ECCENTRIC OFFICIAL STATISTICS 8 1
compared with 1880, and secured a more complete return of the officers, firm members, and clerks and their salaries." For some reason about which we will venture no surmise, Mr. Steuart ends the citation in the middle. He does not quote the rest of his own sentence, which establishes the writer's contention and disproves his own, viz.: "than was reported at previous censuses; therefore the average annual wages per employ^, as obtained from the reports of the two censuses, are not comparable, nor should the amounts be used to ascertain the percentage of increase" (see Compendium, Eleventh Census, Pt III, p. 668).
Mr. Steuart says also : " It is asserted that the presentation of S484.49 as the average per capita wages for all classes of employes, or $444.83 as the average for the employes exclusive of officers, firm members, and clerks, in comparison with $346.91 for all classes in 1880, is erroneous ; also that the division of the total wages by the average number employed is not the correct method of obtaining the average wages." This seems another palpable misrepresentation and evasion. What the writer contended, as will be seen by reference to his article, was that average annual earnings (not rates of wages) could be obtained only by dividing the total wages by the total number of employes. This Mr. Steuart admits farther along in his article, saying: "But it must be remembered that the wages reported were paid to have certain posi- tions of employment filled, and the number of those positions is probably the true divisor for the total wages ; the average comes nearer this number than the total or greatest number. But as carefully explained by the census reports, the average given for wages is not the true average yearly earnings per workman" (p. 628).
How carefully this is explained in the census reports may be seen by reference to the same, or to p. 527 of the writer's first article, in which the remarks of the census are more fully quoted than space will here permit. In the census are given tables purporting to show " aver- age annual earnings," which Mr. Steuart admits the figures given do not represent. Mr. Steuart also acknowledges that the only true way to find average annual earnings is to ascertain the total wages and the total num- ber of workers and divide the one by the other. This has been precisely the writer's contention in his criticism of Colonel Wright's contribution to the Atlantic Monthly. In that article, after quoting the fallacious sta- tistics of the Aldrich report. Colonel Wright says : " It is often con- tended that the increase in rates of wages does not indicate the true social condition of the wage-earner ; that rates of wages belong to
