Interviews and Focus Groups

Matt Bhagat-Conway

Interviews and focus groups

  • Interviews and focus groups let you dive deep into a topic
  • Much more time consuming and costly than surveys
  • But information content per respondent is higher

When to use interviews and focus groups

  • When you are trying to answer a qualitative question - i.e. not how much, how far, how different, etc.
  • When you want to dig into causes and motivations
  • When there is a poor existing understanding of a subject
  • When you want topics to emerge organically

Interviews

  • Generally one-on-one interactions with partcipants
  • Varying levels of structure

Advantages of interviews

  • Can ask directly about why people behave how they do
  • Can follow up on subject-specific information
  • Can produce large and detailed information on specific topics

Challenges with interviews

  • Interviews take a long time
  • It can be hard to recruit participants
  • Lots of data is produced, which can be a challenge for analysis
  • Recruitment can be difficult
  • Interviewer can inadvertently solicit certain answers

Preparing for interviews

  • Three types of interviews
    • Structured
    • Semi-structured most common by far
    • Unstructured

Structured interviews

  • Interviewer has a script that they stick to
  • Relatively rare - why not just do a survey instead?

Semi-structured interviews

  • Interviewer has a loose script, but they can deviate from it/elaborate on questions/etc
  • Feels much more natural, while still ensuring comparability between interviews
  • Interviewer can adjust the script as they realize deficiences, or ask for additional information from short answers

Unstructured interviews

  • Basically a conversation
  • May have a prompt
  • Useful when you’re not looking for specific information
    • e.g. in oral history or ethnography

Inductive vs. deductive

  • Many interviews are inductive—the interviewer isn’t coming in with hypotheses, rather the interview is creating new ideas
  • Can also be deductive—testing a hypothesis (Brinkmann and Kvale 2018)

Writing interview questions

  • Questions should be open-ended, and invite longer responses
  • Fully structured interviews are rare, follow-ups based on responses are common
    • Many interview guides include “probes”—pre-written follow up questions to encourage interviewees to share more

Writing interview questions

  • Avoid leading questions
    • Can be subtle
    • Interviewers may present body language etc. that are difficult to control

Writing good interview questions

  • Ask difficult/embarassing questions towards the end (Harvard Sociology, n.d.)
    • There are ethical issues involved if you are asking potentially embarassing questions
    • Think about this in your consent and as you design your study

Writing interview questions

  • Level of structure may depend on purpose of study
  • When looking for differences between participants or groups, more structure/standardization are called for (Brinkmann and Kvale 2018)

Recording interviews

  • Recording can be helpful for later coding/transcription
  • But it can also change the dynamics
  • Some interviewees may not wish to be recorded
  • IRB has regulations about how you store/use recordings

Transcribing interviews

  • Transcribing interviews can facilitate coding
  • It can be time consuming or costly
  • Some new automated/AI options available
    • May be low quality or not meet IRB guidelines

Virtual vs in-person

  • Many interviews are now happening online
  • This is convenient, eases respondent burden and makes recording easier
  • On the other hand you lose some body language, etc.
  • May be harder to make respondents feel comfortable

Focus groups

  • Small groups of people (often around 8)
  • Members meet as a group for a discussion, possibly multiple times
  • Generally somewhat but not completely homogeneous
    • Consensus is not the goal, but comfort is
  • Very common in market research, but also common in academic research
  • Efficient (no need for many interviews)

Most insights in this section are from Hennink (2014)

Focus groups

  • Focus groups are very useful for inductive/brainstorming research

Focus group setup

  • Generally one moderator and one notetaker
  • Participants may be split by demographics or positions
  • Generally 1–2 hours
  • May or may not meet multiple times
  • Scheduling can be a challenge

Focus group dynamics

  • Group dynamics are the defining aspect of focus groups
  • They can be positive or negative
  • Some participants may be more or less comfortable speaking in groups
  • They may reduce expression of extreme views
    • This may be good or bad

Focus group dynamics

  • May more easily allow discussion to proceed beyond interview questions
  • Less role of moderator/interviewer

“Groupthink”

  • “Groupthink” is when group members try to avoid conflict
  • They quickly reach a consensus without critical evaluation
  • Differing opinions may be unsaid

(BYU Marketing Lab 2020)

Focus group guides

  • Guide may be more or less structured, like interviews
  • Academia tends towards more stuctured
    • Especially when comparing between focus groups is the goal

Focus group guides

  • Each question generally takes 5–10 minutes
  • Don’t overload a focus group
    • Leave time for groups to follow other paths

The role of the moderator

  • The moderator asks the questions
  • Should also ask follow-up questions,
  • Make sure all materials are covered,
  • Watch the time,
  • and promote equal engagement among group members

Confidentiality

  • You can’t guarantee confidentiality in focus groups, though you can encourage it
  • This may dissuade some folks from talking
  • If members already know each other, they may share less for two reasons
    • Perceived lack of confidentiality
    • Assumption other group members already know how they feel

Group activities

  • Can be useful, but also time-consuming
    • Listing items in a category
    • Ranking items
    • Sorting items
    • Vignettes/scenarios and reactions to them
    • Sketch mapping

Focus groups vs interviews

Focus groups

  • Capture range of views
  • Discuss issues
  • Exploration
  • Identify norms
  • Community-level information
  • Evaluate group interaction

Interviews

  • Individual perspectives/narratives
  • Personal/sensitive information
  • Detailed/in-depth information
  • Complex issues/processes that require long descriptions
  • Geographically dispersed participants
    • Or difficult to schedule

Modified from Hennink (2014)

Pretesting

  • Interview guides and focus group guides should be pretested (Harvard Sociology, n.d.)
  • i.e. conducted with the research team, friends, colleagues to get feedback on the guide itself
  • See what data you might be missing

The Delphi method

  • The Delphi method is an alternative to focus groups that tries to avoid “groupthink” and bias
  • It is conducted asynchronously in several rounds, with participants reviewing and commenting on others’ previous answers in each round

How many interviews or focus groups do I need?

  • You’re going to miss the simplicity of \(\frac{s}{\sqrt{n}}\)
  • Saturation is a common goal Brinkmann and Kvale (2018)
    • The point of saturation is when new interviewees don’t contribute things you didn’t already know
  • Qualitative research uses purposive sampling: the people in the sample are chosen on purpose due to some characteristic relevant to the study (Hennink 2014)
  • Many researchers recruit too many subjects, and spend too much time interviewing and not enough time synthesizing

Mixed methods research

  • “Mixed methods” is research that uses multiple methods, generally a mixture of quantitative and qualitative research
  • The methods complement in each other in different ways

Qualitative research before quantitative

  • Qualitative research can generate hypotheses, ideas for survey questions

Qualitative research after quantitative

  • Qualitative research can help understand why observed patterns exist in quantitative data
  • It can provide additional insights for subpopulations too small to support quantitative research

References

Brinkmann, Svend, and Steinar Kvale. 2018. Doing Interviews. London : SAGE Publications Ltd, 2018. https://catalog.lib.unc.edu/catalog/UNCb9763257.
BYU Marketing Lab. 2020. “How to Avoid Groupthink in Focus Groups.” BYU Marketing Lab, November 12. https://www.byumarketinglab.com/blog/2020/11/11/how-to-avoid-groupthink-in-focus-groups.
Harvard Sociology. n.d. Strategies for Qualitative Interviews. https://sociology.fas.harvard.edu/files/sociology/files/interview_strategies.pdf.
Hennink, Monique M. 2014. Focus Group Discussions. Oxford University Press, Incorporated. https://ebookcentral.proquest.com/lib/unc/detail.action?docID=1630622.

Creative Commons License
This work by Matthew Bhagat-Conway is licensed under a Creative Commons Attribution 4.0 International License.