88 THE AMERICAN JOURNAL OF SOCIOLOGY
any degree of accuracy, they were in a measure excusable for filling out their schedules as they might think the case to have been. If, as the returns show, the workers were at work, there would be but few instances in which they would be found at their place of residence, and even where they were found, cases where the information could be given offhand must have been exceedingly rare. Yet from these nec- essarily unreliable returns we find the most eminent of our statisticians endeavoring to demonstrate that which common observation shows is not true.
Comparing the returns made by the enumerators as to the numbers engaged in manufacturing industry with the returns made by special agents, we find that the former are, in many instances, grossly inade- quate, especially as relates to females and children ; and, comparing both classes of returns with the reports of factory inspectors, we find that the reports of special agents also fall far short of showing the fact.
In a former paper attention was called to the fact that, among other omissions, the manufacturing tables report no females and but three children employed in the cigar and tobacco industry in Jersey City, while the state factory inspector reported 2,500 females and 260 chil- dren in the single establishment of Lorillard & Co. In the occupa- tion tables we find 723 females reported in this industry and city. Mr. Steuart asserts that some of the largest sugar refineries, paper mills, cigar factories, and establishments engaged in other industries absolutely refused or willfully neglected to furnish the information required for the census, and expresses the opinion that they undoubt- edly offset the inclusion of the minor industries referred to and tended to counterbalance the results of the more thorough canvass.' That is, the omission of the poorly paid workers in tobacco factories tends to offset the largely increased enumeration of the higher-paid wage-earn- ers in the building trades and makes the wage statistics comparable. Mr. Steuart's statement of the refusal of the manufacturers to furnish the information required by the census confirms the truth of the asser- tion of Edward Atkinson, quoted in the writer's second article (July, 1897), regarding the census of 1880 : " If the questions had been put in such a way that the profits of the different arts investigated would have been disclosed, manufacturers would have returned no answer
'Regarding the omissions referred to, Mr. Steuart remarks in the Census (Com- pendium, Pt. 11, p. 704): "The estal^lishments thus neglecting to comply with the requirements of the law are, with but few exceptions, unimportant, and their omission from the tables can have but slight eSect upon the totals."
