Structured interviews double predictive validity
0.51. Versus 0.28. That’s the difference process makes.
Organisational psychology research consistently shows that structured interviews, with standardised questions, behavioural frameworks, and scoring against defined criteria, achieve predictive validity correlations of 0.51 to 0.63 with job performance. Unstructured interviews achieve correlations of 0.20 to 0.38.
In practical terms: an unstructured interview is roughly half as good at predicting job performance as a structured one. For Government Agencies whose legal framework requires selecting the best candidate, this is not abstract. It is a structural failure in selection methodology.
The merit principle, properly applied, requires structured assessment. Many Agencies endorse the principle but implement processes that do not achieve it in practice. Structure is not an add-on to merit, it is the implementation of merit.