eliminated are the ones which it required the most time to answer correctly from the books. The fact is that it was quite impossible for the ordinary special agent, at the censuses of 1900 and 1905, to fill the schedules completely from actual bookkeeping figures within the limit of time which he was allowed. He was practically forced to resort to estimates with respect to certain of the questions, and this naturally tempted him to accept estimates for other questions which could have been more readily answered from actual bookkeeping data.

These condensations in the manufactures schedule are principally in the following three respects:

1. At the censuses of 1900 and 1905 the schedule called for the average number of men, the average number of women, and the average number of children under 16 employed during each month of the year. In the office these monthly averages were combined to give the annual average. Inasmuch as payrolls do not distinguish age and do not in all cases distinguish sex, the mere segregation by age and sex involved great difficulty and practically necessitated estimates. Moreover, to obtain the average number of employees in any single month would have required the examination of all the pay-rolls for that month, giving regard to the number of days on which each person was employed, and adjusting the cases where a single pay-roll extended from one month into another. All this was quite impossible in any reasonable length of time. At the present census we are asking for the distinction between men, women, and children only for one single date in the year. The distribution as between men, women, and children for that date will be assumed, as it safely can, to be fairly typical of the distribution for the year as a whole. With a view to ascertaining the annual average number of employees of all classes combined, we ask, not the average number for each month, but the number employed on the fifteenth day of each month. The average for these twelve days will give substantially the same result as an average calculated from twelve monthly averages; and the data can be obtained from the actual pay-rolls in but a small fraction of the time which would have been required to calculate monthly averages.

2. We have eliminated entirely the question calling for the number of employees classified according to their rates of wages or earnings. This was done with much regret, for correct information showing how many employees in the country as a whole, or in a given state, or in a given industry, obtain specified rates of wages would be the most valuable possible form of wage-statistics. We were forced to the conclusion, however, that the amount of labor involved in getting this information correctly from actual pay-rolls would be prohibitive in view of our limited appropriations. There is every reason to believe that at prior censuses the information in most cases was not taken from actual records, but was based on more or less inaccurate estimates of the operator of the factory.

3. At the censuses of 1900 and 1905 the schedule called for a distinction between those materials consumed which were strictly raw materials and those which were partly manufactured. This distinction was intended to permit the calculation of the so-called net value of manufactures by deducting from the gross combined value of the product of manufacturing establishments the value of those materials which represented the product of some other manufacturing establishment. As was pointed out in the text of the report of 1900, however, it is not possible to make use of this system of eliminating duplications except with regard to the country as a whole. When applied to individual states or cities or to individual industries, it loses all significance. The books of manufacturing establishments ordinarily do not directly segregate the cost of raw materials proper from that of partly manufactured materials, and the special agent was in most cases forced either to accept estimates of the value of the two classes of materials or to work through a maze of details of separate purchases in order to segregate the one class from the other. It has seemed to us that the value of the information secured did not justify the encouragement of the practice of making estimates, which was bound to extend from the separate classes of materials to the totals. At the present census, therefore, the only distinction which is made with respect to raw materials is that between fuel and other kinds of materials. It

seems to us that the fairest measure of the contribution that manufactures make to the wealth of the country, and the fairest comparison between different states or localities and different industries with respect to their significance in manufacturing, are secured, not by deducting from the gross value of the product the value of partly finished materials, but by deducting the value of all materials. We propose to show this net value of product generally throughout the presentation of the statistics, in addition to the gross value.

Changes have been made in the general mining schedule similar to those in the manufactures schedule. The special schedules for leading individual manufacturing and mining industries have also, in most cases, been materially condensed and simplified.

METHOD OF PRESENTING STATISTICS

It would be inappropriate at the present time to discuss the exceedingly tentative plans which we have in mind with respect to the method of tabulating and publishing the statistics of the census of 1910. In a general way it may be said that it seems desirable that the statistics should be published in a larger number of small volumes instead of a small number of very bulky volumes. It will perhaps be possible to separate the text and analytical tables from the general and detailed tables, and also to a greater extent to publish the material regarding separate subjects in separate volumes. We hope to be able to present an analysis of the statistics of families, which was not done at the census of 1900, and in certain other directions to present and discuss the statistics more fully than was done at that time.

THE OUTLOOK FOR AMERICAN STATISTICS

WALTER F. WILLCOX
Cornell University

This anniversary season, when nine organizations engaged in studying diverse aspects of man's social life are gathered at the metropolis of America for fraternal co-operation and mutual inspiration, naturally invites attention to the field and the outlook of the several societies. Statistics as a subject, however, is larger and more impersonal than the association created to develop it and thus a better theme for our annual reunion.

But why American statistics? Because statistics, like history and unlike economics or sociology, invites or demands a national rather than an international or universal treatment. Statistics is connected with and dependent upon the state, not merely by derivation of the word and history of the thing, but also by a rigid necessity. The original statistical inquiries which have been made by private agencies are insignificant in comparison with those which have been organized by government. In this field the aim of private citizens must be almost confined to a further interpretation and utilization of official returns with due regard to the probable error of the figures. The outlook for statistics, then, depends mainly upon the attitude of government toward the subject.

The first branch of statistical work to develop in the modern world was the statistics of deaths. A little later came that of births and of marriages. This branch of statistics, which in English is usually known as vital statistics but in France and other countries is more often termed demography, was established as a national system in England and Wales in 1837, and, although the historical connection has not been traced, there is little doubt that the establishment of the English registration system in 1837 was an important influence leading to the organization of this society in 1839, twenty-six years before any other of the associations with which we meet today was founded, and to the center

ing of its interest from the start upon vital statistics. The origin of our society at the capital of New England is another evidence of the connection I am suggesting, for New England during more than half a century and until recent years has been the main American nursery of vital statistics. Demography is the oldest branch of statistics; it has developed to a comparatively exact and scientific system; its methods have been subjected to long and searching criticism; its results are more unquestionable, if not more important, than those reached in any other branch. Hence it is the natural and appropriate gateway through which to approach the larger field, and the theme may be narrowed for the present to the outlook for American vital statistics.

The influence of the frontier as a capital fact, perhaps the capital fact, in our national history is now recognized and accepted. The well-nigh insuperable obstacles to securing registration, even of deaths and much more of births and of marriages, in a population living under frontier conditions, or even in the settlements thinly spread over the face of the country for many hundreds of miles east of the frontier, have prevented the rise of an effective American demand for good systems of registration. This is illustrated by the difficulty in tracing the ancestry of the most distinguished American of the nineteenth century, Abraham Lincoln. His biographers tell us, "There are hundreds of families in the West bearing historic names and probably descended from well-known houses in the older states or in England which, by passing through one or two generations of ancestors who could not read or write, have lost their continuity with the past as effectually as if a deluge had intervened." The limitations suggested by this quotation have been even more effective as a bar to the development of public records of deaths, births, or marriages. Canada and South Africa likewise have had little success in transplanting vital statistics from the mother country to the colony, and if the experience of Australia and New Zealand has been different, this must be ascribed in the main to the massing of the population of those colonies in large cities. We may even ask what evidence there is that the regis1 Nicolay and Hay, Abraham Lincoln, I, 1.