ISO 5725-4-2020 PDF

Full title and description

Accuracy (trueness and precision) of measurement methods and results — Part 4: Basic methods for the determination of the trueness of a standard measurement method (ISO 5725-4:2020). This part of ISO 5725 specifies practical, basic methods for estimating the bias (trueness) of a measurement method and the laboratory bias when a standardized measurement method is applied, and gives guidance for design, performance and analysis of bias-estimation experiments.

Abstract

ISO 5725-4:2020 provides basic, routine methods to estimate the bias of a measurement method and laboratory bias when measurements are made according to a standardized method. It is restricted to measurement methods that produce a single numerical result on a continuous scale and applies when an accepted reference value (conventional true value) can be established (for example by use of a suitable reference material, measurement standard, reference method or prepared known sample). The document gives practical procedures, brief guidance for personnel designing or analysing bias studies, and notes limitations (it does not address cases where bias depends on multiple interacting properties; comparison of two methods is covered in another part of the series).

General information

• Status: Published (International Standard)
• Publication date: March 2020 (published 26 March 2020)
• Publisher: International Organization for Standardization (ISO)
• ICS / categories: 17.020 (Metrology and measurement in general); 03.120.30 (Application of statistical methods)
• Edition / version: Edition 2 (2020)
• Number of pages: 26

Scope

ISO 5725-4:2020 applies to standardized measurement methods that yield a single continuous numeric result and where a conventional accepted reference value is available to substitute the true value. It specifies basic experimental and analytical procedures to estimate the systematic error (bias) of a measurement method and of individual laboratories when the method is applied. The standard is intended for cases where bias can be estimated for one property at a time and does not treat situations in which bias depends on levels of other influencing quantities. It assumes measurements are made in accordance with the standardized method and is complementary to other parts of the ISO 5725 series that address precision, alternative designs and use of accuracy values in practice.

Key topics and requirements

• Definition and scope of trueness (bias) versus precision in measurement accuracy.
• Requirements for an acceptable reference value (reference material, reference method, prepared known sample or measurement standard) to serve as a substitute for the true value.
• Design of basic experiments for bias estimation (selection of test items, number of replicates, and laboratory considerations).
• Calculation methods for estimating bias and laboratory bias, including recommended statistical formulae and handling of measurement data.
• Guidance on routine, practical approaches suitable for day-to-day use in laboratories.
• Limitations and applicability: continuous-scale single-value results only; not applicable when bias is affected by other properties or interferences.
• References to related vocabulary and statistical standards (for example ISO/IEC Guide 99 — VIM and ISO 3534 series).

Typical use and users

ISO 5725-4 is used by testing and calibration laboratories, standards developers, method validation teams, accreditation bodies, metrologists and statisticians involved in measurement quality assessment. Typical uses include validation of a standardized method’s trueness, routine checks of laboratory bias against a reference value, preparation of method validation reports, and supporting accreditation evidence for measurement performance.

Related standards

ISO 5725-4 is part of the ISO 5725 series on measurement accuracy. Closely related parts and documents include ISO 5725-1 (general principles and definitions), ISO 5725-2 (basic method for determination of repeatability and reproducibility), ISO 5725-3 (intermediate precision and alternative collaborative designs), ISO 5725-5 (alternative methods for precision), and the parts addressing use in practice and comparisons between methods (other parts of the series). Normative references also include ISO/IEC Guide 99 (VIM) and the ISO 3534 statistics vocabulary.

Keywords

trueness, bias, laboratory bias, accuracy, precision, repeatability, reproducibility, measurement method, reference value, reference material, method validation, ISO 5725, metrology, statistical procedures

FAQ

Q: What is this standard?

A: ISO 5725-4:2020 is the part of the ISO 5725 series that specifies basic methods for determining the trueness (bias) of a standardized measurement method and for estimating laboratory bias.

Q: What does it cover?

A: It covers experimental design and statistical procedures to estimate bias when a suitable accepted reference value is available. It applies to measurement methods giving a single numeric result on a continuous scale and provides practical guidance for routine bias estimation.

Q: Who typically uses it?

A: Testing and calibration laboratories, method developers, accreditation assessors, metrologists and statisticians use this standard to validate and document the trueness of measurement methods and to monitor laboratory bias.

Q: Is it current or superseded?

A: ISO 5725-4:2020 is the current edition (Edition 2, published March 2020) and replaced earlier editions (for example the 1994 edition).

Q: Is it part of a series?

A: Yes — it is part of the ISO 5725 series (Accuracy — trueness and precision) and is intended to be used together with other parts of the series (e.g., ISO 5725-1, -2, -3, -5 and the parts addressing use in practice and method comparisons).

Q: What are the key keywords?

A: Key keywords are bias, trueness, laboratory bias, accuracy, precision, measurement method, reference material, repeatability, reproducibility, and ISO 5725.