Interviews and focus groups in a Capella IRB application: guides, recording, consent, and confidentiality limits

What does the board want to see in an interview guide or focus-group protocol?

The guide is an instrument, and Capella treats it like one. The IRB application asks for "data collection tools (interview protocol, surveys, questionnaires, observation protocols, etc.)," and the board reads the questions themselves, not a description of the topics. Semi-structured does not mean unwritten: list the opening script, each planned question, the probes you expect to use, and the closing. If your questions are adapted from a published instrument or a prior study's guide, Capella's SOPs require written permission to use it, and the permission letter must state clearly that you may modify it if you have. The SOPs also allow the board to request an expert review of instruments or interview guides when there is concern over their appropriateness for the population, which is more likely when the topic is sensitive or the participants are vulnerable.

Two things the board checks that people forget. First, that the guide in the application is the guide in the plan: if Scientific Merit Review saw eight questions and the application carries eleven, the file has drifted and will come back to be reconciled. Second, that the questions match the consent. If the guide asks about disciplinary history and the consent says the interview covers "your experience of the program," the consent is inaccurate.

Recording: audio, video, transcription, and where it all lives

Capella's guidance on virtual data collection says that if you use video conferencing for interviews or focus groups, you should "still only audio-record your participants." Plan the study that way from the start: the consent describes audio recording, the platform is set to record audio, and the application says how the recording is captured and stored. If you intend to take field notes instead of recording, say so; if a participant may decline recording and still take part, say what happens then (notes only, or the interview does not proceed).

Transcription is part of the data path, and the board wants it named. If a third-party service or an automated transcription tool will touch the audio, the data-security plan has to say which one, what it receives (audio with names spoken, or audio after you have already pseudonymised the file name), what the service's terms allow, and when the audio is deleted from it. Capella's SOPs require plans "to manage, store, and destroy the data once it has been collected," and Capella's policy defines research materials to include "audio recordings, transcripts, data files" and requires those materials to be maintained for seven years after completion of a study. Resolve the apparent tension in your plan explicitly: many approved files destroy the audio once the transcript is verified and de-identified, and retain the de-identified transcripts for the required period. Whatever you choose, write it as a sequence with named points of destruction, because "recordings will be kept securely" is the sentence that triggers a revision request.

What must the consent say for an interview study?

Capella requires informed consent before any research-related interaction, obtained in accordance with 45 CFR 46.116 and 46.117, and asks you to use its consent template. The basic elements that matter most in qualitative work:

1. That this is research, its purpose, what the participant will be asked to do and how long it takes, which for an interview means the expected length of the conversation and whether there may be a follow-up.
2. The reasonably foreseeable risks. For interviews that is usually discomfort in discussing the topic and the risk of a breach of confidentiality; name both.
3. A statement "describing the extent, if any, to which confidentiality of records identifying the subject will be maintained" (46.116(b)(5)). This is where the recording, the transcript, the pseudonym, and the destruction plan are summarised in plain language.
4. That participation is voluntary and refusal or withdrawal carries no penalty (46.116(b)(8)). In a workplace or clinical setting, say what that means concretely.
5. Whether identifiers could be removed and the data used for future research, or not (46.116(b)(9)).
6. Whom to contact about the research and about rights as a participant.

For telephone interviews or interviews where the consent form would be the only document linking a person to the study, the board may waive the signature requirement under 45 CFR 46.117(c) and allow verbal consent documented by the researcher, usually with a written information sheet provided beforehand. Ask for that waiver in the application rather than assuming it, and attach the verbal-consent script. Capella's SOPs confirm its IRB may allow the waiver under that provision.

Why can a focus group never promise confidentiality?

Because you are not the only person in the room. You can promise what you will do with the recording and the transcript; you cannot promise what another participant will say at work the next morning. Capella's SOP definition makes the distinction sharp: confidentiality means only the researcher can identify the responses of individual participants, and in a focus group that is false by design. The board expects the consent to say so in words a participant will understand, and expects the protocol to contain ground rules that ask participants to keep what they hear in the room, while making clear this is a request, not a guarantee.

Practical consequences follow for design. Focus groups with coworkers from the same unit raise the stakes of every disclosure; groups mixing supervisors and subordinates make candour unlikely and pressure likely, and the board will ask how you composed the groups. Use first names or chosen pseudonyms on the recording, tell participants they can decline any question, and say in the consent that they should not share anything they would not want repeated. Recruitment text should not describe the group in a way that reveals who else is eligible (a group "for nurses who reported the incident" identifies everyone in it).

What are the limits of confidentiality you must state in advance?

Beyond the focus-group limit, there is the researcher's own. Capella's SOPs note that state and local laws vary in their "provisions for mandatory reporting," and that researchers must consult their own state's requirements when designing the study. If your professional licence or your state's law obliges you to report certain disclosures, abuse, neglect, imminent harm, the consent must say that confidentiality will be maintained except where the law requires disclosure. A clinician interviewing patients or colleagues about care failures should expect the board to ask what happens if someone describes a reportable event, and the honest answer belongs in the consent and the protocol.

Anonymity is not available in face-to-face qualitative work; you have met the person. What you can offer is confidentiality with a named mechanism: a pseudonym assigned at recruitment, a key stored separately from the transcripts with a destruction date, identifying details altered or omitted in quotations, and the site never named in the findings, which Capella's SOPs require of every study.

Which review path does an interview or focus-group study take?

The federal expedited categories include research "employing survey, interview, oral history, focus group, program evaluation, human factors evaluation, or quality assurance methodologies" where the risk is minimal, and that is where most Capella interview studies sit. Some interview studies qualify for exemption category 2, but recordings of identifiable voices and topics touching employability, reputation, or legal liability move a file off that path quickly; sensitive topics can move it to the full committee, which Capella's own overview reserves for studies involving "sensitive topics, deception, or greater than minimal risk of harm." The board, not the applicant, determines the level; our explanation of exempt, expedited and full-board review at Capella covers how it decides.

The four documents and what brings each one back

None of these are research problems. They are file problems, visible before the file is filed, which is the argument for having the whole dossier read and version-matched before submission. How we build and carry a Capella file from determination to approval letter describes that reading, and our step-by-step article places it in Capella's sequence.

What to do next

If your project interviews people or runs focus groups, send us where it stands: the draft guide, the population, the setting, and whether you will record. A consultant reads it the way Capella's board will and writes back with what the file needs. We draft every document, file the application, and answer every reply until approval; the study stays yours and the board's decision is its own. Request the free application review, or read the questions we are asked most often.