Montgomery and Loftis (1987) have listed several situations for which the t-test does not accurately reproduce Type I errors, and should therefore be avoided. Characteristics common to water quality data (skewness or other non-normality, presence of outliers and less-thans) also reduce the power of the t-test, in relation to nonparametric alternatives. Thus if one is interested in reaching correct decisions when trends or differences exist, and not just when they do not, the t-test should not be considered “robust” (in the sense of being generally applicable) when its assumptions are violated. Further, t-tests assume that differences in means are relevant (the mean is a good measure of central tendency), and that data groups differ by some additive amount. When all of these assumptions are recognized, and in light of the availability of truly robust and comparatively powerful non-parametric alternatives, we believe there is little applicability of the t-test for detecting trends or differences in water quality variables. 相似文献
Complications arise when ratios are used to present environmental data because ratios are an unbounded, multiplicative scale that can lead to asymmetrical (skewed) data distributions. Enantiomeric ratios (ERs), historically used in discussions of chiral signatures, often are published as mean ER+/-single-value standard deviation. Application of statistical summaries, such as the widely used sample mean and standard deviation, to skewed ratio data is misleading and often inappropriate. Comparison of statistically summarized ER and enantiomer fraction (EF) data (which are based on a bounded, additive scale) for a range of hypothetical values reveals substantial discrepancies when conversion between ER and EF formats is used. These discrepancies are largest when the ratio data are greater than one and have large variability, because the data are more skewed. In many cases, the use of fractions instead of ratios can help to minimize misrepresentation of environmental data, including chiral data. The use of nonparametric statistical summaries, e.g., median and percentiles, provides a more robust indicator of the typical value and spread for both ER and EF data. 相似文献
The most commonly used method in environmental chemistry to deal with values below detection limits is to substitute a fraction of the detection limit for each nondetect. Two decades of research has shown that this fabrication of values produces poor estimates of statistics, and commonly obscures patterns and trends in the data. Papers using substitution may conclude that significant differences, correlations, and regression relationships do not exist, when in fact they do. The reverse may also be true. Fortunately, good alternative methods for dealing with nondetects already exist, and are summarized here with references to original sources. Substituting values for nondetects should be used rarely, and should generally be considered unacceptable in scientific research. There are better ways. 相似文献
ABSTRACT: The distribution of sediment physical characteristics, sediment phosphorus (P) pools, and laboratory‐based rates of P release from the sediments were used to identify regions and dosage for alum treatment in Wind Lake, Wisconsin. Using variations in sediment moisture content, we identified an erosional zone at depths < 1.4 m and an accumulation zone at depths > 2.6 m. Mean concentrations of porewater P, loosely‐bound P, iron‐ and aluminum‐bound P, and mean rates of P release from sediments under anoxic conditions were high in the accumulation zone compared to sediment P characteristics in the erosional zone, indicating focusing of readily mobilized sediment P pools from shallow regions and accumulation to deep regions. We determined that a future alum treatment for control of internal P loading would be most effective at depths > 2.6 in the accumulation zone. The mean rate of anoxic P release from sediments encountered in the accumulation zone (8.3 mg m‐2 d‐1) was used in conjunction with a summer anoxic period of 122 d, and a treatment area of 1.6 km2 to estimate an internal P load of 1,600 kg to be controlled. Our results suggest that an understanding of the distribution of sediment P pools and P fluxes in lakes provides a strategy for estimating alum dosage and application areas. 相似文献