Observation
Observation
I. Social Observation and Social Case StudiesHoward S. Becker
BIBLIOGRAPHYRosalie Hanl⊙y Wax
I SOCIAL OBSERVATION AND SOCIAL CASE STUDIES
The term “case study” comes from the tradition of medical and psychological research, where it refers to a detailed analysis of an individual case that explicates the dynamics and pathology of a given disease; the method supposes that one can properly acquire knowledge of the phenomenon from intensive exploration of a single case. Adapted from the medical tradition, the case study has become one of the major modes of social science analysis.
The case studied in social science is typically not an individual but an organization or community. Case studies have been done of such widely varying phenomena as industrial towns (E. Hughes 1943), urban neighborhoods (Gans 1962), factories (Dalton 1959), mental hospitals (Goffman 1961), and the interconnections of slums, politics, and rackets (Whyte 1943). Case studies of individuals are, of course, also undertaken by social scientists, especially in the form of the life history; but such studies, although often made by an earlier generation of sociologists and psychologists (Thomas & Znaniecki 1918-1920; Shaw 1930; Conwell 1937) are now relatively rare (but see H. Hughes 1961 and Williamson 1965).
The social scientist making a case study of a community or organization typically makes use of the method of participant observation in one of its many variations, often in connection with other, more structured methods such as interviewing. Observation gives access to a wide range of data, including kinds of data whose existence the investigator may not have anticipated at the time he began his study, and thus is a method well suited to the purposes of the case study.
Aims of the case study
The case study usually has a double purpose. On the one hand, it attempts to arrive at a comprehensive understanding of the group under study: who are its members? what are their stable and recurring modes of activity and interaction? how are they related to one another and how is the group related to the rest of the world? At the same time, the case study also attempts to develop more general theoretical statements about regularities in social structure and process.
Because it aims to understand all of the group’s behavior, the case study cannot be designed singlemindedly to test general propositions. In contrast to the laboratory experiment, which is designed to test one or a few closely related propositions as rigorously and precisely as possible, the case study must be prepared to deal with a great variety of descriptive and theoretical problems. The various phenomena uncovered by the investigator’s observations must all be incorporated into his account of the group and then be given theoretical relevance.
So stated, the aims of the case study can scarcely be realized; it is Utopian to suppose that one can see, describe, and find the theoretical relevance of everything. Investigators typically end up by focusing on a few problems that appear to be of major importance in the group studied—problems that touch on many aspects of the group’s life and structure. Thus a community study (E. Hughes 1943) may come to focus on the problems of industrialization and cultural contact, or a study of an urban neighborhood may focus on the relation between ethnicity and social class (Gans 1962).
The comprehensive goal of the case study, however, even though it is not reached, has important and useful consequences. It prepares the investigator to deal with unexpected findings and, indeed, requires him to reorient his study in the light of such developments. It forces him to consider, however crudely, the multiple interrelations of the particular phenomena he observes. And it saves him from making assumptions that may turn out to be incorrect about matters that are relevant, though tangential, to his main concerns. This is because a case study will nearly always provide some facts to guide those assumptions, while studies with more limited data-gathering procedures are forced to assume what the observer making a case study can check on.
The aims of the case study and the kinds of problems it ordinarily poses for study suggest particular techniques of data gathering and analysis. After describing these, we will consider the uses, scientific and otherwise, which may be made of observational case studies.
Techniques of observation
In gathering data, the participant-observer engages in a number of different activities. One can distinguish several possible modes of proceeding, depending on the degree to which one is a participant as well as an observer (Gold 1958). At one extreme, the observer may not participate at all, as when he hides behind a one-way screen in an experimental room; at the other, he may be a fullfledged participant, living in the community under study or holding a full-time job in the organization he studies and subject to the same life chances as any other member of the group. The particular techniques he uses are shaped by the demands of playing these different roles; a hidden observer cannot openly interview other participants, while a known observer may find that certain group secrets are systematically kept from him.
The observer places himself in the life of the community so that he can see, over a period of time, what people ordinarily do as they go about their daily round of activity. He records his observations as soon as possible after making them. He notes the kinds of people who interact with one another, the content and consequences of the interaction, and how it is talked about and evaluated by the participants and others after the event. He tries to record this material as completely as possible by means of detailed accounts of actions, maps of the location of people as they act (Whyte 1943), and, of course, verbatim transcriptions of conversation.
The problem of bias
The observer has the problem of trying to avoid seeing only those things which accord with his explicit or implicit hypotheses (Zelditch 1962). This kind of bias can occur in several ways. The observer, interacting with those he studies on a long-term basis, comes to know them as fellow human beings as well as research subjects; thus, he can hardly help acquiring feelings of friendship, loyalty, and obligation, which may make him wish to protect some members of the group by not seeing those events which would render them liable to criticism. Some persons or factions may see his research as dangerous and try to keep him from seeing certain aspects of group activity (Dalton 1959). Finally, he may feel that certain events are so distasteful or personally dangerous (for example, the activities of homosexual networks or violent gang conflict) that he is unwilling or afraid to remain close enough to the participants to see what actually happens.
Bias can be avoided by carefully rendering a complete account of all events observed; by seeking to cover all varieties of events by some kind of primitive sampling device (making observations at different times of the day or year, deliberately seeking out members of different groups in the community or organization, and so forth); and by formulating tentative hypotheses as the field work proceeds and then deliberately searching for negative cases (Lindesmith 1947). These topics are more fully treated in the next section of this article.
Types of data
The observer is especially alert for incidents of anything defined as conflict or “trouble” by the community or organization being studied. Such incidents enable him most quickly to discover the expectations that guide interaction; when expectations are violated, trouble follows. By seeing what kinds of actions produce conflict, the observer can infer the existence of implicit expectations, which then become part of his analytic model of the group under study.
He is also alert to nuances of language, such as special meanings given to ordinary words, for these signal the existence of situations, events, and persons the members of the group think distinctive enough to merit being singled out linguistically and thus give a clue to the characteristic problems and responses of the group. By inquiring into the meaning and usage of an unusual term, by investigating instances of its use and seeing when it applies and when it does not, he adds to his analytic model (Becker & Geer 1957).
The observer does not confine himself to observation alone. He may also interview members of the group, either alone or in groups. In the first case, he can inquire into the social background and earlier experiences of a participant as well as into his private opinions about current affairs. In the latter, he is in effect “tapping” the ordinary kinds of communications current in a group, seeing what members will say when in the company of other members. The difference between private opinion and public communication may provide important clues to group norms (Gorden 1952).
The observer will also find it useful to collect documents and statistics (minutes of meetings, annual reports, budgets, newspaper clippings) generated by the community or organization. These can furnish useful historical background, necessary documentation of the conditions of action for a group (as in a codified set of rules), or a convenient record of events for analysis (as, for instance, when a college newspaper reports the marriages of students, specifying their position in the campus social structure). In every case, the observer must inquire carefully into how the documents he works with are created: by whom, following what procedures, and for what purposes? For it is clear that documents cannot be taken at face value but must be interpreted in the light of such considerations (Kitsuse & Cicourel 1963).
The observer may also create his own statistics for the solution of particular problems. Thus, one may observe the number of times people in an office ask one another for advice (Blau 1955), or one may keep accurate records of one’s own piecework production in a machine shop, to be used as an indication of what is possible for the average group member (Roy 1952).
Techniques of analysis
It is a truism to say that procedures of analysis and proof take their form from the problem one is trying to solve. It is more important to indicate the variety of problems typically encountered in analysis of observational material and the means by which they may be solved.
Observational materials, since they are usually gathered over a long period of time, can be analyzed sequentially. That is, analysis need not await completion of data gathering but can go on concurrently with it; results of early analyses may be used to direct further data-gathering operations. Different problems arise at different stages of the research.
Choice of problem
In the beginning the researcher may not be sure what problem is most deserving of study in the community or organization he is working in; he devotes his first analytic efforts to uncovering worthwhile problems and hypotheses that will prove most useful in attacking them (Geer 1964). Researchers frequently discover that the problem they set out to study is not as important as, or cannot be studied except in the context of, some other problem they had not anticipated studying. Thus, Vidich and Bensman (1958) found that the problem of the relationships between the rural communities and the various agencies and institutions of American mass society that affect rural life could be understood only if one also investigated how the community and its members were able to function in spite of the fact that their immediate social environment demonstrably negated their basic beliefs.
In selecting problems, hypotheses, and concepts, the investigator works from concrete findings made early in the research. Typically, he discovers that a given event has occurred, perhaps only once, and asks what the significance of such an event might be. It may be an incident of conflict or the kind of linguistic nuance already referred to. Whatever it is, the investigator must first ascertain that the event actually is what it seems to be and then trace out its possible theoretical implications. The first problem requires him to consider whether people may have been consciously or unconsciously deceiving him; this can be checked by an assessment of whether the event that arouses his curiosity is one that was concocted for his benefit or whether it would have occurred in the same way even if he had not been there. For instance, a statement volunteered by an informant who does not know what the observer is after may be given more weight than one that has been influenced by the observer’s leading questions. Similarly, an event that occurs in an ordinary institutional context, subject to all the constraints of that context, can be given more weight than one that occurs without being observed by other members of the group.
The observer then traces the possible theoretical implications of his finding by considering what class of events it might be representative of, utilizing such theory as is available about that class of events to deduce further propositions. For instance, if one hears a worker in a service profession categorize members of his clientele, he may apply the proposition that such a categorization will be based on the problems clients of various kinds pose for the worker in trying to realize his occupational goals. (Teachers, for example, distinguish pupils according to how hard they are to teach and discipline; doctors distinguish patients according to how easily they can be cured, whether they pay on time, and so on.) Working from this, the observer begins to look for the basic problems implied by the set of categories and the way the problems impinge on workers at different career stages. Obviously, a large number of theories may be applied to discrete observations in order to draw out their implications and use them to direct further observations.
Quasi-statistical method
At a later stage, the observer, having decided, at least provisionally, what he will study in the situation at hand and what theoretical apparatus he will use, is concerned with whether his initial findings hold for the entire community or organization. His data will usually not, unless expressly gathered for the purpose, be sufficiently systematic to be amenable to statistical manipulation. But he can generate what have been called “quasi statistics” (Barton & Lazarsfeld 1955), that is, such imprecisely sampled and enumerated figures as his data contain. Such data are often quite adequate for the points he wishes to make.
In particular, quasi statistics may allow the investigator to dispose of certain troublesome null hypotheses. A simple frequency count of the number of times a given phenomenon appears may make untenable the null hypothesis that the phenomenon is infrequent. A comparison of the number of such instances with the number of negative cases—instances in which some alternative phenomenon that would not be predicted by his theory appears—may make possible a stronger conclusion, especially if the theory was developed early enough in the observational period to allow a systematic search for negative cases. Similarly, an inspection of the range of situations covered by the investigator’s data may allow him to negate the hypothesis that his conclusion is restricted to only a few situations, time periods, or types of people in the organization or community.
The technical problem in creating quasi statistics lies in making sure that one has in fact inspected all the relevant cases. A number of workers have devised schemes for doing this. (A representative scheme is described in Becker & Geer 1960.) The common feature of these schemes is the reduction of the body of data by making an abstract of the field notes that have been accumulated, breaking them down into small units, and classifying each unit under all the analytic categories to which it might be relevant. When the investigator desires to analyze all the material on a given point, he sorts his units (which may be reproduced on keysort cards, for convenience), takes out those items which are irrelevant, and frames a conclusion that takes account of all the relevant evidence remaining.
One of the greatest faults in most observational case studies has been their failure to make explicit the quasi-statistical basis of their conclusions. Even though the investigator uses faulty sampling and enumeration procedures, his evidence may nevertheless be sufficient to warrant the conclusions he draws if he explicitly states what the evidence is and shows how his conclusions are related to it. In particular, the conclusions may appear extremely plausible (Polya 1954) if they are supported by several kinds of evidence at once. Thus, the conclusion that medical students make use of a perspective based on the values of clinical experience and medical responsibility gains great plausibility when it is shown not only that use of the perspective is frequent and appears in a wide range of situations but also that students’ characterizations of patients depend heavily on the same criteria (Becker et al. 1961, pp. 338-340).
Construction of models
As a result of the early stages of analysis, the researcher acquires a number of limited models of parts of the organization or community, propositions which describe one kind of interaction between one pair of statuses in one kind of situation. The final stage of a case study consists in a progressive refinement of these part-models (accomplished by continual checking against evidence already available in the field notes or newly gathered in the field) and their integration into a model of the entire organization or community. The model provides answers to the theoretical questions of the study and shows the contribution of each part of the analyzed structure to the explanation of the phenomenon in question.
Models of the community or organization that result from case studies are not to be confused with mathematical models. Rather, they have the same relation to the group studied that the natural history of a process (such as the race-relations cycle or the process of becoming a drug addict) has to any specific set of events said to embody it. In a natural-history analysis of a process we strip away the historical uniqueness of a number of instances of the same phenomenon, leaving as our result the generic steps in the process—those steps that would always occur if the same result were to be found. Similarly, in a case study of a social structure we strip away what is historically unique and concentrate on the generic properties of the group, viewed as an example of a particular kind of structure. Relations between the essential characteristics of that kind of structure are stated by verbal generalizations. For instance, one might study a prison or school with a view to discovering what the characteristic statuses and forms of interaction are in an institution in which one class of participants is present involuntarily. The result would be a model that might also be applied to other institutions having that characteristic, such as mental hospitals.
The problem of reliability
The reliability of such an analysis is sometimes questioned in an equivocal way that plays on the meaning of “reliability.” The question is put thus: would another observer produce, with the same analysis, the same total model, were he to repeat the study? The answer is of course that he would—but only if he used the same theoretical framework and became interested in the same general problems, for neither the theoretical framework nor the major problem chosen for study is inherent in the group studied. Nevertheless, given the same basic framework— for instance, a sociology based on conceptions of social structure, culture, and symbolic interaction —the same fundamental parts of the group studied would be found in a second study, even if the major problems chosen for study were quite different. For instance, one might study a medical school to discover how the students are changed by their experience in it; this would be a problem in the theory of adult socialization. Or one might, with equal justice, choose to use the medical school as the arena for a study of how specialists cooperate in a common task, a problem in the “politics” of complex organizations. In either case, a complete study would necessarily describe the same basic relationships among students, among faculty, between students and faculty, between both and patients, and so on. Admittedly, the theoretical use to which the analysis was to be put would shape the kind of structural model built, and a model built for one purpose might slight or ignore important elements in the other; but the two could be combined, so that neither would contain any element denied in the other.
The use of observational studies
Every case study allows us to make generalizations about the relations of the various phenomena studied. But, as has often been pointed out, one case is after all only one case. Suppose that some of the most important factors involved in understanding the particular theoretical problems posed by the case are so invariant in it that we are unaware of their importance. How is one to discover their importance?
The problem can be handled (or can in principle be handled) by gathering a large number of cases and “partialing out” the effects of various influences. In any case, the problem is not a real one if we take a long-term view of the development of theory. Each study can develop the role of a different set of conditions or variables as these are found to vary in the setting of the study. Over a series of studies, the comparison of variations in conditions and consequences can provide a highly differentiated theory of the phenomenon under study. As a simple example, a community study might locate six social classes in a community. A later study, in a somewhat different community, discloses only five, the upper class failing to divide between “old” and “new” wealth; comparison of the two may show variations in the histories or ecological positions of the communities that might account for the difference, and the hypothesis can be checked out in yet another study.
Comparative analysis: an example
To take another example, some studies of prisons (Sykes 1958; Cressey 1961) have revealed elaborate organizations of inmates around matters of deprivation; wherever inmates were deprived of something—material possessions, sex, autonomy—they developed social units and practices designed to deal with the deprivation as best they could under prison conditions. Because these early studies were all of men’s prisons, they could not discover what a later study of a women’s prison revealed: that the informal organization of the prison varied according to the kind of people recruited, because deprivations differ according to what it is one values and, therefore, misses when deprived of it. Women apparently set far less store by autonomy than men, do not miss it, and do not develop a sub rosa government; they are, however, very dependent on intimate affectional ties, miss their families intensely, and develop homosexual liaisons as their form of informal organization (Ward & Kassebaum 1965). Other studies might show the influence of age, region, and other factors on the organization of prison life. A series of comparisons, based on variations in the phenomenon, show the influence of each factor; each succeeding study can be built on the contributions of its predecessors.
Developing theory by comparative analysis is necessarily a protracted process. Comparative findings take many years to establish, for each study, by itself, may take several years and, for maximum effect, studies should be built on one another rather than being done simultaneously. The result can be a detailed understanding of the operation of a large number of factors and conditions as they interact to produce different results.
One useful strategy is to state the findings of each study as universal propositions, even though it is obvious that they are provisional. By doing so, the investigator makes it possible to identify exceptions to his propositions and to proceed most efficiently with fruitful comparisons (Lindesmith 1947).
Practical use of research findings
The observational study of an institution or community can be (and often is) used by various people in various ways, depending on their position in, or relation to, the group and their interest in its functioning. It does not differ in this from other kinds of research, but it does differ, typically, in the range and number of variables considered and the distance beneath the surface of events that the research reaches. Studies are often undertaken with the subsidiary—if not primary—purpose of providing guidance to administrators and others who may wish to intervene in the organization or community in order to change some condition thought to be inefficient, distasteful, or inimical to group welfare. The observational study is useful in identifying and specifying such problems and in finding their origins and consequences at various levels and in various parts of the group.
Clues to intervention. The observational study also makes it possible to go beyond the problem as originally conceived by those group members who wanted help and to discover other problems that, from some viewpoint other than theirs, require or warrant intervention. For instance, the officers of an autocratically governed trade union may not think the absence of organizational democracy a problem, but some of the members or an outside observer might take a different view. The farther beneath the surface the study penetrates, the more likely it is to discover problems that have not been labeled as such by the leaders of the group.
Whatever the problems identified, the wide range of the case study makes it likely that it will contain hints or suggestions as to the crucial points of possible intervention. Many studies diagnose the “causes” of a problem and yet are not useful for social action, because the causes discovered are not accessible to manipulation by the people involved. Thus, even though the generalization that the cause of teen-age vandalism lies in the early childhood experiences of the vandal might be true, knowing this is of little value. It is more useful to know, through close observation, that (as might be the case) vandalism takes place more frequently in unlit and unwatched places or becomes more frequent as the certainty of apprehension declines, for these matters are more subject to remedial action by police and other agents of social control.
Ethical problems of the researcher
The published report of an observational study may be used, either by members or outsiders, to embarrass or even endanger the organization or community studied or, at least, its leaders. Every group maintains fictions about itself—they may, perhaps, be necessary for the continued existence of the group—which present it as better in some ways than unprejudiced research will reveal it to be. A town may feel that its government is more broadly representative than it is; a hospital may think its treatment of patients more successful than it is. A case study is bound to reveal the discrepancy between the operating reality and the image believed in and presented to the rest of the world by members. When the results of the study are published, the discrepancy is publicly attested to in a way that members cannot ignore. Their enemies may make use of the opportunity to embarrass or attack them. The members may ask that the findings be withheld or may attempt to coerce the researcher into suppressing them.
The investigator therefore faces an ethical dilemma. Science requires frank and unfettered reporting, and the matters group members complain of may be important aspects of group functioning, whose suppression would emasculate the report and strip it of scientific significance. On the other hand, the investigator surely has some obligation not to bring harm to those who have allowed themselves to be studied; he may, indeed, have promised them that they will not be harmed. In making the promise, he may have meant merely that he would not expose any individual to ridicule or retribution—most sociologists probably regard this as a fixed ethical principle—but he now finds that he is being asked to respect the same niceties in the case of a group.
The solution to the dilemma depends in part on the investigator’s own ethical commitments. However, he can avoid some of the difficulties inherent in the research relationship by striking a clear bargain with those he studies before he begins his work, taking care to alert them to the full range of unpleasant possibilities they may be exposing themselves to. He can also attempt to educate those most likely to take offense at the final report, explaining to them as the study proceeds what its consequences are likely to be and helping them to find a workable way of living with the published study.
Howard S. Becker
[See alsoField Work; Interviewing.]
BIBLIOGRAPHY
Barton, Allen H.; and Lazarsfeld, Paul F. (1955) 1961 Some Functions of Qualitative Analysis in Social Research. Pages 95-122 in Seymour M. Lipset and Neil J. Smelser (editors), Sociology: The Progress of a Decade. Englewood Cliffs, N.J. : Prentice-Hall.
Becker, Howard S.; and Geer, Blanche 1957 Participant Observation and Interviewing: A Comparison.Human Organization 16, no. 3:28-32. → A discussion of the Becker and Geer article, by M. Trow, appears on pages 33-35; a rejoinder by Becker and Geer appears in Human Organization, Volume 17, pages 39-40.
Becker, Howard S.; and Geer, Blanche 1960 Participant Observation: The Analysis of Qualitative Field Data. Pages 267-289 in Richard N. Adams and Jack J. Preiss (editors), Human Organization Research: Field Relations and Techniques. Homewood, III.: Dorsey.
Becker, Howard S. et al. 1961 Boys in White: Student Culture in Medical School. Univ. of Chicago Press. BLAU, PETER M. (1955) 1963 The Dynamics of Bureaucracy: A Study of Interpersonal Relations in Two Government Agencies. 2d ed., rev. Univ. of Chicago Press.
Conwell, Chic 1937 The Professional Thief: By a Professional Thief. Annotated and interpreted by Edwin H. Sutherland. Univ. of Chicago Press. → A paperback edition was published in 1960.
Cressey, Donald R. (editor) 1961 The Prison: Studies in Institutional Organization and Change. New York: Holt.
Dalton, Melville 1959 Men Who Manage: Fusions of Feeling and Theory in Administration. New York: Wiley.
Gans, Herbert J. 1962 The Urban Villagers: Group and Class in the Life of Italian-Americans. New York: Free Press.
Geer, Blanche 1964 First Days in the Field. Pages 322-344 in Phillip E. Hammond (editor), Sociologists at Work: Essays on the Craft of Social Research. New York: Basic Books.
Goffman, Erving (1961) 1962 Asylums: Essays on the Social Situation of Mental Patients and Other Inmates. Chicago: Aldine.
Gold, Raymond L. 1958 Roles in Sociological Field Observations. Social Forces 36:217-223.
Gorden, Raymond L. 1952 Interaction Between Attitude and the Definition of the Situation in the Expression of Opinion. American Sociological Review 17: 50-58.
Hughes, Everett C. 1943 French Canada in Transition. Univ. of Chicago Press.
Hughes, Helen (Macgill) (editor) 1961 The Fantastic Lodge (The Autobiography of a Girl Drug Addict). Boston: Houghton Mifflin.
Kitsuse, John I.; and Cicourel, Aaron V. 1963 A Note on the Uses of Official Statistics. Social Problems 11:131-139.
Lindesmith, Alfred R. 1947 Opiate Addiction. Evanston, III.: Principia Press.
Polya, Gyorgy 1954 Mathematics and Plausible Reasoning. Volume 2: Patterns of Plausible Inference. Oxford Univ. Press; Princeton Univ. Press.
Roy, Donald 1952 Quota Restriction and Goldbricking in a Machine Shop. American Journal of Sociology 57: 427-442.
Shaw, Clifford R. (editor) 1930 The Jack-roller: A Delinquent Boy’s Own Story. Univ. of Chicago Press.
Sykes, Gresham M. 1958 The Society of Captives: A Study of a Maximum Security Prison. Princeton Univ. Press.
Thomas, William I.; and Znaniecki, Florian (1918-1920) 1958 The Polish Peasant in Europe and America. 2 vols. 2d ed. New York: Dover.
Vidich, Arthur J.; and Bensman, Joseph 1958 Small Town in Mass Society: Class, Power and Religion in a Rural Community. Princeton Univ. Press.
Ward, David A.; and Kassebaum, Gene G. 1965 Women’s Prison: Sex and Social Structure. Chicago: Aldine.
Whyte, William F. (1943)1961 Street Corner Society: The Social Structure of an Italian Slum. 2d ed., enl. Univ. of Chicago Press.
Williamson, Henry [pseud.] 1965 Hustler! (Autobiography). Edited by R. Lincoln Keiser. Garden City, N.Y.: Doubleday.
Zelditch, Morris, JR. 1962 Some Methodological Problems of Field Studies. American Journal of Sociology 67:566-576.
II PARTICIPANT OBSERVATION
The phrase “participant observation” is reserved for those forms of research in which the investigator devotes himself to attaining some kind of membership in or close attachment to an alien or exotic group that he wishes to study: for the anthropologist this may be a small community within a foreign society; for the sociologist it may be a street gang, a class of medical students, or the inmates of a hospital, insane asylum, or prison.
Where observation as a technique suggests detachment, participant observation implies involvement and thereby socialization or, more frequently, resocialization. Participant observation is both logically and temporally prior to the more formal and ritualized techniques of investigation of social activities.
In becoming a member of an alien group, the student submits to some degree of resocialization. He may learn another language or argot or new forms of etiquette and morality; he plays various novel roles, assumes obligations, and is granted privileges. If talented and successful, he will in time become capable of thinking and acting with the perspective and basic assumptions belonging to two quite different groups—the one in which he was reared and, to some degree, the one he is studying. At times he will also be able to assume a critical objectivity peripheral to both. What he eventually produces out of this tension will depend on his sophistication and training and on his ability to recognize what he has experienced and learned and to communicate this in terms that will illumine significant areas of the social sciences.
Participant observation is a lengthy process; many (and sometimes most) of the important decisions and insights of the field worker come during the earlier and often painful process of social initiation and not during the later stages of collecting and analyzing data. Every honest field worker proceeds toward knowledge along a thorny path filled with errors and misconceptions; flexibility, humor, patience, and integrity can be more important than specific methods of recording data. Field situations are becoming increasingly complex as more and more groups and factions strive desperately to find an advantageous place in their vision of the new world (Burridge 1960; Maquet 1964; Wax et al. 1964).
Participant observation is essential to almost all branches of the social sciences that depend to any degree on understanding or meaning (Vidich 1955). Little can be accomplished in appraising or analyzing data if neither the language in which they are expressed nor the point of view they reflect are understood by the analysts. Scholars sometimes overlook the fact that the efficiency of scheduled questionnaires and sample surveys depends not so much on elegant statistics as on a sample of respondents whose homogeneity is further accentuated by the very structure of the inquiry (Riesman & Benney 1956). Like Moliere’s bourgeois, these investigators have been participating and observing all their lives in middle-class or mass society; unlike that enterprising person, some have not yet become aware of this fact. Although participant observation is only one of the many instruments or approaches to the study of social activities, it is sometimes the only possible approach (Evans-Pritchard 1940; Goffman 1961).
The works of Peter Freuchen, Jaime de Angulo, Meyer Fortes, James West (pseud.), Howard S. Becker, and also those of Whyte (1943), Turnbull (1961), and Goffman (1961), frequently reveal problems and situations where participant observation has proved a most illuminating and efficient method. Each in its own manner provides unique pictures of an alien and unfamiliar world view: the perspective, the social dynamics, or the central and sometimes secret moralities of a culture. Field research of the future will see a further elaboration of participant observation in studies of situations embracing several groups or aggregates, each of which has its particular and distinct interests, biases, and, sometimes, cultures. This area was called to the attention of social scientists by Hughes and Hughes (1952), and it has been explored by several scholars (see Bibliography).
A considerable amount of pedagogic emphasis is still being placed on the so-called dangers of bias. Thus, a recently published textbook on sociological research by Riley (1963) lists a biased viewpoint as one of the limitations of participant observation. However, it is precisely the “bias” of the participants that the researcher wishes to become capable of assuming and understanding. The observer who establishes himself and remains in a role external to the group being studied is not so much unbiased as incompetent or unenterprising. The disciplined and reliable researcher is not a technical virtuoso, a machine, or a man without a viewpoint, but rather a trained person who enters the field with the expectation that he will be obliged to do many things in spite of his personal preferences, prejudices, or inclinations. The training of the participant-observer and the techniques he employs are designed to assist him in experiencing the significant events and recording the full range of data, despite the idiosyncrasies of his background. In scientific research generally, the problem is not simply one of eliminating error or bias but of assessing its nature, degree, and cause. No technique of research is free from error, and perhaps one virtue of participant observation is that the kind of data it yields allows the biases, inadequacies, and predilections of the researcher to be clearly perceived. The bias and error of sample surveys and other quantitative procedures are sometimes more subtle and more troublesome for science because the over-all perspective reflects the attitudes of management and authority maintained around the conference table. Such investigative techniques frequently lack insight into the activities, attitudes, and basic axioms of the population to be surveyed.
Malinowski performed the archetypical study in participant observation, living in a Trobriand village for almost three years and striving to learn and describe all aspects of the society and the native’s “vision of his world.” He seems also to have been the first social scientist to perceive and communicate to his students that this kind of experience and procedure—with its unparalleled opportunities for observing the complexity and intricacy of tribal life—constituted a new instrument for social scientific research. Indeed, it may be that the develop ment and efflorescence of the school of British social anthropology rested as much on the vistas opened by the appreciation of this new instrument as on the conceptual clarity later introduced by Radcliffe-Brown.
The self-conscious adoption of participant observation by sociologists stems from the perception that ethnographic techniques are extremely effective in the investigation either of the less obvious aspects of one’s own society (Hughes 1960) or of the study of the various religious, ethnic, or status groups inhabiting a particular community, city, or nation. Thus, in 1883 we find Beatrice Webb setting out to see for herself what the British working class was like, just as some sixty years later we see Whyte (1943) insinuating himself into a street corner gang. On the other hand, an ardent and vocal appreciation of the singular virtues of the method appeared in North America only recently, apparently by serendipity. Among a cluster of Chicago graduate students engaged in specific studies of occupations or professions were several who were already resocialized members of exotic groups. Encouraged by E. C. Hughes to make use of their fortuitously acquired data, they were so impressed by the powers and potentialities of the method that, like Malinowski, they became enthusiastic proselytizers (Becker & Geer 1960). Si multaneous and subsequent investigations in which these and other young social scientists entered into and observed unusual or neglected groups have attracted considerable attention.
Despite a theoretical orientation that generally emphasizes empirical rather than social observation, some American anthropologists have used participant observation with notable skill (Lowie, Herskovits, Oliver LaFarge, Dorothy Eggan, and Margaret Mead, to name only a few). Like any other social research technique, participant observation must be flexibly adapted to the procedures and values of the group being studied. It is not to be naively confused with the notion of “living like a native.”
Rosalie Hankey Wax
[Directly related are the entriesCulture, article onCultural Relativism; Field Work. Other relevant material may be found inAnthropology, article onCultural Anthropology; Errors, article onNonsampling Errors; and in the biography ofMalinowski.]
BIBLIOGRAPHY
Becker, Howard S.; and Geer, Blanche 1960 Participant Observation: The Analysis of Qualitative Field Data. Pages 267-289 in Richard N. Adams and Jack J. Preiss (editors), Human Organization Research: Field Relations and Techniques. Homewood, 111.: Dorsey.
Becker, Howard S. et al. 1961 Boys in White: Student Culture in Medical School. Univ. of Chicago Press.
[Bohannan, Laura] 1954 Return to Laughter, by Elenore Smith Bowen [pseud.]. New York: Harper: → paperback edition was published in 1964 by Doubleday.
Burridge, Kenelm 1960 Mambu: A Melanesian Millennium. London: Methuen.
Evans-Pritchard, E. E. (1940) 1963 The Nuer: A De scription of the Modes of Livelihood and Political Institutions of a Nilotic People. Oxford: Clarendon.
Evans-Pritchard, E. E. (1951) 1956 Social Anthropology. London: Cohen & West; Glencoe, III.: Free Press.
Goffman, Erving (1961)1962 Asylums: Essays on the Social Situation of Mental Patients and Other Inmates. Chicago: Aldine.
Herskovits, Melville J. 1948 Man and His Works: The Science of Cultural Anthropology. New York: Knopf.
Hughes, Everett C. (1960) 1962 Introduction: The Place of Field Work in the Social Sciences. Pages v-xv in Buford H. Junker, Field Work: An Introduction to the Social Sciences. Univ. of Chicago Press.
Hughes, Everett C ; and Hughes, Helen M. 1952 Where Peoples Meet: Racial and Ethnic Frontiers. Glencoe, III.: Free Press.
Lowie, Robert H. 1959 Robert H. Lowie, Ethnologist: A Personal Record. Berkeley: Univ. of California Press.
Maquet, Jacques J. 1964 Objectivity in Anthropology. Current Anthropology 5:47-55.
Riesman, David; and Benney, Mark 1956 The Sociology of the Interview. Midwest Sociologist 18:3—15.
Riley, Matilda W. (editor) 1963 Sociological Research. 2 vols. New York: Harcourt.
Roth, Julius A. 1962 The Treatment of Tuberculosis as a Bargaining Process. Pages 575-588 in Arnold M. Rose (editor), Human Behavior and Social Processes: An Interactionist Approach. Boston: Houghton Mifflin.
Turnbull, Colin M. 1961 The Forest People. New York: Simon & Schuster; London: Chatto & Windus. → A paperback edition was published in 1962 by Doubleday.
Vidich, Arthur J. 1955 Participant Observation and the Collection and Interpretation of Data. American Journal of Sociology 60:354-360.
Wax, Murray; Wax, Rosalie H.; and Dumont, Robert V. JR. 1964 Formal Education in an American Indian Community. Spencer, Ind.: Society for the Study of Social Problems.
Wax, Rosalie H. 1957 Twelve Years Later: An Analysis of Field Experience. American Journal of Sociology 63:133-142.
Whyte, William F. (1943)1961 Street Corner Society: The Social Structure of an Italian Slum. 2d ed., enl. Univ. of Chicago Press.
observation
ob·ser·va·tion / ˌäbzərˈvāshən/ • n. 1. the action or process of observing something or someone carefully or in order to gain information: she was brought into the hospital for observation | detailed observations were carried out on the students' behavior. ∎ the ability to notice things, esp. significant details: his powers of observation. ∎ the taking of the altitude of the sun or another celestial body for navigational purposes.2. a remark, statement, or comment based on something one has seen, heard, or noticed: he made a telling observation about Hugh.PHRASES: under observation (esp. of a patient or a suspected criminal) being closely and constantly watched or monitored: he spent two nights in the hospital under observation.DERIVATIVES: ob·ser·va·tion·al / -shənl/ adj.ob·ser·va·tion·al·ly / -shənl-ē/ adv.