1、Designation: E 1697 05Standard Test Method forUnipolar Magnitude Estimation of Sensory Attributes1This standard is issued under the fixed designation E 1697; the number immediately following the designation indicates the year oforiginal adoption or, in the case of revision, the year of last revision
2、. A number in parentheses indicates the year of last reapproval. Asuperscript epsilon (e) indicates an editorial change since the last revision or reapproval.1. Scope1.1 This test method describes a procedure for the applica-tion of unipolar magnitude estimation to the evaluation of themagnitude of
3、sensory attributes. The test method coversprocedures for the training of panelists to produce magnitudeestimations and statistical evaluation of the estimations.1.2 Magnitude estimation is a psychophysical scaling tech-nique in which panelists assign numeric values to the magni-tude of an attribute.
4、 The only constraint placed upon thepanelist is that the values assigned should conform to a ratioprinciple. For example, if the attribute seems twice as strong insample B when compared to sample A, sample B shouldreceive a value which is twice the value assigned to sample A.1.3 The intensity of att
5、ributes such as pleasantness, sweet-ness, saltiness or softness can be evaluated using magnitudeestimation.1.4 Magnitude estimation may provide advantages overother scaling methods, particularly when the number of pan-elists and the time available for training are limited. Withapproximately1hoftrain
6、ing, a panel of 15 to 20 naiveindividuals can produce data of adequate precision and repro-ducibility. Any additional training that may be required toensure that the panelists can properly identify the attributebeing evaluated is beyond the scope of this test method.2. Referenced Documents2.1 ASTM S
7、tandards:2E 253 Terminology Relating to Sensory Evaluation of Ma-terials and ProductsE 1871 Practice for Serving Protocol for Sensory Evalua-tion of Foods and Beverages2.2 ASTM Publications:3Manual 26 Sensory Testing Methods, 2nd EditionSTP 758 Guidelines for the Selection and Training ofSensory Pan
8、el Members2.3 ISO Standards:4ISO 11056:1999 Sensory AnalysisMethodologyMagnitude Estimation MethodISO 4121:1987 Sensory AnalysisMethodologyEvaluation of Food Products by Methods Using ScalesISO/DIS 5492:1990 Sensory AnalysisVocabulary (1)ISO 6658:1985 Sensory AnalysisMethodologyGeneralGuidanceISO/DI
9、S 8586-1:1989 Sensory AnalysisMethodologyGeneral Guide for Selection, Training and MonitoringSubjectsPart 1: Qualifying Subjects (1)ISO 8589:1988 Sensory AnalysisGeneral Guidance forthe Design of Test Rooms3. Terminology3.1 Definitions:3.1.1 external modulusnumber assigned by the panelleader to desc
10、ribe the intensity of the external reference sampleor the first sample of the sample set. The external modulus issometimes referred to as a “fixed modulus” or just the“modulus.” In this case the reference is said to be modulated.3.1.2 external reference sample for magnitude estimationsample designat
11、ed as the one to which all others are to becompared, or to which the first sample of a set is to becompared, when each subsequent sample in the set is comparedto the preceding sample. This sample is normally the firstsample to be presented.3.1.3 internal modulusnumber assigned by the panelist todesc
12、ribe the intensity of the external reference sample or thefirst sample of the sample set. The internal modulus issometimes referred to as a “non-fixed modulus.” When aninternal modulus is used, the reference is sometimes said to beunmodulated.3.1.4 internal reference sample for magnitude estimations
13、ample present in the experimental set, which is presented tothe panelist as if it were a test sample. The value assigned tothis sample(s) can be used for normalizing panelists data. If anexternal reference is used, the internal reference(s) are nor-mally identical to it.1This test method is under th
14、e jurisdiction ofASTM Committee E18 on SensoryEvaluation of Materials and Products and is the direct responsibility of Subcom-mittee E18.03 on Sensory Theory and Statistics.Current edition approved April 1, 2005. Published April 2005. Originallyapproved in 1995. Last previous edition approved in 199
15、9 as E 1697 95 (1999).2For referenced ASTM standards, visit the ASTM website, www.astm.org, orcontact ASTM Customer Service at serviceastm.org. For Annual Book of ASTMStandards volume information, refer to the standards Document Summary page onthe ASTM website.3Available from ASTM Headquarters, 100
16、Barr Harbor Drive, PO Box C700,West Conshohocken, PA 1942829593.4Available from American National Standards Institute (ANSI), 25 W. 43rd St.,4th Floor, New York, NY 10036.1Copyright ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.3.1.5 magnitud
17、e estimationprocess of assigning values tothe intensities of an attribute of products in such a way that theratios of the values assigned and the panelists perceptions ofthe attribute are the same.3.1.6 normalizingprocess of multiplying each panelistsraw data by, or adding to the logarithm of each p
18、anelists rawdata, a value which brings all the data onto a common scale.Also referred to as rescaling.3.1.7 Stevens Equation or the Psychophysical PowerFunctionR 5 KSn(1)where:R = the panelists response (the perceived intensity),K = a constant that reconciles the units of measurementused for R and S
19、,S = the stimulus (chemical concentration or physicalforce), andn = the exponent of the power function and the slope ofthe regression curve for R and S when they areexpressed in logarithmic units.In practice, Stevens equation is generally transformed tologarithms, either common or natural:lnR 5 lnK
20、1 nlnS (2)3.2 Reference Terminology E 253 for general definitionsrelated to sensory evaluation.4. Summary of Test Method4.1 Panelists judge the intensity of an attribute of a set ofsamples, presented in random order, on a ratio scale. Forexample, if one sample is given a value of 50 and a secondsamp
21、le is twice as strong, it will be given a value of 100. If itis half as strong it will be given a value of 25. There are threeprocedures that can be used.4.1.1 Panelists are instructed to assign any value to describethe intensity of the first sample (external reference, which mayor may not be part o
22、f the sample set). Panelists then rate theintensity of the following samples in relation to the value of theexternal reference.4.1.2 The external reference is pre-assigned a value (modu-lus) to describe its intensity by the panel leader. Panelists ratethe intensity of the following samples in relati
23、on to the externalreference and the modulus.4.1.3 Panelists rate the intensity of each subsequent samplein relation to the preceding sample. The first sample of the setmay or may not have a modulus.4.2 Individual judgments can be converted to a commonscale by normalizing the data. Three normalizing
24、methods canbe used: internal standard normalizing, external calibrationand, if a modulus is not used, no standard normalizing (methodof averages). See 11.4 and Appendix X2-Appendix X4.4.3 Results are averaged using geometric means. Analysisof variance or other statistical analyses may be performed a
25、fterthe data have been converted to logarithms.5. Significance and Use5.1 Magnitude estimation may be used to measure andcompare the intensities of attributes of a wide variety ofproducts.5.2 Magnitude estimation provides a large degree of flex-ibility for both the experimenter and the panelist. Onc
26、e trainedin magnitude estimation, panelists are generally able to applytheir skill to a wide variety of sample types and attributes, withminimal additional training.5.3 Magnitude estimation is not as susceptible to end-effectsas interval scaling techniques. These can occur when panelistsare not fami
27、liar with the entire range of sensations beingpresented. Under these circumstances, panelists may assign anearly sample to a category which is too close to one end of thescale. Subsequently, they may “run out of scale” and be forcedto assign perceptually different samples to the same category.This s
28、hould not occur with magnitude estimation, as, in theory,there are an infinite number of categories.5.4 Magnitude estimation is one frequently used techniquethat permits the representation of data in terms of StevensPower Law.5.5 The disadvantages of magnitude estimation arise prima-rily from the re
29、quirements of the data analysis.5.5.1 Permitting each panelist to choose a different numeri-cal scale may produce significant panelist effects. This disad-vantage can be overcome in a number of ways, as follows. Theexperimenter must choose the approach most appropriate forthe circumstances.5.5.1.1 E
30、xperiments can be designed such that analysis ofvariance can be used to remove the panelist effects andinteractions.5.5.1.2 Alternatively, panelists can be forced to a commonscale, either by training or by use of external reference sampleswith assigned values (modulus).5.5.1.3 Finally, each panelist
31、s data can be brought to acommon scale by one of a variety of normalizing methods.5.5.2 Logarithms must be applied before carrying out dataanalysis. This becomes problematic if values are near thresh-old, as a logarithm of zero cannot be taken (see 11.2.1).5.6 Magnitude estimation should be used:5.6
32、.1 When end-effects are a concern, for example whenpanelists are not familiar with the entire range of sensationsbeing presented.5.6.2 When Stevens Power Law is to be applied to the data.5.6.3 Generally, in central location testing with paneliststrained in the technique. It is not appropriate for ho
33、me use ormall intercept testing with consumers.5.7 This test method is only meant to be used with panelistswho are specifically trained in magnitude estimation. Do notuse this method with untrained panelists or untrained consum-ers.6. Conditions of Testing6.1 The general conditions for testing, such
34、 as the location,preparations, presentation and coding of samples, and theselection and training of panelists are described in the stan-dards for general methodology, such as ISO 6658, ISO/DIS8586-1, ISO 8589, ASTM STP 758 or those describingmethods using scales and categories, for example, ISO 4121
35、and ASTM Manual 26, and for specific serving protocols inPractice E 1871.E16970527. Selection and Training of Panelists7.1 Refer to ISO 8586-1 or ASTM STP 758 for all thegeneral considerations concerning the selection and training ofpanelists. Refer to ISO 11056 for considerations specific tomagnitu
36、de estimation.7.2 As is true for all methods of sensory evaluation, thepanel leader will have to make judgments as to the level ofproficiency required of the panelists. The objectives of the test,the availability of panelists, the costs of securing additionalpanelists and of additional training shou
37、ld all be considered inthe design of a training program. Panelists generally reach astable level of proficiency in the method itself after three tofour exercises in assigning magnitudes.7.3 Estimating the areas of geometric shapes has provenvery useful for introducing panelists to the basic concepts
38、 ofmagnitude estimation. A set of 18 figures composed of sixcircles, six equilateral triangles and six squares ranging in sizefrom approximately 2 cm2to 200 cm2has been used success-fully for training panelists (see Table 1).7.4 Prior to presenting the figures, the panel leader instructsthe candidat
39、e in the principles of the method. This instructionshould include, but is not necessarily limited to the followingthree points.7.4.1 If the attribute is not present, the value 0 should beassigned.7.4.2 There is no upper limit to the scale.7.4.3 Values should be assigned on a ratio basis: if theattri
40、bute is twice as intense, it should receive a rating twice aslarge.7.5 Panelists have a tendency to use “round numbers” suchas 5, 10, 20, 25, and so forth. This should be pointed outexplicitly during training. Panelists should be encouraged,“given permission,” to use all numbers. Panelists are alsoi
41、nfluenced by the ratios mentioned in training. Therefore, careshould be taken to mention a variety of different ratios, forexample, 3:1 and13 , 7.5, 2.4, not just 2:1 and12 .7.6 Assigning Codes to the FiguresThe figures are pre-sented singly, centered on an 8.5 3 11 in. sheet of white paper.The pane
42、list states his magnitude estimate; the estimation isrecorded. The 8.5-cm square is presented first with the instruc-tion to assign it a value between 30 and 100. The balance of thegeometric figures should be shuffled prior to each test so thatthe type of geometric figure and the size of the areas d
43、o notform a particular pattern.7.7 Comparing the ResultsAfter completing the full set ofshape estimates, panelists should be allowed to compare theirresults with the averaged results of the group. If this is notpractical, the results from a previous group can also be used.The objective is to provide
44、 positive feedback, that is, toreassure the panelists that they understand the exercise. Careshould be taken not to create the impression that there is a“right” answer. Unless their results are very different, depar-tures from the group results should be explained as ordereffects, that is, their res
45、ponses are affected by the order inwhich they evaluate the samples. They should be reassured thatdespite individual order effects, the groups results will beaccurate.7.8 If the panelists results are very different, review theprinciples of the method again. If the panel leader judges thata panelist c
46、annot be trained in the method, the training shouldbe discontinued at this point and the panelist excused.7.9 Once the panel has successfully completed the areaestimation exercise, further training should be carried out withthe commodity or type of test substance to be used in the maintrial(s). This
47、 gives the panelist experience in applying magni-tude estimation to attributes characterizing the test sample.7.10 The panel leader may need to design exercises fortraining panelists to properly identify the attributes to beevaluated. The need for this will depend on the objectives andrequirements o
48、f the test.8. Number of Panelists Required8.1 As is true for other forms of scaling, the number ofpanelists necessary for a given task depends on the complexityof the task, how close together the various test samples are inthe attribute being evaluated, the amount of training thepanelists have recei
49、ved, and the importance to be attached tothe decision based on the test results (c.f., ISO 8586-1). Issuesof statistical power need to be resolved based on the varianceassociated with a particular evaluation and the magnitude ofthe differences that need to be detected.9. Reference Samples9.1 External ReferencesThe panel leader specifies to thepanelists that the reference sample has a value of, for example,30, 50, 100 or whatever seems appropriate to the panel leader.The leader instructs the panelists to make his or her subsequentjudgments relative t
