Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
20 CHAPTER 5- PRECISION STATEMENT FOR AASHTO T180 5.1 Comparison of the ILS and PSP Precisions In preparing a precision statement for AASHTO T180, the precision estimates from ILS and PSP will be statistically compared. This is to examine if they can be combined or should be reported separately. In the following sections the results from statistical comparison of the precision estimates are discussed: An F-test on variances was conducted to compare the repeatability and reproducibly standard deviation from the ILS and PSP data. The comparison of the standard deviations of maximum dry density from Table 3-7 and Table 4-4 indicates that both single-operator and multilaboratory precisions of the ILS coarse-graded materials are larger than those of the PSP fine-graded soil. This is expected since compaction of coarse-graded materials encounters more resistance than compaction of fine-graded materials, which results in more variability in maximum dry density measurements. The results of the F-test are shown in Table 5-1. As indicated in the table, the repeatability of the maximum dry density from ILS is significantly larger than that from PSP (compare computed F value of 1.31 with the critical F of 1.28). Although not quite significant, the reproducibility of the maximum dry density from ILS is larger than that from the PSP (compare computed F of 1.25 with critical F of 1.28). Therefore, the precision estimates for maximum density would be presented separately in the precision statement of AASHTO T180. The results of an F test on comparison of variability of the optimum moisture content values from PSP and ILS from Table 3-7 and Table 4-4 is provided in Table 5-1. The comparison of single-laboratory and multilaboratory variability of the optimum moisture content measurements indicates that both within and between-laboratory precisions of the ILS samples are significantly smaller than those of PSP samples (compare computed F values of 1.47 and 2.26 with the critical F value of 1.33). The reason for this could be the high clay content of PSP blends. Due to the difference in the repeatability and reproducibility statistics of ILS and PSP, the precision estimates for optimum moisture content would be presented separately in the precision statement of AASHTO T180.
21 Table 5-1. Summary of Statistics for comparison of maximum dry density (lb/cu. ft.) and optimum moisture content (%) of PSP and ILS samples Compare Degrees of Freedom Critical F Repeatability Reproducibility Computed F (S Decision r) Computed F (S Decision R) Maximum Dry Density 87 & 635 1.28 1.25 Accept 1.31 Reject Optimum Moisture Content 635 & 87 1.33 1.47 Reject 2.26 Reject Note: The critical F values are for 95% level of confidence for one tailed test. 5.2 Proposed Precision Estimates for AASHTO T180 Table 5-2 provides the precision estimates for maximum dry density and optimum moisture content measurements resulted from the ILS and PSP data. As indicated earlier, based on the significant difference between the repeatability and reproducibility statistics of maximum density and optimum moisture content computed from ILS and PSP data, the precisions from the two sources will be presented separately in a proposed precision statement of AASHTO T180 presented in Appendix E. Table 5-2. Repeatability and reproducibility standard deviations of maximum dry density and optimum moisture content of sandy clay (PSP) and sand and gravel (ILS) materials Property Material Repeatability Standard Deviation Reproducibility standard Deviation Maximum Dry Density Sandy Clay 0.65 1.55 Sand and Gravel 0.73 1.76 Moisture Content Sandy Clay 0.37 0.74 Sand and Gravel 0.30 0.49
