Accuracy vs Precision: Why Laboratory Results Require Both

In analytical testing, data integrity underpins every technical, commercial, and regulatory decision.
Two concepts are often used interchangeably —
Yet they represent fundamentally different dimensions of data quality:
Accuracy and precision.
Definitions
• Accuracy — proximity of a result to the accepted reference value
• Precision — agreement between repeated measurements under defined conditions
In simple terms:
Accuracy is hitting the target.
Precision is consistency in where the shots land.
Operational Interpretation
1️⃣ Precise but inaccurate
Consistent results with systematic bias
→ Often linked to calibration drift or method bias
2️⃣ Accurate but imprecise
Correct on average, but highly variable
→ Indicates poor repeatability or uncontrolled conditions
3️⃣ Accurate and precise
Correct and reproducible
→ The requirement under ISO/IEC 17025
Why This Matters
In high-stakes industries such as mining, small analytical deviations translate into real consequences:
• Metal accounting — minor grade bias can result in significant financial exposure
• Regulatory compliance — results must include stated measurement uncertainty
• Contractual integrity — decisions rely on certified, defensible data
A result without context of uncertainty is incomplete.
Control Mechanisms in High-Performing Laboratories
✔ Accuracy assurance
Routine use of Certified Reference Materials traceable to recognised standards
✔ Precision monitoring
Systematic inclusion of duplicates, replicates, and control samples
✔ Process control
Defined SOPs, instrument calibration, and maintenance systems
✔ Transparent reporting
Clear expression of measurement uncertainty
e.g. 2.15% Cu ±0.05% (k=2, 95% confidence level)
Two Questions That Define Data Quality
• How close are results to the reference value? → Accuracy
• How consistent are repeated measurements? → Precision
Both parameters are non-negotiable.
Analytical data is not judged by approximation —
but by its ability to withstand technical scrutiny.
In laboratory science, reliability is not assumed.
It is demonstrated.