Introduction to quantitative systematics. Bulletin of the AMNH ; v. 93, article 5

Supplemental Materials
Journal Title
Journal ISSN
Volume Title
New York : [American Museum of Natural History]
"Inasmuch as the frequency distributions of many biological data approach the normal probability curve as developed by statisticians it is often possible for the taxonomist to adopt various statistical measures based on this type of distribution in the systematic analyses of biological material. The ones most commonly employed by the systematist to analyze normally distributed data are those of the central tendency (mean, median, and mode), of variation (standard deviation), of variability (coefficient of variability), and of reliability (standard errors). The mean is an expression of the average tendency in the sample and serves as a point on the variation scale from which the measure of variation can be oriented. The median and mode are measures of central tendency used primarily in comparing the frequency distribution of the sample with the normal curve to reveal possible skewness. The standard deviation is the measure of variation with which the systematist estimates the range of variation in the population from which a particular sample was taken. The population range is calculated as the mean of the sample plus and minus three standard deviations of the sample (M [plus or minus] 3 S.D.) which gives the systematist the range of variation within which would occur approximately 100 per cent of the total population represented by that sample. The coefficient of variability enables the systematist to establish the relationship between the variation and the mean size of the sample, giving the relative variability which can then be used in making comparisons. The standard errors indicate the reliability of the preceding measures and are used to show the systematist the theoretical range of variation of any of these measures, within which range would be found the same measures of additional samples drawn from the same population. With these statistical tools and a knowledge of the biology and distribution of the population, the systematist should be able to establish more accurately the classification status of his samples"--P. [387].
p. 347-388 : ill. ; 26 cm.
Includes bibliographical references (p. 388).