| Abstract: |
Supports the article by Wood et al (1992). Byers suggest that it may be possible to assess the magnitude of the selected mortality acting on a trait such as stature by examining shifts in the distribution of the trait among individuals who die at later (presumably post-selection) ages. This statistical method lacks in statistical power, and also makes a strong presumption that the trait of interest is normally distributed at the outset of the selection process. Otherwise, measures of skew and kurtosis by themselves are uninformative about selection. <p> Mentions that Goodman (CA 34:281-288) has challenged their claims, suggesting that the interplay between stature, mortality, and forces of stress makes their model too simplistic. Also, Goodman states that if it can shown that selection on stature is unlikely, Wood et al.'s argument is moot. <p> This article uses Wood et al.'s example of stature to make an attempt to explain statistically that the possibility of selection should be research. If the coefficients neither are significantly different from zero nor have a tendency toward positive or negative values, the case for selection is less plausible. In th end, Byers states that if numerous measurements are taken a strong case can be made for whether selection was working on the sample population which would add confidence to interpretations.
|
