Can Participants Withdraw Their Data After Learning They Were Deceived?

What happens if participants learn the truth and no longer want their data used?

A participant agrees to a study believing one account of what will happen. Afterward, researchers reveal that a key part of that account was false or incomplete. The participant now knows something they could not have known when making the original decision.

It is reasonable to ask whether that new information should allow them to reconsider use of their data.

Some ethics frameworks explicitly provide such an opportunity after deception or altered consent. But “participants can always withdraw their data after debriefing” is too broad as a universal rule. Requirements differ, and practical withdrawal can become impossible once data can no longer be linked to the individual.

Sometimes yes, and some ethics frameworks explicitly require the opportunity

Post-debrief withdrawal can restore a choice participants could not make earlier

Deception changes the informational basis of the original decision

A participant who agrees to research involving deception cannot evaluate information that has intentionally been withheld or falsified. That departure may be ethically justified, but it means the participant's original decision was made without the complete truth about the study.

Debriefing changes that situation. Once participants know what actually happened, they may evaluate their participation differently.

A post-debrief opportunity concerning the data can therefore serve an important ethical function: it allows participants to make a choice with information that was unavailable when they initially agreed to participate.

APA expressly requires data withdrawal to be permitted after integral deception is explained

APA Standard 8.07 states that psychologists explain deception integral to the design and conduct of an experiment as early as feasible, preferably at the conclusion of participation and no later than the conclusion of data collection. The standard then states that psychologists permit participants to withdraw their data.

For research governed by the APA Ethics Code, this makes the withdrawal opportunity part of the professional ethical requirements associated with deceptive experimentation.

Researchers should not generalize this provision into the claim that every regulatory system uses the same rule. APA is a professional ethics code, not a universal research statute.

TCPS 2 also links debriefing with renewed consent

Canada's TCPS 2 takes a particularly explicit approach. Article 3.7B states that when debriefing is required following an alteration to consent, participants must have an opportunity to refuse consent and request withdrawal of their data or human biological materials whenever possible, practicable, and appropriate.

The policy explains that participants should be offered a genuine opportunity to consent or refuse after learning why ordinary consent requirements were altered.

This reflects the idea that debriefing can do more than supply information. It can create a point at which participants exercise meaningful choice with fuller knowledge of the research.

The U.S. Common Rule does not state the same blanket post-debrief withdrawal rule

Under the Common Rule's general waiver or alteration provision, an IRB must find that additional pertinent information will be provided to participants after participation whenever appropriate. The regulation does not state, in the same provision, a universal requirement that every participant subsequently be permitted to withdraw already collected data.

Researchers should therefore distinguish requirements arising from the Common Rule from additional obligations imposed by professional codes, institutional policies, study protocols, consent materials, or other applicable standards.

Withdrawing from further participation and withdrawing existing data are different actions

This distinction is easy to miss.

A participant can stop taking part in future research procedures. That does not necessarily answer what happens to information already collected. Depending on the applicable rules, consent language, protocol, and data status, previously collected data may be removed, retained, or no longer technically removable.

Researchers should use precise language when explaining these options rather than treating “withdraw from the study” as though it automatically answers both questions.

Data must still be identifiable for individual removal to be possible

Suppose participants complete an anonymous survey and the study is deliberately designed so that no code, identifier, IP address, or other link connects responses to individual participants. Once a response has been submitted, researchers may have no reliable way to determine which record belongs to someone who later asks for removal.

Similarly, identifiable data may later be irreversibly anonymized or combined in a way that makes individual removal impossible or impracticable.

Researchers should therefore plan the relationship between debriefing timing, identifiers, and withdrawal procedures before collecting data. Offering withdrawal after deception is not meaningful if the study design makes removal impossible before participants ever receive the information needed to decide.

Anonymous and confidential data are not the same

Researchers sometimes say data are anonymous when they actually mean confidential.

Confidential data may still be linked to a participant through a name, study code, contact record, or separate linkage file while access is restricted. Such data may remain individually retrievable.

Truly anonymous or irreversibly anonymized data cannot be linked back to the participant using information available to the research team. Once that link is absent, individual withdrawal may no longer be technically possible.

This distinction matters when designing a meaningful post-debrief withdrawal process.

Delayed debriefing can complicate withdrawal

If researchers wait until the end of data collection before revealing deception, they should consider what will happen to the data in the meantime.

Will records remain identifiable until participants are debriefed? Will researchers begin analysis? Will the linkage key be destroyed? Could results be disseminated before participants receive the opportunity required by the applicable framework?

This is one reason delayed debriefing should be planned alongside data management rather than treated solely as a communication issue.

Withdrawal should be a real choice, not a discouraged formality

If participants have a right or approved opportunity to request removal, researchers should not design the debrief to pressure them into keeping their data.

Statements emphasizing how much the study needs their response, how withdrawal will damage the research, or how difficult removal will be can undermine the voluntariness of the choice.

Participants can be told accurately what withdrawal entails and any genuine limits. The explanation should not turn renewed consent into a sales pitch for the dataset.

Researchers should explain what can and cannot be withdrawn

Complex studies may contain several kinds of material: questionnaire responses, recordings, biological specimens, derived variables, interview transcripts, or data already incorporated into analyses.

The practical withdrawal process may differ across these materials. Researchers should know what the governing framework and approved protocol permit and communicate the scope accurately.

TCPS 2, for example, expressly refers to requests to withdraw data or human biological materials following debriefing whenever possible, practicable, and appropriate.

Post-debrief withdrawal is not a punishment for using deception

The rationale is not that deceptive research is presumed wrongful and participants must be offered compensation in the form of data deletion.

Rather, where the governing framework provides the option, it recognizes that participants now possess information they lacked when making the original decision. The opportunity allows them to reconsider use of their contribution with a more complete understanding of the study.

Why data management and debriefing need to be designed together

Now reverse the order. If researchers irreversibly anonymize every record immediately after submission and only reveal the deception weeks later, they may be unable to identify the participant's record when the post-debrief choice finally occurs. That conflict should be resolved during protocol design, not discovered after participants ask for withdrawal.

Common misconceptions about withdrawing data after deception

Design the withdrawal pathway before collecting the first deceptive data

If your governing framework provides participants an opportunity to reconsider use of their data after debriefing, build that opportunity into the data-management plan from the beginning.

Ask how the participant's record will be located, when identifiers will be removed, what materials can be withdrawn, and when removal ceases to be technically possible.

The cleanest approach is to align the ethics protocol, debriefing script, consent materials, withdrawal procedure, and data-management plan. Otherwise, a beautifully worded right to withdraw may encounter the less beautiful reality of a dataset with no remaining identifiers.

Before promising post-debrief withdrawal, make sure it can actually happen

Frequently asked questions about withdrawing data after deception

Post-debrief withdrawal works only if ethical permission and technical possibility align

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.