Transcription of Focus Groups - Make Sure Your Recording is Up To Scratch

Things to consider before conducting the focus group

If you are conducting a focus group and need to produce a transcript of it later on for analysis then you need to consider the transcription aspect right from the start.

What could be more frustrating than moderating a really productive and information-rich session only to find that your recording is unusable or the transcript comes back from your transcriptionist looking like gobbledegook.

Transcription of even a one-to-one interview is a tricky business and a professional transcriptionist will take around four hours per hour to transcribe a clearly recorded one-to-one interview. A focus group takes even longer because of the variety of different voices and accents, the fact that people talk over each other and the need to work out who is speaking. However, there is quite a bit you can do to make the transcription easier, and an easier transcription will take less time and therefore should cost less if you’re having it professionally transcribed.

If you’re providing refreshments preferably have the food before or after the discussion. Not only is transcription of people eating somewhat unpleasant, it’s also considerably more difficult to understand people, as you’ll know if you’ve ever listened to someone talking with their mouth full.

Do make sure you have a good enough recording system with external microphones. Putting a Dictaphone in the middle of the table is just not going to do the job. These are designed to be held close to the mouth for one person to record dictation into. For a focus group you will need a system with external microphones, ideally one per participant. If that’s not practical then one good quality microphone at the very least and preferably two or three. Do also make sure the group is meeting in a quiet self-contained room with no distractions.

Sometimes you won’t need to know who’s speaking, so long as each time a new person speaks this is indicated on the transcript. This will usually be cheaper as the transcriptionist does not have to distinguish the different voices, so it’s worth considering in advance whether you need the information on who’s speaking at any particular time.

If it is important to know who is talking then you will ideally not have more than five people of the same sex or eight people altogether, unless their voices or accents are very distinctive.

Conducting the focus group

Firstly lay down the ground rules at the beginning of the focus group. Explain why the tape recorder is there and check no one objects. Have a back-up plan or policy ready if anyone does object, or make sure everyone is notified in advance.

You will obviously need to explain what you want from the participants, why the group is being conducted, what you are researching etc. but if you don’t want this transcribed switch the recorder on after you’ve done that, or tell the transcriptionist where on the tape to start. (This is easier with digital transcription as audio tape counters are not very precise.)

Explain that people talking over each other will be problematic for the recording so ask that they please refrain from doing so. Don’t be afraid to remind them of this as when people get passionate, excited or angry they will talk over each other!

Before going any further get everyone in the group to introduce themselves. Ask them to say specific things so that you get more than a name. For example, ‘Hello, can you tell me your name, where you live, your job, who else lives with you and your favourite food’. This would be appropriate for market research. For a group of scientists you might say ‘can you say your name, just for the purposes of the tape, where you work and a sentence or two about your current research’. The point is that the transcriptionist has a chance to ‘learn’ the different voices.

It’s hard to strike the perfect balance between getting all the information and letting a true discussion develop. If the purpose of the group is to develop a discussion e.g. a group of company directors discussing government policy or a group of scientists giving opinions on your company’s latest product, then you won’t want to be continuously interjecting to remind them not to talk over each other, but if your primary reason for using a focus group rather than one-to-one interviews is financial e.g. market research, then the primary goal is probably to get opinions from as many people as possible and an in-depth discussion is not really the aim. In this case you can be firmer with moderating the group. You could, for instance, ask them to say their name every time they speak e.g. ‘Mike. I think it tastes like sawdust actually’. ‘Hi, I’m Lisa. I think it’s a cross between a digestive biscuit and sandpaper.’ ‘Mike again. Yeah, sandpaper’s a good description.’ The transcript would then be produced as follows:

Mike: I think it tastes like sawdust actually Lisa: I think it’s a cross between a digestive biscuit and sandpaper. Mike: Yeah, sandpaper’s a good description.

If it’s important to stick to a discussion plan the don’t be afraid to gently reign people in if they go off track e.g. ‘That’s a very interesting point Mike, but what we’re really discussing here is the price. Of course taste is extremely important and we’ll certainly be discussing it later on. For now though, what do you think of the price?’

After the group – preparing for transcription

Make a list of all the names (with correct spellings) and genders, and other relevant info e.g. Dr Sandy Shaw, Blogs Hospital NHS Trust, Qualified 10 years – Female. Give this list to the transcriber along with the names of the moderator and anyone else who is present and makes any comment e.g. someone who’s helped to organise the group or is participating in the research.

Make sure you tell your transcriptionist exactly what you want e.g. you almost certainly won’t need a ‘verbatim’ transcription including every ‘um’ or ‘er’ and every repeated word. If you do need this it will probably be more expensive. Don’t be confused into thinking that it won’t be an accurate transcript if it’s not verbatim. ‘Intelligent verbatim’ is exactly what’s said, but leaving out all the ‘speech fillers’, hesitations etc. and this is probably more practical in these situations. Sometimes you might want the answers edited to tidy up the grammar, and sometimes for market research you might want each comment categorised into a table to make the results more quantitative. Each of these ways of transcribing will be differently priced and some transcriptionists will have more experience in market research categorisation, so think carefully about what you need and make sure you find the appropriate person to carry out the work for you.