Periodic evaluation of trends in land treatment and
water quality serves to trackprogress andprovide
information for potential refinements. Statistical
analysis with formal hypothesis testing strength-
ens the quantitative evaluation ofprogress.
A detailed preliminary analysis using scatter plots and statistical tests of assump-
tions and the properties of the data set such as the distribution, homogeneity in
variance, bias, independence, etc., precede formal hypothesis testing and statisti-
cal analysis. From the objective and the properties of the data set, the appropriate
statistical test may be chosen to determine a trend, impact, or causality.
For trend detection, some of the appropriate tests include Student's t-test, linear
Failure to observe improve-
regression, time series, and nonparametric trend tests. For an assessment of
ment may mean that the
impact, a careful tracking of treatment is required and the two sample Student's
t-test, linear regression, and intervention time series are appropriate statistical
problem is not carefully
tests. Evidence from experimental plot studies, edge-of-field pollutant runoff
monitoring, and modeling studies may be used to support the conclusion of
action is not directed properly,
the strength of the treatment is
Failure to observe improvement may mean the problem was not carefully
inadequate, the monitoring
documented, management action was not directed properly, the strength of the
treatment was inadequate, the monitoring program was not sensitive enough to
program is not sensitive
detect change, orasufficient time has not elapsed to develop the expected changes.
enough to detect change, or
A mid-course evaluation, if conducted early enough, provides an opportunity for
more time is needed.
modifications in project goals or monitoring design.
Changes in sampling design may not be worthwhile unless a sufficiently long time
series can be monitored in a consistent fashion. A power analysis may determine
that too many samples are being taken and the number could be reduced to save
money if the monitoring objectives can be met with fewer samples. If some
variables are unneeded (they no longer support objective or no longer support a
modified objective) or some stations do not provide sufficient additional informa-
tion, then they can be dropped.