Environmental Engineering Reference
In-Depth Information
measured lead from the attached device was three times as expensive as RDT
sampling. The measured lead from an attached device (COMP) scored the lowest in
terms of cost because of the amount of time involved in the procedure, which made
it expensive. However, in terms of social interest the most expensive sampling
procedure (COMP) provided the most ideal method of determining lead in drinking
water. (Recall that COMP involves attaching a lead measuring device at the con-
sumer
s drinking water tap.)
Not only is the EU Report
'
flawed in interpreting the statistics, but it also failed to
take into account the highest social opportunity cost or highest social loss associ-
ated with total lead in drinking water. For that, a procedure that can detect the
highest amount of lead should have been chosen. Based on the information pro-
vided in the EU Report (see Fig. 2, EU Report, p. 15), a 6 h stagnation sampling
protocol should have been used since this is the period of stagnation that is
equivalent to the equilibrium lead concentration (i.e. when lead concentration
approaches the saturation level). Since this information is included in the EU Report
in Fig. 2, we cannot understand why this information was ignored in the EU Report
in recommending the appropriate protocol.
As far as reproducibility is concerned, sampling protocols were assessed by
analyzing the relative range which is equivalent to the (max
-
min) /mean. A
−
relative range of zero is ideal (max
min = 0). The 30MS and the FF sampler
performed the best under this criterion while RDT performed the worst. The poor
performance of RDT (in terms of reproducibility) was due to the fact that
stag-
nation time is not controlled for the RDT sample, whereas stagnation time is
controlled for both the FF and 30MS samples
“
(EU Report, p. 52). However,
stagnation time is not controlled for the COMP sample as well; the COMP sample
can be viewed as a group of individual samples with widely varying stagnation
times. Therefore, out of all the protocols, RDT can be best compared to the COMP
in terms of stagnation times since samples for the RDT protocol captured widely
varying stagnation times (as noted above, results from the RDT protocol could be
improved if samples drawn were truly random). The EU Report has also failed to
show the results of
”
for the COMP sample and useful information
is lost, such as comparing reproducibility for COMP with the other protocols.
“
“
reproducibility
”
in the EU Report does not indicate anything about being able to
reproduce the results of a sampling protocol repeatedly over different times, but
rather it is simply a statistic that shows the range of extremes compared to the mean.
We can summarize our critique of the EU Report as follows:
The RDT and FF protocols together capture more elements of consumer behavior
than just a single sampling protocol alone. In fact, statistical results from the EU
report show that the average of RDT and Full Flushed provided the most represen-
tative sample when compared to other protocols. The average of RDT and FF out-
performed the EU
Reproducibility
”
'
s recommended 30MS sampling and RDT. In terms of practicality,
cost-effectiveness and consumer acceptance, RDTand FF protocols were evaluated as
the two best protocols while the 30MS was judged to have been the worst. All these
factors point toward two methods (RDT and FF) which are cost-effective and
Search WWH ::
Custom Search