Department of Defense (DoD) members sought to improve the inter-rater reliability of the DoD Human Factors Analysis and Classification System, (DoDHFACS). DoDHFACS differs from the original system developed by Wiegmann and Shappell (2003), based on the work of Reason (1990), by further analyzing mishaps and hazards to a more granular level – arriving at specific “nanocodes.” The steps involved in the effort included determining which of the 147 “nanocodes” were rarely/never used and collapsing nanocodes and rewriting definitions to arrive at 109 nanocodes. Next, a stepwise checklist to guide investigators through consideration of nanocodes was created. Student investigators were guided to continually test checklists and generate results to gauge inter-rater reliability. They were asked to offer constructive criticism to hone checklist questions. While inter-rater reliability results are encouraging (Fleiss’ Kappa of .847 at the broadest level), additional work is necessary to realize the goal of an optimally reliable human factors taxonomy.
King, R. E.
(2015). A Comprehensive Effort to Arrive at an Optimally Reliable Human Factors Taxonomy. 18th International Symposium on Aviation Psychology, 37-42.