A Note on Participant Observation

This document aims to provide some guidelines for filling out and reviewing IRB applications for research involving participant observation.

“Participant observation” refers to research in which the investigator observes and documents while actively participating in the community or organization under investigation in a range of ways.  The investigator often combines the role of “researcher” with other social roles (such as volunteer, friend, employee, etc).  Participant observation is commonly used in ethnographic research but might also be used in other qualitative studies.  

Research relying on participant observation poses special challenges for IRB review:

  • First, it might include information-gathering activities that do not fit into the categories provided in the federal rule on human subjects protection. These activities may include observations of private or quasi-private behavior, informal conversations, and analysis of social media.  
  • Second, the researcher must inhabit multiple roles, complicating the standard researcher/subject relationship.

This sort of research can be difficult for IRB members to classify (is it exempt, exempt with limited review, etc?).  It also poses challenges related to consent and confidentiality.  

Can participant observation be “exempt”?:

Most social research at Carleton poses minimal risk to subjects (that is, no more risk than one would encounter in daily life) and is therefore ruled “exempt” from review.  However, in order to be “exempt,” the research must fall into one of specified categories of research activities (surveys, interviews, observation of public behavior, benign behavioral intervention).  

Participant observation will usually involve some activities that don’t fall into those categories.  Moreover, it may be hard to judge the risk of the project where the researcher doesn’t know exactly what topics will arise in the course of the investigation.

Therefore, projects involving participant observation often will not be ruled “exempt.”

What goes on the application:  If the project is not ruled exempt, the researcher will need to address the issue of consent on the IRB application form.  Even if the project is ruled exempt, it may be subject to “limited review”:  If the researcher may record sensitive information and the subjects’ identities cannot be masked, the IRB application must address how the researcher will handle such sensitive information.

Consent:

In thinking about informed consent, the IRB member should take into account the investigator’s experience and history with the community under investigation:

Experienced ethnographers aim to establish a relationship of trust with the community under investigation.  This relationship may take many months or years to evolve.  Once the ethnographer establishes rapport, using formal consent documents for participant-observation activities will generally be unnecessary and counterproductive.*  The subjects will have a good sense of what the project is about and what will go into the research report. In addition, an experienced ethnographer also tends to revisit consent throughout research, reminding or discussing the research project with informants as new situations arise or new kinds of informants are encountered.

In such a situation, as long as the research poses minimal risk, it will generally be appropriate to waive formal documentation of consent.  Sec. 117()(1)(ii) of the rule explicitly allows for a waiver where the risk is minimal and the research involves “no procedures for which written consent is normally required outside of the research context.”  

*Typically ethnographers will still obtain informed consent when conducting formal interviews.   

What goes on the application:  Researchers should explain the nature of their relationship to the subjects, some representative examples of the kinds of research activities they’ll engage in, and how they know subjects are or are not OK with being studied in this way.

Novice researchers will need to demonstrate that they have a well-considered plan for negotiating consent.  Different situations will require different consent procedures.

For example:

  • In a small community, such as a small village, one might explain the project in a public meeting and secure the collective consent of the community.  
  • In a large organization such as a government agency, business, or professional association, one might secure the consent of the institutional representative, who may impose certain requirements or expectations of how the researcher will interact with subjects.  
  • Observations of private or semi-private behavior--in homes, churches, meetings, etc.--may require the researcher not merely to get official consent but to establish a relationship of trust with subjects.

Once the subjects understand and are comfortable with the range of research activities and topics to be investigated, the researcher should not need to get separate consent for every conversation, etc.  Again, waiving documentation of consent may be perfectly acceptable as long as the project poses minimal risk.

What goes on the application: Researchers should explain in the application how they plan to negotiate consent.  If they need the IRB to waive documentation of consent, they should say so and explain why.

What if the project evolves?:

It’s quite common for researchers to change their focus or information-gathering strategy after they’ve secured consent.  If the change is significant, it may be necessary to have another conversation with subjects about the new direction of the project.  

A change in research protocol should always be reported to the IRB chair, to determine whether a new application needs to be submitted.

Confidentiality:

Usually the main risk of research involving participant observation is that private information might become public, putting the subject at risk of financial, legal, or reputational harm.  This is why researchers usually try to anonymize their information.  

If the subjects can’t be identified from the researcher’s notes, the IRB will consider the research record “de-identified.”

If the subjects cannot be de-identified--if for example the project involves video tapes that will be preserved, or the research concerns a small community so that subjects can’t be hidden in the crowd--then the researcher will need to explain how the research records will be secured.  

If the subjects’ identities will be made public and the research involves sensitive information, then the project is not “minimal risk” and a full IRB review will likely be required.  

Final note:

In all cases, the IRB can approve a research project as long as the benefits of the research outweigh the risks and the risks have been minimized as much as possible.  However, the IRB must always ensure that both the researcher and Carleton College as an institution have engaged thoughtfully and responsibly with the range of ethical risks posed by the project.