Laserfiche WebLink
vegetation distribution (no habitat affinities) , then the proportion of <br /> individuals of a species encountered in a particular vegetation type would <br /> correspond to the proportion of the sampling effort devoted to th^a-t <br /> vegetation type. For example if 30% of the small manmal live traps were <br /> established in upland sagebrush, then 30% of the total number of individuals <br /> of all species captured over all sampling transects would have been caught <br /> in the upland sagebrush habitat if the species were distributed randomly. <br /> Conversely, if 90% of all Peromyscus maniculatus (deer muse) captures <br /> occurred in the upland sagebrush type, this would indicate a definite <br /> affinity of P. maniculatus for the upland sagebrush vegetation type. <br /> The Chi-square test (Snedecor and Cochran 1967) was used to detexmine <br /> if the differences between the observed number of captures for a given <br /> species in a given vegetation type and the number of captures expected on <br /> the basis of a random distribution was significant. The Chi-square <br /> statistic, X2, is given by: <br /> X2= iE 1 <br /> where: <br /> Oi = observed number of captures (or encounters) for habitat i <br /> Ei = expected number of captures in habitat i based on a random <br /> distribution among habitats. <br /> The calculated Chi-square value was compared to a Chi-square table to determine <br /> if a significant difference between observed and expected number of captures <br /> occurred at the 95% confidence level. If a significant difference did exist, <br /> individuai Chi-square values for individual habitats were examined to provide <br /> an indication of those habitats for which a species shows an affinity or an <br /> avoidance. <br /> -i5- <br /> J U N 1 1 1980 <br />