STATISTICS Laboratory question


Does the method perform locally as expected?


What is the method’s precision in our laboratory?


Why are results more variable than expected?


Is precision acceptable near a decision limit?


Likely study type Precision verification


Precision evaluation Troubleshooting study


Concentration-focused precision study


Table 2. Common reasons for performing a precision study.


mean square contains both true between- group variation and the within-group scater that is already present in each group mean. For a balanced design with the same number of replicate measurements in each group, the calculation can be expressed conceptually as follows: Repeatability variance = MSwithin Between-group variance = (MSbetween − MSwithin) / n Within-laboratory variance = repeatability variance + between-group variance SD = square root of variance CV (%) = SD / mean × 100


Here, n is the number of replicate measurements in each group. If the calculated between-group variance is negative, it is usually treated as zero because a negative variance has no practical meaning. This can occur when the observed between-group differences


A precision study should start with the laboratory question, not with the spreadsheet or calculation. The design should define what will be repeated and what will be allowed to vary. A single-run replicate study estimates repeatability


Typical design feature Several days and replicate


measurements


Broader design including routine sources of variation


Deliberate inclusion of


suspected sources of variation Materials selected near clinically


relevant concentrations Main output Local repeatability and


within-laboratory estimates Variance components and


concentration-specific SD/CV Identification of where variation is arising Concentration specific SD and CV


are no larger than expected from replicate scater alone. The important statistical point is


that variances add; CVs do not. Within- laboratory imprecision is obtained by combining variance components, taking the square root to return to an SD, and then expressing that SD relative to the mean if a CV is required.


Repeatability variance Between-group variance


69% 76%


Designing the study: what is allowed to vary? A precision study should start with the laboratory question, not with the spreadsheet or calculation. The design should define what will be repeated and what will be allowed to vary. A single-run replicate study estimates repeatability. A five-day design estimates a short-term form of within-laboratory imprecision. A fuller evaluation may deliberately include different runs, operators, reagent lots, analysers or sites. The choice of material also maters.


31% Level 1 Material


Fig 2. Contribution of repeatability and between-group variance to within-laboratory variance in the worked example. The ploted percentages are based on variance components, not CVs.


18 WWW.PATHOLOGYINPRACTICE.COM August 2026 24% Level 2


Commercial control material is often convenient because it is stable and available in sufficient volume, but it may not always behave like patient material. Pooled patient material may be more clinically realistic, but can introduce issues of volume, stability, storage and governance. The material used should therefore be documented as part of the study, not treated as a minor technical detail. Concentration level is equally


important. Precision should be assessed at more than one level, particularly where performance is clinically important at low concentrations, abnormal concentrations or near a decision limit. The result at one concentration should not be assumed to


Contribution to within-laboratory variance (%)


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56