The Potential for Sex Bias in Physical Employment Test Design

Abstract

Physical employment tests often simulate job-related tasks. The implementation of such tests can be constrained by cost, ease, and simplicity, which may unintentionally create a sex bias. PURPOSE: To investigate potential sex bias in prototype UK Royal Navy (RN) physical employment tests. METHODS: One hundred and fifty RN personnel volunteered (men n=75, women n=75). Physical characteristics, mean ± SD were (men vs. women): age 32 ± 8 vs. 29 ± 6 years; stature 1.77 ± 0.06 vs. 1.65 ± 0.05 m; body mass 81.0 ± 10.7 vs. 67.5 ± 9.8 kg. Five tests were developed with cylindrical, sand-filled Powerbags that simulated manual handling tasks performed onboard RN warships. Tests included two variants of an Aft Casualty Carry (35.0 kg and 41.5 kg, 10 m), Fore Casualty Carry (27.0 kg, 10 m), Foam Drum Carry (2 × 21.6 kg, 60 m), and Damage Control Timber Carry (28.4 kg, 50 m). Loads and distances were derived from equipment, infrastructure, Subject Matter Expert opinion, and anthropometric norms. Tests were performed as quickly as possible. A sub-sample of women (n=19) also performed a modified Aft Casualty Carry test using a Barbell (circumference 0.08 m) instead of a Powerbag (circumference 0.91 m). Cut-scores were developed for all tests. RESULTS: All men (100%) passed all tests. Compared to men the pass rates of women were above 80% in three tests (Fore Casualty Carry (88%), Foam Drum Carry (92%), Damage Control Timber Carry (89%)), but lower than 80% in both Aft Casualty Carry tests (29% 35.0 kg test; 8% 41.5 kg test). The lifting phase accounted for the majority of Aft Casualty Carry failures (49% 35.0 kg test; 77% 41.5 kg test). Within the sub-sample (n=19), 100% of women successfully lifted the Barbell Aft Casualty Carry loads, but markedly less lifted the same loads in the Powerbag version of the test (58 % and 32 % in the 35.0 kg and 41.5 kg tests, respectively). CONCLUSIONS: The larger circumferential girth of the Powerbag in the Aft Casualty Carry test was not representative of the anthropometric dimensions of a casualty, and appeared to create a sex bias against women. This highlights the potential risk of sex bias if physical employment test implementation considerations are prioritised above the faithful replication of simulated job-based tasks.