Revealing Latent Strategy Structures from Expert Critical Care Nurses

Abstract

It is common knowledge that operators controlling dynamic environments use different types of strategies and choose strategies for different reasons. We have been using a Threat-Strategy Interview (TSI) procedure based on the threat and error management view to elicit strategies. However, there is a need for a method to uncover strategies and sub-strategies from subject matter expert (SME) interviews. This study explored an algorithmic procedure for the generation of strategy organization from participant sorts of strategies, tactics, and actions mentioned by SMEs during structured TSIs. Clusters that emerged from N x N agreement matrices of rater judgments on strategies collected from pediatric intensive care unit nurses allowed for the identification of strategy structures that were similar to those that emerged after lengthy discussion but did so in considerably less time. Through the use of this agreement matrix, we hope to uncover the latent structure of strategies from experienced nurses and, ultimately, to teach those strategies to novice nurses.