94 / 154
© 2005 AOAC INTERNATIONAL
AOAC O
FFICIAL
M
ETHODS OF
A
NALYSIS
(2005)
I
NTERLABORATORY
C
OLLABORATIVE
S
TUDY
Appendix D, p. 11
Appendix 2. Critical values for the Grubbs extreme
deviation outlier tests at the 2.5% (2-tail), 1.25% (1-tail)
rejection level, expressed as the percent reduction in the
standard deviations caused by removal of the suspect
value(s) (
see
text for calculating the Grubbs statistics)
L = number of laboratories at a given level (concentration)
L
One highest or
lowest
Two highest or
two lowest
One highest
and one lowest
4
86.1
98.9
99.1
5
73.5
90.3
92.7
6
64.0
81.3
84.0
7
57.0
73.1
76.2
8
51.4
66.5
69.6
9
46.8
61.0
64.1
10
42.8
56.4
59.5
11
39.3
52.5
55.5
12
36.1
48.5
51.6
13
33.8
46.1
49.1
14
31.7
43.5
46.5
15
29.9
41.2
44.1
16
28.3
39.2
42.0
17
26.9
37.4
40.1
18
25.7
35.9
38.4
19
24.6
34.5
36.9
20
23.6
33.2
35.4
21
22.7
31.9
34.0
22
21.9
30.7
32.8
23
21.2
29.7
31.8
24
20.5
28.8
30.8
25
19.8
28.0
29.8
26
19.1
27.1
28.9
27
18.4
26.2
28.1
28
17.8
25.4
27.3
29
17.4
24.7
26.6
30
17.1
24.1
26.0
40
13.3
19.1
20.5
50
11.1
16.2
17.3
Source: Both tables were calculated by R. Albert (October 1993) by computer
simulation involving several runs of approximately 7000 cycles each for each
value, and then smoothed. Although the table of
Appendix 1
is strictly
applicable only to a balanced design (same number of replicates from all
laboratories), it can be applied to an unbalanced design without too much error,
if there are only a few deviations.
Appendix 1. Critical values for the Cochran maximum
variance ratio at the 2.5% (1-tail) rejection level, expressed as
the percentage the highest variance is of the total variance
L = number of laboratories at a given level (concentration)
r = number of replicates per laboratory
L
r = 2
r = 3
r = 4
r = 5
r = 6
4
94.3
81.0
72.5
65.4
62.5
5
88.6
72.6
64.6
58.1
53.9
6
83.2
65.8
58.3
52.2
47.3
7
78.2
60.2
52.2
47.3
42.3
8
73.6
55.6
47.4
43.0
38.5
9
69.3
51.8
43.3
39.3
35.3
10
65.5
48.6
39.9
36.2
32.6
11
62.2
45.8
37.2
33.6
30.3
12
59.2
43.1
35.0
31.3
28.3
13
56.4
40.5
33.2
29.2
26.5
14
53.8
38.3
31.5
27.3
25.0
15
51.5
36.4
29.9
25.7
23.7
16
49.5
34.7
28.4
24.4
22.0
17
47.8
33.2
27.1
23.3
21.2
18
46.0
31.8
25.9
22.4
20.4
19
44.3
30.5
24.8
21.5
19.5
20
42.8
29.3
23.8
20.7
18.7
21
41.5
28.2
22.9
19.9
18.0
22
40.3
27.2
22.0
19.2
17.3
23
39.1
26.3
21.2
18.5
16.6
24
37.9
25.5
20.5
17.8
16.0
25
36.7
24.8
19.9
17.2
15.5
26
35.5
24.1
19.3
16.6
15.0
27
34.5
23.4
18.7
16.1
14.5
28
33.7
22.7
18.1
15.7
14.1
29
33.1
22.1
17.5
15.3
13.7
30
32.5
21.6
16.9
14.9
13.3
35
29.3
19.5
15.3
12.9
11.6
40
26.0
17.0
13.5
11.6
10.2
50
21.6
14.3
11.4
9.7
8.6
Cochran statistic = (largest individual within-laboratory variance)/(sum of all the
within-laboratory variances).