« PreviousContinue »
LETTER OF TRANSMITTAL.
Department Op Commerce,
Bubeau Op The Census, Washington., D. C, October 19, 1918. Sib: I transmit herewith a report on Marriage and Divorce. This report is the result of a collection of statistics for the year 1916 only, and represents the first collection of statistics of marriage and divorce by this bureau since the publication of a two-volume report in 1909. The report published in 1909 contained statistics collected by the Bureau of the Census which covered the 20-year period from 1887 to 1906, and also statistics collected by the former Department of Labor, now the Bureau of Labor Statistics, which covered the preceding 20 years from 1867 to 1886.
This report was prepared under the supervision of Mr. William C. Hunt, chief statistician for population, assisted by Mr. Arthur E. Seymour, who had immediate charge of the collection of the data. Respectfully,
Sam. L. Rogebs,
Director of the Census.
Hon. William C. Redfield,
Secretary of Commerce.
MARRIAGE AND DIVORCE: 1916.
The statistics relating to marriage and divorce herein presented are for the calendar year 1916 only.
Two investigations with respect to marriage and divorce have been made previously by the Federal Government—one in 1887-88 by the Department of Labor, covering the 20 years from 1867 to 1886, inclusive, and the other in 1906-7 by the Bureau of the Census, covering the 20 years from 1887 to 1906, inclusive. These two investigations together covered a consecutive period of 40 years.
The need for the collection of the statistics of marriage and divorce for the years which have elapsed since 1906 was brought to the attention of the Director of the Census in April, 1914, by a letter received from the International Committee on Marriage and Divorce, signed by the Rt. Rev. Alfred Harding, Bishop of Washington, and the Rev. Francis Miner Moody, its executive secretary. As stated in this communication, the urgency of the matter rested largely in the growing demand for an amendment to the Federal Constitution giving Congress power to legislate on all questions of marriage and divorce; and for the intelligent consideration of the proposed measure all the facts were needed.
In February, 1915, there were also received through Dr. Moody several petitions addressed to the President urging the passage of the joint resolution providing for the collection of the statistics of marriage and divorce, then pending in Congress. These petitions, with the letter of Dr. Moody transmitting them, were referred by the President to the Bureau of the Census for reply and whatever action seemed practicable.
The intent of these letters and petitions, so far as they applied to the collection of the statistics of marriage and divorce, was to secure legislation providing for such collection for the seven years ending in 1913, and annually thereafter; but the condition of the work of the Census Bureau at that time did not warrant immediate action, and the writers were so informed.
Prior to the receipt of these letters and petitions, however, the Bureau of the Census had given the matter most careful consideration, and in January, 1914, the Director of the Census addressed a communication to the Secretary of Commerce, in which the following statements were made:
The Bureau of the Census collected in 1906-7 the statistics of marriage and divorce in the United States covering the 20 years from 1887 to 1906, inclusive. This was the second investigation of this character conducted by the Federal Government, the first
having been made in 1887-88 by the (then) Department of Labor for the 20 years from 1867 to 1886, inclusive.
In initiating the steps for securing the necessary legislation for the second investigation, the importance and need of these statistics was deemed sufficiently great to warrant the sending by the President on January 30, 1905, of a special message to Congress, and this was immediately followed by the passage of a joint resolution, approved February 9, 1905, giving the Director of the Census authority to collect these statistics from January 1, 1887; and the investigation thus provided for was made to cover the full 20-year period ending in 1906.
These two investigations by the Federal Government together furnished, therefore, substantially all the available statistics, particularly as to divorce, for a consecutive period of 40 years from January 1, 1867, to December 31, 1906. At the close of the year 1916 another decade will have elapsed and the steadily increasing prevalence of divorce, as shown by the results of the two investigations covering the 40-year period ending in 1906 as well as by later partial statistics, suggests the advisability of requesting from Congress at an early day authority for a similar collection by the Bureau of the Census of the statistics of marriage and divorce in the United States for the 10 years ending in 1916, and thereafter annually.
The law now provides for the collection annually by the Bureau of the Census of the statistics of deaths and births, and similar authority should be given with respect to marriage and divorce, thus providing for the collection yearly of the vital statistics of the country, to such extent immediately as may be practicable, but ultimately in accordance with standardized forms and methods, as is now being rapidly done with respect to deaths and for which a beginning also has been made with respect to births.
The results of the investigation conducted by the Bureau of the Census for the 20 years from 1887 to 1906 showed a total of 945.625 divorces granted, as compared with a total for the preceding 20 years of only 328,716; or a number for the first 20 years hardly more than one-third of that recorded for the second 20 years. Of the 945,625 divorces from 1887 to 1906, 352,263 were granted during the first 10 years of the period and 593,302 during the second 10 years, representing an increase for the last decade over the former of 241,099, or 68.4 per cent.
The collection of the statistics by the Bureau of the Census in 1906-7 was made mainly by regular special agents and detailed clerks, who obtained the information at the county seats of the various counties. In 765 of the smaller and more remote counties, however, the statistics were furnished by the court clerks, and in 206 counties in the Southern states the special agents of the bureau ordinarily employed to collect statistics of the cotton crop were engaged to secure the required information. It is believed, however, that, if a similar investigation should be undertaken for the 10 years ending in 1916, by proper provision beforehand the statistics could be almost wholly furnished through the court clerks, either by their own employment or that of their deputies or other responsible persons, working at piece-price rates.
The present condition of the work of the Bureau of the Census, and the amount of work outlined for the fiscal year 1914-15, probably does not warrant the attempt to secure now an appropriation for carrying on in 1914-15 any portion of the work with respect to the collection of the marriage and divorce statistics, even if authorized; it does, however, seem feasible in the near future to take the necessary steps toward securing during the present session of Congress legislative authority for the work, in order that there may be included in the estimates for the fiscal year 1915-16 a sum sufficient to cover the probable cost of the collection for the eight years from 1907 to 1914 and to provide in later estimates a further sum to cover the cost for each of the two remaining years of the decade (1915-1916), at a total annual cost of not more than $20,000 probably.
The systematic collection of the marriage and divorce statistics by the Federal Government, as a part of the vital statistics of the country, is of sufficient importance and value in itself, irrespective of the demand for the latest complete statistics—which is constant and will become increasingly so as time goes on—to warrant immediate consideration, and I have taken the liberty, therefore, to bring the matter to your attention at this time, and to inclose for your information a copy of the joint resolution authorizing the collection of the statistics of marriage and divorce from January 1,1887, a copy of the special message of President Roosevelt concerning the need for these statistics, and a copy of a proposed joint resolution to provide for the collection of these statistics for the 10 years ending in 1916, and annually thereafter, and I shall be glad to furnish you with any additional information that you may wish concerning this most important question.
No action was taken in the matter at the Sixtythird Congress, but the necessary legislation was requested by the Secretary of Commerce at the first session of the Sixty-fourth Congress. A joint resolution (S. J. Res. 107) was passed by the Senate June 3, 1916, but there was no action regarding it in the House of Representatives.
The purpose of this joint resolution was to provide for the collection and publication of the statistics of marriage and divorce for the nine years from 1907 to 1915, inclusive, thus bringing these statistics substantially up to date, and to provide for their collection and publication each year thereafter. The collection annually of these statistics would tend to standardize the inquiries, to improve the records from which they are derived, and thus to reduce proportionately the expense of collection, and to make the statistics constantly available rather than after long intervals, as has heretofore been the case.
Although the House of Representatives took no action with regard to the joint resolution above referred to, it did include in the legislative, executive, and judicial appropriation bill a sum of money for the collection of the statistics of marriage and divorce, and when the money thus appropriated became available in July, 1917, the Bureau of the Census had an amount sufficient to cover the cost of collection of these statistics for the entire period of 10 years from January 1, 1907, to December 31, 1916. It was decided, however, because of war conditions, not to cover the preceding years but to limit the work to 1916, and thus make a beginning with respect to their collection each year as a part of the annual work of the bureau.
With this understanding the necessary schedules and instructions were prepared and the work was begun the latter part of July, 1917. The statistics of marriage and divorce for the year 1916 were collected wholly by correspondence and refer to 48 states (in
eluding the District of Columbia), comprising a total of 2,980 counties or equivalent subdivisions from which returns were expected. For South Carolina, with 44 counties, and for 9 unorganized counties in South Dakota and Texas no statistics of marriage and divorce are available.
The statistics of marriage consist simply of a statement of the number of marriages performed during the year. For 27 states the number of marriages in each county was obtained from printed reports or state records without cost to the Federal Government, while for the remaining 21 states, representing a total of 1,649 counties, the data were obtained by correspondence direct with the county or court clerks or other officers having custody of the marriage records. The inquiries as to divorce in 1916 were much more comprehensive than for marriage and included substantially the same items as for the 20-year period 1887-1906, except that the investigation for 1916 did not cover all applications for divorce, as was the case for the period 1887-1906, but was restricted to divorces granted only. For 3 states—Nebraska, New Jersey, and Wisconsin—it was possible to obtain the required information as to divorce from state officials, but for the remainder of the country, representing 2,795 counties, it had to be obtained direct from the county or court clerks.
The schedules for divorce were mailed from Washington the latter part of July, 1917, and by the end of October—three months later—complete returns had been received from three-fifths of the counties and by the end of the following December from more than seven-tenths of all the counties. The work of securing the returns from the remaining counties was a much slower process, requiring much extra effort, both by mail and telegraph, and the last returns obtained were not received until sometime in July, 1918. It should be said, however, that the delay in sending the returns was due in many cases to lack of time because of service on the exemption boards created in connection with the operation of the selective-service law and also too, in other cases, because the amount of compensation offered was considered too small. As at the former investigation, special agents employed by the Bureau of the Census to collect the cotton reports did the work in certain southern counties where the court clerks failed to make the returns as requested. These statements apply to the divorce statistics mainly, as the filing of the return as to the number of marriages in counties where the information could not be secured from printed reports or state records was but an incident of the work for which no great amount of time was required.
All things considered, it may be said that this method—here tried for the first time—of obtaining marriage and divorce data by correspondence through the mails has proved entirely practicable, returns as to marriage having been received for 2,874 counties and those as to divorce for 2,885 counties, out of the total of 2,980 counties from which returns were expected. The counties for which no returns were received are mainly in the Southern states and represent for the most part those in which the number of marriages or divorces was not large and, as a matter of fact, comprise only a relatively small proportion of the total number of marriages celebrated or divorces granted in the entire country during the year 1916, probably not much more than 2 per cent in the case of marriages and not much more than 1 per cent in the case of divorces. It would have been possible probably to have secured eventually, through correspondence, a return for both marriages and divorces for all of the missing counties, but it was not deemed worth while, on account of the comparatively small numbers involved, to delay the completion of the work for this purpose, and so the collection of the statistics was brought to a close substantially at the end of June, 1918.
The returns as to marriage and divorce were secured at a total cost of $15,000, of which less than 8500 represents the cost of securing the returns as to marriage.
The decision to confine the collection of the statistics of marriage and divorce to the single year 1916 was influenced in part, aside from war conditions, by the desire to determine, as a matter of actual experience, whether or not it was practicable to secure these returns wholly by mail instead of detailing employees of the bureau to travel over the country for this purpose, as had been the practice heretofore, and also
Scope.—The statistics of marriages, as before stated, consist simply of a statement as to the number of marriages celebrated during the calendar year 1916. It is evident, moreover, that until an annual collection of the marriage statistics by a central office is provided for as a component part of the vital statistics for the entire country, it will not be possible either (1) to attempt to have the several states adopt uniform methods for the registration and return of marriages by the county officers to a state office of record, from which the Federal statistics can be readily compiled; or (2) to have the marriage statistics include, in addition to tho number of marriages celebrated, detailed statistics as to color, sex, age, nativity, marital condition, whether first or second marriage, and the like, thus embracing data which are almost universally collected by the statistical bureaus of most civilized countries.
Number of marriages.—The whole number of marriages reported for the year 1916 (see Table 1, p. 9) was 1,040,778, representing the returns received from 2,874 counties, out of a total of 2,980 counties from
whether the information sought would be furnished by the county and court officials at a reasonable cost; and thus by this means have definite data available for the use of Congress in the consideration of the request for legislative authority for their collection annually. Nearly every important country 1 except the United States makes annual compilations of the statistics of marriage and divorce, but in this country, as heretofore stated, such statistics have thus far been compiled on two occasions only, each compilation covering retrospectively a period of 20 years.
There seems to be no question as to the necessity for the systematic collection of the statistics of marriage and divorce as a part of the vital statistics of this country, and if collected annually by the Bureau of the Census the same methods could be applied to the securing of these statistics as have been so successfully used in perfecting the returns and in extending the area of registration with respect to deaths and births, so that as time goes on it would be possible for the Federal Government to secure all of these returns through state offices of record and not from the various county and court officials as must now be largely done. The collection of these statistics each year would also ensure greater expedition in the completion and return by the county and court clerks of the marriage and divorce schedules. This work would become a part of the regular office routine and the compensation offered, although not large, would be sufficient to pay for the trouble of making out the schedules from time to time, as the records became available, so that the complete returns could be sent in soon after the end of each calendar year.
which returns were sought. There are 106 counties (including 10 new counties), therefore, from which no returns as to the number of marriages in 1916 were received. Of these, 9 are in states for which the number of marriages was obtained from state records, being counties for which returns were lacking and could not be supplied by the state officials. The remaining 97 counties are distributed among 18 different states in which the returns as to marriages were sought from the county officials, but could not be obtained through correspondence, although the requests were made repeatedly.
For 1906, as shown by the former investigation, there was reported a total of 853,290 marriages, with 35 missing counties, and for 1896 a total of 613,873 marriages, with 56 missing counties. These figures, when compared with those reported for 1916, show an apparent increase in the number of marriages between
'England and Wales, Scotland, Ireland, Canada, Germany, Norway, Sweden, Denmark, Netherlands, Belgium, France, Switzerland, Italy, Russia, Finland, Austria, Hungary, Rumania, Bulgaria, Serbia, and Japan.
1906 and 1916 of 187,488, or 22 per cent, and between 1896 and 1906 of 239,417, or 39 per cent. Excluding the marriages reported for 1906 and 1896 for the counties for which no returns were secured for 1916, the number of marriages for 1906 would be reduced to 838,451, a difference of 14,839, and for 1896 to 602,542, a difference of 11,331; and on this basis the increase between 1906 and 1916 would be 202,327, or 24.1 per cent, and between 1896 and 1906, 235,909, or 39.2 per cent.
The number of marriages reported for 1906 and 1896 for the 106 counties for which there were no returns for 1916—namely, 14,839 and 11,331—constituted in each case less than 2 per cent of the total number of marriages reported. There was no return of marriages for 35 counties in 1906 and for 56 counties in 1896.
The number of marriages reported from each county is shown in Table 29 (p. 39). The number of counties in 1916 from which marriage returns were sought, the number of counties for which no report was made, and the total number of marriages reported, with comparative figures for 1906 and 1896, respectively, are shown by states and geographic divisions, in Table 1 (p. 9),- which also shows the increase for the two 10-year periods. The increases given in this table are not based upon the total number of marriages reported for 1906 and 1896, but upon the number exclusive of those for the counties for which the returns are lacking for 1916. This is necessary because of the large number of missing counties for 1916, particularly in some of the Southern states.
In using these figures, it must be remembered, of course, that the number of marriages reported for 1916, as well as for each of the two years with which comparison is made, represent the returns for a single year only, and that the number so reported is dependent upon or influenced by several considerations, such as no report whatever of marriages for a considerable number of counties, incompleteness in the returns for counties reporting, due to lack of proper registration methods, fluctuations in the number of marriages celebrated from year to year, due to economic conditions, and the like, so that comparisons of increase in the number of marriages during the two 10-year
intervals are at the most only roughly indicative of the actual facts of the case.
Marriage rates.—The number of marriages per 10,000 population in 1916, based upon estimated population, is compared in Table 2 (p. 10) with similar rates for 1906, also based upon estimated population, and for 1900 and 1890, based upon enumerated population as shown by the Federal censuses taken in those years. The population used in each case is exclusive of the population of the counties for which the number of marriages was lacking.
The rates as given in Table 2 are higher for 1916 than for 1906 in 26 states, while for 44 states the rates for 1906 were higher than those for 1900, thus making 18 states in which the rate increased in 1906 as compared with 1900 and decreased in 1916 as compared with 1906. As stated above, the number of marriages celebrated from year to year fluctuates with economic conditions, so the high rates in 1906 as compared with 1916 and 1900 would indicate an unusually large number of marriages in that year. Although 20 states showed a smaller rate in 1916 than in 1906, there are 6 states which showed an exceptionally high rate in 1916—Montana, 176; Maryland, 150; Florida, 149; Vermont, 145; Arkansas, 143; and Arizona, 142. These high rates may be due in part to actual increases in the number of marriages, or to a better system of return and registration, or in some cases to too low estimates of population.
All of the states except South Carolina require every marriage solemnized to be reported to some official specified by law, and for nearly two-thirds of the states there is legal provision for the state registration of marriages, even though in some of these states this provision of law is not fully carried out. It is undoubtedly true that the methods of registration are much better now than formerly and that because of this the number of marriages reported in many states is larger, irrespective of other conditions which may have been operative in the same direction.
Marriage rates based upon marriageable population are more indicative than those based upon the entire population, but the necessary data are not available for 1916.