Structural validation of RN-HELM-360, a three-viewpoint leadership assessment for the Royal Navy

RN-HELM-360 is a 360° leadership assessment for the Royal Navy. It shows a selection board how one leader is seen from three viewpoints: Peers, Subordinates and Seniors. The three viewpoints are kept separate and never combined into one score. The questionnaire is fixed. It has five effectiveness questions (B1–B5) and six Skills questions (C1–C6, one of which is an integrity item). The project asks whether the Skills questions give the board something beyond a general “good leader” impression, and whether the answer holds up in new raters. Study 1 uses about 300 raters. It tests the internal structure of the eleven questions. A rule fixed in advance sorts the Skills questions into one of three outcomes: A: a distinct Skills score. B: largely captured by the overall score. C: in between. Study 2. Study 2 uses about 200 new raters. It tests whether the Study 1 structure replicates, using seven predictions that can fail: The structure fits acceptably. The Skills score stays “in between”. The effectiveness questions stay clean. The structure is the same for Peers, Subordinates and Seniors. The effectiveness score relates to an established effectiveness measure. It is distinct from servant-leadership style. Answers are stable when repeated. If any prediction fails, we report that it failed. How the two fit together. Study 1 finds the pattern and Study 2 tests it. Study 1 cannot count as a blind test, so Study 2 is the primary confirmation of the structure. Neither study shows that HELM predicts later career outcomes. That is planned separately, as a shadow-mode comparison at the March executive command selection board, with HELM kept out of the real decision. Materials. The registrations will include the locked analysis plan, the analysis code and the output tables. They will not include raw data, because the data contain personal information on serving personnel and are held in a restricted area.