1、Designation: E1488 09 E1488 12 An American National StandardStandard Guide forStatistical Procedures to Use in Developing and ApplyingTest Methods1This standard is issued under the fixed designation E1488; the number immediately following the designation indicates the year oforiginal adoption or, in
2、 the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. Asuperscript epsilon () indicates an editorial change since the last revision or reapproval.1. Scope1.1 This guide identifies statistical procedures for use in developing new test methods
3、 or revising or evaluating existing testmethods, or both.1.2 This guide also cites statistical procedures especially useful in the application of test methods.2. Referenced Documents2.1 ASTM Standards:2E177 Practice for Use of the Terms Precision and Bias in ASTM Test MethodsE178 Practice for Dealin
4、g With Outlying ObservationsE456 Terminology Relating to Quality and StatisticsE691 Practice for Conducting an Interlaboratory Study to Determine the Precision of a Test MethodE1169 Practice for Conducting Ruggedness TestsE1402 Guide for Sampling DesignE2282 Guide for Defining the Test Result of a T
5、est MethodE2489 Practice for Statistical Analysis of One-Sample and Two-Sample Interlaboratory Proficiency Testing ProgramsE2554 Practice for Estimating and Monitoring the Uncertainty of Test Results of a Test Method in a Single Laboratory Usinga Control Sample ProgramE2586 Practice for Calculating
6、and Using Basic StatisticsE2587 Practice for Use of Control Charts in Statistical Process ControlE2655 Guide for Reporting Uncertainty of Test Results and Use of the Term Measurement Uncertainty in ASTM Test Methods2.2 ISO Standards:ISO 17025 General Requirements for the Competence of Testing and Ca
7、libration Laboratories3ISO Guide to the Expression of Uncertainty in Measurement33. Terminology3.1 Definitions:DefinitionsFor a more extensive list of terms in E11 standards, see Terminology E456.3.1.1 bias, nthe difference between the expectation of the test results and an accepted reference value.
8、 E1773.1.1.1 DiscussionStatistical procedures include the sampling considerations or the experiment design for the collection of data, or both, and thenumerical and graphical approaches to summarize and analyze the collected data.3.1.2 coeffcient of variation, CV, nfor a nonnegative characteristic,
9、the ratio of the standard deviation to the mean for apopulation or sample. E25863.1.3 component of variance, na part of a total variance identified with a specified source of variability.1 This guide is under the jurisdiction of ASTM Committee E11 on Quality and Statistics and is the direct responsi
10、bility of Subcommittee E11.20 on Test Method Evaluationand Quality Control.Current edition approved May 15, 2009Aug. 1, 2012. Published July 2009 September 2012. Originally approved in 1992. Last previous edition approved in 20082009as E1488 08a.E1488 09. DOI: 10.1520/E1488-09.10.1520/E1488-12.2 For
11、 referenced ASTM standards, visit the ASTM website, www.astm.org, or contact ASTM Customer Service at serviceastm.org. For Annual Book of ASTM Standardsvolume information, refer to the standards Document Summary page on the ASTM website.This document is not an ASTM standard and is intended only to p
12、rovide the user of an ASTM standard an indication of what changes have been made to the previous version. Becauseit may not be technically possible to adequately depict all changes accurately, ASTM recommends that users consult prior editions as appropriate. In all cases only the current versionof t
13、he standard as published by ASTM is to be considered the official document.Copyright ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States13.1.4 control chart, nchart on which are plotted a statistical measure of a subgroup versus time of sampling al
14、ong with limitsbased on the statistical distribution of that measure so as to indicate how much common, or chance, cause variation is inherent inthe process or product. E25873.1.5 observation, nthe process of obtaining information regarding the presence or absence of an attribute of a test specimen,
15、or of making a reading on a characteristic or dimension of a test specimen. E22823.1.6 observed value, nthe value obtained by making an observation. E22823.1.7 precision, nthe closeness of agreement between independent test results obtained under stipulated conditions. E1773.1.8 proficiency testing,
16、 ndetermination of laboratory testing performance by means of interlaboratory comparisons. E24893.1.9 repeatability, nprecision under repeatability conditions. E1773.1.10 repeatability conditions, nconditions where independent test results are obtained with the same method on identical testitems in
17、the same laboratory by the same operator using the same equipment within short intervals of time. E1773.1.11 repeatability limit r, nthe value below which the absolute difference between two individual test results obtained underrepeatability conditions may be expected to occur with a probability of
18、 approximately 0.95 (95 %). E1773.1.12 repeatability standard deviation, sr, n the standard deviation of test results obtained under repeatability conditions.E1773.1.13 reproducibility, nprecision under reproducibility conditions. E1773.1.14 reproducibility conditions, nconditions where test results
19、 are obtained with the same method on identical test items indifferent laboratories with different operators using different equipment. E1773.1.15 reproducibility limit, R, nthe value below which the absolute difference between two test results obtained underreproducibility conditions may be expecte
20、d to occur with a probability of approximately 0.95 (95 %). E1773.1.16 reproducibility standard deviation Sdeviation, sR R, , nthe standard deviation of test results obtained underreproducibility conditions. E1773.1.17 ruggedness, ninsensitivity of a test method to departures from specified test or
21、environmental conditions. E11693.1.18 ruggedness test, na planned experiment in which environmental factors or test conditions are deliberately varied inorder to evaluate the effects of such variation. E11693.1.19 standard deviation, nof a population, , the square root of the average or expected val
22、ue of the squared deviation ofa variable from its mean of a sample x, the square root of the sum of the squared deviations of the observed values in the sampledivided by the sample size minus 1. E25863.1.20 state of statistical control, nprocess condition when only common causes are operating on the
23、 process. E25873.1.21 statistical procedures, nthe organized techniques and methods used to collect, analyze, and interpret data.3.1.21.1 DiscussionStatistical procedures include the sampling considerations or the experiment design for the collection of data, or both, and thenumerical and graphical
24、approaches to summarize and analyze the collected data.3.1.22 test determination, nthe value of a characteristic or dimension of a single test specimen derived from one or moreobserved values. E22823.1.23 test method, na definitive procedure that produces a test result. E22823.1.24 test observation,
25、 nsee observation. E22823.1.25 test result, nthe value of a characteristic obtained by carrying out a specified test method. E22823.1.25.1 DiscussionStatistical procedures include the sampling considerations or the experiment design for the collection of data, or both, and thenumerical and graphical
26、 approaches to summarize and analyze the collected data.3.2 For all other formal definitions of statistical terms, see Terminology E456.4. Significance and Use4.1 The creation of a standardized test method generally follows a series of steps from inception to approval and ongoing use.In all such sta
27、ges there are questions of how well the test method performs.4.1.1 Assessments of a new or existing test method generally involve statistical planning and analysis. This standardrecommends what approaches may be taken and indicates which standards may be used to perform such assessments.E1488 1224.2
28、 This standard introduces a series of phases which are recommended to be considered during the life cycle of a test methodas depicted in Fig. 1. These begin with a design phase where the standard is initially prepared. A development phase involves avariety of experiments that allow further refinemen
29、t and understanding of how the test method performs within a laboratory. Inan evaluation phase the test method is then examined by way of interlaboratory studies resulting in precision and biasstatistics which are published in the standard. Finally, the test method is subject to a monitoring phase.4
30、.3 All ASTM test methods are required to include statements on precision and bias.34.4 Since ASTM began to require all test methods to have precision and bias statements that are based on interlaboratory testmethods, there has been increased concern regarding what statistical experiments and procedu
31、res to use during the developmentof the test methods. Although there exists a wide range of statistical procedures, there is a small group of generally acceptedtechniques that are very beneficial to follow. This documentguide is designed to provide a brief overview of these procedures andto suggest
32、an appropriate sequence of carrying out these procedures.4.5 Statistical procedures often result in interpretations that are not absolutes. Sometimes the information obtained may beinadequate or incomplete, which may lead to additional questions and the need for further experimentation. Information
33、outsidethe data is also important in establishing standards and in the interpretation of numerical results.5. Summary of Guide5.1 Outlined below is a suggested sequence of four phases useful in the development of a test method. A flowchart is providedin Fig. 1. Such a sequence of analyses may need t
34、o be modified in specific situations. The assistance of a qualified statistician isrecommended at each review phase.5.2 Design Phase:5.2.1 This phase includes the formalization of the scope and the significance and use sections. It may include determining thepurpose and describing a general approach
35、 to the test method but usually does not involve statistical studies.3 See the Form and Style Manual for ASTM Standards that specifies, when possible, precision statements shall be estimated based on the results of an interlaboratory testprogram.FIG. 1 Sequence of StepsE1488 1235.2 Development Phase
36、Design Phase:5.2.1 This phase includes the formalization of the scope and the significance and use sections. It may include determining thepurpose and describing a general approach to the test method but usually does not involve statistical studies.5.2.2 Studies may be conducted to evaluate the basi
37、c performance of the method. The draft test method is prepared and samplingrequirements and the test result (see Guide E2282) are clearly defined.5.2.3 A flow chart is extremely valuable to identify the sequence of operations involved in a test method, for example, thesampling steps required to obta
38、in the test specimens, definition of the test determination, how a test result is to be computed, andrunning the tests on the specimens.5.3 Validation PhaseDevelopment Phase:5.3.1 The test method is examined for such concerns as its stability, ruggedness, statistical control and the contributions to
39、variability. The completion of this phase should result in preliminary estimates of precision and the identification and suggestedways to estimate potential contributors to uncertainty.5.3.2 Evaluation of Short Term Control of Test MethodA test method must exhibit an ability to provide consistent re
40、sults atleast over short time periods. Preliminary studies or a pilot test should be conducted to evaluate the short term stability of the testmethod. A small series of repeated tests should be conducted.5.3.3 Analysis of VariabilityStatistically designed experiments conducted in one or two laborato
41、ries can be used to assess therelative magnitudes of different sources or potential contributors to variability of the test results. Such studies can provide estimatesof intermediate measures of precision.5.3.4 Ruggedness TestA ruggedness test (see Practice E1169) is a statistically designed experim
42、ent that helps identifyproblems in running the test method, clarifies errors, and points out possible environmental conditions, which may adversely affectthe test method or point out need for tightening requirements. The ruggedness test can assist in locating ways of reducingvariability in the test
43、method.5.3.5 Preliminary Estimates of PrecisionFrom the various studies conducted in accordance with 5.4.25.3.2-5.3.45.4.4,preliminary estimates of repeatability standard deviations should be developed and published in this test method. Until aninterlaboratory study is performed, these estimates gen
44、erally are considered to be provisional. Information on how a lab shoulddevelop uncertainty estimates should also be provided.5.3.6 Statistical ControlA test method must show capability of performing in a consistent way over time. The use of controlcharts (see ManualGuide E26557)3 to monitor a propo
45、sed, or existing, test method over time is one recommended way to examinethe controllability or stability of a test method. This statistical control should be demonstrated in one or two laboratories usinghomogeneous material (test specimen).5.4 Evaluation Phase:5.4.1 The test method is subjected to
46、interlaboratory studies to provide estimates of within-laboratory repeatability andbetween-laboratory reproducibility. Additional information is supplied from proficiency studies when conducted.5.4.2 Interlaboratory Study (ILS)In accordance with ASTM Form and Style Manual, whenever feasible, an inte
47、rlaboratorystudy must be conducted. This procedure will provide specific estimates of variation anticipated when using the test method.5.4.3 Protocol for the ILS, Practice E691, provides a guide for developing the ILS for the test method. A first step is the writingof an ILS Protocol, which will set
48、 out what needs to be done before the test specimens (or test materials) are distributed to theparticipating laboratories.5.4.4 Precision StatementsUsing the estimates of variation obtained in the interlaboratory test, one may prepare precisionstatements using Practices E691 and E177 or equivalent p
49、rocedures.5.5 Monitoring Phase:5.5.1 After a test method is approved and in use it is important to ensure that the published precision and bias statistics for thetest method remain achievable and consistent over time or amongst different groups conducting the tests.5.5.2 Monitoring Within a Single LocationIt is important for any laboratory or organization that will use a particular testmethod over time that a means of monitoring to ensure the method results using quality control samples are stable and in control.Regular evaluation o
