What is TRIANGULATION? (by Norman K. Denzin, denzin@vmd.cso.uiuc.edu Triangulation is the application and combination of several research methodologies in the study of the same phenomenon. The diverse methods and measures which are combined should relate in some specified way to the theoretical constructs under examination. The use of multiple methods in an investigation so as to overcome the weaknesses or biases of a single method taken by itself is sometimes called multiple operationalism. The insistence on a multiple operational orientation in the social sciences is commonly associated in the field of psychology with the work of Donald T. Campbell and his associates (Brewer and Collins 1981). Two outgrowths of Campbell's works have included the multitrait-multimethod matrix technique (Campbell and Fiske 1959) and the invention of the cross-lagged panel correlational technique (Pelz and Andrews 1964). The use of multiple measures and methods so as to overcome the inherent weaknesses of single measurement instruments, has, however, a long history in the physical sciences. The concept of triangulation, as in the action of making a triangle, may be traced to the Greeks and the origins of modern mathematics. 1. The Need for Triangulation The social sciences rely, in varying degrees, on the following research methods: social surveys, experiments and quasiexperiments, participant observation, interviewing, case study and life history constructions, and unobtrusive methods (Denzin 1989) (see Survey Research Methods; Participant Observation; Interviews in Sample Surveys; Case Study Methods). Each of these methods have built-in weaknesses which range from an inability to enter realistically the subject's life-world in experiments and surveys, to the problems of reflecting change and process in unobtrusive methods, the controlling of rival interpretive factors in participant observation and life histories, or an excessive reliance on paper and pencil techniques in surveys and interviewing. The realities to which sociological methods are fitted are not fixed. The social world is socially constructed, and its meanings, to the observers and those observed, is constantly changing. As a consequence, no single research method will ever capture all of the changing features of the social world under study. Each research method implies a different interpretation of the world and suggests different lines of action the observer may take towards the research process. The meanings of methods are constantly changing, and each investigator brings different interpretations to bear upon the very research methods that are utilized. For those reasons, the most fruitful search for sound interpretations of the real world must rely upon triangulation strategies. Interpretations which are built upon triangulation are certain to be stronger than those which rest on the more constricted framework of a single method. 2. The Hermeneutics of Interpretation What is sought in triangulation is an interpretation of the phenomenon at hand that illuminates and reveals the subject matter in a thickly contextualized manner. A triangulated interpretation reflects the phenomenon as a process that is relational and interactive. The interpretation engulfs the subject matter, incorporating all of the understandings the researcher's diverse methods reveal about the phenomenon A hermeneutic interpretation does not remove the investigators from the subject matter of study but rather places them directly in the circle of interpretation (see Hermeneutics). While it is commonplace in the social sciences to place the investigator outside the interpretive process, hence asking the research methods to produce the interpretation that is sought, the hermeneutic interpretation dictates that "what is decisive is not to get out of the circle [of interpretation] but to come into it the right way" (Heidegger 1962 p. 195). Triangulation is the appropriate way of entering the circle of interpretation. The researcher is part of the interpretation. 3. Types of Triangulation While it is commonly assumed that triangulation is the use of multiple methods in the study of the same phenomenon, this is only one form of the strategy. There are four basic types of triangulation. Data triangulation, involving time, space, and persons is the first type. Investigator triangulation consists of the use of multiple, rather than single observers. Theory triangulation consists of using more than one theoretical scheme in the interpretation of the phenomenon. Methodological triangulation, using more than one method, may consist of within-method or between-method strategies. Multiple triangulation exists when the researcher combines in one investigation multiple observers, theoretical perspectives, sources of data, and methodologies (Denzin 1978 p. 304). 4. A Case of Multiple triangulation The social sciences must move beyond investigations that triangulate only by data source, or by research method. Multiple triangulation must become the goal and aim of these disciplines. There are, however, few outstanding illustrations of this commitment. Perhaps Thomas and Znaniecki's publication, The Polish Peasant in Europe and America (1918, 1919, 1920) remains the classic in the social sciences. This five-volume work, which sought to build a social psychology within the nascent field of sociology, utilized personal, historical, religious, and economic documents from and about Polish society, as it was disintegrating and undergoing transition prior to the First World War. The work consists of five documentary volumes which offer a study of the social organization and evolution of the peasant primary groups (family and community) under the influence of industrialization and immigration to America and Germany. Volumes 1 and 2 study the peasant family, the Polish marriage and class system, economic life, religious attitudes, and include correspondence between members of six family groups. Volume 3 is the autobiography of a peasant immigrant. Volume 4 examines the dissolution of primary groups in Poland and Volume 5 is based on studies of the Polish immigrant in America. Thomas and Znaniecki's investigation used triangulated data, investigators, theories, and methods. Life histories, autobiographies, and family letters were at the core of their study, yet, in an unparalleled fashion, the research utilized participant observation, interviews, quasicomparative experiments on a grand scale, unobtrusive methods (letters), and surveys. Theoretically, the work wove its way (often implicitly) through the theories of Freud, James, Marx, Spencer, Durkheim, Mauss, Weber, Tonnies, Simmel, Hegel, Mead, Cooley, and Comte. This study, still a classic and in need of reinterpretation, illustrates the scope and volume that multiple triangulation may assume. Smaller in size, but illustrative and pivotal in importance, stands Geetz's (1972) study on the "Balinese Cockfight." This investigation, based on description and interpretation, also triangulated data, investigators, theory, and methods. 5. Problems in Designing Multiple-triangulated Investigations There are at least four basic problems to be confronted in such research. These are: (a) locating a common subject of analysis to which multiple methods, observers, and theories can be applied; (b) reconciling discrepant findings and interpretations; (c) novelty, or the location of a problem that has not been investigated before; (d) restrictions of time and money. The location of a common subject of analysis can only be resolved through a clear understanding of the question the investigator wishes to answer. Divergent and discrepant findings are to be expected, for each inspection of the phenomenon is likely to yield different pictures, images, and findings. These differences are not to be ignored, but should be reported so that future investigators can build upon such observations. Novel or new problems are often, upon inspection, not new, but merely manifestations of familiar topics previously examined from different perspectives and questions. Restrictions of time and money are the least problematic, for if investigators are thoroughly committed to understanding a problem area they will persist in examining it, even under difficult circumstances. Triangulation is the preferred line of research in the social sciences. By combining multiple observers, theories, methods, and data sources, social scientists can begin to overcome the intrinsic bias that is bound to come from single-method, single-observer, single-theory investigations. It must be noted that the method of triangulation is not without its critics (see Fielding and Fielding, l986; Lincoln and Guba, l985; Patton, l980; Silverman, l985; and Denzin, l989 for a review of these critics and their positions). See also: Research Paradigms in Education; Research Methodology: Behavioral Sciences Bibliography Brewer M B, Collins B E 1981 Scientific Inquiry and the Social Sciences: A Volume in Honor of Donald T. Campbell. Jossey-Bass, San Francisco, California Campbell D T, Fiske D W 1959 Convergent and discriminant validation by the multitrait-multimethod matrix. Psychol. Bull. 56: 81-105 Denzin N K 1989 The Research Act: A Theoretical Introduction to Sociological Methods, 3rd edn. Prentice-Hall, Englewood Cliffs, N.J. Fielding, N.G., and J.L. Fielding l986. Linking Data. Sage, Beverly Hills, CA. Geertz C 1972 Deep play: Notes on the Balinese cockfight. Daedalus 101: 1-37 Heidegger M 1962 Being and Time. Harper, New York Lincoln, Y.S., and E. G. Guba l985 Naturalistic Inquiry. Sage, Beverly Hills, Ca. Patton, M.Q. l980 Qualitative Evaluation Methods.Sage, Beverly Hills, CA Pelz D C, Andrews F M 1964 Detecting causal priorities in panel study data. Am. Sociol. Rev. 29: 836-48 Silverman, D. l985 Qualitative Methodology and Sociology. Gower, Brookfield, Vt. Thomas W I, Znaniecki F 1918, 1919, 1920 The Polish Peasant in Europe and America: Monograph of an Immigrant Group. Gorham Press, Boston, Massachusetts N. K. Denzin