03 · What You Need to Know
The instrument can reveal ethical issues that the protocol summary cannot
A protocol tells reviewers what you intend to do. An instrument can show what participants will actually encounter.
WHO's recommended format for research protocols explicitly states that instruments used to collect information, including questionnaires, focus-group discussion guides, observation recording forms, and case-report forms, must be provided within its Research Ethics Review Committee framework.
That WHO requirement is not automatically the rule at every institution. It illustrates why instruments can be integral to ethical review rather than mere technical appendices.
A research instrument is more than a questionnaire
The term can include many materials used to expose participants to content, elicit information, record observations, or structure data collection.
Depending on the study, instruments may include:
- paper or online questionnaires;
- structured or semi-structured interview guides;
- focus-group discussion guides;
- observation forms or coding sheets;
- psychological or educational tests;
- experimental tasks and stimuli;
- vignettes, scenarios, images, audio, or video shown to participants;
- case-report or data-collection forms;
- diaries or experience-sampling prompts;
- screening instruments;
- forms used to extract information from records.
What matters ethically is not the label. It is what the instrument asks, shows, records, or causes the participant to do.
The questions can change the risk of the study
Two surveys of identical length can create very different ethical concerns.
A questionnaire about ordinary study habits may involve little beyond time and inconvenience. A questionnaire asking about self-harm, illegal drug use, sexual behavior, workplace misconduct, immigration status, experiences of violence, or other sensitive matters may create emotional, social, legal, privacy, or confidentiality concerns depending on the population and setting.
If reviewers see only "participants will complete a 20-item survey," they cannot necessarily determine which of those studies you are proposing.
The instrument therefore helps substantiate the risk description in the ethics application.
Interview guides can matter even when interviews are flexible
Qualitative researchers sometimes hesitate to submit interview guides because semi-structured interviews are intentionally adaptive. Follow-up questions depend on what participants say, and a conversation cannot be scripted completely in advance.
That flexibility does not make the guide irrelevant.
A committee can review the planned domains, core questions, prompts, and foreseeable sensitive topics without requiring the researcher to predict every conversational follow-up. The application can explain that probing will vary according to participants' responses while defining the approved scope of inquiry.
This distinction matters. Ethical review need not convert a semi-structured interview into a rigid questionnaire, but methodological flexibility also does not necessarily authorize unlimited questioning about topics never disclosed in the application.
Focus groups create risks beyond the questions themselves
A focus-group guide may contain relatively ordinary questions while the group format creates an additional confidentiality problem: participants hear one another's responses.
Researchers can ask participants to respect confidentiality, but they generally cannot guarantee what other group members will later disclose. The instrument, procedure, consent materials, and risk description should reflect that limitation.
This is a useful reminder that instruments are reviewed in context. Ethical acceptability depends not only on what is asked but how, where, and in whose presence it is asked.
Experimental stimuli may be ethically important even when they collect no data
An instrument does not have to contain questions to matter ethically.
Images, videos, vignettes, simulated interactions, false feedback, emotionally provocative material, or other stimuli may create distress, deception, embarrassment, or behavioral effects. If participants are exposed to the material as part of the research, reviewers may need to understand its content and how any associated risks are managed.
For deception research, the submitted materials may also need to be considered alongside the consent and debriefing procedures.
Observation instruments can reveal what information researchers intend to record
"We will observe classroom behavior" leaves open an important question: what exactly will observers record?
An observation form may show whether researchers record only predefined classroom interactions or also document names, sensitive disclosures, disciplinary events, characteristics of people who are not participants, or other information that changes the privacy implications of the study.
The same principle applies to record-extraction forms. A study described as using existing records may still collect substantially more information than necessary unless the extraction variables are clearly defined.
Validated instruments are not automatically ethically neutral
A standardized or previously validated scale may have strong psychometric evidence, but validation does not answer every ethics question.
The instrument may ask sensitive questions, require specialized interpretation, carry licensing restrictions, be inappropriate for a particular population, or generate responses that trigger a safety procedure. A measure developed in one cultural or clinical context may also have different implications elsewhere.
Do not assume that "validated questionnaire" means "no ethical issues to consider."
Instrument content should match the consent information
Participants should not consent to one type of study and encounter a substantially different one when the questionnaire opens.
If participant information says the interview concerns general experiences of remote work but the guide contains detailed questions about mental health diagnoses, harassment, illegal activity, and family conflict, the consent process may not have adequately prepared participants for what participation entails.
This does not mean consent materials need to reproduce every question. They should, however, describe the nature of participation and reasonably foreseeable risks accurately enough for the applicable consent requirements.
The instrument should collect what the study actually needs
Instrument design can also affect data minimization and privacy.
If age categories are sufficient for the analysis, is an exact birth date necessary? If the study does not use participants' names, why does the questionnaire ask for them? If an open-text field invites disclosure of identifiable third parties, does the research need that information?
Collecting unnecessary sensitive or identifying information can create risk without corresponding scientific value. The instrument therefore connects directly to the broader question of why an ethics committee needs to understand how the research design operates.
Online survey platforms can collect more than the visible questions
A survey may appear anonymous to the participant while the platform records information such as email addresses, account identifiers, IP-related information, device data, timestamps, or other metadata depending on configuration and service.
Do not describe a survey as anonymous merely because you did not type "Name:" into the questionnaire.
Researchers should understand what their selected platform collects, what settings are enabled, where data are processed or stored, and whether the resulting information is consistent with statements made in the ethics application and participant materials.
Changing questions after approval can change the approved research
Once an instrument forms part of the approved study, modifying it may constitute a change to the research.
For research subject to the U.S. Common Rule, IRBs must have procedures ensuring that proposed changes are reported and that changes to approved research are not initiated without prior IRB review and approval except where necessary to eliminate apparent immediate hazards. OHRP notes that institutions may establish procedures for handling minor changes and that purely administrative corrections can be treated differently from changes to the research.
Other review systems have their own amendment procedures.
Watch Out
Do not decide that a new question is "minor" solely because it adds one line to the instrument. A single question about a substantially more sensitive topic, a new identifier, or a new category of personal information can alter the risk profile or approved scope of the study.
Version control matters for instruments too
If an approved questionnaire is revised three times, the research team needs to know which version participants are actually supposed to receive.
Use the document identification system required by your institution, which may include version numbers, dates, file names, or approval stamps. Keep approved copies separate from drafts.
This becomes especially important when instruments are among the supporting documents accompanying an ethics application. The existence of a questionnaire in the submission is not useful if nobody can later determine which questionnaire was approved.
