Each group experienced difficulty during the first few minutes of Exercise 1. This
difficulty appeared to be a result of two factors. First, the groups were still familiarizing
themselves with the codes used to describe each job. Second, they were trying to figure
out how to cognitively frame the task. Based on the discussion that took place, the
groups used both similarity and relatedness as a clustering scheme, often drifting back
and forth between the two concepts without a clear distinction. In addition, each group
explicitly discussed the typical workflow process as a method to cluster jobs. The
dynamics of the third group were particularly interesting. After an initial period of
indecision, several SMEs adopted a relatedness clustering approach, while several
others adopted a similarity approach. Neither group of individuals explicitly used those
terms, however from the discussion it was apparent that the concepts of similarity and
relatedness were behind their reasoning. Because of this difference, there was a split in
the group for all of Exercise 1. We did not interfere, and instead let the differences play
out to see if or how they would be resolved – they never were. The two factions were
unable to find common ground or a common understanding because they formed
different fundamental conceptualizations of the problem. This ineffectiveness was
noticeably frustrating for some participants.