[Evaluation of proficiency test with
low number of participants:a case study on the testing of lead and nitrite in water].
WEI SHENG YAN JIU = JOURNAL OF HYGIENE RESEARCH 2021;
50:653-659. [PMID:
34311839 DOI:
10.19813/j.cnki.weishengyanjiu.2021.04.019]
[Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Subscribe] [Scholar Register] [Indexed: 06/13/2023]
Abstract
OBJECTIVE
The gross error of the assigned value-classical statistical method was developed that all participants in proficiency testing activities could obtain unbiased evaluation in case of a low number of participants(& lt; 17).
METHODS
This developed method was employed to evaluate the testing result of nitrite-N and lead in water from participants along the "Belt and Road", which was organized by CNCA in 2019.
RESULTS
Use the specified value gross error-classical statistical method to evaluate the feedback result of 29(15+14) participants with non-normal distribution, 4 out of 15 participants in "Lead in Water" get a "unsatisfied", 5 out of the 14 participants in "Nitrite in Water(as Nitrogen)"get a "unsatisfactory".
CONCLUSION
When the robust statistics and classical statistical method are unable to make objective evaluations respectively, this method could mildly remove outliers and avoid the extreme evaluation of "all satisfied"or "all dissatisfied", which truly reflected the real ability of each participant and could be applied to the proficiency testing activity with low number of participants.
Collapse