1、ASTM D3780 8 0759530 0033387 2)Designation:D 3980-88 Standard Practice for Interlaboratory Testing of Paint and Related Materials This standard is issued under the fixed designation D 3980;the number immediately following the designation indicates the year of original adoption or,in the case of revi
2、sion,the year of last revision.A number in parentheses indicates the year of last reapproval.A superscript epsilon(E)indicates an editorial change since the last revision or reapproval.1.Scope 1.1 This practice covers the interlaboratory testing of paint and related materials.The information present
3、ed is intended to assist task groups in the effective planning of round-robin test programs.1.2 The various parts appear in the following order:Part A-Definitions 3 Part C-Statistical Analysis Part%Analysis Presentation 1.3 This practice,in spite of its length,is not intended to cover all aspects of
4、 statistical design and analysis.For example,the analysis of variance used in the practice as the basis for estimating the precision is abridged in comparison to that normally used for establishing the significance of experimental factors.The publications listed in the appendix should be consulted f
5、or further information or explanations.For convenience,comparable practices developed by other ASTM technical committees are also shown.Sections Part B-Planning 4t0 11 12 to 16 17 to 20 2.Referenced Documents 2.1 ASTM Standards:D 968 Test Methods for Abrasion Resistance of Organic Coatings by the Fa
6、lling Abrasive Teste?D 3793 Test Method for Low-Temperature Coalescence of Latex Paint Films E 180 Practice for Determining the Precision of ASTM Methods for Analysis and Testing of Industrial Chemicals3 PART A-TERMINOLOGY 3.Definitions 3.1 The statistical terms used in this practice are defined,as
7、nearly as possible,in every day language.For convenience,the terms are listed in order of use in this practice.3.2 experimental design-the complete plan for con-ducting an investigation or an interlaboratory study of materials,processes,or test methods.It is conveniently set forth in the form of a s
8、ingle comprehensive table showing number of laboratories,materials to be tested,test methods,number of operators,testing conditions,time of testing,and-I T h i s practice is under the jurisdiction ofASTM Committee D I on Paint and Related Coatings and Materials and is the direct responsibility of Su
9、bcommittee W1.20 on Sampling,Statistics,etc.Current edition approved May 27,1988.Published December 1988.Originally published as D 3980-81.Last previous edition D 3980-87.Annical Book 0jAST.M Standards,Vol 06.01.Annital Book oJASTMStandards,Vol 15.05.degree of replication and repetition.The expenmen
10、tal design should provide the basis for an efficient statistical analysis of the results(see Part By Planning).3.3 replication-the execution of multiple chemical or physical determinations on the same specimen or physical measurements on the same panel at the same time,as closely as possible.Two rep
11、licates are called duplicates,three are called triplicates,etc.The individual results are not suitable for estimating precision but are used only to ensure by comparison with similar measurements that there is not a gross error among them(see 15.1.3.3).3.4 repetition-The execution of multiple physic
12、al mea-surements on the same panel or chemical determinations at different times,or physical measurements on different panels,areas of large structures,or specimens of a liquid sample within a short time interval,to establish the precision of a method.In most cases the measurements are the mean of r
13、eplicates as defined in 3.3.3.5 population-the totality of observations on or deter-minations of a certain property or component obtained by the same procedure;theoretically,an infinite collection of measurements on a given item of interest but,practically,a large number of measurements of the item.
14、Thus,when statistical procedures are used to determine whether or not two(or more)materials or test methods differ significantly with respect to some measurable property,in effect,the determination is whether these materials belong to the same or to different populations.3.6 sample-a randomly select
15、ed subset of a population intended to be representative of it so as to enable obtaining an estimate of the property or composition of the entire population.The reliability of such an estimate can be expressed in terms of confidence limits(see 3.19).Through tests of significance(see 3.21),it is possi
16、ble to state,with a specified degree of confidence,whether two or more samples are drawn from the same or from different populations.3.7 average-a typical numerical value that attempts to summarize or reflect the location of a group of observations by a single number.While it is a measure of central tendency,it does not provide information on the variability of the individual observations.The following are different types of averages:arithmetic mean,weighted mean,algebraic mean,geometric mean,ha
