Participant observation is one of the qualitative types of observation, and it mainly includes overt and covert ways to do the research. Ethical issues in such a qualitative research are often less visible and more subtle than issue in survey or experimental research. (Morse, 1994) And also the reactivity of informants in participant observation can be multiple, which depends on the role of researcher and the relationship between researcher and informants.
The concept of covert participant observation is one which can drive discussion and create controversy. It is ultimately the responsibility of the researchers to inform their research subjects in order to be able to obtain voluntary consent. However in order to carry out effective and authentic research, there should be an element of covert research so that research subjects can be studied in its innate environment. Maanen (1978) has noted that there are four types of participant-observer role and these four types can be divided into overt style and covert style. However, all forms of participant observation involved in covert research to some extent.
This essay starts by introducing four key terms of participant observation, comparing these different roles which researchers might choose. Then outline the reactivity of informants to different roles of researchers acted and several ethical issues including informed consent, deception and protection of privacy in participant observation, which can both reflect the advantages and disadvantages of participant observation. In the remainder of the essay, it tends to analyses the covert research in depth, points out that the covert research inevitably exists in all forms of participant observation to some extent.
The researcher roles of participant observation
Participant observation is defined as "the researcher attempts to participate fully in the lives and activities of subjects and thus becomes a member of their group, organization or community; this enables researchers to share their experiences by not merely observing what is happening but also feeling it". (Gill and Johnson 2002, cited in Saunders et al 2008p.289) It can be a principal research method as a valuable tool and also gathering data in combination with other methods like interviews, collection of documents and so on. (Tomlinson, 2009) Participant observation involves interaction between researchers and researched, which means researchers do the research not only observing around, but also immersing in the activities of people being studied to some extent.
What participant observers actually do depends on which role they would be cast. Norris (2002) adapts from Van Maanen's theory of four types of participant-observer role and analyses these four types in context of his research on police culture:
This role likes a "spy". The researcher attempts to enter into the specific group and become one member in it. However researcher is inside the group, his/her true identity as a researcher is not known by other members; others usually regard him/her as a colleague or friend. In addition, this role is also categorized as "complete participant" because researcher takes part in activity and researcher's identity is concealed. (Gill and Johnson 2002, cited in Saunders et al 2008)
This role is called "voyeur" as a figure of speech. In this role, researchers do not take part in the specific group, just stand outside and notice the phenomenon. However, like the role of "spy", researchers also would not reveal their observing activity. Gill and Johnson (2002) treat this role as "complete observer" because researcher only does observation without participates in activity and researcher's identity is concealed. (Gill and Johnson 2002, cited in Saunders et al 2008)
"Fan"----observer as participant
In this role, researcher attends to observe things without take part in the specific group; although the researcher is not a member of group, his/her identity is still revealed by all. The researcher is mainly an interviewer in this role, because there is much observation but very little participation involved in it. (Bryman and Bell, 2003) The role is also called "observer as participant", which means researcher only observes the activity and others know researcher's objective. (Saunders et al, 2008)
"Member"----participant as observer
In the role of "member", researcher participates in the specific group and other members are aware of both researcher's true statue and the research purpose. This "participant as observer" role makes researcher openly engaging in regular interaction with people and participating in their daily lives. (Bryman and Bell, 2003) Thus, researchers can get information frankly to enhance their understanding.
Although different roles have different features, there will be a movement between roles. As Norris (2002) has written in his research of police culture:
"My predominant research role was also that of the 'fan'. When I deliberately placed myself in a position to overhear private conversations between officers I felt like a 'voyeur'; when I excused myself to the toilet, hurriedly to scribble down notes, I felt like the 'spy'; but when I was attending incidents on the street, passively listening and watching, I was the 'fan'. However, I was cast in the role of a police officer (a member)." (2002p.127)
Reactivity is a phenomenon that occurs when individuals alter their performance or behavior due to the awareness that they are being observed. (Heppner 2008, cited in Wikipedia Website 2009)
On the one hand, the complete participant and complete observer are classified as covert research role. In such a covert research, people under observed would perform naturally and not adjust their behaviours, because they do not know their colleague or friend is a researcher and recording their behavious. That is to say, covert roles can reduce reactivity. (Bryman and Bell, 2003p.320)
For instance, a researcher wants to know the employees' satisfaction about their boss. He/she probably needs to communicate with other employees in daily life and listens the contents employees talking about together sometimes. If people know the researcher's position and the research objective, they probably do not complain their boss and talk real feeling too much; even they will intentionally performance better, because they think their performance may be recorded. Therefore, the information will be difficult to gather and the gathered information may be less authentic.
Saunders (2008p.288) holds another example which complete observer studying consumer behaviour in supermarket. Researcher does the observation like a "voyeur", and people under observed do not know there is an observer, even do not know there is an observation in process at all. The consumer behaviour would be observed by the researcher being stood at a suitable position. In that case, reactivity is not a problem.
On the other hand, observer as participant and participant as observer are two kinds of overt role. In observer as participant, researcher is not a fully functioning member of the group and known as a researcher by others. Thus he/she can only focus on the research role and take the notes whenever they want. Actually notes are important and too risky to keep only on memory. (Bryman and Bell, 2003) For covert researchers, it is difficult and probably in some circumstances impossible to take notes. (Bryman and Bell 2003:320) When people realize that they are being researched and their behavour or words will be recorded, they may perform cautiously to the researcher and research objects, even this caution is unconscious.
The participant as observer takes part in group, he/she can gradually gain the trust of other members; as the identity was clear, he/she can ask more to enhance own understanding. (Saunders et al, 2008) However, it is normally that informants may hold back some things or avoid answering the question directly because of their own benefit. In that case, the reliability and validity of the research would become unsure.
Participant observer is different from the researcher views outside the interaction. It is a research method based on interaction between researchers and researched, which would bring more subtle ethical issues.
Do informants fully understand the purpose and features of the research and what it means for them to participant in the research? Do informants have enough information about the risks and benefits of attending research? Informed consent which is important in any types of research should be adhered to. In informed consent, firstly, researchers should make sure that research subjects are made aware of and the purpose of the research are introduced to informants. Then, let informants consider all aspects from a position of knowledge and freely make a decision about whether or not participating in the research. (Norris, 2002) Finally, make clear to the informants at the outset if researchers do not know in advance the questions that the informants might be asked, or what potential risks might be involved in the future. (Morse, 1994)
Actually, informed consent is not always easy to complete. Norris (2002) had an informed consent from the police, there may be no problem; but other people did not know he was a researcher and this should be where the problem was. However, Norris (2002p.130) mentioned that when he arrived at the scene of a fight, suddenly to declare to all those present that he was a researcher would have been impossible; and sometimes to state his purpose of presence would create problem of "observer effect". Thus, Norris chose to keep silent about his role.
Language and cultural factors would also be barriers to do informed consent. (Morse, 1994) Informant real understands of the research and what it means to take part in the research depends on these factors to big extent. Sometimes an informant may agree to participant in a study, but the researcher found it hard to communicate, it may result from he/she overestimates the informant's language capabilities or ignores the tradition of informant.
For instance, a researcher is waiting for bus at the station and chatting with people who also waiting for the same bus. Someone may begin talking something related to the researcher's study. It seems impossible for researcher knowing in advance what kinds of questions might ask informants exactly. At first researcher might continue to talk casually, until want to ask some specific questions, researcher should reveal the identity. However, even researcher promise to keep confidentiality, the people should be informed that they have right to refuse further discussion.
Sometimes researchers keep secret of their identity and purpose, while sometimes they state these things openly as an essential. There are no formal rules for researchers revealing the status while in casual conversation with others; but when someone asks "what you are doing" directly, researcher may need to judge how exactly to handle a given situation and provide a truthful response as possible as they can. (Mack et al, 2005)
The participant observation is inevitably deceitful to some extent. Because it may become hard to obtain reliable data if people know they are observed and there is a researcher around them. In order to do the researches in an innate environment, researchers make the research role invisible and gradually minimize the distance between themselves and informants.
In Norris's (2002) study, he would dress in a CID style and other officers who did not know of his real role often took him for CID officer. This kind of deception makes researcher similar to the research environment, and builds trust through becoming familiar.
The process of research evolves and ethical issues would emerge when the relationship between researcher and informants changed. Complete participant takes part in activity, spying on people who probably become his/her colleague or friend. People share the information with colleague or friend because of trust, and they may not share these things if they know the colleague or friend's true status is a researcher. Later, informants know there is a researcher around them, they still might be "forgetting" they are being observed when the researcher becomes more a part of social scenery and less a novelty. (Cassell 1980, cited in Morse 1994p.344) It is a common thing that informants forget the researcher's real role and think him/her as a friend.
Maintaining confidentiality means: a) ensuring that informants can never be linked to the data they provide, in which researcher must not record identifying information or private data such as names and addresses of them during participant observation; b) even do not disclose personal characteristics that could allow others to guess the identities of people who played a role in the research; c) ask people's permission when it is unsure whether information they provide is appropriate for research field notes. (Mack et al, 2005)
As Norris (2002) pretended as a police officer, he entered people's houses and witnessed most private of scenes such as a husband-and-wife quarrel, an attempted suicide victim and so on. In fact, these people were no granted the right to control information about themselves. As researchers are become party to people's private information, they have more responsibility to maintain confidentiality.
Participant observation researchers should protect relevant information of informants, no matter the observation is formal or not. However, it may be reasonable in some instances such as for follow observation needed to record the basic information of informants. (Mack et al, 2005) This information may be documented and shared with other research staff; however, it still should secure these files with limited access.
Four forms of participant observation could be divided into two kinds----covert research and overt research. However, researcher's role should not be static and often respond sensitively to the changing situation.
In order to carry out effective authentic research, there should be an element of covert research so that research subjects can be studied in its innate environment. The covert role mainly has two advantages. First, there is no problem of access, because researcher dose not have to seek permission to gain entry to a specific group; second, reactivity is not a problem, because people do not know they are under research and less likely to adjust their behaviour. (Bryman and Bell, 2003) And the disadvantages of covert research are also obvious, which includes the problem of taking notes, the problem of not being able to use other methods and especially ethical problems. (Bryman and Bell, 2003)
The role researchers take as "spy" and "voyeur", which are surely the covert research. But the overt roles of "fan" and "member" which people know researcher is a researcher no matter whether he/she takes part in community or not, are still can be covert to some extent. For example, when the trusts build between researcher and informant gradually, then informant would treat researcher as a friend and phone researcher sometimes. They may talk about something that researcher think is relevant to his/her study. In that case, researcher would take notes but could not tell informant that he/she was writing notes as they talked on the phone, because it probably would have hurt informant's feelings. However researcher would tell that this conversation really helped him/her to understand something better, it becomes a kind of covet research to some extent. (Morse, 1994)
Increasingly social scientists begin to discourage covert observation and hold that researchers must reveal their identities. However much observed behaviour may be stigmatized, illegal or embarrassing, such data may not be obtainable in any other way when people know of the existence of the researcher. Therefore, if all researchers avoid using covert research, it may not be possible to ensure the reliability and validity of research.
In short, there are several different ways of classifying the roles of participant observation. And one widely recommended classification divides participant observer into four roles as follow: 1) spy, which active in research process but covert; 2) member, which active in research process and overt; 3) voyeur, which do not active in research process and covert; 4) fan, which do not active in research process but overt. (Van Maanen 1978, cited in Tomlinson 2009) It makes distinctions along activity and visibility two factors. Thus the former two are more tend to participant form, and the latter two are more towards observer form.
There are advantages and disadvantages of all forms of participant observation which are to do with reactivity and ethics. Reactivity may be not a problem in using a covert role to do research, it dues to people are not aware of being researched; while in overt research, people may adjust their behaviour, because they know there is a research in process. And ethical issues in participant observation are related to the characteristics of it. The main ethical issues include informed consent, deception and protection of privacy. On the one hand, researchers should let informants freely make decision about whether or not take part in research basic on fully understand the overall of the research; on the other hand, researchers should avoid deceiving the informants as possible as they can and protect the informants' privacy.
As the researcher's stance is not always static, all forms of participant observation involve some aspect of covert research actually. In order to carry out effective authentic and legitimate research, it needs the covert research to create an innate environment.
- Bryman, A and Bell, E. (2003) Business research methods, New York: Oxford University.
- Morse, J. (1994) Critical Issues in Qualitative Research Methods, Printed in the United States of America.
- Mack, N. et al. (2005) Qualitative Research Methods: A data collector's field guide, USA: Family Health International.
- Norris, C. 'Some Ethical Considerations on Fieldwork within the Police' In Hobbs and May ed. (2002) Interpreting the Field: Accounts of Ethnography.
- Saunders, M. et al. (2009) Research Methods for Business Student, London: FT Prentice Hall.
- TOMLINSON, J. 2009. Ethnography and Participant Observation, lecture notes distributed in 5330M Research Methodology. Leeds University Business School, 27th December.
- Wikipedia Website. (2009) Reactivity (psychology) [on line]. [Accessed in October 2009]. Available from: http://en.wikipedia.org/wiki/Reactivity_(psychology).