GB/T 6379.3-2012Accuracy (trueness and precision) of measurement methods and results -- Part 3: Intermediate measures of the precision of a standard measurement method (English PDF)

Foreword

GB/T 6379 "measurement methods and results of Accuracy (trueness and precision)" is divided into the following sections, which correspond to the expected structure and International standards.

--- Part 1. General principles and definitions (ISO 5725-1:1994, IDT)

--- Part 2. Determine the standard methods of measurement repeatability and reproducibility of the basic method (ISO 5725-2:1994, IDT)

--- Part 3. Standard Methods of measurement precision intermediate measurement (ISO 5725-3:1994, IDT)

--- Part 4. method of determining basic accuracy of a standard measurement method (ISO 5725-4:1994, IDT) Alternative method to determine a standard measurement method precision. --- Part 5 (ISO 5725-5:1998, IDT)

--- Part 6. practice of accuracy values \u200b\u200b(ISO 5725-6:1994, IDT) this is Part 3 GB/T 6379's.

This section drafted in accordance with GB/T 1.1-2009 given rules.

This section identical with the international standard ISO 5725-3. "measurement methods and results of Accuracy (trueness and precision) of Part 31994 1994 edition of ISO 5725-3 technology to modify the standard methods of measurement precision intermediate measure "and in ISO 2001-10-15 released. minutes single. To ISO 5725-3:1994 error made the following changes and corrections.

--- Revised 9.4 on the design based on a complete set of test standard deviation of repeatability, reproducibility and intermediate precision standard deviation standard deviation Estimates are not accurate account.

--- The Table D.5 in Appendix D sI (T) value of the first 6 levels by 9.545 × 10-3 corrected to 8.020 × 10-3.

GB/T 6379 Part 1 to Part 6 as a whole instead of GB/T 6379-1986 and GB/T 11792-1989. Mark Registration precision of the original concept to be expanded, increasing the accuracy of the concept, referred to as accuracy; except repeatability conditions and reproducibility conditions, the increase the intermediate precision conditions.

This part of the application by the National Standardization Technical Committee on statistical methods (SAC/TC21) and focal points.

This section drafted by. China National Institute of Standardization, Academy of Mathematics and Systems Science, Shenzhen City, measuring Detection Co.

Secretary of Hainan Province Product Quality Supervision and Inspection, Wuxi City Product Quality Supervision, Inspection, Guangzhou Exit Inspection and Quarantine.

Introduction

0.1 GB/T 6379 with two terms "trueness" and "precision" to describe the accuracy of measuring methods. Accuracy refers to the large number of tests (Arithmetic) average of the results and the true value or accept the degree of consistency between the reference value; while precision refers to the degree of consistency between test results.

0.2 GB/T 6379.1 various amounts given in the above general considerations in this section will not be repeated. It must be emphasized that, GB/T 6379.1 should be GB/T 6379 all other parts (including this section) combine to read, because the GB/T 6379.1 gives the group the definitions and general provisions.

0.3 Many different factors (in addition to the difference between the assumed identical samples outside) can lead to variation in measurement results, these factors include.

a) the operator;

b) the use of the device;

c) calibration equipment;

d) the environment (temperature, humidity, air pollution, etc.);

e) the batch of reagents;

Interval f) different measurements.

Often much in a short time by the same operator using a phase measurements made by the different operators and measurements carried out on different devices with equipment to measure produced great variation.

0.4 precision two conditions, namely repeatability conditions and reproducibility conditions are necessary, and in many practical situations, to describe the measurement side Variation law useful. Under repeatability conditions, the factors listed in 0.3 a) ~ f) are kept constant, no variation in reproducibility Under conditions which are changes that can cause variation of the test results. Therefore, the repeatability and reproducibility condition precision the two extreme cases. The former describes the smallest variation in the test results, which describe the maximum variation in test results. In both extreme conditions of One or more intermediate between the conditions also exist, i.e., factor a) ~ f) being changed, they can be used in certain circumstances.

Precision is usually standard deviation.

0.5 intermediate section discusses a measure of precision measurement. Since the values \u200b\u200bof these metrics in the two kinds of measurement electrode Between the end of metric precision (repeatability standard deviation and reproducibility standard deviation), said intermediate precision.

To the necessity of these intermediate precision measure of illustration, consider the current and a workshop related to the operation of the laboratory, in the real Mortem indoor three shifts working system, measured by different operators performed on different devices. Thus operators and equipment that can cause the test results of some of the factors of variation. In assessing the precision of the method must be considered.

Intermediate precision measures defined in section 0.6 first applied to the following scenario. In order to measure the internal laboratory methods to improve, standard Registration and control, the need for intermediate precision estimates; between a specially designed laboratory studies also estimate the need for intermediate precision meter. However, due to the 1.3 and 9.1 of the reasons, the interpretation and application of these measures should be cautious.

0.7 four factors most likely to influence the measurement precision of the method are.

a) Time. Continuity measurement time interval is large or small.

b) Calibration. between successive sets of measuring whether the same device after recalibration.

c) Operator. continuous measurement is completed by the same operator.

d) Equipment. whether to use the same device in the measurement (or the same group of agents).

0.8 below, before the introduction of the M factor different intermediate precision condition (M = 1,2,3, or 4), to take account of the laboratory measurement conditions (when Room, calibration, operator and equipment) change.

a) M = 1. only one of four different factors;

b) M = 2. There are two different four factors;

c) M = 3. three different four factors;

d) M = 4. all four factors are different.

Different intermediate precision conditions produce different intermediate precision standard deviation, denoted by sI (·), corresponding to the specific conditions in parentheses Clearly marked, for example, sI (TO) represents a different time, a different operator intermediate precision standard deviation.

0.9 For measurements under intermediate precision conditions, 0.7 listed factors have one or more different under repeatability conditions, those who Su is assumed to be constant.

The standard deviation of test results obtained under repeatability conditions, generally less than the difference between the standard test results obtained under intermediate precision conditions.

Under normal circumstances, in chemical analysis, standard deviation under intermediate precision conditions will be standard deviation under repeatability conditions 2 to 3 times. Of course, it it should not be greater than the reproducibility standard deviation.

For example, in a total of 35 laboratories to determine the copper ore copper content involved in collaborative trials, found that both the use of gravity electrolysis Assay or by Na2S2O3 titration method, a factor (different time but the same operator and equipment) of intermediate precision conditions Under the resulting standard deviation under repeatability conditions than the standard differential is 1.5 times greater.

Measurement methods and results of Accuracy (trueness and precision) Part 3. Methods of measurement precision intermediate standard measure

1 Scope

1.1 Since the middle section defines four laboratory observation conditions (time, calibration, operator and equipment) changes arising precision measure. These measures may be in the middle of a laboratory test to determine in which can also be generated through inter-laboratory tests. In addition, GB/T 6379 this part.

a) discussion of intermediate precision measures defined meanings;

b) provide guidance for the practical work on the intermediate precision measure to estimate the interpretation and application;

c) does not provide any measure of an intermediate estimate of precision measurement error;

d) does not involve how to determine the correct measurement method itself, but discusses the relationship between trueness and measurement conditions between.

1.2 section applies to measurement methods involved especially continuous volume measurement, and only takes a value for each measurement as a measurement result as a result, although this value may be the results of a set of observations.

1.3 determine the nature of these intermediate precision metrics that quantify the capability measurement method under specified conditions, repeat the test results.

Statistical methods described in Section 1.4 based on the following premise. can be combined "similar" measurement conditions in the information to obtain an intermediate finish Density measurement more accurate information. As long as the alleged "similar" is indeed "similar", this premise that is effective. But through inter-laboratory research when the study to estimate the intermediate precision measure, this premise is difficult to be met. For example, to make a joint information from different laboratories makes sense, need by controlling all participating test lab "time" the impact (effect) or "operator" influence (effect), so they are "similar", it is very difficult. Therefore, when the intermediate precision results obtained in the Institute laboratory to be careful. Laboratory studies also dependent on the above-mentioned Premise, but this time because the actual impact analysis by a factor of more understanding, but also know how to control it, so this premise Easier to implement.

1.5 In addition to techniques described in this section, there are other estimates and confirmed the middle of a laboratory precision measurement technology, for example, control Drawing (see GB/T 6379.6). No statement in this section is provided in a particular laboratory estimate of intermediate precision measurement the only way.

Note. This section relates to experimental design, such as set design knowledge. Appendix B and Appendix C gives the relevant basics.

2 Normative references

The following documents for the application of this document is essential. For dated references, only the dated version suitable for use herein Member. For undated references, the latest edition (including any amendments) applies to this document.

GB/T 3358.1-2009 Statistics Vocabulary and symbols - Part 1. General statistical terms and terms used in probability (ISO 3534-1:2006, IDT) GB/T 6379.1-2004 measurement methods and results of Accuracy (trueness and precision) - Part 1. General principles and definitions (ISO 5725-1:1994, IDT) (Trueness and precision) - Part 2 Accuracy GB/T 6379.2-2004 measurement methods and results. the standard measurement to determine the party Method repeatability and reproducibility of the basic method (ISO 5725-2:1994, IDT) ISO 3534-1:1993 Statistics - Vocabulary and symbols - Part 1. Probability and general statistical terms ISO Guide 33. 1989 certified reference material use ISO Guide 35. General and statistical principles setting 1989 standard materials