Internal validation of an exploratory human/information technology cause analysis model.

摘要

Despite significant capital investment in increasingly complex information technology (IT) systems, IT systems continue to fail. Although there are several human performance technology (HPT) and IT cause analysis and failure cause models and taxonomies, these models do not integrate HPT and IT perspectives. This integration may provide the necessary direction to form a more comprehensive picture of the causes of performance issues when IT systems are involved. This more comprehensive picture may allow for more appropriate sets of interventions to resolve the performance issues, reducing IT system failures. The proposed human/IT system (H/ITS) cause analysis model, which combines the strengths of HPT and IT models and taxonomies, was developed and internally validated using a three-phase, multiple-method design: (a) literature review to analyze existing HPT and IT models and taxonomies on which to base the model, (b) selection of the model components and creation of the integrated H/ITS cause analysis model, and (c) three-round Delphi study using participants with expertise in HPT models and dual expertise in HPT and IT. The model was revised after each round, based on participants' responses. Internal validity was determined by the extent to which consensus was achieved regarding the model and each round's questions. Overall, the model was found to be comprehensive and potentially useful, including some components that were new or not typically considered by the participants in their practice. However, the initial versions of the H/ITS cause analysis model did not provide participants with sufficient information about the intended context and application of the model. Support or explanatory materials should accompany HPT models, particularly new HPT models. When performance issues are related to specific areas such as IT systems, HPT practitioners may find similarly specific HPT models useful. Study participants proposed that the H/ITS cause analysis model could facilitate conversations between HPT practitioners and their clients to more fully diagnose performance issues and their root causes. Future research should address further development and external validation of the H/ITS cause analysis model and the development of other specialized cause analysis models focused on areas such as human resources or project management.