Guidelines for Preparing Focus Group Data

There is only one requirement when preparing focus group data: Each speaker or needs to have a unique identifier and this identifier should be applied consistently throughout the entire transcript.

ATLAS.ti offers two default patterns to recognize speakers:

  • name: (pattern 1)
  • @name: (pattern 2)

As speaker IDs are used as codes, the second version has the advantage that all speakers are sorted underneath each other in the alphabetically sorted code list. Further, it is unlikely that ATLAS.ti finds anything but speakers. Searching for pattern one, ATLAS.ti also finds things like "for example:....."

If your data is not transcribed using the @, you can still change code names before the automatic coding process begins.

Custom Pattern

You can also define your own pattern including the use of regular expressions.

This might be useful if you have previously prepared transcripts and speaker IDs have not been used consistently.

For instance, the regular expression: To+(m|n) matches "Toom," "Ton," or "Toon" to find different spellings of the name 'Tom'.

([A-Z]+): recognizes identifiers consisting of letters from A to Z.

([A-Z]+[0-9]+): recognizes identifiers that consist of a combination of letters and numbers

For more information, see Regular Expressions (GREP).

You can use this functionality to auto code any data that has a specific structure. It does not necessarily have to be a speaker. It could be a date, a number, or any other type of identifier.

Example transcripts

The speaker IDs do not have to be in bold letters. Bold letters are used here only to make it easier for you to see, which ones are the speaker IDs.

Example A (pattern 1)

Alex: I don't know, I'm the sort, I don't really struggle to make friends cos  everyone tells me I've got a big mouth, and I don't stop talking [laughs]

Tom: So is that how, is that how you met just, just through you striking up a conversation?

Deb: I'm trying to think exactly [laughs] I think that's what it was, we were both in the same research methods class ...

Example B (pattern 1)

Alex:

I don't know, I'm the sort, I don't really struggle to make friends cos  everyone tells me I've got a big mouth, and I don't stop talking [laughs] .

Tom:

So is that how, is that how you met just, just through you striking up a conversation?

Deb:

I'm trying to think exactly [laughs] I think that's what it was, we were both in the same research methods class.

Example C (pattern 1)

It is also possible to add further information to each speaker like their gender, age group, educational level etc. This way, also this information will be automatically coded.

Alex: gender male: age group 1: education high school:

I don't know, I'm the sort, I don't really struggle to make friends cos  everyone tells me I've got a big mouth, and I don't stop talking [laughs]

Tom: gender male: age group 2: education high school:

So is that how, is that how you met just, just through you striking up a conversation?

Deb: gender female: age group 2: education some college:

I'm trying to think exactly [laughs] I think that's what it was, we were both in the same research methods class.

Example D (pattern 2)

@Alex: I don't know, I'm the sort, I don't really struggle to make friends cos  everyone tells me I've got a big mouth, and I don't stop talking [laughs].

@Tom: So is that how, is that how you met just, just through you striking up a conversation?

@Deb: I'm trying to think exactly [laughs] I think that's what it was, we were both in the same research methods class.

If your data is not transcribed yet, we recommend using pattern 2 as it is unlikely this pattern will find sections that are not speaker units.

Recommendation

For readability, you may want to consider starting each speaker unit on a new line.

Additionally, you may want to enter an empty line between speaker units.

Both is not required. When a pattern is recognized the chosen code(s) are applied from the first letter of the pattern to the start of the next recognized pattern. Therefore, it does not matter whether a new unit starts on a new line or whether there is an empty line in between.