Form 18 ESTIMATES OF POPULATION FOR THE SEVERAL STATES AND THE DISTRICT OF COLUMBIA, FOR COUNTIES, AND FOR CITIES OF OVER 30,000 MADE BY THE BUREAU OF THE CENSUS OF THE DEPARTMENT OF COMMERCE IN PURSUANCE The estimates for cities of 30,000 and over are excluded from those for the counties in which such cities are located JULY 1, 1917 PREFACE. The estimates of population contained herein relate to cities which, according to the official estimates of population made by the Bureau of the Census prior to the registration of June 5, 1917, had more than 30,000 inhabitants, and to counties or portions of counties which lie outside of such cities (except in the cases of Massachusetts, Connecticut, and Rhode Island, for which States the estimates are made by "divisions," each of which consists of a number of towns)—that is to say, the figures given for any county or division are exclusive of the population of any city or cities having more than 30,000 inhab itants which are situated in that county or division. The name of each city is indented under that of the county in which it is located. These estimates have been made by the Bureau of the Census, at the request of the Provost Marshal General of the War Department, solely for use as a basis for the apportionment of the forthcoming draft. The method employed may be described briefly as follows: The total registration, as shown by the preliminary telegraphic returns, 9,659,382, was divided by the total population of continental United States, exclusive of Alaska, as heretofore estimated by the Census Bureau, 103,635,300, in order to obtain the proportion which the registrants represented of the total population. The resulting figure, 9.32 per cent, was assumed to represent the proportion which the registrants in each geographical unit represented of the total population thereof. Since the State totals, as shown by the corrected returns received by mail, varied more or less from those given in the preliminary telegraphic returns, both the total registration for the United States and the total population estimated herein differ slightly from the figures above given. The assumption that the proportion which the registrants in any county or city represent of its total population is the same throughout the United States may not be true, but the only way to ascertain the true proportion for each geographical unit would be to make an actual enumeration of its population. The proportion in 1910 varied considerably in different parts of the country, being greater as a rule in the newer than in the older sections and greater in cities than in rural localities. Because of the abnormal shiftings of population which have taken place among certain parts of the country in recent years, and particularly since the outbreak of the war gave so great an impetus to the manufacture of munitions and other commodities needed by the belligerents, the proportions which men 21 to 30 years of age, inclusive, represent of the total population of many localities have changed greatly since 1910. For this reason the assumption that the proportions existing in 1910 still prevail would result in very inaccurate estimates for some localities. Since, therefore, it was impossible in any event to estimate precisely the population of cities, counties, and States, on the basis of the registration, the simplest and speediest method-that based on the assumption that the proportion which the registrants represent of the total population is the same throughout the country-was adopted. One reason for the employment of this method was that the Census Bureau had only a very short time in which to prepare the estimates; but another, and still more cogent, reason was that in this manner there was obtained the fairest possible basis for the apportionment.of the draft, since the localities whose population estimates.may be exaggerated are those in which there is an excess of men 21 to 30 years of age, inclusive, while the ones whose population is understated are those in which the proportion of men of these ages is smaller than the average. Although substantially complete telegraphic returns have been received for all States, the returns for a great many cities and counties, which should have reached the office of the Provost Marshal General by mail before this time, are still missing. Telegraphic requests were made for the totals for these missing counties and cities, but some of the replies were not received in time to make use of them in compiling these estimates. It was necessary, therefore, in the cases of most of the cities and counties for which the returns were still missing, to make the estimates on the assumption that the ratio between actual and estimated registration in these places was the same as that prevailing in localities comparable in size and, so far as was known, in other respects, for which returns had been received. The names of cities and counties for which returns were received too late to be used in computing the estimates are indicated by footnotes. The estimates for Alaska, Hawaii, and Porto Rico, in which the registration has not yet taken place, were made in the same manner in which the official estimates of population have been made by the Census Bureau heretofore, namely, on the assumption that the annual numerical increase in each geographical unit since 1910 has been the same as the average numerical increase between 1900 and 1910. |