M3 Interviewing and Ethics in Qualitative Research

When to Use Interviews

1. Understanding people’s perspective

2. Exploring lived experiences

3. Understanding how or why something happened

4. Need rich detail and context

5. Triangulate other data sources

Why we Use Interviews

1. Structured, open-ended questions

2. Rich data

3. Observation of non-verbal cues

4. Building trust

Open Ended Questions

Interviews let you ask participants directly and let them answer in their own words, giving flexibility while staying focused on your topic.

Rich Data

Get detailed, nuanced information, stories, examples, and explanations that surveys or records alone can’t provide.

Observation of Non Verbal Cues

Body language, facial expressions, and tone of voice can give extra meaning to what participants say.

Building Trust

The personal interaction allows the researcher to establish rapport, making participants more comfortable sharing honest or sensitive information.

How to Conduct Interviews

• In-person (ideal)

• Video-call (second best)

• Phone

• Email (least ideal)

In Person

Ideal, allows you to observe body language, build stronger rapport, adapt your questions in real time

• Usually gives richest data

Video Call

Second best, allows some visual cues and interaction

• Can feel less personal than in-person

  • Certain body language may be lost

Phone

No visual cues, rely on tone of voice

• Data less rich

Email

Should be avoided. Responses may be short, superficial, limited in asking follow-ups on the spot

• No non-verbal cues

• No personal interaction

Sampling for Interviews

When sampling, decide who to talk to to get the most relevant and meaningful information

• Identify people with direct experience

• Consider roles and impacts

• Think about access

• Focus on a sub group

• Decide on the perspective you need

Identify People with Direct Experience

Ask yourself

• Who has experienced this phenomenon?

These people are most likely to provide the best/richest insight

Consider Roles and Impacts

Ask yourself

• Who controlled this phenomenon? Who was impacted by this phenomenon? Who was benefited/harmed by this phenomenon?

This helps to capture multiple perspectives

Think About Access

Consider practical limitations like location, availability, and willingness

• You can only interview people you’re able to reach

Focus on a Sub-Group

Narrow your sample to a specific group

• Makes the study more manageable and targeted

Decide on the Perspective You Need

Ask yourself

• Whose experiences are understudied? Who is closest to the issue? What has previous research covered?

Purposeful Sampling

Deliberately choosing participants who are most relevant to the research question, rather than selecting randomly (i.e. as with quantitative research).

• Goal: get information-rich cases

Purposeful Sampling Names

May also be called

• Judgmental sampling, selective sampling, subjective sampling

Why Purposeful Sampling

In interviews, we care more about depth and relevance

• Rather than representing a whole population

• Non-random sampling is appropriate in these cases

Purposeful Sampling Types

These are purposeful sampling types for interviews:

• Maximum Variation

• Homogenous

• Critical Case

• Theory-Based

• Snowball Sampling

• Extreme or Deviant

• Typical Case

• Opportunistic

• Convenience Sampling

• Expert Sampling

Maximum Variation

Selecting participants that are as different from each other as possible

• Documents diverse perspectives or variations in experiences

Homogenous

Selecting participants that are very similar to each other

• Simplifies analysis, focuses on shared experiences, great for group interviews

Critical Case

Selecting participants where “if it happens here, it will happen anywhere.”

• Particularly hard to justify

• Allows generalization to other cases and maximum applicability

Theory Based

Selecting participants based on how well they represent a theory (or the opportunity to explore a theory)

• Tests, elaborates, or illustrates a theoretical concept

Snowball Sampling

Asking participants to suggest other potential participants

• Find new participants through existing connections

  • Great for establishing trust in potential participants from the get-go

  • Useful for contacting hard-to-reach groups

Extreme or Deviant

Selecting unusual or outlier participants

• Learn from rare or highly unique examples of the phenomenon

Typical Case

Selecting participants who are average or “normal”

• Highlights common or typical experiences

Opportunistic

Selecting participants based on who you know or easily access

• Allows flexibility and takes advantage of new leads or connections

Convenience Sampling

Selecting participants based on proximity or availability

• Saves time, effort, and resources

• Less rigorous, but practical

Expert Sampling

Selecting participants with high expertise on the topic

• Produces deeper, more informed insights than non-experts

Interview Best Practices

1. Record the Interview

2. Set Expectations

3. Listen Actively

4. Keep the Interview on Track

5. Give Yourself Time

Record the Interview

Recording lets you review details later and makes transcription easier for analysis.

• Always ask permission first

Set Expectations

Tell participants how many questions you have and how long it will take.

• Gives an idea of how long their answers should be and how long it will take

Listen Actively

Smile, nod, and make eye contact to show engagement.

• Don’t fill in pauses—wait a beat before moving to the next question.

• Don’t interrupt before they have finished talking

Keep the Interview on Track

Let participants answer fully.

• Use follow-up questions or gently redirect them if they go off-topic or miss details.

“You briefly mentioned…”

“I’ll jot that down. Can you explain…”

Give Yourself Plenty of Time

Don’t schedule interviews right before other obligations.

• Avoid being rushed, which can cut the interview short or make participants feel pressured.

Responses

The interviewer’s role is to listen, clarify, and gently guide

• Paraphrase and summarize to gauge your own understanding and seek clarification

• Ask follow-ups when more detail or examples are needed

• Validate interviewee’s experiences

Interview Challenges

1. Finding an appropriate sample

2. Getting off-track or going over time

3. Finding a quiet space/hearing each other

4. Distrust

5. Building rapport

Finding an Appropriate Sample

It can be hard to identify participants who have the experience or perspective you need.

• Solution: Use purposeful sampling to reach the right people.

Getting Off-Track or Going Over Time

Participants may digress, or conversations may run longer than planned.

• Solution: Use gentle redirection and schedule enough time so you’re not rushed.

Finding a Quiet Space

Background noise, poor acoustics, or technical problems (video/phone) can make recording and transcribing difficult.

• Solution: Choose quiet locations, test equipment beforehand, and use good-quality microphones if possible.

Distrust

Participants may be hesitant to share openly.

• Solution: Build rapport (e.g. make small talk, show active listening, reassure them), explain confidentiality, and maintain a non-judgmental, respectful tone.

Knowing When to Stop

You may be unsure of when to stop conducting new interviews

• Solution: Saturation

Saturation

A point where you’re no longer getting new information or themes from participants.

• If you did another interview or survey, the answers would be very similar to what you’ve already collected.

Saturation Example

When you can anticipate what people will say, it’s a strong sign you’ve reached saturation.

• Example: In a study about public transportation, after 15 interviews, participants all describe the same barriers (e.g., delays, safety concerns). Additional interviews just repeat these points.

Saturation and Context-Specific

To prove saturation, you don’t have to show that your results would apply to everyone everywhere

• Only that they are sufficient for your specific research project and its goals

Example:

once multiple students keep mentioning the same reasons (delays, cost, convenience), you’ve likely reached saturation for that university’s student population. You don’t have to show this is true for all universities worldwide.

Ethics

The inquiry around what is right and what is wrong.

Encompasses:

• What we should study

• How should we study it

• Who should be involved

• What should happen with the data

• What should be done with the results

What we should study

Choose topics that are important, meaningful, and safe to research.

• Avoid studying topics that could unnecessarily harm participants or groups.

NEJM Policy

1984 NEJM policy

• Very strict, zero-tolerance approach

• Reviewers could not have any personal ties to research funders at all

• Assumption: any tie = unacceptable risk of bias

2022 NEJM policy

• More flexible, “managed conflict” approach

• Reviewers may have personal/financial ties

• BUT they must:

  • Earn less than $10,000 from companies whose work they review

  • Disclose these ties

How should we study it

Use methods that are respectful, fair, and minimize risk.

• Ensure the process protects participants’ well-being (e.g., informed consent, confidentiality).

Core Principles of Human Ethics

1. Respect for persons

• Treat people as autonomous individuals

• Protect those with reduced autonomy (e.g., children, older adults, patients)

2. Beneficence

• Maximize benefits and minimize harm

• Avoid unnecessary risk

3. Justice

• Fairness in who bears the risks and who receives the benefits

• No exploitation of vulnerable groups

Respect for Persons

Shows up in research as:

• Proper informed consent

• Voluntary participation

• Right to withdraw

• Privacy and confidentiality

Beneficence

Shows up in research as:

• Scientifically sound methodology

• Careful assessment of risks vs benefits

• Safety measures and monitoring

Justice

Shows up in research as:

• Fair selection and recruitment of participants

• Inclusion/exclusion criteria that are justified

• Equitable access to benefits of research

Justifying Human Research

Research is ethically acceptable only if all of the following are met:

• ✔ Scientifically sound (otherwise participants are exposed to risk for no reason)

• ✔ Benefits outweigh harms

• ✔ Informed consent/assent is adequate

• ✔ Fair participant selection

Free and Informed Consent

Means that a person:

• Voluntarily agrees to participate (free from pressure, coercion, or undue influence)

• Has been given clear, complete, and understandable information

• Has the capacity to make the decision

• Knows they can withdraw at any time without penalty

How to Get Consent from Kids

A minor may provide informed consent if they have

• 1. Adequate decision-making capacity

  • Verbal, understand all aspects of the study

• 2. Consent provided by a parent or legal guardian

  • Then the child provides assent (an age-appropriate agreement)

    • If child says no, participation should not continue

How to Get Consent from People Who Cannot Provide Consent

1. Next of kin

2. Legally authorized representative

3. Spouse

4. Court-appointed decision maker

Next of Kin

A close family member may provide consent if legally permitted

• Typically used when no formal legal documentation exists

• Must act in the participant’s best interests

Legally Authorized Representative

Someone with legal documentation, such as:

• Power of Attorney for Personal Care

• Substitute decision-maker designation

Has the formal authority to consent on the person’s behalf

Spouse

A spouse may automatically act as the decision-maker in some jurisdictions

• Authority depends on local law and institutional policy

• Must prioritize the participant’s values and wishes

Court-Appointed Decision-Maker

A judge may appoint a guardian or representative

• This person has clear legal authority to make decisions

• Often used when there is conflict or no available family

What should happen with the data

Data should be:

• Stored securely

• Anonymized

• Analyzed

• Kept

• Decide whether it can be shared

Stored Securely

Protect data from unauthorized access

• Use password-protected files, encryption, locked cabinets

• Limit access to only the research team

Anonymized

Remove direct identifiers (names, addresses, student numbers)

• Replace with codes or pseudonyms

• Store the key linking codes to identities separately

Analyzed

Ethics doesn’t stop after collection

• Cleaned datasets, transcripts, coded files also need protection

• Analysis files can still contain sensitive info

Kept

Most institutions require data retention (often 5–7 years)

• Allows for verification, audits, or follow-up studies

• Data must remain secure for the entire retention period

Decide whether raw data can be shared

Not all data can ethically be shared

• Especially true for:

  • Small samples

  • Rare conditions

  • Qualitative interviews

  • Audio/video data

What should be done with the results

1. Share the results with participants

2. Publish or present

Share the results with participants

Participants contributed their time, data, and experiences

• Ethically, they deserve to know what came out of the research

• This can be:

  • A summary report

  • A presentation

  • A plain-language handout

May come in the form of triangulation (member checking)

Triangulation

• Member-checking means sharing findings with participants and asking:

  • “Does this reflect your experience?”

• Participants can:

  • Confirm accuracy

  • Clarify misunderstandings

  • Challenge interpretations

Publish or Present

Research should contribute to knowledge, not sit in a drawer

• Sharing results through:

  • Journal articles

  • Conferences

  • Reports to organizations or communities