17 laboratories took part in the experiment from 10 European countries. The laboratories have been given numerical codes. For the purposes of this report they have also been assigned letter-codes (because single-character codes are needed in the graphs).
Samples of three coarse aggregates were prepared and distributed by Partner 3. The aggregate used for Level 1 was an 8/11 mm crushed gravel which meets the requirements of Category B in Table A below. This was a production material. Level 2 was another 8/11 mm gravel, prepared by mixing rounded gravel into the aggregate used for Level 1, to give a material corresponding to Category D in the table. Level 3 was a mixture of three crushed gravels chosen to give an 8/16 mm aggregate corresponding to Category C.
The samples were prepared, for each level of the experiment, as laboratory samples from one bulk sample. A rotary divider was used to reduce each bulk sample in several stages until sub-samples of the size required for laboratory samples for the experiment had been obtained. Hence the measures of repeatability and reproducibility given by the experiment are consistent with the definitions of r1 and R1 used here.
|Percentage of crushed or broken particles|
(including totally crushed or broken particles)
|Percentage of totally rounded particles||Category|
|90 to 100||0 to 3||B|
|50 to 100||0 to 10||C|
|50 to 100||0 to 30||D|
|No requirement||No requirement||E|
Where a participant failed to report a determination, the missing value is shown as "-.-" in the data tables.
The test method requires test results to be rounded to the nearest 1 %. However, for the purpose of the cross-testing experiment, the test results were calculated to the nearest 0.1 %. This was to prevent rounding of the data affecting the assessment of the repeatability and reproducibility of the test method.
Only three participants (Laboratories 76, 106, and 148) carried out the test correctly. They first separated their test portions into "crushed particles" and "rounded particles", and weighed these groups so that they could calculate the percentages of "crushed particles" and "rounded particles". Then they separated out the "totally crushed particles" and the "totally rounded particles" and weighed these groups, so that they could calculate the percentages of "totally crushed particles" and "totally rounded particles".
Thirteen other laboratories carried out the test incorrectly. They first separated their test portions into "crushed particles" and "rounded particles". Then they separated out the "totally crushed particles" and the "totally rounded particles", so that they had four groups of particles, which they then weighed. This produced absurd results, for example, for some test portions, the percentage of "totally crushed particles" exceeded the percentage of "crushed particles", or the percentage of "totally rounded particles" exceeded the percentage of "rounded particles".
The other laboratory (128) derived the percentage of "crushed particles" correctly, but the percentage of "rounded particles" incorrectly.
However, it was possible to calculate the correct results from the masses reported, so corrected data for all laboratories are given here.
When so many participants misunderstand the test method, the fault must lie with the method description, not with the participants. The test method description could be clarified, for example, by referring to the separation of the particles into "crushed" and "rounded" groups as "Stage 1" of the test, and the further separations as "Stage 2", and then describing the calculations for "Stage 1" and "Stage 2" separately.
Laboratory averages are used to calculate the reproducibility of the test method, and to assess laboratory biasses. Between-test-portion ranges are used to calculate the repeatability of the test method, and to assess the repeatability of tests from individual laboratories. The averages and ranges are shown in the histograms, and the laboratory averages are plotted in the Mandel plots.
The averages and ranges are also used to test for stragglers and outliers. Where these have been found, they are indicated throughout using a single question mark (?) to indicate a straggler, and a double question mark (??) to indicate an outlier.
Standardised values of the averages and ranges are shown in the Mandel plots. These figures are used to identify laboratories that give rise to large laboratory biasses, or large between-test-portion ranges, in more than one level of an experiment. The horizontal lines in these graphs show the critical values of the "h" and "k" statistics at the 5 % and 1 % significance levels, taken from the ISO standard on precision (ISO 5725, 1994).
One of the Mandel plots shows the standardised statistics for the determination of the percentage of crushed particles. Because Cc + Cr = 100, the standardised statistics for the determination of the percentage of rounded particles would be identical to those for the percentage of crushed particles, but with the signs of the laboratory biasses reversed.