Participant observation was introduced into anthropology at the beginning of the twentieth century when Bronislaw Malinowski (1884-1942) challenged the traditional paradigm of researchers conducting their studies from the veranda of a missionary station, by taking accounts from individuals rather than observing situations firsthand (Wax and Cassell 1979). He exhorted his colleagues to conduct fieldwork in situ, using participant observation. This technique was used by Malinowski in his studies of the Trobriand Islands (Malinowski 1922, 1935, 1948) "to grasp the native’s point of view, his relation to life, to realize his vision of his world" (Malinowski 1922, p. 25, emphasis in original). Participant observation involved a field trip of one or two years, working in the native language as a member of the community being studied. Yet Malinowski’s diary illustrates the difficulties in living up to his own demands as he had identified the ideal conditions in which to conduct participant observation, while many problems needed to be resolved in the field (Malinowski 1967).
This observational approach was taken up by many anthropologists in classic studies including E. E. Evans-Pritchard (1902-1973) on the Nuer (1940) and Margaret Mead (1901-1978) in New Guinea (1977). Since the 1960s, anthropologists have "come home" to research their own societies using participant observation to examine urban settings. Harry Wolcott (1973, 1982) studying elementary schools perceived the principal as if he were the chief of a small tribe. Sociologists have taken a similar approach, studying schools (Ball 1981; Burgess 1983), factories (Pollert 1981; Beynon 1973), hospitals (Roth 1963), and new religious movements (Barker 1984, 1987; Zablocki 2001). In sociology, the work of Robert E. Park (1864-1944) and the Chicago school (Park 1952) used observational methods to study homeless men, street-corner gangs, and delinquents. Like the social anthropologists, the Chicago sociologists were strangers in their own society; they were involved but also detached. An observational approach was also used in community and locality studies in the United States and in Britain (Lynd and Lynd 1929, 1937; Warner and Lunt 1941, 1942; Warner 1959; Frankenberg 1957; Stacey 1960; Kluckhohn 1940).
The hallmarks of participant observation involve the researcher living in the community being studied, participating with individuals, observing and talking with them and interpreting the situations observed. The researcher is the main instrument of data collection, and shares in the lives and activities of those being studied by learning their language and interpreting their behavior (Becker 1958). Participant observation involves examining social behavior as it occurs rather than as it is reported through interviews and questionnaires.
Much has been written on the roles used by the participant observer (Adams and Preiss 1960; Bryman and Burgess 1999; Gans 1999). Participant observation can be formal or informal, concealed or revealed, and can involve complete participation and complete observation. These ideal types have been extended by R. Gold (1958) and Buford Junker (1960) into four major roles: (1) the complete participant, (2) the participant as observer, (3) the observer as participant, and (4) the complete observer.
The complete participant rarely reveals that research is being conducted because the researcher does not wish to influence the conduct of the activities being studied (Festinger et al. 1956; Humphreys 1970; Homan 1978). However, in these circumstances it is difficult for the participant observer to pose questions. This puts the researcher in the role of spy and makes it impossible to distinguish everyday roles from research roles. There is also a danger of "going native" by failing to question the activities observed (Murray 2003). Because this role infringes the principle of "informed consent," it is rarely used. The participant as observer role involves researcher and researched being aware that their relationship stems from research activity (Roy 1970). The researcher is involved in the social situation but also detached (Cohen 2000). This role is most frequently used. The observer as participant consists of the observer making the research purpose clear from the start of the investigation, but there is no intense relationship with those researched (Schatzman and Strauss 1973; Hong and Duff 2002). Finally, the complete observer role entirely removes the researcher from any form of participation so that the purposes of the research are not revealed. All four roles are used interchangeably and can assist or impede data collection.
The role of the participant observer is influenced by his or her membership characteristics (Delaney 1988). Age and gender will influence access to groups (Whyte 1955; Patrick 1973). Gender, ethnicity, and social class will also influence the perspective from which data are collected and in some instances access will be granted to, or withheld from, certain individuals in a social setting (Golde 1970; Wax 1979; Roberts 1981; Easterday et al. 1977; Liebow 1967; Bell 1999). Participant observation also involves selection (Arcury and Quandt 1999). Many participant observers use informants in their studies to take them into social situations, explain the context in which observations occur, and provide a perspective on the social world (Casagrande 1960; Gallmeier 1988). Participant observers need to consider how informants are selected as they influence data collection (Cohen 2000).
Participant observers record data and keep detailed field notes (Lofland 1971, 1974; Schatzman and Strauss 1973; Burgess 1982, 1984). Substantive field notes provide a record of the observations, conversations, and interviews that take place (Humphreys 1970). Methodological field notes illustrate the research process and record the personal impressions of the researcher (Geer 1964; Murray 2003), the impact of roles upon data, the selection of informants and relationships with them, and an analysis of research experience. Analytic field notes record concept development throughout an investigation and contribute to data analysis and the written narrative that constitutes the article or monograph produced (Glaser and Strauss 1967; Glaser 1978; Atkinson 1990; Ashworth 1995).
Several problems have been identified when conducting participant observation. Researchers must always remember that they are located in a social setting for the purposes of social science. They are involved and yet detached. This will help them to overcome the risks of overidentifying with other participants and "going native" in the setting by no longer questioning the actions and activities that are observed. The researcher needs to collect data that are reliable and valid (Shaffir and Stebbins 1991). Ethical problems are also frequently raised for the researcher through being placed in a marginal role with the result that stress and anxiety have to be managed throughout a study. This is frequently the case in covert studies where the researcher is unable to take notes or to use a range of other methods of research and often violates principles of informed consent, privacy, and confidentiality (Burgess 1989; Lauder 2003). The participant observer therefore needs to manage the study by being aware of the problems encountered in the research process by engaging in critical self-reflection of the research experience (Bourdieu 2003) and by bringing the study to a successful close.