284 / 363
1700
Martin & Campos-Gim
É
nez:
J
ournal of
AOAC I
nternational
V
ol.
98, N
o.
6, 2015
Results and Discussion
Part 1
Sixteen laboratories initially agreed to participate in the
collaborative study. One laboratory dropped out during part 1
due to issues related to availability of resources. One laboratory
did not qualify for the second part of the collaborative study due
to results out of the range of expected levels for both practice
samples.
Results for the two practice samples can be found in Table 1.
The nonqualified laboratory’s results were not taken into account.
Repeatability was 2.1 and 2.5%, respectively; reproducibility was
5.3 and 6.0%, respectively. HorRat values were below 1, at 0.59
and 0.68. As this precision estimate is calculated on nonblinded
duplicates, it therefore cannot be used for the actual collaborative
study. Several minor comments were addressed to the Study
Director and were taken into account in the final version of the
method, which is presented in this paper.
Part 2
Fourteen laboratories sent a complete set of results. The
statistical evaluation can be found in Table 2. Precision results
(repeatability and reproducibility) are well within the limits
stated in the
Standard Method Performance Requirements
(SMPR
®
) 2012.009 (10). Repeatability ranged from 1.3 to
3.3%, and reproducibility ranged from 4.1 to 7.0%. HorRat
values were all below 1, from 0.33 to 0.69.
Laboratory 14 was identified as an outlier for the product
“Adult Nutritional Powder Milk Protein Based.” Laboratory 13
was an outlier for the products “Infant Formula Powder Soy
Table 3. Full set of data, part 1. All results are given in mg/100 g
Lab No.
Adult nutritional RTF
high fat
SRM 1849a
Child formula powder
Adult nutritional powder
milk protein based
Infant formula powder
soy based
1
2.05
2.09
7.01
6.80
5.61
5.84
2.51
2.46
5.14
5.22
2
2.00
1.91
6.66
6.78
5.93
6.03
2.64
2.63
4.86
4.82
3
2.26
2.18
7.66
7.34
6.22
6.12
2.80
2.78
5.56
5.48
4
1.71
1.80
7.01
6.93
6.00
5.84
2.66
2.64
4.82
5.23
5
1.96
1.89
6.90
6.83
5.88
5.66
2.49
2.48
4.67
4.97
6
1.98
2.07
6.91
7.08
5.75
5.91
2.64
2.66
5.15
5.08
7
1.99
2.00
6.82
6.70
5.77
5.84
2.53
2.61
4.85
4.90
8
2.14
2.15
7.11
7.43
6.64
6.17
2.68
2.87
5.35
5.21
9
2.13
1.99
6.56
6.52
5.68
5.62
2.52
2.48
4.83
4.93
10
2.13
2.08
6.36
6.77
5.78
5.96
2.55
2.60
4.74
5.15
11
2.18
2.17
6.82
6.79
5.76
5.74
2.44
2.39
4.85
4.98
12
2.22
2.34
6.66
6.71
5.56
5.82
2.59
2.51
4.91
4.83
13
2.02
2.13
7.81
7.61
6.26
6.77
2.77
2.85
6.23
a
5.19
a
14
2.16
2.27
7.12
7.09
6.01
6.18
2.34
a
2.67
a
5.20
5.26
a
Outlier identified by the Cochran test.
Table 4. Full set of data, part 2. All results are given in mg/100 g
Lab No.
Infant formula RTF
milk based
Adult nutritional powder
low fat
Adult nutritional RTF
high protein
Infant elemental powder
Infant formula powder part
hydrolyzed soy based
1
0.536
0.558
7.96
8.01
1.54
1.52
6.35
6.25
3.68
3.63
2
0.533
0.530
8.26
8.02
1.49
1.46
6.22
6.35
3.95
4.01
3
0.589
0.592
8.69
8.67
1.67
1.68
7.18
7.22
4.19
4.17
4
0.554
0.548
8.21
8.00
1.58
1.58
6.09
6.78
3.92
3.94
5
0.547
0.529
7.73
7.39
1.37
1.41
6.74
6.28
3.46
3.54
6
0.558
0.550
8.24
8.18
1.61
1.59
6.87
6.69
3.81
3.83
7
0.502
0.515
7.96
8.05
1.55
1.48
6.57
6.61
3.95
3.81
8
0.540
0.550
8.12
7.90
1.62
1.59
6.62
7.00
3.71
3.73
9
0.527
0.529
7.92
7.84
1.54
1.50
6.38
6.27
3.58
3.66
10
0.573
0.556
8.41
8.25
1.59
1.57
6.93
6.85
3.91
3.99
11
0.534
0.548
7.75
8.07
1.62
1.68
6.64
6.58
3.81
3.80
12
0.585
0.577
7.59
7.57
1.69
1.66
6.33
6.34
3.63
3.69
13
0.560
a
0.670
a
7.13
a
8.55
a
1.64
a
1.45
a
7.00
7.59
4.18
4.03
14
0.554
0.554
8.41
8.55
1.62
1.66
6.60
6.85
4.06
4.05
a
Outlier identified by the Cochran test.
Candidates for 2016 Method of the Year
283