This standard specifies the use of a three-point comparison method to identify the difference between two samples. This standard is applicable to identifying subtle differences between samples, and can also be used to select and train evaluators or check the ability of evaluators. GB 12311-1990 Sensory analysis method three-point test GB12311-1990 standard download decompression password: www.bzxz.net
This standard specifies the use of a three-point comparison method to identify the difference between two samples. This standard is applicable to identifying subtle differences between samples, and can also be used to select and train evaluators or check the ability of evaluators.

National Standard of the People's Republic of China
Sensory analysis method three-point test
Sensory analysis method--Trlangular testGB 12811-90
This standard refers to the international standard ISO 4120-1983 "Sensory analysis methodology--three-point test". 1 Subject content and applicable examples
This standard specifies the use of a three-point comparison method to identify the difference between two samples. This standard is suitable for identifying subtle differences between samples, and can also be used to select and train evaluators or check the ability of evaluators. 2 Reference standards
GB10220 General introduction to sensory analysis methods
GB10221.1~10221.4 Sensory analysis terms GB3358 Statistical terms, terms and symbols
3 Method summary
At the same time, the evaluator provides a group of three samples, two of which are exactly the same, and the evaluator picks out a single sample. 4 Equipment
The person in charge of inspection shall select equipment according to the nature of the product and the number of samples. The equipment used shall not affect the inspection results. Standardized equipment that meets the inspection needs shall be used first.
5 Sampling
Sampling shall be carried out according to the sampling standards of the products being picked. If there is no such standard or the sampling standard is not fully applicable, the sampling method shall be agreed upon by the parties concerned.
6 General conditions for inspection
6.1 Environment
Should meet the conditions required by GB 10220.
6.2 Evaluators
6.2.1 Conditions
Should meet the conditions specified in GB10220, and all evaluators should have the same qualifications and inspection capabilities. 6.2.2 Number of evaluators
The number of evaluators is determined according to the level of inspection. Usually there are more than 6 experts, or more than 15 preferred evaluators, or more than 25 primary evaluators. At 0.1% significant water, more than 7 experts are required. 6.2.3 Inspection Person in Charge
The inspection person in charge should not generally participate in the inspection. If he participates, he should not know the sample number. 6.3 Preparation
The inspection person in charge can make a preliminary introduction to the relevant issues and sample properties that will not affect the evaluation. When it involves the inspection of station dyes, a non-station dye sample and a station dye sample for comparison should be prepared. The State Bureau of Technical Supervision approved on April 10, 1890 and implemented on December 1, 1990
Test steps
7.1 Preparation of samples to be inspected
GB 12311--90
7.1.1 Provide sufficient amounts of samples A and B, with three inspection samples in one group. 7.1.2 Prepare equal numbers of sample groups from laboratory samples according to the following six combinations:
BBABAB. ABBAABABABAA
7.1.3 The manner in which the samples are presented should not allow the assessors to draw conclusions about the nature of the samples. The various test sample groups should be prepared in the same manner (same equipment, same containers, same number of products and same array form (triangle, straight line, etc.)). 7.1.4 The humidity of the test samples in any sample group should be the same and, if possible, the temperature of all other sample groups in the test series should be the same.
7.1.5 The containers containing the test samples should be numbered, usually with a three-digit number randomly selected. The number should be different for each test. 7.2 Test Techniques
7.2.1 The assessors should be informed of the test to such an extent that their conclusions are not biased. 7.2.2: The groups of samples prepared in 7.1.2 are randomly assigned to the assessors. 7.2.3 The assessors shall examine the groups of test samples in the prescribed order, which shall be the same for the same series of tests. When evaluating the same group of three test samples, the assessors shall have the opportunity to repeat the test on each test sample. The person in charge of the inspection can tell the assessor the number and volume of samples to be provided when necessary. When the number of assessors is less than a multiple of 6, the following two methods can be adopted.
a: Discard the redundant sample groups;
b. Provide 6 groups of samples for each assessor for repeated inspection. 7.2.4 There are two inspection techniques as follows, and the person in charge can choose one. a.
"Forced choice" requires the assessor to point out the difference between one sample and the other two even if the assessor states that there is no difference.
b. When the assessor cannot identify the difference, the answer "no difference" is allowed. If the accuracy of the test results is to be considered, "forced choice" should be used. The answer format is shown in Appendix B.
8 Expression and interpretation of results
"Forced choice"
Count the number of correct answers, and then refer to the table below to determine whether there is a significant difference between the samples. See Appendix C for an application example. When the value of n in the table is greater than 100, the minimum number of correct answers is calculated according to the following formula and the nearest integer value is taken. X=0.47142 n+(2n+3)
Wu Zhong: z:
αs.0, 001
2 = 3. 10
Number of answers
GB 12811--80
The minimum number of correct cases required to determine the significant difference in a three-point test at different significance levelsThe minimum number of correct answers required at different significance levels
Number of cases
The minimum number of correct cases required at different significance levels
8.2 “No difference”
Depending on the purpose of the test, the answer of “no difference” can be handled in different ways. 0.1%
Number of answers
Different significance levelswww.bzxz.net
Minimum number of correct answers
8.2.1 Ignore the number of "no difference" answers, that is, subtract these numbers from the total number of answers of the evaluation team. Consider the following ways
GB 12311—90
8. a. Attribute one-half of the "no difference" answers to the correct answer. b. Attribute the "no difference" answers to the incorrect answer. c. Attribute the "no difference" answers to the incorrect answer. d. Consider separately.
When the no difference answers account for a large proportion, it means that the difference between the two samples is below the perception threshold of the evaluators. It may be that the test method is flawed, or some evaluators may have undergone some physiological changes or lack enthusiasm for the test they participated in. 9 Test report
The test report shall include the following contents:
Explanation of the sample information,
Test parameters adopted, especially the number of three test sample groups; the number of tests and the number and qualifications of the evaluation participants; test conditions, especially whether "strong selection" was adopted; test results and their statistical interpretation;
Test date, time and conditions;
Name of the person in charge of the test.
GB 12311--90
Appendix A
Extension of three-point test
(Supplement)
A1 In certain cases, the person in charge of the test may decide to extend the three-point test to provide some additional contents. For example; specify the nature of the difference
Specify the intensity or degree of the difference
The development of such difference over time, etc.
A2 It is possible to provide the following information according to 6.3. The number of typical samples prepared should be limited (2 or 3), but they should be representative. Use the extended three-point test answer sheet, see Appendix 2. The person in charge of the test should consider not overloading the evaluator and causing sensory fatigue, and the questioning method should not bias the answer. The extended test conducted should be explained in the test report. Appendix B
Answer format
(Supplement)
Simple three-point test ('forced choice') Sample
Test three products in the specified order. Circle the sample number you think is single. You must make a choice
Test sample number
B2 Extended three-point test ('forced choice') Sample
Test three products in the specified order. Circle the sample number you think is single. You must make a choice
Test sample number
Describe the difference
E3 Extended two-point test (allow the answer "no difference") Date
Name of evaluated product
Name of assessor
GB 12311—90
Test three samples in the specified order. If there is a difference, mark the sample number that you think is different
Mark the difference in intensity that you perceive
Appendix ℃
Simple three-point test example "Forced Choice" (reference)
Name of assessor
C1" The manufacturer wants to know whether a new product with a changed formula is similar to the original product by sensory analysis. 12 assessors sign and evaluate.
C2 Prepare two batches of samples, one batch of old formula (A) and one batch of new formula (B). Each assessor is required to make only one assessment. 18 samples of formula A and 18 samples of formula B must be prepared, divided into six groups, two samples in each group, and the combinations are as follows: 1 row ABB
2 rows AAB
2 rows ABA
2 rows BA.4
2 rows BBA
2 rows BAB
C3 The evaluators randomly evaluate these groups, and the person in charge of the inspection selects the 5% significant level. The number of correct answers of the evaluators is 8. According to Table 1, there are 8 correct answers among the 12 answers, so it is determined that the two products are different at the 5% significant level.
Additional Notes:
This standard is proposed by the Ministry of Agriculture of the People's Republic of China. This standard is drafted by the Analysis and Testing Center of the Chinese Academy of Agricultural Sciences. The main drafters of this standard are Cui Shuwen and Li Weige.
Tip: This standard content only shows part of the intercepted content of the complete standard. If you need the complete standard, please go to the top to download the complete standard document for free.