The COVID-19 pandemic has forced many researchers to switch to virtual sessions, but I was driven to using virtual methods for interviews and focus group sessions several years ago and I have not looked back since. This article briefly describes why conducting virtual focus groups may be a good move for your research and how MAXQDA can help you analyze your virtual focus group data.
Why Focus Groups?
Focus group studies have enjoyed broad adoption as a relatively low-cost and moderately effective means to pose open-ended questions to small groups of people selected to represent a target population to test ideas, products, concepts, or scenarios, elicit reactions and sentiments, or spark innovation. Focus groups can allow participants to build on each other’s responses in a constructivist and generative manner that allows discovery beyond the ability of individual interviews or surveys.
Focus groups have been described as collaborative co-discovery sessions, and are most similar to participatory design, generative brainstorming, and facilitated card sorting. Focus groups have been used extensively in marketing, public relations, political campaigns, product design, and computer user experience.
I was essentially driven to hold virtual focus groups through a combination of cost, convenience, and comfort. “Cost” of travel and accommodation is often a major cause of concern for administration and research managers, and even short trips can rack up considerable costs in a short time. Virtual focus group sessions greatly reduce the cost, and can often facilitate conducting more studies, being more inclusive, or including more topics.
When I say “convenience”, I mean that I can actually facilitate multiple focus groups in a day, often with people who are in different cities. For example, if I want to hold a session with nurse bed managers, I can hold the session with participants from ten different hospitals in five different cities without having to fly them all across the country. Hospitals also welcome me not being on-site – they are very focused on infection control, and the fewer people wandering around with potentially contaminated clothing, the happier they are!
By “comfort”, I mean the participant’s ability to attend while maintaining their confidentiality, working within their environmental constraints, and making it as easy as possible for them to participate. My typical participants are often healthcare workers (who work shifts), patients (who often have special needs), or their carers. Using virtual focus groups, my participants can attend the session in a conference room, a spare examination room, their car, a library room, or at home. For patients, this has an added element – the ability to attend without a strong sense of stigma. This is especially so when the group is related to topics for which there may be social bias, such as sexually transmitted diseases, mental health, or liver disorders.
Virtual Focus Group Logistics
Focus groups typically consist of the following:
- 6-10 members
- with an agenda constructed to address 5-10 topics or questions
- in sessions that often last around 1-2 hours.
These are not hard and fast numbers, and for virtual sessions, I have found that groups of fewer than five, or more than eight participants, are harder to facilitate. Too few and it feels like a meeting, and too many and some voices become silent.
Using virtual sessions makes certain assumptions about the participant’s level of computer literacy, and these assumptions deserve reflection. For participants who are highly familiar with conferencing platforms, using chat boxes, raising hands, or sharing files, it may be very fruitful to use all the rich features offered by the platform you choose.
However, it is still worth checking that all participants are at that level. Participants who are unfamiliar with conferencing platforms may have great difficulty in navigating the platform and be distracted by the technology instead of fully participating in the session.
Be prepared for unexpected interruptions (by a pet or child, for example)! One of the considerations when selecting a venue for holding an in-person focus group session is the ability to sequester participants from external distractions, such as noise, people wandering into the session, etc. With virtual sessions, unexpected interruptions are far more likely, and it pays to (a) have a plan for how to deal with them, and (b) brief participants that unexpected interruptions are likely, how the group should deal with them, and not to worry if it happens to them.
Generating Focus Group Data with MAXQDA
Virtual focus group sessions will often generate audiovisual content, including video recordings of the participants, images of whiteboards, images of models or constructions, or found objects, etc., and MAXQDA can accommodate all these data artifacts. However, it is important to be mindful of privacy and to seek individual permission to record participants.
It is wise to include a statement about any intended recording in the invitation sent to participants, but to also explicitly ask for permission prior to starting the recording, and then confirm again once recording has started so that this is captured in the recording and transcript. I typically read the preamble from a script to avoid drift and misstatements (but try not to become too robotic).
Bear in mind that while video is highly rich from an analysis point of view, it takes far greater bandwidth. It is a good idea to test that each participant has sufficient internet speed and working video and audio equipment (can be a cellphone) ahead of the formal session. I try to stick to two basic principles regarding video:
- the facilitator should always show video unless technology prevents this, and
- if one person cannot be on video, then rather have everyone on audio-only. This avoids those without video being forgotten or treated differently.
Importing Data into MAXQDA 2020.1
Although MAXQDA 2020.1 has very effective editing tools, it may be necessary to clean the audio, video, or still images prior to import. For example, it may be necessary to blur backgrounds in images, or filter out unwanted noise, or edit out unwanted image artifacts. Cleaning and preparing data may also be required to insert speaker changes manually and to replace any unwanted colons in the text. Colons will be interpreted by MAXQDA as speaker changes.
For example, an import of this blog article as a transcript would result in the sentence “One caution on auto-transcription: Many audio transcription services …”, being interpreted as a speaker change to a new speaker named “auto-transcription”, who started a new paragraph with “Many audio transcription services …”.
To import the files, click on Import (Figure 1 Item 1) > Focus Group Transcripts (2). You will be able to import transcript files that have embedded timestamps (3) or without timestamps (4). MAXQDA also supports a range of auto-transcription and audiovisual editing systems (5). You can also import still images such as photos or screenshots, audio clips, or video files (6).
Figure 1 Importing Session Data
It is worth noting that timestamps are not necessarily required for analysis of sequential events in a focus group, and simply using speaker changes in the text may be all your project requires. In this case, if the speakers are identified in the transcript with a colon following each speaker, MAXQDA will import the text, create speaker-specific documents and document groups, and pre-code the text sections ready for analysis.
For example, in Figure 2, click on Transcripts without Timestamps (1), and the resulting import will take the speakers identified in the text (2), and create documents for each speaker under the document group named for the imported file. In this case, I have one document per interview, but these could have as easily been done with a single document for the entire session that included all speakers.
Unlike how this project was done, it may be more common to use one document per session or topic, rather than per participant. Using a single transcript per session or topic will import those as the name of the document group, and so allow comparisons between codes, participants, and topics. Note that MAXQDA has auto-coded the text segments with a new code for each associated speaker (4).
Figure 2 Import sans Timestamps
Data Analysis & Reporting with MAXQDA
Once the data have been cleaned and imported, you are in a position to begin coding. Because text segments were auto-coded with a speaker code and placed within-session document groups during the MAXQDA import, you will be in a position to use MAXQDA’s Analysis, Mixed Methods, and Visual Tools menus to identify and display where there are commonalities or differences between the speakers. I have found autocoding with MAXQDA to be a significant time-saver in the analysis step and very useful in quickly identifying and comparing outliers.
For example, you can use MAXQDA’s Interactive Quote Matrix to compare multiple participants on multiple codes. In Figure 3, I am comparing participants #6 and #9 for their responses related to what I coded as “Primary Care”.
Figure 3 Interactive Quote Matrix
Likewise, it is possible to use MAXQDA’s Code Matrix Browser to see the code prevalence per participant, (Figure 4). Using this tool, I can quickly identify that Participant #1 talked about Patient Flow and Cost, while the rest did not, and that Participant #4 said very little about any of these topic elements. I may want to compare what Participant #9 and Participant #2 said regarding Space Utilization since they both had much to say about this element.
Figure 4 Code Matrix Browser
Many of the considerations that apply to providing virtual training are also pertinent to conducting virtual focus group sessions. You can read on that topic in the MAXQDA blog “Delivering MAXQDA Training or Consulting Remotely: Technical Requirements, Tips & Checklist. The most important of these may be the advice to test all technology well before the session.
My checklist specific to virtual focus groups includes:
- Does the communication software support auto transcription?
- Does it identify speaker changes, or will I need to do this manually?
- Has connectivity for each participant been tested prior to the session for both audio and video?
- Do I have enough storage space for concurrent video feeds?
- If I need to switch to audio-only, will I still have acceptable data for the research purposes?
- Do I have an alternate date in case the entire session needs to be aborted?
Virtual focus groups have significant cost, convenience, and flexibility advantages over traditional face-to-face sessions, but rely heavily on the availability of reliable data connectivity, and high-speed connectivity if the video is to be used. As long as the participants are familiar with the level and type of technology to be used (phone, computer, conferencing software), the virtual experience can be almost as rich as traditional settings, and some kinds of participants may find virtual sessions to be more convenient and have a greater sense of physical and psychological safety.
About the Author
Ready to try MAXQDA 2020.1 for yourself? If you are a MAXQDA 2020 user, you can install this update with the new features and improvements completely free! If you are new to MAXQDA or have an older version of the software, download the free trial to test MAXQDA 2020.1 for 30 days with no obligation: