We started this analysis by using dendrogram method [34,
p. 220]. This led to the reduction of initial 64 nodes into
five clusters. We then went back to the original data and
seek for further understanding of the context (as discussed in
the previous section). This exercise revealed that all five clusters
actually described processes/interactions that created/
overcame barriers during the maintenance stage of ISO
14001. Secondly, further analysis of the data uncovered that
these processes linked key actors, which was a distinguishable
feature. As a result we labelled these five clusters (transforming
and value adding, administering and improving, understanding
and accepting, communicating and learning, availability) and
similarly named the key actors (external environment, EMS
team, EMS, organisation, resources and skills) and created
the first version of the model (presented in this paper in its
final version, see Fig. 2). Yet, as Miles and Huberman [34]
point out ‘‘clusters, like the results of other conclusion drawing
tactics, must be verified’’. This process is described below.