ASTM D4467-94(2001)
(Practice)Standard Practice for Interlaboratory Testing of a Textile Test Method That Produces Non-Normally Distributed Data (Withdrawn 2010)
Standard Practice for Interlaboratory Testing of a Textile Test Method That Produces Non-Normally Distributed Data (Withdrawn 2010)
SCOPE
1.1 This practice covers design and analysis of interlaboratory testing of a test procedure in the case where the resulting test data are discrete variates or are continuous variates not normally distributed. This practice applies to all such interlaboratory tests used to validate a test procedure.
1.2 Analysis of interlaboratory test results permits validation that the process of using the test method is in statistical control and provides the information required to write statements on precision and bias as directed in Practice D2906. It also gives the information for determining the number of specimens per unit in the laboratory sample as required in Practice D2905.
1.3 Precision statements for non-normally distributed data can be written as a function of the level of the property of interest without an interlaboratory test if the underlying distribution is known and statistical control can be assumed.
1.4 If the underlying distribution is unknown, the precision of the test method can only be approximated. There are no generally accepted methods of making approximations of this sort.
1.5 If statistical control cannot be assumed, then a meaningful precision statement cannot be written and the test method should not be used.
1.6 This practice is intended for use with data from test methods that cannot be properly modeled by a normal distribution. See Practices D2904 and E691 for applications that can be modeled by a normal distribution.
1.7 This practice includes the following sections: SectionsScope1Referenced Documents 2Terminology 3Significance and Uses 4General Considerations 5Basic Statistical Design 6Pilot-Scale Interlaboratory Test 7Full-Scale Interlaboratory Test 8Missing Data 9Outlying Observations 10Interpretation of Data 11Plotting Results 12Keywords 13Pilot-Scale and Full-Scale Interlaboratory Tests Annex A1Calculation of Chi-Square Annex A2
1.8 This standard does not purport to address all of the safety concerns, if any, associated with its use. It is the responsibility of whoever uses this standard to consult and establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use.
WITHDRAWN RATIONALE
This practice covers design and analysis of interlaboratory testing of a test procedure in the case where the resulting test data are discrete variates or are continuous variates not normally distributed. This practice applies to all such interlaboratory tests used to validate a test procedure.
This practice is being withdrawn because D13 no longer has the expertise to maintain and statistical standards are being maintained by Committee E11.
Formerly under the jurisdiction of Committee D13 on Textiles and the direct responsibility of Subcommittee D13.93 on Statistics, this practice was withdrawn in February 2010 with no replacement.
General Information
Relations
Standards Content (Sample)
NOTICE: This standard has either been superseded and replaced by a new version or withdrawn.
Contact ASTM International (www.astm.org) for the latest information.
Designation: D4467 – 94 (Reapproved 2001)
Standard Practice for
Interlaboratory Testing of a Textile Test Method That
Produces Non-Normally Distributed Data
This standard is issued under the fixed designation D4467; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision.Anumber in parentheses indicates the year of last reapproval.A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
1. Scope
Outlying Observations 10
Interpretation of Data 11
1.1 This practice covers design and analysis of interlabora-
Plotting Results 12
tory testing of a test procedure in the case where the resulting
Keywords 13
Pilot-Scale and Full-Scale Interlaboratory Tests Annex A1
test data are discrete variates or are continuous variates not
Calculation of Chi-Square Annex A2
normallydistributed.Thispracticeappliestoallsuchinterlabo-
ratory tests used to validate a test procedure. 1.8 This standard does not purport to address all of the
safety concerns, if any, associated with its use. It is the
1.2 Analysis of interlaboratory test results permits valida-
responsibility of whoever uses this standard to consult and
tion that the process of using the test method is in statistical
establish appropriate safety and health practices and deter-
control and provides the information required to write state-
mine the applicability of regulatory limitations prior to use.
ments on precision and bias as directed in Practice D2906.It
also gives the information for determining the number of
2. Referenced Documents
specimens per unit in the laboratory sample as required in
2.1 ASTM Standards:
Practice D2905.
D123 Terminology Relating to Textiles
1.3 Precision statements for non-normally distributed data
D2904 Practice for Interlaboratory Testing of a Textile Test
can be written as a function of the level of the property of
Method that Produces Normally Distributed Data
interest without an interlaboratory test if the underlying distri-
D2905 PracticeforStatementsonNumberofSpecimensfor
bution is known and statistical control can be assumed.
Textiles
1.4 If the underlying distribution is unknown, the precision
D2906 Practice for Statements on Precision and Bias for
of the test method can only be approximated. There are no
Textiles
generally accepted methods of making approximations of this
D4646 Test Method for 24-h Batch-Type Measurement of
sort.
Contaminant Sorption by Soils and Sediments
1.5 If statistical control cannot be assumed, then a mean-
D4853 Guide for Reducing Test Variability
ingful precision statement cannot be written and the test
E456 Terminology Relating to Quality and Statistics
method should not be used.
E691 Practice for Conducting an Interlaboratory Study to
1.6 This practice is intended for use with data from test
Determine the Precision of a Test Method
methods that cannot be properly modeled by a normal distri-
E1169 Practice for Conducting Ruggedness Tests
bution.SeePracticesD2904andE691forapplicationsthatcan
be modeled by a normal distribution.
3. Terminology
1.7 This practice includes the following sections:
3.1 Definitions:
Sections
Scope 1 3.1.1 test method, n—a definitive procedure for the identi-
Referenced Documents 2
fication,measurement,andevaluationofoneormorequalities,
Terminology 3
characteristics, or properties of a material, product, system, or
Significance and Uses 4
General Considerations 5 service that produces a test result.
Basic Statistical Design 6
3.1.2 For definitions of textile and statistical terms used in
Pilot-Scale Interlaboratory Test 7
this practice and discussions of their use, refer to Terminology
Full-Scale Interlaboratory Test 8
Missing Data 9 D123, and Terminology E456.
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
ThispracticeisunderthejurisdictionofASTMCommitteeD13onTextilesand contact ASTM Customer Service at service@astm.org. For Annual Book of ASTM
is the direct responsibility of Subcommittee D13.93 on Statistics. Standards volume information, refer to the standard’s Document Summary page on
Current edition approved June 15, 1994. Published August 1994. Originally the ASTM website.
published as D4467–85. Last previous edition D4467–85. DOI: 10.1520/D4467- Withdrawn. The last approved version of this historical standard is referenced
94R01. on www.astm.org.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.
D4467 – 94 (2001)
3.2 Definitions of Terms Specific to This Standard: be run without a previous pilot-scale test but with the under-
3.2.1 assignable cause—afactorwhichcontributestovaria- standing that unsatisfactory results would require another
tion and is feasible to detect and identify. full-scale test.
3.2.2 interlaboratory testing—the evaluating of a test 4.8 Interlaboratory tests of the type discussed in this prac-
methodinmorethanonelaboratorybyanalyzingdataobtained
tice are used to locate and measure the sources of variability
from one or more materials that are as homogeneous as associated with a test method when the test method is used to
practical.
evaluate a property of one or more materials, each of which is
3.2.3 random cause—one of many factors which contribute ashomogeneousaspracticalwithrespecttothatproperty.Such
to variation but which are not feasible to detect and identify
interlaboratory tests provide no information about the sources
since they are random in origin and usually small in effect. of variability associated with the sampling of the stream of
3.2.4 state of statistical control—a condition in which a
product from a manufacturing process, a shipment, or material
process, including a measurement process, is subject only to in inventory. Estimation of such sampling errors requires an
random variation.
entirely different type of experiment which is not specified
presently in an ASTM Committee D-13 standard.
4. Significance and Use
4.1 The planning of interlaboratory tests requires a general
5. General Considerations
knowledgeofstatisticalprinciples.Interlaboratorytestsshould
5.1 Overview—This section covers various aspects of allo-
be planned, conducted, and analyzed after consultation with
cating specimens to the participating laboratories.
statisticians who are experienced in the design and analysis of
5.2 Sampling of Materials—Select a source of samples of
experiments and who have some knowledge of the nature of
material in such a way that any one portion of the material,
the variability likely to be encountered in the test method.
within which laboratories, operators, days, and other factors
4.2 The instructions of this practice are specifically appli-
are to be compared, will be as homogeneous as possible with
cable to the design and analysis of the following tests:
respect to the property being measured. Otherwise, increased
4.2.1 Pilot-scale interlaboratory tests and
replication will be required to reduce the size of the difference
4.2.2 Full-scale interlaboratory tests.
which can be detected.
4.3 Procedures given in this practice are applicable to
5.3 Randomization of Specimens:
methods based on the measurement of the following types of
5.3.1 Complete Randomization—Randomize the selection
variates:
of specimens for each laboratory sample; divide all the
4.3.1 Ratings (grades or scores), such as those resulting
randomized specimens of a specific material, after labeling,
from comparisons with AATCC gray scales,
into the required number of groups, each group corresponding
4.3.2 Percent of observations with a specific attribute,
to a specific laboratory.
4.3.3 Counts of attributes, such as number of nonconformi-
5.3.2 Stratification—In some cases it is advantageous to
ties,
followastratifiedpatternintheallocationsofthespecimensto
4.3.4 Any data not normally distributed which the analyst
laboratories.Forexample,ifthespecimensarebobbinsofyarn
cannotorprefersnottotransform,suchasflammabilitydataor
from different spinning frames, it is better to allocate to each
percent extractables.
laboratory equal numbers of specimens from each spinning
4.4 Interlaboratory testing is a means of determining the
frame.Insuchcases,thespecimenswithineachspinningframe
consistency of results when the same material is tested under
arerandomizedseparatelyratherthanallofthespecimensfrom
varying conditions such as: operators, laboratories, equipment,
all of the frames.
or environment. An interlaboratory test should do the follow-
5.4 Order of Tests—In many situations, variability among
ing:
replicate tests is greater when measurements are made at
4.4.1 Show if the test method distinguishes between levels
different times than when they are made together as part of a
of the property being tested,
group. Sometimes trends are apparent among results obtained
4.4.2 Show if the test method is in statistical control;
consecutively. Furthermore, some materials undergo measur-
statistical control being the presence of only random variation,
able changes within relatively short storage periods. For these
4.4.3 Detect operators, laboratories, and equipment out of
reasons, treat the dates of testing, as well as the order of tests
statistical control.
carried out in a group as controlled, systematic variables.
4.5 An initial single-laboratory preliminary test of a test
5.5 Selecting the Measure of Average Performance—Data
procedureisnecessary,usuallyincludingruggednesstesting,to
are summarized for presentation and analysis by use of some
determine the feasibility of the method and to determine the
measure of typical performance. For textile testing, there are
method’ssensitivitytovariableswhichmustbecontrolled.See
usually three choices:
GuidesD4853orE1169foradiscussionofruggednesstesting.
4.6 A pilot-scale interlaboratory test may be needed to 5.5.1 Arithmetic Average—The arithmetic average is the
measure of choice when the data are symmetrically distributed
identify sources of variation, to establish clarity of instructions
of the proposed operating procedures, and to obtain estimates or are from a Poisson distribution.
as to the number of test results per operator per material to be 5.5.2 Median—The median (midpoint, fiftieth percentile) is
used in the initial full-scale interlaboratory test. the preferred measure when the data are asymmetrically
4.7 A full-scale interlaboratory test is usually made after a distributed. When the distribution is symmetrical, the arith-
pilot-scale test. If the task group prefers, a full-scale test may metic average and the median are equal.
D4467 – 94 (2001)
5.5.3 Proportion—Aproportion,whichmaybeexpressedas
p = proportion of the observations having a specific at-
a fraction (decimal) or percent, is the measure to use when the
tribute, expressed as a decimal fraction, and
data are counts of items having a particular attribute out of a
wheretheothertermsintheequationareasdefinedin5.6.1.
specified number of items.
5.7 Gain of Statistical Information—More statistical infor-
5.6 Number of Replicate Specimens—The number of speci-
mationcanbeobtainedfromasmallnumberofdeterminations
mens tested by each operator in each laboratory for each
onalargenumberofmaterialsthanfromthesametotalnumber
material may be calculated from previous information or from
ofdeterminationsdistributedoverfewermaterials.Inthesame
a pilot run. This number of specimens or replications (mini-
way, a specific number of determinations per material will
mum of two) depends on the relative size of the random error
yield more information if they are spread over the largest
and the smallest effect to be detectable.Areplicate consists of
number of laboratories possible. For the recommended mini-
onespecimenofeachconditionandmaterialtobetestedinthe
mum design, see 6.2. If experience with the pilot-scale inter-
statistical design.
laboratory test casts doubt on the adequacy of the starting
5.6.1 Symmetrical Non-Normal Distributions—Calculate design,estimatethenumberofdeterminationsneededtodetect
the number of observations required in each mean using Eq 1 the smallest differences of practical importance.
(Note 1):
5.8 Multiple Equipment (Instruments)—When multiple in-
2 2
struments within a laboratory are used on an interlaboratory
n 5 ~ts/E! 516~s/E! (1)
test, tests should be made on all equipment to establish the
where:
presence or absence of the equipment effects. All types of
n = number of observations in each mean,
equipment allowed by a test method should be tested to allow
t = 4=specified value in Tchebychev’s inequality (Note
greatestflexibility.Ifanequipmenteffectispresentandcannot
2),
be eliminated by use of pertinent scientific principles, known
s = standard deviation for individual observations ob-
standards should be run and appropriate within-laboratory
tained from previously conducted studies, and
quality control procedure should be used.
E = smallest difference it is of practical importance to
detect, expressed in the same units of measure as the
6. Basic Statistical Design
averages and standard deviation.
6.1 It is advisable to keep the design as simple as possible,
NOTE 1—With a balanced design, half of the total observations in the
yet to obtain estimates of within- and between-laboratory
experimentwillbeineachofthetwosamplemeansusedtodeterminethe
variationunconfoundedwithsecondaryeffects.Provisionsalso
possibleeffectofeachfactorbeingevaluatedattwolevels;onethirdofthe
should be made for estimates of significance of variation due
total observations will be in each of the three sample means used to
to: materials-by-laboratories interactions, and operators-by-
determinethepossibleeffectofeachfactorbeingevaluatedatthreelevels;
materials interactions.
and so on. The required value of n refers to such means.
6.2 Include in the basic statistical design the following:
NOTE 2—Tchebychev’s inequality states that in all cases at least
(1−1/t) of the total observations, n, will lie within the closed range x¯ 6
6.2.1 A minimum of three materials spanning the range of
ts , when t is not less than 1. For t=4, at least 93.75% of all
interest for the property being measured,
observations will fall within x¯ 6 4s. For symmetrical distributions, the
6.2.2 At least ten laboratories unless the test method cannot
observed percentage is usually well above the minimum percentage
be used in that many laboratories,
specified by Tchebychev’s inequality.
6.2.3 Arecommended minimum of two operators per labo-
5.6.2 Asymmetrical Distribution Except Poisson or
ratory, and
Binomial—Calculate the number of observations required in
6.2.4 Atleasttwospecimensofeachmaterialtobetestedby
each mean using Eq 2 (Note 2):
each operator in a designat
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.