding sufficiently well advanced so that I can state specific hypotheses in a form amenable to precise scientific answer Thinking through the answers to these questions on a specific problem makes it clear that before we can begin to use the powerful technique of laboratory experimentation effectively, we must already have gained much understanding by means of other techniques, such as observation of uncontrolled situations and field experiments under less strictly controlled conditions. From the controversy in above, it is obvious that there is no definite decision whether laboratory experiments are the best way to conduct the research in the field of social psychology. Some scientists believe that laboratory experiments do not help to proceed with getting exact and effective final outcome in the research of one or another issue of social psychology; those are discourse analysts. Others keep the opinion that laboratory experiments is the best way to empirically identify the nature of a social problem as when using cognitive approach towards dealing with the problem is a pledge of its successful outcome. I will further gradually review each approach in order to define whether laboratory experiments are applicable and effective for current development of social psychology. Practical examples will be used in the following discussion in order to better shape the distinction between these to approaches. For the past two decades, psychological social psychologists have largely focused on the latter tenet, emphasizing inside-the-head phenomena, and have left the former tenet to sociological social psychologists. As it was observed, [psychological] "social psychology and cognitive social psychology are today nearly synonymous. The cognitive approach is now clearly the dominant approach among social psychologists, having virtually no competitors" (Gergen, 1992). Moreover, the nature of cognitive theorizing shifted from an emphasis on "warm" cognition and motivated reasoning to an emphasis on "cold" cognition. Consequently, memory processes, logical inference, and cognitive biases became key topics of interest in cognitive social psychology. This shift reflected the advent of information processing models in cognitive psychology in the late 1990s and their rapid adoption by social psychologists, documented in a seminal volume on person memory. The resulting theoretical approach, known as social cognition, has been characterized as social psychology under the paradigm of information processing. It emphasized information encoding, storage, and retrieval and drew heavily on computer metaphors. On the positive side, the information processing approach has rapidly advanced our understanding of the cognitive processes underlying many social phenomena (for reviews, see Devine et al 1994). Moreover, its emphasis on detailed process models has changed the field's standards for what counts as appropriate evidence: "It was no longer enough to detail a theoretical model and use the results as confirmation of the model. If one stated what one thought the process was, one had to demonstrate the intervening steps" (Taylor 1998, p. 74). To obtain relevant process data, social psychologists now include response latencies, recall and recognition tests, think-aloud protocols, and similar measures in their experiments, and they increasingly employ mediational analyses to test the theoretically specified process assumptions. Perhaps the best-known experimental studies of productivity in social psychology are the classical "alone-and together" experiments. The experimental technique was the simple procedure of measuring the output of the individual alone, without the presence of others, and then comparing it with his output on the same task in a group situation. Some studies, for example, measured the individual's speed of tapping in isolation and compared it with the speed of tapping in a group of subjects all engaged in the same activity. In general these studies demonstrated that an individual is more productive in a group setting than he is in isolation. However, they were not really measuring the productivity of a group as we think of the definition of a group in current research. In the "together" situations there was parallel individual behavior rather than interdependent behavior of members with differentiated functions. Thus the alone and together experiments were a transition stage from the study of individual psychology to the study of the psychology of real groups where the members are more highly interdependent and have a sense of belonging to an existing group. The classical field experiments on group productivity are the comprehensive experiments conducted in the Hawthorne plant of the Western Electric Company. These experiments studied a real group in a factory where the members were, in fact, highly interdependent in their behavior. The original purpose of the experiment was to determine the relationship between physical conditions of work and the productivity of the group. Many months of careful measurement showed no significant relationship between output and such physical factors as temperature and illumination. The experiments did demonstrate in a dramatic way, however, the importance of social factors in group productivity. From a methodological point of view the most interesting of these findings was what we might call the "Hawthorne effect." In order to control more precisely the physical factors affecting production, the experimenters had set up a special experimental room for a small group of girls wiring relays. This bank-wiring room was separated from the rest of the factory and the girls working in it received special attention from both outside experimenters and the management of the plant. Careful studies of this bank-wiring group showed marked increases in production which were related only to the special social position and social treatment they received. Thus it was the social aspects of the experimental conditions set up for measurement which produced the increases in group productivity. A more recent series of field experiments on group productivity has been conducted in the sewing plant of the Harwood Manufacturing Corporation in Marion, Virginia. These experiments were a logical development of the work of Lewin, Lippitt, and White on the effect of autocratic and democratic atmospheres on groups of children. The autocracy-democracy experiments at Iowa were laboratory experiments designed specifically to explore the effects of different styles of leadership and of social atmosphere and organization on the behavior of the groups. They showed among other things that the democratic clubs of children were more productive in mask making than the autocratic clubs. In considering the significance of the greater productivity of the experimental democratic groups, two problems immediately arose: First, are these semi-qualitative judgments of papier-mch masks really adequate measures of productivity Second, would democratic techniques induce greater productivity in a tough situation with adults such as in a factory It is interesting to note that both these problems pointed to a field experiment as the best methodology for a next step in studying leadership techniques in relation to group productivity. As a result of all these experiments, three conclusions seem justified: First, the application of skillful democratic techniques of leadership in industrial settings can result in extremely marked increases in group productivity which will persist over long periods of time. Second, the procedure used, such as the difference between group decision and pacing cards, does make a difference, but this difference is not independent of the style of leadership. Third, in such experimental settings the style of personal leadership of the face-to-face group is probably the most important variable, and it differs markedly from one experimenter to another. In all these experiments it had been necessary to make sure that the social conditions and the style of leadership used by the experimenter were not contradicted by the supervisors once the group was back in the work setting. For example, if an experimental group of operators decided to use pacing cards, it was necessary to instruct that supervisor not to press for faster production during one of the periods when they had planned to work slowly. The more general implications of these facts were clear: One cannot hope to change the level of productivity of the total factory When discussion cognitive approach toward social psychology it is important to mention the works of Michael Billing who's interest in the study of ideology has continued throughout his recent works. The emphasis is made on looking at extreme forms of ideology towards looking at the influence of ideology on common-sense, or everyday patterns of thinking. For example, he conducted a study based upon family discussions about the Royal Family (Talking about the Royal Family, Routledge, 1992). His latest book (Banal Nationalism, Sage, 1995) has examined the way that contemporary life is infused with nationalist assumptions and symbols, which often pass unnoticed. In 1996 this book received the Myers Center Award for "outstanding work on intolerance in North America". Contradictory as for the laboratory experiments' ineffectiveness is enhanced by discourse analysts who believed that laboratory experiments alone are not sufficient, no matter what the state of development of social psychology may be -- now or in the future. Without the constant checking against field situations and without the continual supplying of ideas from the field to the laboratory, the laboratory investigator runs the risk of having his research become fruitless, sterile, and unimportant. Another statement of discourse analysts is that without the rigorous and precise explorations of the laboratory, the field investigator risks the danger of building up an insecure structure with tentative hypotheses resting on inconclusive results. The controlled laboratory experiment is not an attempt to duplicate, in miniature, a real-life situation. It is rather an attempt to set up the pure case. In other words, it is an attempt to take a factor or cluster of factors which we have good reason to believe are important and systematically vary them in a context where other factors are well controlled. By such means we may begin to build up a body of knowledge concerning the precise functional relationships between these factors and the behavior of human beings in social situations. The laboratory experiment can give us the empirical laws of behavior. It does not immediately tell us how to apply these laws to the real-life situation. To permit application we must proceed to the diagnosis and measurement of factors which operate in real-life situations, and to the determination of which laws do and do not apply. We must, however, know the laws before we can apply them. Discourse analysts speak about measures of independent variables. The role in experiments of measures of independent variables and proposed mediating variables is the topic of current study. In the causal sequence assumed in interpreting an experimental result, the independent variable and proposed mediating variable are presumed to mediate the effect of the experimental treatment on the dependent measure. Measures of independent variables and mediators provide checks on the assumptions that the experimental treatment successfully manipulated those variables and are unquestionably useful. A separate, controversial issue is whether such measures are necessary in experiments. If no plausible alternative explanations exist, data from such measures are not needed. Plausible alternative explanations are not eliminated by data from such measures. Alternative explanations, critical for assessing construct validity (Mills, 1969), are distinguished from different general theoretical accounts of a finding. There are very good reasons for collecting measures of independent variables and mediating variables in social psychology experiments, and we have no wish to question the usefulness or to discourage in any way the inclusion of such measures in experiments. But despite the usefulness of measures of independent variables and mediators, occasions occur when experiments are conducted that do not include such measures. Good measures of independent variables or mediators may not be feasible for a variety of reasons. Such measures may draw attention to the purpose of the study, thereby creating suspicion that would invalidate the test of the experimental hypothesis. They may contaminate responses to the dependent measure, unless taken after the dependent measure by which time they may be less appropriate (e.g., checks on mood manipulations). Sometimes the inclusion of a measure of the independent variable can actually undermine the success of the manipulation of the independent variable by raising doubt about the veracity of the instructions designed to manipulate the independent variable. The divergence of views concerning the necessity of including measures of independent variables and mediating variables in experiments has important consequences for experimental social psychology. Experiments that do not include such measures will sometimes be considered flawed even when they are otherwise well designed. Experimenters may be constrained to include measures of independent variables or mediators in experiments even if they have reason to believe the inclusion of such measures may compromise the procedure of the experiment. We believe that social psychologists need to examine carefully the issue of whether measures of conceptual independent variables and mediators are necessary in experiments. In a social psychology experiment, what is of interest is evidence of a causal relation between a conceptual independent variable and a conceptual dependent variable. Sometimes interest also exists in the operation of a mediating variable that links the independent and dependent variables, that is, a conceptual variable that is influenced by the conceptual independent variable and that, in turn, influences the conceptual dependent variable. In an experiment, the conceptual independent variable, an abstract concept, is operationalized in a concrete way by the experimental treatment. The conceptual dependent variable, also an abstract concept, is operationalized in a concrete way by the dependent measure. The experimental treatment and dependent measure are chosen because the researcher believes that an effect of the experimental treatment on the dependent measure will provide evidence of a causal relation between the conceptual independent variable and the conceptual dependent variable. Researchers of discourse approach assume that the experimental treatment has an effect on the dependent measure as the result of the following causal sequence: The experimental treatment is perceived by the experimental participants, differences in the perceived treatment create differences on the conceptual independent variable, differences on the conceptual independent variable create differences on the conceptual dependent variable, and differences on the conceptual dependent variable create differences on the dependent measure. In conclusion I would like to say, that the difficulties of making rapid progress in social psychology are numerous. At present there are a lot of laboratories in the country that are adequately equipped to do this kind of research. There are also very few social psychologists in this country who find that they have the time to do such research. The organization of social science research units and research teams which have adequate facilities and abilities will greatly speed this development. Bibliography: 1. Mills, J. (1969). "The experimental method." In J. Mills (Ed.), Experimental socialpsychology (pp. 409-448). New York: Macmillan. 2. Mills, J. (1977). "Criteria for publication." Personality and Social Psychology Bulletin, 3, 523. 3. Petty, R. E., & Cacioppo, J. T. (1981). Attitudes and persuasion: Classic and contemporary approaches. Dubuque, IA: Brown. 4. Sigall, H., & Michela, J. (1976). "I'll bet you say that to all the girls: Physical attractiveness and reactions to praise." Journal of Personality, 35, 12-20. 5. Billig, Micheal. Arguing and Thinking: A Rhetorical Approach to Social Psychology. Cambridge: Cambridge University Press, 1987. 6. Gergen, K. (1992). Social construction and moral action. In D. Robinson (Ed.), Social discourse and moral judgement (pp. 9-27). San Diego: Academic Press. 7. Harre, R. (1980). Social being: A theory for social psychology. Totowa, NJ: Littlefield, Adams & Company. 8. Devine PG, Hamilton DL, Ostrom TM, eds. 1994. Social Cognition: Impact on Social Psychology. San Diego: Academic. 9. Taylor SE. 1998. The social being in social psychology. In Handbook of Social Psychology, ed. D Gilbert, ST Fiske, G Lindzey, 1:58 98. New York: McGraw-Hill. 4th ed. Read More