In proficiency testing (PT), a provider prepares a homogeneous material, often hemp or cannabis flower, an oil, or a spiked matrix, splits it and sends identical blind portions to participating labs. Each lab analyses it with its routine method. The provider compares the results with an assigned value and scores each lab, commonly with a z-score.
In practice
A z-score between -2 and +2 is usually satisfactory; beyond ±3 is unsatisfactory and requires corrective action. ISO/IEC 17025 labs are expected to take part regularly. PT is one of the best tools against lab-to-lab bias: if one lab reads consistently high in PT rounds, that shows up. Some regulators publish PT outcomes or use them in audits. For consumers, asking a lab about its PT record is reasonable.
Units and formula
z = (lab result - assigned value) / standard deviation for proficiency assessment. |z| ≤ 2 satisfactory, 2 < |z| < 3 questionable, |z| ≥ 3 unsatisfactory.
Don't confuse with
- Accreditation: the overall assessment; PT is one input.
- Split sample retest: an informal check between two labs, useful but not scored.
