Framework for user acceptance: Clustering for fine-grained results

In an attempt to uncover intra-group behavior similarities, we developed an open multi-level framework for understanding the process of technology acceptance by its users. We partitioned our population into groups by clustering at several levels and then for each level it was divided into subgroups with a measurement layer added to uncover subgroup influence. Thus, by intersecting the resulting clusters of the set of models, the population was divided into subgroups that have similarities in the factors measured by the cluster layer models. Subsequently we tested our framework in a university hospital setting; personality and prior technology background models were used in clustering via the Five Factor Model and the Technology Readiness Index. UTAUT was used in the measurement layer. Our hypothesis that the subgroups have differing degrees of explained variance and different predictors was confirmed. Our framework was open, because any model that results in a taxonomy of the population can be used to obtain meaningful clusters.