1、Designation: E2282 14 An American National StandardStandard Guide forDefining the Test Result of a Test Method1This standard is issued under the fixed designation E2282; the number immediately following the designation indicates the year oforiginal adoption or, in the case of revision, the year of l
2、ast 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 The purpose of this guide is to provide guidelines foridentifying the elements that comprise the test result of a testm
3、ethod and to illustrate how these elements combine into thetest result.1.2 Types of measurement scales used for expressing obser-vations and test results are discussed.1.3 No system of units is specified in this standard.2. Referenced Documents2.1 ASTM Standards:2E177 Practice for Use of the Terms P
4、recision and Bias inASTM Test MethodsE456 Terminology Relating to Quality and Statistics2.2 ISO Standard:3ISO 35342 StatisticsVocabulary and Symbols, Part 2:Applied Statistics3. Terminology3.1 DefinitionsFor a more extensive list of terms in E11standards, refer to Terminology E456.3.1.1 binary scale
5、, nnominal scale with only two possiblecategories.3.1.2 characteristic, na property of items in a sample orpopulation which, when measured, counted or otherwiseobserved, helps to distinguish between the items.3.1.3 interval scale, ncontinuous scale or discrete scalewith equal sized scale values and
6、an arbitrary zero.ISO 353423.1.4 nominal scale, nscale with unordered labeled cat-egories or ordered by convention. ISO 353423.1.5 observation, nthe process of obtaining informationregarding the presence or absence of an attribute of a testspecimen, or of making a reading on a characteristic ordimen
7、sion of a test specimen.3.1.5.1 DiscussionObservation is also associated with theattribute or measurement information obtained from the pro-cess. The term “observed value” is preferred for this secondusage.3.1.6 observed value, nthe value obtained by making anobservation.3.1.7 ordinal scale, nscale
8、with ordered labeledcategories. ISO 353423.1.8 ratio scale, ncontinuous scale with equal sized scalevalues and an absolute or natural zero point. ISO 353423.1.8.1 DiscussionRatio scales consist of only non-negative values.3.1.9 scale, nsystem of reference values for acharacteristic. ISO 353423.1.10
9、test determination, nthe value of a characteristic ordimension of a single test specimen derived from one or moreobserved values.3.1.11 test method, na definitive procedure that producesa test result.3.1.11.1 DiscussionExamples of test methods include, butare not limited to: identification, measurem
10、ent, and evaluationof one or more qualities, characteristics, or properties.ASTM Regulations 2.2.63.1.12 test observation, nsee observation.3.1.13 test result, nthe value of a characteristic obtainedby carrying out a specified test method.3.1.13.1 DiscussionThe test method specifies that one or anum
11、ber of individual observations be made, and their averageor another appropriate function, (such as the median or thestandard deviation), be reported as the test result. It can alsorequire standard corrections to be applied, such as correction ofgas volumes to standard temperature and pressure. Thus,
12、 a testresult can be a result calculated from several observed values.In the simple case, the test result is the observed value itself.ISO 353421This guide is under the jurisdiction of ASTM Committee E11 on Quality andStatistics and is the direct responsibility of Subcommittee E11.20 on Test MethodE
13、valuation and Quality Control.Current edition approved Oct. 1, 2014. Published November 2014. Originallyapproved in 2003. Last previous edition approved in 2013 as E2282 13. DOI:10.1520/E2282-14.2For referenced ASTM standards, visit the ASTM website, www.astm.org, orcontact ASTM Customer Service at
14、serviceastm.org. For Annual Book of ASTMStandards volume information, refer to the standards Document Summary page onthe ASTM website.3Available from International Organization for Standardization (ISO), 1, ch. dela Voie-Creuse, Case postale 56, CH-1211, Geneva 20, Switzerland, http:/www.iso.ch.Copy
15、right ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States13.1.14 test specimen, nthe portion of a test unit needed toobtain a single test determination.3.1.14.1 DiscussionWhen used for a physical test, this issometimes called “test piece.” For a ch
16、emical test, it issometimes called test portion or test sample. For optical andother tests, it is also sometimes called test sample. In inter-laboratory evaluation of test methods and other statisticalprocedures, it is best to reserve the word sample for the wholeamount of material involved and not
17、the individual testspecimens, pieces or portions being tested.3.1.15 test unit, nthe total quantity of material (containingone or more test specimens) needed to obtain a test result asspecified in the test method. (See test result.)3.2 Definitions of Terms Specific to This Standard:3.2.1 binary test
18、 result, na test result for which the finalvalue is one of two possible categories.3.2.2 ordinal test result, na test result where the finalvalue is reported as one of the scale results on an ordinal scale.4. Significance and Use4.1 All test methods have an output in the form of a testresult. This g
19、uide provides information on the construction oftest results from more elemental measurements.4.2 A well defined test result is necessary before anyprecision statements can be made about the test method.4.2.1 Form and Style for ASTM Standards, Section A21,requires that every test method shall contai
20、n a statementregarding its precision, preferably as a result of an interlabo-ratory test program. Reporting of such studies is described inPractice E177, which illustrates the development of test resultsfrom observations and test determinations.4.2.2 Precision statements for ASTM test methods areapp
21、licable to test results. They are not applicable to testdeterminations or observations, unless specifically and clearlyindicated otherwise.5. Scales5.1 The test method must clearly identify the scale formeasuring the test observations and reporting the test results.Measurement scales are classified
22、into various types. Theprimary classification is into numerical or categorical scales.Numerical scales, also known as quantitative scales, areestablished in terms of a defined numerical range with specifiedscale divisions. Categorical scales, also known as qualitativescales, are defined in terms of
23、words, but the categories may beassigned numbers for purposes of data analysis.5.2 Measurement scales may be sub-classified into ahierarchal system denoted as nominal, ordinal, interval, andratio scales as follows:5.2.1 A nominal scale is an unordered categorical scale.Examples include blood types (
24、A, B, O) or categories of defecttypes.5.2.1.1 A binary scale is the special case of a nominal scalewith only two categories. An example is the presence orabsence of some condition in a test specimen or in conductinga test method, such as a pipe or glass breaking after an impact,a cigarette igniting
25、a piece of fabric, or a light bulb turning on.5.2.2 An ordinal scale is an ordered categorical scale. Anexample is a rating scale comprising four categories: poor, fair,good, and excellent.5.2.2.1 Worded categories may be assigned numbers, suchas 1 = poor, 2 = fair, 3 = good, 4 = excellent.5.2.2.2 T
26、he differences in categories, whether in numbers orlabels, are not uniform and are often arbitrary or subjective.5.2.3 An interval scale is a numeric scale with an arbitraryzero. Such scales may consist of negative and positivenumbers, rounded to a defined number of significant figures.An example is
27、 the Celsius scale for temperature where 0C isdefined as the freezing point of water.5.2.3.1 Differences are meaningful on an interval scale. Adifference of 10 degrees Celsius is the same change intemperature throughout the scale.5.2.4 A ratio scale is a numeric scale with an absolute zero,and all v
28、alues are non-negative numbers. Examples are thelength of an item or the temperature as measured on the Kelvinscale.5.2.4.1 Ratios, as well as differences, are meaningful on aratio scale.5.3 Other Types of Scales:5.3.1 A number of special types of scales may be con-structed or utilized. These may in
29、volve non-linear scales suchas logarithmic or power scales. Other situations may involvecensored numerical responses where values that would bebelow a lower limit, or above an upper limit, are not reportednumerically.6. Developing the Test Result6.1 A test method may have three distinct stages: (1)
30、thedirect measurement or observation of dimensions or properties,or the occurrence of an event; (2) the arithmetical combinationof observed values to obtain a single determination; and (3) thearithmetical combination of a number of determinations toobtain the test result of a test method.6.2 Observa
31、tion:6.2.1 An observation or observed value should be inter-preted as the most elemental single reading or correctedreading obtained in the process of making a test or measure-ment.6.2.2 An observation may be a classification into one of twocategories or a numerical value on a continuous scale. Anob
32、servation may involve a direct reading (for example, azero-adjusted micrometer reading of the thickness of a test stripat one position along the strip) or it may require the interpo-lation of the reading from a calibration curve.6.3 Test Determination:6.3.1 For a quantitative test method, a test det
33、erminationmay be described as the process of calculating from one ormore observations a property of a single test specimen, or asthe value obtained from the process. Thus, a test determinationmay summarize or combine one or more observations.6.3.2 For a qualitative test, such as a binary procedure,
34、thetest determination may be the total number of items falling intoone of the two classifications following repetition of the basicprotocol.E2282 1426.3.3 In some cases the protocol may require observations tobe made under several test conditions and then calculating atest determination from these o
35、bservations.6.3.4 Test methods for chemical properties are often basedon comparison of the response for a sample to responses of aknown standard substance at varying concentrations. Thestandard curve relates response to concentration of the sub-stance of interest. Responses belonging to the standard
36、 curve,and to test specimens, are test observations. The test determi-nation is calculated from the response, or average of multipleresponses, for the test specimen using an equation for thestandard curve.6.3.5 Examples of a Test Determination:6.3.5.1 The measurement of the density of a test specime
37、nmay involve the separate observation of the mass and thevolume of the specimen and the calculation of the ratiomass/volume. The density calculated from the ratio of one pairof mass and volume observations made on one specimen is atest determination.6.3.5.2 The determination of the thickness of a te
38、st speci-men strip may involve averaging micrometer caliper observa-tions taken at several points along the strip.6.3.5.3 A set of three cigarettes is ignited on a fabric. Eachcigarette is observed to burn or not burn. The number of burnsout of the set of three is a test determination.6.4 Test Resul
39、t:6.4.1 In general, the test method should describe not onlythe manner in which each test determination is to be made, butalso the number of test determinations to be made and howthese are to be combined to provide the test result.6.4.2 Test Results Reported on Binary or Ordinal Scales:6.4.2.1 Binar
40、y Test ResultsIn some situations allobservations, test determinations, and combinations of testdeterminations may lead to a final result that falls into one oftwo categories. See examples in 6.4.3.6 and 6.4.3.7.6.4.2.2 Ordinal Test ResultsIn some situations the obser-vations or test determinations i
41、nitially reported on a quantita-tive scale are then reported on the basis of a series of ordinalscale values. See example in 6.4.3.8.6.4.3 Examples of a Test Result:6.4.3.1 The test method on density might require that themass and volume observations of a specimen be combined togive a test determina
42、tion of density (6.3.5.1) and the testdetermination of each of five specimens be averaged to give atest result.6.4.3.2 The test method for paper thickness may require thatthe determination of strip thickness in 6.3.5.2 be made on tenstrips and that the ten test determinations be averaged to givethe
43、test result.6.4.3.3 The test method for a tensile strength test of papermay specify that a tensile strength determination be performedon each of ten specimens and that the ten tensile test determi-nations be averaged to get the test result.6.4.3.4 In a chemical analysis, the method may call for thep
44、reparation of a single solution from the sample (test sampleunit), and measurement on three aliquots (specimens) of thesolution. The average of the three analytical test determinationswould then be the test result. Alternatively, the method maycall for test determinations to be made on different pre
45、parations(specimens) of the sample. The average of the determinationswould then be the test result for the test sample6.4.3.5 In rubber testing, the method may describe not onlythe shape of the test specimen to be taken from a sheet ofrubber, but also the preparation of the sheet, including com-poun
46、ding and curing. For example, one rubber test methodspecifies that four sheets be individually compounded andcured and three specimens tested from each sheet. The testresult is then defined as the average of the four medians, eachmedian being the middle determination, in the order ofmagnitude, of th
47、e three values obtained from a sheet.6.4.3.6 In examining for presence of cancer in a patient, acollection of different measurements might be obtained and thevarious test determinations combined to just report whether theindividual has or does not have cancer. In this case, the testresult is binary.
48、6.4.3.7 A collection of measurements on a product such aslength, width, density, tensile strength, and number of defectspresent can be accumulated. The final test result could be toaccept or reject. In this case, the test result is binary.6.4.3.8 The total weight of a dozen eggs is measured on anoun
49、ce or gram scale. That weight is then converted to a scaleranging from small to jumbo, an ordinal scale.7. Keywords7.1 observation; test determination; test method; test resultASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentionedin this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the riskof infringement of such rights, are entirely their own responsibility.This standard is subject to revision at any
