8. Reporting Analytical Data

Effective Date: 07/2008
Point of Contact: Quality Engineer for Regulatory Programs

Data are generated using appropriate procedures or test plans that are prepared in accordance with Section 7 of this Plan. All data shall be traceable to the M&TE and procedure (including procedure revision) or test plan used, and if the reported results are quantitative, a valid calibration.

Data are generated by the analyst and certified to be correct to the best of the analyst's knowledge, with the certification documented by a dated signature or initials on the data. If the document is paginated (e.g., page 1 of 3) or if the document indicates the range of time associated with the activity (such as an analytical run), the analyst's signature may appear on either the first or last page of the document. If the document does not contain page numbers or does not indicate the range of time associated with the activity, the analyst's signature or initials and the date are required on each page of the document. If the results are printed by the instrument and include the identification of the analyst and date, then additional signatures are not required.

If errors are found during the generation of the data, the analyst corrects the errors by placing a single line through the incorrect information, adding the correct information if necessary and initialing and dating the correction. An explanation is required if the reason for the correction is not obvious.

Significant Figures

Significant figures reflect the accuracy and precision of a given result. A result should always be rounded to the number of figures that are consistent with the confidence that can be placed on it. Thus, the number of significant figures is the number of digits remaining after the data is rounded. Reported values should contain only significant figures. A value is made up of significant figures when it contains all digits known to be true and one last digit in doubt (e.g., if a value is reported as 18.8 mg/l, the 18 must be firm while the 0.8 is somewhat uncertain, but presumably better than one of the values 0.7 or 0.9 would be).

Final zeros after a decimal point are always meant to be significant figures. For example, 9.800 g would be considered as having 4 significant figures. Zeros before a decimal point with nonzero digits preceding them are assumed to be significant. With no preceding nonzero digit, a zero before the decimal point is not considered significant. If there are no nonzero digits preceding a decimal point, the zeros after the decimal point but preceding other nonzero digits are not considered to be significant. These zeros only indicate the position of the decimal point. Final zeros in a whole number may or may not be significant (e.g., in a conductivity measurement of 1,000 µmho/cm, there is no implication by convention that the conductivity is 1,000 ± 1 µmho. Rather, the zeros only indicate the magnitude of the number).

A good measure of the significance of one or more zeros interspersed in a number is to determine whether the zeros can be dropped by expressing the number in exponential form. If they can, the zeros may not be significant (e.g., no zeros can be dropped when expressing a weight of 100.08 g in exponential form; therefore, the zeros are significant. However, if a weight of 0.0008 g can be expressed in exponential form as 8 x 10⁻⁴ g, then the zeros are not significant).

Once the number of significant figures obtainable from a given type of analysis is established, data resulting from such analyses are reduced according to set rules for rounding (Handbook for Analytical Quality Control in Water and Wastewater Laboratories, EPA-600/4-79-019, March 1979) which are summarized below.

Data reported shall conform to the significant figure guidance provided above. An exception is permitted when vendor-supplied software or in-house spreadsheets make application of the policy impractical. In such cases, any limitations associated with the results shall be documented in the client report.

Rounding Off Numbers

Rounding off shall be applied only as described as follows:

If the figure following those to be retained is less than 5, the figure is dropped, and the retained figures are kept unchanged (e.g., 11.443 is rounded off to 11.44).
If the figure following those to be retained is greater than 5, the figure is dropped, and the last retained figure is raised by 1 (e.g., 11.446 is rounded off to 11.45).
If the figure following those to be retained is 5, and if there are no figures other than zeros beyond the five, the figure 5 is dropped, and the last-place figure retained is increased by one if it is an odd number or it is kept unchanged if an even number. As an example, 11.435 is rounded off to 11.44, while 11.425 is rounded off to 11.42 (Handbook for Analytical Quality Control in Water and Wastewater Laboratories, EPA-600/4-79-019, March 1979).
When reporting radiochemical uncertainties, always round up.
When reporting less than results, always round up (i.e., rounding at the detection limit should not result in a reported value falling below the true detection limit).

Numerical Values and Data Qualifiers

Inorganic or organic results shall be reported as numeric values with appropriate data qualifiers if above the instrument detection limit (IDL). If the value is less than IDL, it can be reported as undetectable.

Radiochemical results shall be reported based on calculated concentration or activity values (whether negative, positive, or zero) using the appropriate blank for each nuclide.

When data flags are used, they shall be defined in the client report. The following data flags shall be used:

U = Analyte was analyzed but not detected
J = Estimated quantity used for detection below EQL but above IDL, or for tentatively Identified compounds
B = Analyte found in associated laboratory blank as well as sample
E = Analyte exceeds upper quantitation limit (UQL)
N = Tentatively identified compound
H = Analysis was performed after the allowed analytical holding time had elapsed or analysis date is not available
X = A significant quality assurance deficiency was associated with the reported result
Y = User to define; definition stated with results
Z = User to define; definition stated with results

Analytical Data Reviews

A Staff Member other than the person performing the work and knowledgeable of the area being reviewed, reviews the data before reporting results. The review consists of the following elements:

evaluation of method and QC performance
compliance with technical and QC requirements as defined in project documents
accurate transcriptions
correct calculations
overall consistency and reasonableness of data
appropriate corrective action, if necessary.

Note: A check for accurate transcriptions is accomplished by review of the raw data and the translation into final results. Accuracy of the calculations is verified by a review of computer generated output and verification of hand calculations. Only a fraction (e.g., 10 percent) of the hand calculations need to be verified if there are a large number of analyses performed or a large number of analytes per analysis. Data generated by software shall be verified in accordance with the Software and Safety Software subject areas.

Data selection may be made on the hardcopy printout by marking the data, which may include circled values, checkmarks, and words such as "OK", "do not report", and "report this". These markings are part of the data selection process and are not considered corrections. If a Staff Member makes data selection markings as part of the data review, they shall sign (or initial) and date the hardcopy printout as indicated for the data reviewer below. If the Staff Member makes data selection markings that are not part of the data review, then they shall sign (or initial), date, and make a notation on the hardcopy printout that they have selected data to be reported. This signature shall be distinct from the analyst and reviewer signatures.

Staff correct identified errors in accordance with the following:

Errors found during the generation of analytical data are corrected by placing a single line through the incorrect information, adding the correct information if necessary, and initialing and dating the correction. An explanation is required if the reason for the correction is not obvious. The reviewer's initials or signature applied during data review indicates that the technical reviewer agreed with the correction.
If the error is detected during data review or after the report is sent to the client, the data is forwarded to the originator for correction. If the originator is not available, the correction must be made by someone cognizant of the analytical data and the process used to generate it.
If the error is of an editorial nature, which is determined to be non-quality affecting, a correction can be made with a single line, initial, and date by the individual who identifies the error.
Corrections affecting results are forwarded to the client. The reviewer is responsible for issuing a Quality Problem Report (see the Corrective Actions section) if the quality of the data is inadequate or indeterminate and, for ensuring the final report narrative clearly identifies deviations and deficiencies.

The technical reviewer's signature or initials attesting to the above review appear with the data reviewed and must be clearly distinct from the analyst's signature.

If preliminary data (hard copy or electronic) is being reported to a client, project staff ensure the following actions take place:

the data is marked "Preliminary" or "Draft"
the data is technically reviewed
a copy of the preliminary report is maintained in the project records.

The project staff prepares analytical data reports. The project manager prepares the final report to the client.

The following information must be included in the final report as a minimum:

lab name and address
client name
client sample identification
laboratory sample identification
date of sample receipt, sample preparation, and analysis
method/procedure references
identification of amended test results (if applicable)
QC results
data qualifiers with definitions (if used)
detection limits
results, units, and associated uncertainties for radiochemistry
summary of deviations
information indicating limitations of data (may include estimates of analytical precision, accuracy, detection limits, and similar information or special observations to assist clients' interpretation of the data)
date of report
typed or printed name, including title, and dated signature of project manager
narrative (required whenever customer QC requirements are not met).

When data validation is required by the client, the project manager also includes the following in the final report:

cover letter summarizing the work performed (if more than one analysis was performed)
table of contents for the report
a narrative describing known limitations (always required for a full deliverable report)
preparative or bench sheets and run logs, as applicable to the method
category 1 and category 2 M&TE calibration and continuing calibration information
field and internal laboratory chains-of-custody
traceability of standards to a nationally recognized source
raw data.

The project manager reviews and signs all final reports before submission to the client. The project manager's signature attests to the fact that the report is complete.

Records of all reviews conducted shall be included in the project records.

The project manager transmits the final data report to the client.

If electronic data and/or reports are being submitted to the client, the implementing project manager ensures the following actions take place:

the data and/or report is technically reviewed and the name of the reviewer appears within the document
a hard copy of the electronic data and/or report is maintained in the project records.

Note: Only hard copies of the data and/or report are considered final. As a result, the client must be provided a final hard copy of the data and/or report.

References

Software subject area

Safety Software subject area