Behavioral Observation

Frequently Asked Questions

I. What is observation?

Observation is a qualitative method in which data is collected firsthand by researchers who directly observe the behavior and interactions of participants in their natural environments. These environments are varied, encompassing physical locations such as public parks and museums, as well as digital spaces like social media sites or chat forums. Observation can be applied to study either groups or individuals:

a). Group-level observation focuses on the collective behavior of a social group. For this level, there are two methods of observation:

• During unengaged (non-participatory) observation, the researcher is a detached spectator, observing the group’s behavior and interactions without actively participating or interacting with the individuals being observed. This is the most common method of observation at UC Berkeley.
• During immersive (participatory) observation, the researcher adopts a dual role, acting as both an observer and a participant. This involves embedding oneself in the group, living or working alongside them, and engaging in the group’s daily activities or routines.

b). Individual-level observation (shadowing) involves following a specific participant closely across different locations and situations; over a set period, the researcher tracks the participant’s movements and activities meticulously without intervening in their natural flow of activities. This method is generally viewed as more intrusive than observing group-level behavior.

II. When does observation require IRB review?

All observation meeting the definition of human subjects research requires approval from the Committee for Protection of Human Subjects (CPHS) or the Office for Protection of Human Subjects (OPHS) before any research activities commence.

III. Does observation require Exempt or Non-Exempt review?

The required level of review is determined by the degree of risk, the nature of the setting (i.e., public vs. private), and the planned interactions:

a). Exempt review: This level is reserved for studies that fit into one or more Federal Exemption Categories, generally involving minimal risk to participants (meaning that the probability and magnitude of harm or discomfort anticipated in the research are no greater than those ordinarily encountered in daily life). For observation research, Exempt review is typically applicable only for:

• unengaged (non-participatory) observation in classrooms or other commonly accepted educational settings, for research involving normal educational practices meeting the criteria for Federal Exemption Category 1; and,
• unengaged (non-participatory) observation of public behavior (i.e., behavior taking place in a publicly accessible location in which the participant does not have an expectation of privacy), for research meeting the criteria for Federal Exemption Category 2.

b). Non-Exempt review: This more rigorous review is necessary for all research that does not meet the strict criteria for Exempt status, specifically when the research involves:

• unengaged (non-participatory) observation of behavior that is not deemed “public behavior” (e.g., in-person observations conducted at work sites or online in members-only forums), including observations in classrooms or other educational settings that does not meet the criteria for Federal Exemption Category 1
• immersive (participatory) observation, OR
• individual-level observation (shadowing)

Researchers are strongly encouraged to contact OPHS (ophs@berkeley.edu) early in the project planning phase for definitive guidance on the appropriate review level for their specific observation protocol.

IV. Do I need to obtain formal consent from the people I will be observing?

The ethical principle of informed consent generally requires that potential participants voluntarily agree to take part in research after fully understanding its nature, risks, and benefits. For all research involving group- or individual-level observation, informed consent must be obtained from every participant, where feasible, prior to the observation taking place. Likewise, for observation of minors, child assent and parent permission must be obtained prior to the observation taking place. There are, however, some exceptions to this requirement:

a). Waiving Consent with Disclosure (“Opt-Out” Method): What if it is not possible to obtain informed consent? For research involving group-level observation, obtaining formal informed consent from every participant may sometimes be methodologically or practically unfeasible. However, it is often still possible to inform those being observed and offer them an opportunity to “opt out” of the research. A prime example is a large community meeting where attendees are unknown to the researchers beforehand and come and go throughout the session. In this scenario, securing individual consent prior to the meeting may be impracticable. Under these circumstances, the IRB may approve observations without formal consent, provided researchers implement clear disclosure and opt-out strategies. These strategies may include:

• Announcing the project at group meetings, events, or gatherings, and clearly outlining how individuals can request their data not be used.
• For online research, using the site’s functionality – such as pinned posts or profile bios – to signal the researcher’s presence. Researchers should provide necessary contact information (e.g., email address) so users can easily and confidentially request to be excluded from the study. It is the researcher’s responsibility to adhere to applicable rules and permissions associated with each physical or online site.
• Distributing written materials such as flyers, posters, or brief information sheets in the setting where the observation will occur. These materials should contain a brief, non-technical description of the research goals along with the researcher’s contact details.

For both Exempt and Non-Exempt studies, disclosure strategies must be discussed in the protocol.

Additionally, for Non-Exempt studies, researchers are required to request a formal Waiver of Informed Consentfrom the IRB. This waiver must be justified and submitted as part of the Non-Exempt application, in the Informed Consent section (for observation of adults) and/or the Child Assent & Parent/Guardian Permission section (for observation of minors).

The IRB does not review informed consent materials for Exempt research, so a formal Waiver of Informed Consent is not required for studies determined to be Exempt.

b). Waiving Consent with No Disclosure: What if both formal consent and disclosure/“opt-out” strategies are unfeasible?

In certain group-level observations, providing disclosure or opt-out opportunities may be just as impracticable as obtaining formal consent. Common examples include observing large-scale public events, such as protests, concerts, or artistic performances.

If a researcher believes that disclosure strategies are not practicable, the study protocol must include a detailed justification for this approach. For Non-Exempt research, researchers must also specifically request a formal Waiver of Informed Consent from the IRB.

V. Are there any unique confidentiality or data security considerations with regard to observations?

As with all research, researchers must be thoroughly familiar with the CPHS Data Security Policyand Data Security Guidelines and Matrix, and implement robust security measures to protect the privacy and confidentiality of participants. Additional considerations for observation research include the following:

• Researchers should avoid recording or taking photos of individuals who have not provided their informed consent to participate in the research. If researchers are planning to record while carrying out observations in public spaces where it is not feasible to obtain informed consent, the IRB may require disclosure strategies such as the use of a visible pin or badge to alert the public that active recording is in progress.
• For online observation, researchers should be keenly aware of and strictly abide by the site’s Terms of Service (ToS), as some sites explicitly forbid research or data scraping regardless of IRB approval. Additionally, sub-forums (e.g., subreddits) may have their own rules and expectations outside of ToS.