Basic HTML Version
24
both the micro- and macro-scale by addressing both how well measurements taken within a sampling unit
reflect that unit and the degree to which measurements from a set of sampling units represent the
population of interest.
Representativeness is usually considered a qualitative term. The basic questions to be answered are
whether the individual measurements of the characteristics of interest accurately reflect the conditions in
the sampling unit, and whether an adequate number of units were measured to reflect the population of
interest. It is addressed primarily in the sample design, through the selection of sampling sites and
procedures that reflect the test goals and the environment being sampled, i.e., the Mobile Test Platform.
A review of the results of quality assessment samples such as field duplicates (collocated samples), splits,
or other replicates also is performed. It is ensured in the laboratory through (1) the proper handling,
homogenizing, compositing, and storage of samples and (2) analysis within the specified holding times so
that the material analyzed reflects the material collected as accurately as possible.
Completeness
Completeness is a measure of the amount of valid data obtained from a measurement system as compared
to the amount needed to ensure that the uncertainty or error is within acceptable limits. It is a measure of
how well a sampling and analysis design was implemented. It is expressed as follows:
%C = (M
v
) x 100
(M
p
)
where M
v
= number of valid measurements
M
p
= number of planned measurements
The goal for data completeness is 100%. Events that may contribute to reduction in measurement
completeness include sample container breakage and laboratory equipment failures. Samples are
considered invalid if they are contaminated, fail to meet the data quality objectives or other QA protocols,
are lost through sample destruction, are incorrectly collected or analyzed, and/or if there is insufficient
amount of sample for analysis.
The field and laboratory completeness objectives for each BWTS test are determined during test
development and specified in each specific Test Plan. The general completeness criterion for all field
measurements and sample collection is 90 percent, but will be influenced by factors mentioned above. If
the completeness objectives are not achieved for any particular category of data, the MERC PC will
provide documentation why the objective was not met and how the lower percentage impacted the overall
study objectives. If the objectives of the study are compromised, re-sampling or re-measurement may be
necessary. The respective Senior Researcher/Laboratory Director assures the validity of the analytical
measurements reported, and the PC validates the numbers of valid measurements. The completeness
criterion for all laboratory measurements is 95 percent, unless specified differently in a Test Plan.
Comparability
Comparability
is a qualitative measure of the confidence with which one data set can be compared to
another. The key to comparability is consistency of approach, which applies to both the field portion of
the sampling and the laboratory analysis of the samples. In the field, it is addressed primarily in sampling
design through use of comparable sampling. In the laboratory, comparability is ensured through the use
of comparable analytical procedures and ensuring that project staff are trained in the proper application of
the procedures. Within-study comparability is assessed through analytical performance (QC samples).
The assessment of this DQI determines if analytical results being reported are equivalent to data obtained