Our validity and reliability studies are not a one-time exercise; they are embedded in an ongoing cycle of refinement and continuous improvement across our entire assessment portfolio.
Across our library of more than 450+ assessments, we apply consistent psychometric quality benchmarks. These benchmarks represent our minimum target standards, and when a test does not meet them, it is not considered final. Instead, it enters a structured revision process until it consistently reaches or exceeds our thresholds.
Our Psychometric Quality Targets
We hold every assessment to the following evidence-based standards:
These thresholds reflect widely accepted best practices in educational and psychological measurement. They guide our development decisions and ensure consistent quality across industries, roles, and use cases.
What These Standards Mean in Practice
We treat psychometric quality as a living process, not a fixed checkpoint. For every assessment:
We analyze item-level and test-level performance.
We monitor reliability and validity metrics across diverse samples.
We systematically flag underperforming items.
We revise, replace, or remove items that do not meet standards.
We re-test and re-validate until the assessment consistently meets our benchmarks.
This process applies across our entire assessment ecosystem, from technical skills tests to behavioral and situational judgment assessments.