Page:Popular Science Monthly Volume 86.djvu/492

488 does the wage rate among male clerks rise above $1,250; in an even smaller number of instances do wage rates of female clerks rise above $750.

One of the most unsatisfactory situations which an analysis of wage statistics reveals, is the paucity of the wage figures relating to transportation and commerce. It is in these fields that inquisitorial bodies have the greatest authority; yet it is in these fields, strangely enough, that the wage statistics are least satisfactory. With the exception of the Census volumes for 1913 on express, and telephone and telegraph, and of a special report by the Bureau of Labor on the telephone industry, there is little or nothing of note.

The wages in the railroad industry, employing as it does more than a million and a half persons, are stated only as averages. The excuse for this statement of railroad wages in terms of averages—it requires some excuse, for, though the averages are given by districts and for ten wage-earning occupations and two groups of miscellaneous wage-earners, these again classified by districts and by the class of railroads, they are still averages, and therefore suffer under all of the disqualifications that averages are heir to—seems to be that the length of time and the conditions of the work done by different employees vary so greatly that no classified statement could do justice to the situation. Pursuant of such philosophy the Interstate Commerce Commission has done, under the circumstances, the most misleading thing that it could possibly have done—that is, it has published averages; and the State Railroad Commissions, following the footsteps which, unknown to them, led so directly into this statistical quagmire, also have published nothing but averages.

Granted that, in the case of railroad employees, the classified or group system of stating wages is inaccurate, how much more inaccurate does the average become? Instead of accepting errors at their face value, the average thus obtained compounds and augments error. Nor is this a case in which errors tend to neutralize each other. What avails an average of the wages of switch tenders in Maine and in Ohio? What avails an average wage for "all other employees and laborers," including for the United States nearly a third of a million men? The method carries its own refutation. Except as a basis of comparison from year to year, the figures are meaningless and absurd.

The difficulties lying in the path of obtaining classified wages for railroad employees do not seem to be so great as the protestors claim them to be. Why should not the Interstate Commerce Commission secure from each railroad a statement for the first week in June and December showing the number of employees of each class who had