Qualitative Data – High Volume

Six Sigma – iSixSigma Forums General Forums General Qualitative Data – High Volume

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
  • #53541


    Hi All,

    We have recently run a number of all company meetings and have run ‘Affinity’ type workshops with all staff to garner their views on how we can improve over the next year.

    We now come to the Analysis and I must say it’s been a struggle. We have gathered over 2,200 comments, which do have a certain amount of classification, so we can give some statistical feedback. I have generated ‘word clouds’ for visual impact along with Pareto to show team involvement and buy-in etc. However, the only way I can think to analyse the actual comments is to read each and give a single word summary, but with so many, that would be quite an investment in time, and it breaks my personal golden rule of adding my opinion to data!

    Does anyone have any suggestions or tools which they have used in similar situations?

    Any and all feedback would be much appreciated.
    Kind regards, Angela



    Your team is going to have to distill the comments. That does not mean adding their opinion to them, but they are going to have to be transformed from 2200 items to a manageable number of items. They will have to be prioritized. That means setting aside a very large proportion of what the team heard. That’s also not changing the comments by adding opinion, but it is making a selection of which can be acted on with limited resources.

    You mentioned that you’ve done some classification. That’s a start. You probably need to reduce the 2200 to 200 or less items by reducing redundancy. This requires judgement, but not adding opinion. Have the team affinitize the raw comments into groups of very similar comments. Replace the individual comments with a single comment that best describes the group. It may be one of the actual comments or it could be the team’s best interpretation of the group. On this representative comment, be sure to place the number of comments that were consolidated into it. Something that was mentioned 20 times may be more critical to work on than something mentioned once.

    When you are down to 200 or so, follow the KJ process to distill further and prioritize. A good reference for KJ is in “Commercializing Great Products with Design for Six Sigma”, by Perry and Bacon. Note that most of the KJ references that can be found with internet searches are abridged, incomplete, or just wrong descriptions of the KJ process. Get a good DFSS book with in-depth description of KJ if you want to do it well.

Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic.