Framework Analysis is similar to Thematic Analysis (Braun and Clarke, 2006), but the second is much more widely used.

Can you say more about why you want to use Framework Analysis?

Thanks Mr Morgan

I intend to check variations when pretests (in that case societal campaigns aimed at promoting condom use) are held in 3 different contexts. I use the standard lab (commercial room used by marketers), home setting (as done in some countries) and rock festival (a place where marginality is well tolerated). Considering the importance of genre for this topic, I use on "male only" group and one "girls only" group at each site. At each place, I did the same pretests with the same material, trying to keep the same method to produce the material usually requested for a "pretest report". The collect phase is completed and I have the transcripts ready to analyze.

I am now at the analysis part. I planned to proceed in two steps. For the last part, I intend to correlate the differences observed in the results obtained in each site with the manifestation of self-revelation behavior detectable in the discussions (for this I use thematic analysis based on theory ... a more classical approach in focus group analysis. But I want to identify those differences... so, based on what I found in the litterature, I found comparatives studies that used Framewor Analysis. The "matrix" form resulting from the analysis seems appropriate for the comparative nature of my work.

In fact, in this two-steps work, the focus groups analysis will provide me the material that I will analyse at the last step. The focus group "pretest" technique that I used was built using the actual practices of a real producer of health-promotion campaign (I check the work done in the last 10 years). So I intend to keep the data collect close to the real-life practices... i.e. based on predefined questions, without a real analysis of interactions, focused on checking perceptions, comprehension, and emotive reactions (verbal or non-verbal) to the ads. The choice, I know, seems more "mechanical", but I chose to use the same approach that is used in real-life pretesting of campaign projects, because that it is precisely the often dogmatic nature of these practices (methodolatry) that I wish to question in my conclusion.

i hope to be able to observe that the context, in a pretest focus group, is not an "inert" factor... in many publications, it is presented as a minor (even inimportant) factor, comparatively to the animation, the recruitment process, etc.).

Sorry for my English writing skills, and thanks for your interst.

My understanding of Framework Analysis is that it sets up a matrix where the rows are the sources of the data and the columns are the codes. I don't see any reason why you couldn't use that format with Thematic Analysis (or any other kind of qualitative analysis, for that matter. It approach is, however, most explicit in Framework Analysis.

Also, my impression is that all the qualitative data software programs will provide you with this data format in a spreadsheet. I am not haven't used NVivo, but I know that ATLAS, MAX, and Dedoose will give you this kind of row x column arrangement, and allow you to "filter" by row (i.e., data source).

In terms of "pre-test," I am not very familiar with the use of focus groups in marketing or product design. So is this use pre-testing part of pre-testing a product or pre-testing the focus group design (e.g., the questions you will ask). 

Thanks for your quick answer.

I am still uncertain about Framework Analysis... but the fact that it ask the researcher to work with synthesis of the discussions (not with the exact retranscripts) seems an interesting point (raising of course questions of rigor and validity in the interpretation).

And for my use of the word "pretest" ... My explanation was not clear (in fact, I will have to work on this... even my Director was confused at first!). The fact that i work with a qualitative instrument (focus groups) within a mainly-quantitative discipline (marketing) is surely not helping !!

In my thesis, I study the way commercial pretests are done by those who launch campaigns. These pretests are formative research that promoters use to check their creative projects before launching them and spending a lot of money. Using different places, I will examine if the context of those discussions may have an influence on the way people express their opinions on the ads presented. Can the fact that we are discussing in a rock festival or a commercial building change the way we perceived ads or express our frank opinions ?

I am well aware that focus groups are a more potent and rich instrument that the way it is used in marketing, but I have to choose my battles. This one, targeting my marketer colleages (I work in a governement's agency in the health communication team), is centered on their perception that there is only one "proper way to do focus groups" to pretest ads (an affirmation I have heard many times for marketers trying to justify the recurrent use of a standard technique to conduct focus groups when pretesting any campaign). A "standard use" asking usually for commercial facilities providing rooms with mirrors, webcasting of discussions, hot meals and alcohol for observers behind the glass, etc.

Thanks again for your time sir, and have a nice day!