The luddites were not the first to resist the introduction of new technologies and I have always had some sympathy for Ned and his followers. Their work was being taken away by new machinery and no one paid any particular attention to their survival needs. So this post is a mixture of explanation, frustration and a plea for ideas for new gainful employment for a cadre of qualitative researchers whose work is being displaced by social computing, big data and SenseMaker® in particular.
Now the cause of this is a series of discussions over the years around the differences between SenseMaker® and various forms of qualitative research. The rest of this post assumes some knowledge of the way SenseMaker® works by the way. What I’ve done is list some of the objections or often questions posed as objections and a response. Its not complete and more answers and questions would be appreciated as I want to build this into a document of some type, maybe a full article.
A expert tagging the data and assigning themes to stories can get as good if not better results
(i)It is dubious if they can, when we have compared expert tagging with responded tagging they tend to differ (ii) what about voice and pictures? They are often more useful than text (iii) Expert tagging can’t scale or provide real time feedback (iv) decision makers have become expert in ignoring experts if they don't like the results, in part ‘cause they have had bad experiences (v) we want the subjects research subjects engaged not only in primary but also secondary interpretation, more me this is a moral issue (vi) SenseMaker® approaches also enable peer to peer KM as an added benefit
Its easier to analyse responses if linear scales are used
And its easier to model human systems if we assume they are ants but we are not. One problem is that you are allowing people to run on automatic pilot and they will either gift or game the response. By asking people to balance all positive or all negative qualities they have to think more deeply, they engage in sense-making, they don't know what the right answer is.
Giving non-statisticians access to statistical tools is dangerous
and allowing statisticians the power of interpretation is even more scary
It would be better to have three linear scales than a triad then people could score them all rather than having to do trade offs
(i) there is nothing to stop you having linear scales as well (ii) the whole point is to prevent gifting or gaming (iii) you can always add a MCQ or scale to validate the importance
Well how do you interpret material at the centre of a triad? It could mean they are equally important or equally unimportant
This is a more interesting one and it is valid. Centre scoring can be seen as equally important or equally unimportant. That can be handled by a secondary question with the triad. That said I tend to ignore centre scoring anyway, the interest is in the outliers with abductive research.
But you are imposing a structure with the names you give the labels
Yes we are creating a grammar of meaning to allow research subject and researcher to communicate with each other. If you depend on machine interpretation (which is a useful addition) then you are assuming () algorithms can understand all the subtleties of human language (ii) that the meaning is in the text and (iii) that you only have text. All of those are dubious assumptions. High abstraction signifiers empower the signifier to determine for themselves how their narrative will be interpreted by giving them access to the interpretative framework rather than it being hidden in the researchers mind set.
There are lots of things like cross-tabs that SenseMaker® Explorer doesn’t do
We don’t do that we export to SPSS and other tools.
You can’t call it a correlation if it comes from a triad
No we can’t its evidence of an abductive link, this is abductive not inductive research
SO WHAT TO DO?
So what do with people who just don't get it? Ideas on the back of a post card please to …..
Cognitive Edge Ltd. & Cognitive Edge Pte. trading as The Cynefin Company and The Cynefin Centre.
© COPYRIGHT 2023
Enterprise Architecture seems to be catching up with Agile in terms of interest in complexity ...
An odd and messy sort of day. I am in the middle of sorting out ...
Leave a Reply