01 · The Question
Does entering a public place make someone available for research?
A researcher wants to observe pedestrians in a plaza, passengers waiting at a station, visitors moving through a museum, or spectators at a public event. Nobody has been recruited. Nobody has signed a consent form. The researcher is simply observing behavior that anyone nearby could see.
It is tempting to reason that people who enter public places have already accepted being seen, so researchers should be free to study them.
That conclusion goes too far. People generally expect some degree of ordinary observation in public. That does not mean they have agreed to every form of systematic research, recording, identification, tracking, or dissemination that could technically occur there.
03 · What You Need to Know
Public observation can qualify for special treatment, but only under conditions
Public presence is not implied research consent
Informed consent is a process through which a prospective participant receives relevant information and voluntarily agrees to participate. Merely walking into a park, station, shopping district, or other public setting does not perform that function.
This distinction is important even when a particular research protocol is legitimately permitted to proceed without individual consent. The ethical rationale is not that everyone in the location silently consented. Rather, the applicable framework may recognize that certain forms of public observation can occur without individual consent when specified conditions and protections are satisfied.
TCPS 2, for example, generally treats consent as a manifestation of respect for persons and states that research normally begins after consent has been provided, subject to defined exceptions.
Some public-place observation may not require individual consent
Public observation is not prohibited merely because obtaining consent from everyone would be impossible.
Different ethics systems handle this issue differently. A particularly clear example appears in Canada's TCPS 2. Article 2.3 states that research ethics board review is not required for observation of people in public places when all three specified conditions are satisfied: the research involves no researcher-staged intervention or direct interaction, the people observed have no reasonable expectation of privacy, and dissemination does not permit identification of specific individuals.
| Condition |
What it asks |
Why it matters |
| No staged intervention or direct interaction |
Is the researcher simply observing naturally occurring activity? |
Introducing an intervention changes the relationship between researcher and observed individuals. |
| No reasonable expectation of privacy |
Is the behavior genuinely occurring under circumstances where privacy is not reasonably expected? |
Public location alone does not settle whether the activity is private. |
| No identification in dissemination |
Can specific observed individuals be recognized in publications or other research outputs? |
Identification can create privacy and harm risks that aggregate observation does not. |
These are requirements within a particular Canadian framework, not universal rules for every country or institution. They nevertheless illustrate an important principle: ethical permission for public observation depends on characteristics of the research, not simply the address at which it occurs.
“Public place” itself requires interpretation
Even deciding whether somewhere counts as public may require more thought than expected.
TCPS guidance indicates that this assessment should be made case by case. Relevant considerations include whether a space is open to the public and intended to serve the public, followed by whether the particular observational research satisfies the applicable conditions concerning researcher involvement, expectations of privacy, and identification.
A city park is different from an invitation-only event. A museum open to anyone who purchases admission may differ from a members-only meeting occurring inside it. A publicly accessible hospital corridor may still expose information whose sensitivity makes simplistic assumptions about publicity inappropriate.
The label attached to the building does not necessarily characterize every activity occurring within it.
People can retain privacy expectations in public
This is the point that most directly limits the “they were in public” argument.
TCPS 2 explicitly recognizes that something publicly accessible may nevertheless be regarded as private in a participant's culture and that some groups or activities may carry reasonable expectations of privacy. Its guidance uses religious services and practices as examples where participants may expect observers to respect a degree of privacy.
The distinction between public and private behavior therefore requires attention to context, not merely geography.
Observing a crowd is different from targeting an individual
Imagine a researcher counting how many people cross a plaza during each 15-minute interval. Now imagine the researcher selecting one pedestrian, following that person through the city for three hours, documenting where they go, whom they meet, and what they do.
Both researchers remain physically in public spaces. Yet the informational intrusion is plainly different.
Systematic tracking can aggregate otherwise fleeting observations into a detailed account of someone's movements and associations. Researchers should therefore consider the granularity, duration, persistence, and linkage of observations rather than assuming that each individual observation is ethically trivial because it occurred publicly.
Identifying people can materially change the analysis
A dataset containing only aggregate pedestrian counts differs from a collection of identifiable photographs, facial images, voices, detailed descriptions, or exact trajectories.
TCPS 2 specifically instructs researchers to consider whether dissemination through publications, photographs, audio, or video could identify people observed in public, particularly when a place may be associated with stigma.
Removing names does not necessarily eliminate this issue. Faces, clothing, tattoos, locations, occupations, unusual behaviors, or combinations of details may identify someone.
Recording is not merely observation with better memory
Photography, audio, and video preserve information and may capture much more than the variables a researcher originally intended to study.
The APA Ethics Code illustrates this distinction. It generally requires informed consent before recording research participants' voices or images, but recognizes an exception where the research consists solely of naturalistic observations in public places and the recording is not anticipated to be used in a manner that could cause personal identification or harm.
Whether researchers may record people in public without telling them consequently requires its own analysis.
Risk matters even when the behavior is ordinary
Researchers should consider what participation in the dataset could reveal or cause.
Knowing that someone entered a supermarket may ordinarily reveal little. Knowing that someone entered a particular treatment center, political meeting, religious gathering, shelter, or support group could carry considerably greater implications. The location itself can become data about the person.
Public visibility therefore does not guarantee harmlessness.
Vulnerability can require additional safeguards
The population being observed also matters. A protocol involving ordinary adult pedestrian traffic should not automatically be treated as ethically equivalent to observation centered on children or people whose circumstances make privacy violations particularly consequential.
Researchers working with such populations should examine the additional considerations involved in observing children or vulnerable people rather than relying on the public character of the setting.
Do not convert a methodological problem into an ethical shortcut
Obtaining consent from every person passing through a railway station may be impracticable and may substantially alter the behavior a naturalistic study intends to observe. Those are legitimate methodological considerations.
They do not mean that researchers can independently declare the study exempt from consent or ethics requirements. APA guidance, for example, permits dispensing with informed consent for certain naturalistic observation where the research would not reasonably be expected to create distress or harm, disclosure would not expose participants to specified risks, and confidentiality is protected, or where dispensing with consent is otherwise permitted by law or institutional or federal regulations.
The governing framework should therefore be checked before data collection, not reconstructed afterward to justify what has already been collected.
04 · A Practical Example
Two studies can use the same station and require different ethical judgments
Hypothetical Example
Observing passengers at a railway station
A researcher wants to investigate how station design affects passenger movement during rush hour. The station is open to the traveling public.
Protocol A
From a normal public vantage point, the researcher counts how many passengers choose each of three exits. No interaction occurs, no images or conversations are captured, and no individual can be identified from the dataset or resulting publication.
Protocol B
The researcher selects individual passengers and records video while following them through the station, documenting their faces, companions, movements, and conversations.
Same place, different data
Both protocols occur in the same public station, but the second creates substantially richer and more identifiable records about particular people.
Decision
The researcher evaluates each protocol under the institution's applicable ethics rules rather than assuming that the station's public status provides identical permission for both.
The ethical difference comes from what the researcher does, collects, preserves, and can disclose. “It happened in public” describes the setting. It does not describe the entire protocol.
07 · A Quick Checklist
Before studying people in a public place without consent, verify these points
Before beginning public-place observation, check:
Verify whether the setting is genuinely public under the ethics framework and law applicable to your study.
Determine whether the particular behavior being observed carries a reasonable expectation of privacy.
Identify any researcher interaction, intervention, staging, following, or participation that changes the nature of the observation.
Assess whether individuals can be identified from field notes, photographs, audio, video, timestamps, locations, or other collected information.
Consider whether the location, activity, or observed population makes the information sensitive or potentially stigmatizing.
Collect the minimum amount of individual-level information needed to answer the research question.
Plan dissemination so that research outputs do not unnecessarily expose identifiable individuals or sensitive associations.
Confirm whether consent, a waiver, an exemption determination, or formal ethics review is required by your institution.
Check applicable privacy, recording, surveillance, and data-protection law rather than assuming ethical guidance establishes legal permission.