8. Report
8.1 Report as the color the number of the standard that most nearly matches the specimen. In the event that the color lies midway between two standards, report the darker of the two.
8.2 If, owing to differences in hue between the specimen and the standards, a definite match cannot be obtained, report the range over which an apparent match is obtained, and report the material as "off-hue".
9. Precision
9.1 Color Samples with Pt-Co Color Greater than 25:
9.1.1 These precision statements are based upon an inter-laboratory study in which six platinum-cobalt standards having values of 25, 75, 170, 265, 385, and 475 were prepared in accordance with the instructions given in Section 6 of this test method and were given coded labels. These solutions were tested by one analyst in each of ten different laboratories making a single observation on one day and then repeating the observation on a second day. The analysts were requested to estimate the color to the nearest one unit for solutions below 40 platinum-cobalt, to the nearest five units for solutions between 40 and 100 platinum-cobalt and to the nearest ten units for solutions above 100 platinum-cobalt. Based on the results of this interlaboratory study, the following criteria, calculated according to RR-D02-1007, should be used for judging the acceptability of results at the 95 % confidence level when the results are obtained under optimum conditions where the hue of the sample matches exactly the hue of the standards. Poor precision will be obtained in varying degrees as the hue of the sample departs from that of the standards.
9.1.1.1 Repeatability - Two results, obtained by the same analyst should be considered suspect if they differ by more than:
r = 0.027 (X + 92) platinum-cobalt units
where X is the average of the two results.
9.1.1.2 Reproducibility - Two results, obtained by analysts in different laboratories, should be considered suspect if they differ by more than: R = 0.087 (X + 92) platinum-cobalt units where X is the average of the two results.
9.1.1.3 Table 4 shows precision values for samples with Pt-Co Color of greater than 25.
9.2 Color Samples with Pt-Co Color Less than 25:
9.2.1 The results of two interlaboratory studies were pooled to give precision values calculated according to RR-D02-1007. One study of glycols included 4 samples and 7 laboratories; the other study included 5 samples and 6 laboratories. Based on the pooled repeatability and reproducibility standard deviations, the following criteria should be used for judging, at the 95 % confidence level, the acceptability of results obtained on samples with less than 25 Pt-Co Color.
9.2.1.1 Reapeatability - Two results, each the mean of duplicates, obtained by the same operator on different days should be considered suspect if they differ by more than two platinum-colbalt units.
9.2.1.2 Reproducibility - Two results, each the mean of duplicates, obtained by operators in different laboratories, should be considered suspect of they differ by more than seven platinum-cobalt units.
9.3 Bias - The test procedure has no bias because the value of the test result is defined only in terms of the test method.
10. Keywords
10.1 clear liquids; color; platinum-cobalt color scale