180
10 Interlaboratory Comparisons
clear description of the used scores. However, it is also important to highlight that
the laboratory, when evaluating its participation in the given comparison, should
critically evaluate the extent to which the criteria used correspond to the expected
requirements regarding the scope of the activity.
Example
In the ILC aiming for determination of a selected substance in soil, there were 25
participants. Of these, 20 submitted their results with an assigned uncertainty. In this
case, the results provided by the laboratory were compared to the certified reference
values, since the test item was a CRM. NOTE: the results listed in the table are raw
data as reported by laboratory, without rounding.
The certified content of the substance of interest is 0.51 mg/L ± 0.048 mg/L,
for k 2.
The standard uncertainty of the reference value is u ref 0.024 mg/L.
The calculated standard deviation of the mean value from the results of all
participants is σ 0.034 mg/L.
Results submitted by the participants were statistically evaluated according to
ISO/DIN 13528 standard ‘Statistical Methods For Use in Proficiency Testing by
Interlaboratory Comparisons.’ The results submitted by the laboratories and the
scores for each laboratory are shown in Table 10.1.
Column I: ID of the laboratories participating in the comparison.
Column II and III: the results submitted by the laboratories (value and uncertainty,
respectively).
Column IV–VII: selected scores.
NOTE! Values of the zeta score and E n score were calculated only for those
laboratories that submitted their results with an assigned uncertainty.
10.12.1 Pay Attention to the Results of the Laboratory
with ID 22
When z and z
-scores were used, in both cases their values indicated unsatisfactory
performance of the laboratory 22.
When zeta score and E n score were used, in both cases their values indicated
satisfactory performance of the laboratory 22 (zeta < 2; E n < 1).
With this example, taken from real practice, considering the results reported by
participating laboratories, it can be seen how the selection of adequate evaluation
criteria (scores) is essential for the evaluation of a given laboratory. The example of
laboratory 22 shows that the evaluation of the result on the basis of indicators that
only take into consideration the uncertainty assigned to the reference value is harsher
10 Interlaboratory Comparisons
clear description of the used scores. However, it is also important to highlight that
the laboratory, when evaluating its participation in the given comparison, should
critically evaluate the extent to which the criteria used correspond to the expected
requirements regarding the scope of the activity.
Example
In the ILC aiming for determination of a selected substance in soil, there were 25
participants. Of these, 20 submitted their results with an assigned uncertainty. In this
case, the results provided by the laboratory were compared to the certified reference
values, since the test item was a CRM. NOTE: the results listed in the table are raw
data as reported by laboratory, without rounding.
The certified content of the substance of interest is 0.51 mg/L ± 0.048 mg/L,
for k 2.
The standard uncertainty of the reference value is u ref 0.024 mg/L.
The calculated standard deviation of the mean value from the results of all
participants is σ 0.034 mg/L.
Results submitted by the participants were statistically evaluated according to
ISO/DIN 13528 standard ‘Statistical Methods For Use in Proficiency Testing by
Interlaboratory Comparisons.’ The results submitted by the laboratories and the
scores for each laboratory are shown in Table 10.1.
Column I: ID of the laboratories participating in the comparison.
Column II and III: the results submitted by the laboratories (value and uncertainty,
respectively).
Column IV–VII: selected scores.
NOTE! Values of the zeta score and E n score were calculated only for those
laboratories that submitted their results with an assigned uncertainty.
10.12.1 Pay Attention to the Results of the Laboratory
with ID 22
When z and z
-scores were used, in both cases their values indicated unsatisfactory
performance of the laboratory 22.
When zeta score and E n score were used, in both cases their values indicated
satisfactory performance of the laboratory 22 (zeta < 2; E n < 1).
With this example, taken from real practice, considering the results reported by
participating laboratories, it can be seen how the selection of adequate evaluation
criteria (scores) is essential for the evaluation of a given laboratory. The example of
laboratory 22 shows that the evaluation of the result on the basis of indicators that
only take into consideration the uncertainty assigned to the reference value is harsher