Caste and Race in India c9.9

c9.9

This procedure opens up the second problem, viz. what is the method by which we can detect racial affinities and compare groups? For such comparisons the average has long been in use. Risley principally used the average for his work, rarely bringing in the seriations and still more rarely the absolute measurements.² Seriations, like the frequency curve, only serve the purpose of showing the actual distribution and cannot be conveniently used for purposes of comparison, a fact which explains the rare use of these by Risley when dealing with the types, though they are given in the appendices. The average without the standard deviation is an abstraction which tells us almost nothing. The standard deviation is very useful both as serving the basis of the formula that we have utilized for comparison, as well as giving us an idea of the actual range of variation. "A range of six times the standard deviation usually includes 99 per cent or more of all the observations in the case of distributions of the symmetrical or moderately asymmetrical type."³ We have therefore given in the appendix the means and their standard deviations for selected castes.