1、Designation: E 1601 98 (Reapproved 2003)e1Standard Practice forConducting an Interlaboratory Study to Evaluate thePerformance of an Analytical Method1This standard is issued under the fixed designation E 1601; the number immediately following the designation indicates the year oforiginal adoption or
2、, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. Asuperscript epsilon (e) indicates an editorial change since the last revision or reapproval.e1NOTECaution notes were moved into the text editorially in November 2003.1. Scope1.1 This
3、 practice covers procedures and statistics for aninterlaboratory study (ILS) of the performance of an analyticalmethod. The study provides statistical values which are usefulin determining if a method is satisfactory for the purposes forwhich it was developed. These statistical values may beincorpor
4、ated in the methods precision and bias section. Thispractice discusses the meaning of the statistics and what usersof analytical methods may learn from them.1.2 This standard does not purport to address all of thesafety concerns, if any, associated with its use. It is theresponsibility of the user o
5、f this standard to establish appro-priate safety and health practices and determine the applica-bility of regulatory limitations prior to use.2. Referenced Documents2.1 ASTM Standards:2E 135 Terminology Relating to Analytical Chemistry forMetals, Ores, and Related MaterialsE 691 Practice for Conduct
6、ing an Interlaboratory Study toDetermine the Precision of a Test MethodE 1169 Guide for Conducting Ruggedness TestsE 1763 Guide for Interpretation and Use of Results fromInterlaboratory Testing of Chemical Analysis Methods3. Terminology3.1 DefinitionsFor definitions of terms used in this prac-tice,
7、refer to Terminology E 135.3.2 Definitions of Terms Specific to This Standard:3.2.1 interlaboratory study (ILS)study undertaken to as-certain if a test method is suitable for its intended use. The ILSincludes preparation, testing, and evaluation phases.3.2.2 interlaboratory testmeasures the variabil
8、ity of re-sults when a test method is applied many times in a number oflaboratories.3.2.3 replicate resultsresults obtained by applying a testmethod a specified number of times to a material.3.2.4 resultthe numerical value obtained by applying atest method once to a material.3.2.5 test methodgives d
9、irections for producing a singleresult.3.2.6 test protocolgives instructions to each participatinglaboratory, detailing the way it is to conduct its part of theinterlaboratory test program.3.3 Unless the test method destroys the test portion eachtime it is applied, the protocol for a Plan A test spe
10、cifies, ifpossible, replicate results on a single test portion (which maybe in solution). The protocol for a Plan B test specifies thenumber of test portions of a material and requires duplicateresults (2 only) on each portion (which may be in solution).4. Summary of Practice4.1 Instructions are pro
11、vided for planning and conducting acooperative evaluation of a proposed analytical method.4.2 The following list describes the organization of thispractice:4.2.1 Sections 1-5 define the scope, significance and use,referenced documents, and terms used in this practice.4.2.2 Section 6 helps users of a
12、nalytical methods understandand use the statistics found in the Precision and Bias section ofmethods.4.2.3 Sections 7 and 8 instruct the ILS coordinator andmembers of the task group on how to plan and conduct theexperimental phase of the study.4.2.4 Section 9 discusses the procedures for collecting,
13、evaluating, and disseminating the data from the interlaboratorytest.4.2.5 Section 10 presents the statistical calculations.1This practice is under the jurisdiction of ASTM Committee E01 on AnalyticalChemistry for Metals, Ores, and Related Materials and is the direct responsibility ofSubcommittee E01
14、.22 on Statistics and Quality Control.Current edition approved Oct. 1, 2003. Published November 2003. Originallyapproved in 1994. Last previous edition approved in 1998 as E 1601 98.2For referenced ASTM standards, visit the ASTM website, www.astm.org, orcontact ASTM Customer Service at serviceastm.o
15、rg. For Annual Book of ASTMStandards volume information, refer to the standards Document Summary page onthe ASTM website.1Copyright ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.4.2.6 Sections 11 and 12 discuss the use of statistics toevaluat
16、e a test method and the means of incorporating the ILSstatistics into Precision and Bias statements.4.2.7 The Annex A1 gives the rationale for the calculationsin Section 10.5. Significance and Use5.1 Ideally, interlaboratory testing of a method is carried outby a randomly chosen group of laboratorie
17、s that typifies thekind of laboratory that is likely to use the method. In actuality,this ideal is only approximated by the laboratories that areavailable and willing to undertake the test work. The coordi-nator of the program must ensure that every participatinglaboratory has appropriate facilities
18、 and personnel and per-forms the method exactly as written. If this goal is achieved,the statistics developed during the ILS will be adequate fordetermining if the method is capable of producing satisfactoryprecision in actual use. If the program includes certifiedreference materials, the test data
19、also provide informationconcerning the accuracy of the method. The statistics providea general guide to the expected performance of the method inthe laboratories of those who will use it.6. Statistical Guide for the Users of Analytical MethodsEvaluated in Accordance With This Practice6.1 Standard De
20、viations:6.1.1 Minimum Standard Deviation of Method, sMThisstatistic measures the precision of test results under conditionsof minimum variability. Because it is improbable that a methodin ordinary use will exhibit precision this good, no predictiveindex is calculated for sM. Users adept in statisti
21、cs may wish tocompare sMand the short-term standard deviation of themethod measured in their laboratory. For most methods,short-term variability refers to results obtained within severalminutes. (CautionThe standard deviation of results obtainedon different occasions, even in the same laboratory, pr
22、obablywill exceed sM.)6.1.2 Between-Laboratory Standard Deviation, sRThisstatistic is a measure of the precision expected for resultsobtained in different laboratories. It reflects all sources ofvariability that operate during the interlaboratory test (excepttest material inhomogeneity in tests desi
23、gned to eliminate thateffect). It is used to calculate the reproducibility index, R. UsesRfor evaluating the precision of methods. It represents theexpected variability of results when a method is used indifferent laboratories.6.1.3 Within-Laboratory Standard Deviation, srThis sta-tistic cannot be c
24、alculated in a normal interlaboratory test. It isdetermined only in tests designed to measure variability withinlaboratories. When this statistic is given in a method, it reflectsall variability that may occur from day-to-day within a labo-ratory (for example, from calibration, standardization, or e
25、nvi-ronmental changes). It is used to calculate the repeatabilityindex, r. The user is cautioned that additional sources ofvariation may affect results obtained in other laboratories.6.2 Predictive IndexesFor the following indexes to apply,these conditions must be met: (1) the test materials must be
26、homogeneous; (2) analysts must be competent and diligent; (3)analytical instruments and equipment must be in good condi-tion; and (4) the method must be performed exactly as written.6.2.1 Reproducibility Index, RThis statistic estimates theexpected range of differences in results reported from twola
27、boratories, a range that is not exceeded in more than 5 % ofsuch comparisons. Use R to predict how well your resultsshould agree with those from another laboratory: First, obtaina result under the conditions stated in 6.2, then add R to, andsubtract R from, this result to form a concentration confid
28、enceinterval. Such an interval has a 95 % probability of including aresult obtainable by the method should another laboratoryanalyze the same sample. For example, a result of 46.57 % wasobtained. If R for the method at about 45 % is 0.543, the 95 %confidence interval for the result (that is, one exp
29、ected toinclude the result obtained in another laboratory 19 times out of20) extends from 46.03 to 47.11 %.NOTE 1For those not conversant with statistical concepts, it isimportant to realize that in most such comparisons, the differences will bemuch smaller than the confidence interval implies. The
30、50 % confidenceinterval is only about one third (34.6 %) as wide. Thus, the “average”interval for the above result (one expected to include the result obtainedby another laboratory half the time) extends from 46.4 to 46.8 %. Theobvious implication is that, although half the differences will be more
31、than0.2 %, half will be less than 0.2 %.6.2.2 Repeatability Index, rThis statistic is given in themethod only if the interlaboratory test was designed to measuresr. It estimates the expected range of results reported in thesame laboratory on different days, a range that is not exceededin more than 5
32、 % of such comparisons.7. Interlaboratory Test Planning7.1 Analytical test methods start from a perceived need toservice one or more material specifications.7.1.1 Develop a performance requirement for a methodfrom the material specification(s). Include the following fac-tors: expected ranges of chem
33、ical compositions of the materialsto be covered (methods general scope); specified elements andtheir concentrations (determination concentration ranges); andthe precision required.7.1.2 Prepare a table of the elements and concentrationranges to cover the critical values in the material specification
34、s.Use this information together with knowledge of the charac-teristics of the candidate analytical method to select testmaterials for the interlaboratory program.7.2 Draft MethodThe process of developing methods andtesting them in a preliminary way is beyond the scope of thispractice. All analytical
35、 skill and experience available to thetask group must be exerted to ensure that the method will meetthe project requirements in 7.1 and that it is free of technicalfaults. A preliminary, informal test of a method must be carriedout in several laboratories before the final draft is prepared.Individua
36、ls responsible for selecting the method may findhelpful information in Practice E 691 and Guide E 1169. Theformal interlaboratory test must not start until the task groupreaches consensus on a clearly written, explicitly stated, andunambiguously worded draft of the method in ASTM format,which has co
37、mpleted editorial review.7.3 Test MaterialsAppropriate test materials are essentialfor a successful ILS. The larger the number of test materialsE 1601 98 (2003)e12included in the test program, the better the statistical informa-tion generated. On the other hand, the burden of running a verylarge num
38、ber of materials may reduce the number of labora-tories willing to participate. A method must cover a concen-tration range extending both above and below the specifiedvalue(s). If possible, provide test materials near each limit.Concentration ranges covering several orders of magnitudeshould be test
39、ed with three or more materials.7.3.1 Material composition and form must be within thegeneral scope of the method. If possible, include all materialtypes the scope is expected to cover. Often, only limitednumbers of certified reference materials are available. Usethose that best meet the criteria fo
40、r the test. If they do not coverall concentration levels, find or prepare other materials to fill inmissing values.7.3.2 The quantity of the material must be sufficient todistribute to all laboratories participating in the test with about50 % held in reserve to cover unforeseen eventualities.7.3.3 M
41、aterials should be homogeneous on the scale of thetest portion consumed in each determination as well as amongthe portions sent to different laboratories. Usually certifiedreference materials have been tested for homogeneity, but testmaterials from other sources may have had only a minimalexaminatio
42、n. The use of laboratory-scale melting and castingto produce test materials can sometimes lead to segregation ofone or more components in an alloy. Unless specially gatheredor prepared materials have been subjected to a thoroughhomogeneity test, they require the use of Test Plan B. Itstatistically r
43、emoves the effect of moderate test materialinhomogeneity from the estimates of the ILS statistics.7.3.4 Test material sent to each laboratory must be perma-nently marked with its identity in such a manner that theidentification is not likely to be lost or obliterated.7.3.5 If the test program is to
44、evaluate the accuracy of themethod, at least one test material must be certified for theconcentration of each element. More certified materials pro-vide more complete information on accuracy.7.3.6 Prepare a list of the test materials, their identifyingnumbers, a brief description of material type (f
45、or example,low-carbon steel), and approximate concentration of the ele-ments to be determined. This table becomes part of thedocumentation sent to participating laboratories and providesinformation needed for the research report and the precisionand bias statement.7.4 Number of Cooperating Laborator
46、iesConventionalwisdom holds that the more laboratories participating in anILS, the better. Further, the laboratory types included in thestudy task group should consist of typical users laboratories.There is wide agreement that estimates of precision based uponfewer than six laboratories become incre
47、asingly unreliable asthe number decreases. A test program involving fewer than sixlaboratories does not comply with the requirements of thispractice (Note 2). An effort should be made to enlist at leastseven qualified laboratories before beginning a test program, toallow for attrition. To be qualifi
48、ed to participate, a laboratorymust have proper equipment and personnel with sufficienttraining and experience to enable them to perform the methodexactly as it is written.NOTE 2If all reasonable effort fails to recruit at least six cooperatinglaboratories, up to two of the recruited laboratories ma
49、y each volunteer tosubmit two independent sets of test data as an expedient to provide a totalof at least six sets of data. Minimum requirements for independence arethat two typical analysts, who do not consult with each other about themethod, perform the test protocol on different days. They should useseparate equipment if possible and must not share calibration solutions orcalibration curves.8. Conducting the Interlaboratory Test8.1 Program CoordinatorOne individual (presumably thetask group chairman) will coordinate the entire ILS. A pro-spective ILS progra
