LabCert Program

Program Information

Water Testing

Program Contacts

Administrative Rules

Application Process

Proficiency Testing (PT)

Outreach

Science Services

Proficiency Testing (PT)

fromerly referred to as "Reference Samples"

Review historical laboratory performance.

Reference samples, or proficiency testing (PT) samples, are an integral part of the Wisconsin Laboratory Certification program. Certified and registered laboratories are responsible for ensuring that they meet all of the requirements for the analysis of reference samples. Laboratories are required to pass at least one sample every year for each combination of matrix - technology - analyte (method-analyte for drinbking water)wherever a PT sample is available. A list of Wisconsin's approved PT providers is available. The status of acceptable providers may change. Contact the LabCert program for updated PT sample information.

Enforcement

The WDNR will no longer issue enforcement for successive failures of PTs submitted for accreditation in the aqueous or solid matrices. The WDNR may revoke certifications in the drinking water matrix when a laboratory fails consecutive PTs for any analytical method/analyte or analyte group combination. This action reflects a requirement of the "Manual for the Certification of Laboratories Analyzing Drinking Water".

PT Evaluation

All PT providers use the same criteria when determining whether a result is scored "Acceptable", "Check for Error" or "Not Acceptable".
An "Acceptable" PT result is one where the reported value is within 2 standard deviations of the average reported value for that particular analyte (think of this as the warning limit for PTs).
"Check for Error" is assigned when the reported result is greater than 2, yet less than 3 standard deviations (i.e., the control limit).
A "Not Acceptable" score is when the submitted result exceeds the mean ± 3 standard deviations.

Quantitative regression-based scoring

The lower and upper limits of the acceptable range are calculated using the assigned value and a regression equation. The assigned value is defined as the gravimetric true concentration of an analyte that has been verified by Wisconsin DNR-approved methodologies. Regression equations are found in the USEPA NERL-Ci-0045 "National Standards for Water Proficiency Testing Studies, Criteria Document," December 30, 1998 and the NELAC PT Fields of Testing tables, December 2004. These acceptance criteria were revised, became effective June 1, 2005 and are located on the NELAC website. The mean and standard deviation are calculated using four regression factors (A, B, C, and D):

Mean = (Assigned value × A) + B

Standard deviation (SD) = (Assigned value × C) + D

The acceptable range is calculated using the mean and standard deviation:

Potable water analyte acceptable range = Mean ± 2.0 SD (unless Federal rules require specific acceptance criteria)

Non-potable water analyte acceptable range = Mean ± 3.0 SD (unless Federal rules require specific acceptance criteria)

PT Samples We Cannot Accept

Certain sources and types of PT samples available from the accepted providers are not acceptable for certification and registration. Examples of unacceptable studies are ERA UST samples and "Real World" matrix samples. If you have questions, be sure to contact the Laboratory Certification and Registration Program before ordering the samples.

General Reference Sample Requirements:

Wisconsin has several specific PT sample requirements, which are described below:

  • The study date for the results submitted with applications must be no older than 6 months prior to the date the application is received by the WDNR.
  • Results reported from unapproved methods will not be accepted.
  • Samples for PAHs by liquid chromatography must be at the appropriate concentration levels. Fluorescing analytes (all but the first four or five analytes to elute) should be prepared with true values in the 1-10 ppb range. For the UVdetectable analytes, the true value may be higher.
  • True values cannot be released to the lab before the results are reported.
  • Outliers cannot be included in acceptance limit calculations.
  • Calculated acceptance limits cannot be broader than ± 2.78 standard deviations.
  • Acceptance limits are generally calculated based regression equations developed by the EPA. For some parameters, such as for drinking water (WS studies), specific acceptance criteria (generally ± some percentage of "true" value) are established by law in the Code of Federal Regulations (CFR).
Last Revised: Tuesday December 23 2008