Can Researchers Record People in Public Without Telling Them?

If researchers can observe public behavior, can they also record it?

A researcher can stand in a public square and watch how people move through it. Does that mean the researcher can place a camera there and record everyone who passes? What about recording conversations in a café, photographing participants at a public event, or retaining video that clearly shows people's faces?

Recording may initially seem like a more accurate version of observation. Methodologically, it can be. Ethically, however, a recording changes what the researcher possesses.

Video, photographs, and audio can preserve identifiable information long after an encounter ends. They can be replayed, analyzed for purposes not originally anticipated, copied, linked with other information, shared with collaborators, or exposed through a data breach. Whether recording without prior notice is permissible therefore requires more than asking whether the researcher could legally or ethically stand there and watch.

Sometimes, but permission to observe is not automatically permission to record

A recording creates a different kind of research record

Watching and recording are ethically different activities

Ordinary observation is often transient. A researcher sees something, perhaps records a behavioral code or writes field notes, and the original event disappears.

A camera or audio recorder can preserve far more. Faces, voices, clothing, companions, conversations, exact movements, timestamps, surroundings, and activities may all remain available for later examination. Some of those details may not even have been relevant to the original research question.

This distinction is why a protocol that permits observation without individual consent should not automatically be interpreted as permission to create identifiable recordings.

Some professional ethics frameworks recognize limited exceptions

The American Psychological Association's Ethics Code provides a useful example of how recording may be treated differently from ordinary observation. Standard 8.03 generally requires psychologists to obtain informed consent before recording research participants' voices or images for data collection.

The standard recognizes an exception when 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. It also recognizes a separate situation involving deception where consent for use of the recording is obtained during debriefing.

This is not a universal rule governing every researcher or jurisdiction. It illustrates the narrower point that even where professional guidance permits recording without prior consent, the exception may depend on the nature of the observation and what subsequently happens to the recording.

Public location does not eliminate privacy expectations

A researcher should not reason backward from “public place” to “no privacy.” The ethical distinction between public and private behavior is contextual.

A wide-angle recording of pedestrian flow in a plaza may raise different concerns from deliberately pointing a directional microphone toward two people having a quiet personal conversation in that plaza. The physical coordinates have not changed. The nature of the information being captured has.

Likewise, merely being in a public place does not constitute informed consent to every form of research recording.

Audio can raise different issues from silent observation

Researchers sometimes focus on cameras while overlooking microphones. Yet audio may capture names, relationships, health information, political or religious views, financial circumstances, personal disputes, and other information that cannot be inferred merely by watching someone.

Audio recording can also be subject to specific laws governing interception or recording of conversations. These rules vary considerably among jurisdictions and may depend on factors such as the nature of the conversation, reasonable expectations of privacy, and whose consent is required.

Identifiability is not limited to showing someone's name

A recording can identify people directly through their faces or voices. Identification may also occur indirectly through uniforms, distinctive clothing, tattoos, vehicles, companions, locations, signs, spoken names, or combinations of contextual details.

Researchers therefore need to ask whether people are identifiable from the original research files as well as from anything that will eventually be published or presented.

Those are separate questions. A research team might retain identifiable raw video while publishing only aggregate statistics. That reduces identification through publication, but it does not remove the privacy and security risks associated with possessing the original footage.

Recording can capture people who were never the intended subjects

A camera aimed at a particular activity may also capture bystanders, children, employees, license plates, computer screens, conversations, or people entering nearby premises. A microphone may record speakers far beyond the interaction of interest.

Researchers should therefore examine the recording frame, duration, audio range, and surrounding environment before collection begins. Technical capability is not a reason to collect everything the device can capture.

Data minimization matters

If the research question concerns crowd density, the researcher may not need recognizable faces. If it concerns movement trajectories, the study may not require audio. If manual counts are sufficient, permanent video may be unnecessary altogether.

The principle is straightforward: collect enough information to answer the research question, not every piece of information that the equipment makes available.

Researchers can also consider technical and procedural safeguards such as limiting the field of view, disabling unnecessary audio, reducing image detail when scientifically appropriate, restricting access, separating identifying information, establishing retention periods, and securely deleting recordings when they are no longer required.

Publication creates a second ethical decision

Using a recording for analysis and publishing that recording are not necessarily the same ethical act.

A video may have been collected under conditions that permit research analysis while still requiring additional consideration before identifiable frames appear in an article, conference presentation, repository, teaching material, or public website.

Researchers should therefore specify anticipated uses during protocol development. “For research purposes” is often too vague when the data consist of identifiable images or voices.

Undisclosed recording may also become a covert method

If people are deliberately kept unaware that research recording is occurring, the protocol may raise issues beyond recording itself. The researcher may need to consider autonomy, concealment, deception, debriefing, and whether awareness would make the research impracticable.

The justification for covert observation should therefore be evaluated separately rather than hidden inside a general claim that the recording happened in public.

A camera can collect much more than the research question requires

The lesson is not that one particular camera configuration is universally ethical. It is that the research question should determine what needs to be recorded. The existence of a better camera is not, regrettably for equipment budgets everywhere, an ethical justification for collecting more data.

Common mistakes when recording public behavior

Design the recording around what you actually need to know

Before adding a camera or microphone to an observational protocol, identify exactly what information the recording contributes. Then ask what additional information it captures simply because the technology can.

Before recording people in public, check the entire data lifecycle

Questions about recording people for observational research

Recording is not merely observation preserved on a device

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.