100 yrs marriage div usa

8/13/2019 100 Yrs Marriage Div Usa

1/68

Data from theNATIONAL VITAL STATISTICS SYSTEM

SeriesNumber 4

1 YearsfMarriage and DivorceStatisticsUnitedtates867 967

Analysis of marriage and divorce statistics for the period 1867-1967,including information on data collection procedures throughout the100 years, time trends in national and area totals and rates, and char-acteristics of marriages and divorces.

DHEW Publication No. HRA 74-1902U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE

Public Health Service

Health Resources AdministrationNational Center for Health StatisticsRockville, Md. December 1973


2/68


3/68

N TION L ENTER FOR HE LTH ST TISTI SEDWARD B PERRIN Ph D Actin g Director

PHILIP S LAWRENCE SC D Dep ut y D ir ect orGAIL F FISHER Assis ta nt Director f or H ea lth Sta tis t ics Devek pr nen t

WALT R SIMMONS M A Assist an t Director f or Resea rch a nd Scien tif ic Developm en tJOHN J HANLON M D M ed ica l A dvi sorJAMES E KELLY D D S D en ta l A dv isor

EDWARD E MINTY E xecu ti ve Of fi cerALICE HAYWOOD In form ation Off icer

DIVISION OF VIT L ST TISTI S

ROBERT A ISRAEL M S DirectorJOHN E PATTERSON Assistant Director for Demograph ic Af lt ii rsALICE M HETZEL Ch ief Marr iage and Divorce Sta t ist ics Branch

Vital and Health Statistics-Series 21-No. 24

DHEW Publication No. HRA 74-1302Library of Congress Ca ta log Curd Number 73 6 233


4/68

ONT NTSA Century of Data

Data CoUectionType of Information CollectedTime Trends in Totals andRates

National Totals of MarriagesandDivorces :National Marriage RatesNational Divorce RatesArea Totalsof MarriagesandDivorces Area MarriageRatesArea Divorce Rates

Characteristics of MarrbgesCharacteristics of DivorcesDuration of Marriage Prior toDecree

Duration ofSeparationPlace of MarriageChildren of theDivorcedLegal GromdsParty to Whom Decree was GrantedOther kgalVariables

Personal CharacteristicsReferencesList of DetailedTablesAppendix I Bibliography of Marriage and Divorce Statistical Studies Pub

lishedby the FederalGovemmentAppendix 11 Sources and Methods

SourcesNational Totals andRatesArea TotalsArea RatesChanging AreasSelected YearsLegalGrounds

Page1146679111121314141517171719192

2

21

56

5959595959616161


5/68

SYM O SData not available ----------------------- ---Category not applicable ------------------ . . .Quantity zero --------------------------- -Quantity more than Obutless than 0.05---- 0.0Figure does not meet standardsofreliability or precision ------------------


6/68

1 YEARS OF MARRIAGE ANDDIVORCE STATISTICS: 1867 1967Alexander A. Plateris, Ph.D., Division of T W Statistics

ENTURY OF D TData ollection

The history of marriage and divorce statisticsin the United States is long and varied. The firstyear for which national marriage and divorce dataare available is 1867; thus we possess statis-tics on family formation and disruption for aperiod covering over .100 years. Methods ofcollecting data changed considerably during thatperiod.

The first collection was ordered by Congressin response to petitions from leaders in the fieldsof religion, law, and education who were concernedwith the confusion and lack of uniformity in thedivorce records in the United States. Legislationadopted on March 3, 1887, directed the Com-missioner of Labor to collect statistics onmarriages and divorces occurring in all Statesand territories during a period of 20 years,1867-86. The results of the investigation werepublished in 1889 in the form of a 1,074-pagebook entitled A Report on Mdwiage and Divorcein the United States, 1867 to 1886.1 This reportwas signed by Carroll D. Wright, Commissionerof Labor, and is occasionally referred to as theWright Report.Information was collected by special agentsof the Department of Labor who visited thecourts empowered to grant divorces and gathereddata from their records. An exception was madefor courts in small or distant areas, from whichreturns by mail were obtained. It was found that2,624 courts granted divorces, and informationwas obtained from practically all of them exceptin cases w~ere records had been destroyed by fire or flood.

Marriage statistics were much less complete. For the whole period 1867-86 no marriage datawere received from Maine and South Carolina.The reporting from most other States was so poorthat annual totals were prepared for only 14States Connecticut, Delaware, Illinois, Kansas,Maryland, Massachusetts, Michigan, Minnesota,New Hampshire, New Jersey, Ohio, Rhode Island,Vermont, and Wisconsin), as well as the Districtof Columbia. The incompletenesss of marriagedata was due to the inadequacy of State lawsgoverning marriage licenses. The Wright Reportstated Licenses are granted on various bases andunder various conditions, and there is littlecompulsory law relative to the returns of mar-riages celebrated....Marriage may take placeeither under license or publication of bans... ~Commissioner Wright said further, Only a veryfew States make marriage registration com-pulsory. In some of the older States, where the lawhas for years provided for State registration, I aminformed authoritatively that the number of mar-riages returned will fall short at least 10 per centof the number of marriages celebrated. 3In the years 1902-05 the President andCongress again received many petitions request-ing the legislation authorizing a second collectionof marriage and divorce data, and on January30, 1905, President Thecdore Roosevelt sent aspecial message to Congress indicating the needfor such legislation and stating that the Directorof the Census should be authorized to collectand publish statistics for the years 1887-1906.A congressional resolution to that effect wasapproved on February 9, 1905. The results ofthe second data collection were published by theBureau of the Census in a two-volume reportentitled Maw zge and Divcrrce, 1867-1906.4


7/68

The method of collecting data in 1907 wasvery similar to that used in 1887. Special agentsof the Bureau of the Census gathered this in-formation, except for data returns from 765small and remote counties, which were sent inby mail. In 1906 there were 2,844 counties orequivalent areas in the United States. Informationon marriages was obtained from all but the 41counties in South Carolina and 28 local areas inother States; information on divorceswas obtainedfrom all but six local areas.

The second data collection was consideredhighly satisfactory, and it was decided to repeatthe collection each decade; the entire 10-yearperiod was to be covered. The next collectionwas scheduled to take place in 1917 and to coverthe years 1907-16. However, by 1917 the UnitedStates had entered World War I, and though the1917 collection took place as scheduled, it waslimited to data for 1916. The statistics werepublished by the Bureau of the Census inMa?wiagwand Divorce 1916. s

Shortly after World War I the collection andpublication of marriage and divorce statisticswas resumed on an annual basis. For each year1922-32 the Bureau of the Census published anannual pamphlet, Marnizge and Divorce Inthe 1926 and later issues annulment statisticswere included, tabulated separately from divorcedata. The publication of these annual reports wasdiscontinued during the Depression as an aus-terity measure. For years when data were notcollected, 1907-15 and 1917-21, national mar-riage and divorce totals were estimated bythe Bureau of the Census from the availableState totals.

During the years 1916 and 1922-32 nearly allareas reported divorces; data from a growingnumber of States were obtained from statewidecentral files. Divorces were reported from centralfiles by three States Nebraska, New Jersey, andWisconsin in 1916 and by 16 Alabama, Arkansas,Florida, Iowa, Louisiana, Maryland, Michigan,Mississippi, Nebraska, New Hampshire, NewJersey, Oregon, South Dakota, Vermont, Virginia,and Wisconsin in 1931. The number of Statescentrally reporting marriages increased by threeduring the 1916-31 period and included 30 Statesin 1931--the 16 States with central files ofdivorces and 14 others California, Connecticut,

Delaware, Idaho, Indiana, Kansas, Maine, Mas-sachusetts, New York, North Dakota, Pennsyl-vania, Rhode Island, West Virginia, and Wyoming .Marriage data for South Carolina were obtainedfor the first time in 1922; divorce data for SouthCarolina were not published until 1950 as alllaws pertaining to divorce were repealed in1878 and no divorces were granted in that Stateduring the years 1879-1948.

No marriage or divorce information whatso-ever was collected for the years 1933-36. Nationaltotals for these years were estimated by S. F.Stouffer and L. M. Spencer from data availablefrom selected States. These estimates werepublished in the Amevican Jownd of SociologyJanuary 1939 issue,G

When the publication of the Mawiage andDivorce yearbooks was discontinued, the firstperiod in the collection of statistics came to anend. It was characterized by nationwide datalimited to variables available on the records oflocal and State off ices. When data collection wasresumed, demographically important item,s wereemphasized, and they were obtained from limitednumbers of States. During the first period mostcollected information referred to divorces; after-ward most attention was given to marriages. Amarriage-collection area was established anal-ogous to the present marriage-registration area,and the first publication gave 1939 data from 25States and the District of Columbia: Alabama,Arkansas, California, Connecticut, Delaware,Florida, Idaho, Iowa, Maine, Maryland, Mas-sachusetts, Michigan, Mississippi, Nebraska, NewHampshire, New Jersey, Oklahoma, Oregon,, Penn-sylvania, South Dakota, Utah, Vermont, Virginia,West Virginia, and Wisconsin. New York andRhode Island were added in 1940.

There were seven publications on marriagesoccurring in the marriage-collection area in1939 and 1940. In addition, there was one publica-tion on 1939 divorces in the divorce-collectionarea. Twelve States were included in that area:Delaware, Florida, Iowa, Maryland, Michigan,Mississippi, Nebraska, Oklahoma, Oregon,, SouthDakota, Virginia, and Wisconsin. All these re-ports appeared in the series Vital StatisticsSpecial Repovts In addition to detailed statisticsfor collection areas, totals for the United Statesand each State were published in this series,

2


8/68

including marriages performed during 1937-40,divorces granted during. the same period, andtotals for both marriages and divorces for 8 years,1937-45. Historical statistics for the .50-yearperiod 1887-1937 were published in 1940.

World War II interrupted for a few years thedata collection, except for totals, and the mar-riage- and divorce-collection areas disappeared.The collection of marriage and divorce data wastransferred from the Bureau of the Census tothe newly formed National Office of Vital Statis-tics in the Public HeaIth Service. Publication ofmarriage and divorce data in the series VitulStatistics4pec al Reports was resumed with the1946 data, and one or more reports were publishedannually through 1957 data. Since 1946 marriageand divorce statistics have also been publishedin the annual volumes Vitul Statistics of theUnited States published by the National Office ofVital Statistics and later by the National Centerfor Health Statistics. Annulments were includedwith divorces. Data in both these sources werefrom varying numbers of reporting States, exceptfor national, State, and county totals, which werealmost complete. In addition, two reports dealtwith detailed marriage data from a single areaone with Georgia, where information about theeducation of bride and groom was collected, theother with the District of Columbia.

During the same period a new source ofinformation was utilized: the Current PopulationSurveys of the Bureau of the Census. Data oncharacteristics of newly married persons weregathered and published in several reports. Itmust be noted, however, that marriage totalsobtained from Current Population Surveys weredrastically smaller than those obtained fromvital registration. Thus according to vital regis-tration data 5,324,700 couples were married inthe United States from January 1955 through June1958, but the survey found only 3,692,000 suchcouples.7For the collection of detailed statistics themarriage-registration area MRA was estab-lished in 1957 and the divorce-registration areaDRA in 1958. The former originally included 28States, all of New York State except New YorkCity, and four outlying areas; the latter included14 States and three outlying areas. By 1967 thenumber of States had increased to 38 and the

District of Columbia for the MRA and 22 for theDRA.

When the MR4 was organized the followingStates were included: Alabama, California, Con-necticut, Delaware, Florida, Georgia, Idaho, Iowa,Kansas, Louisiana including the independentregistration area of New Orleans , Maine, Mary-land, Michigan, Mississippi, Montana, New Hamp-shire, New Jersey, New York excluding New YorkCity , Ohio, Oregon, Pennsylvania, Rhode Island,South Dakota, Tennessee, Utah, Vermont, Vir-ginia, Wisconsin, and Wyoming. Also in the MRAwere the outlying areas Alaska, Hawaii, PuertoRico, and the Virgin Islands. In 1957 Nebraska wasadded; in 1959 Kentucky was added and Alaskabecame a State; in 1960 data for- Hawaii wereincluded with the United States; in 1961 theDistrict of Columbia, Indiana, and Massachusettswere added; in 1964, Illinois and North Carolina;and in 1965, West Virginia and New York ,City.

The DRA, when organized in 1958, includedAlabama, Georgia, Idaho, Iowa, Montana, Ne-braska, Oregon, Pennsylvania, South Dakota, Ten-nessee, Utah, Virginia, Wisconsin, and Wyomingas well as Alaska, Hawaii, and the Virgin Islands.Between 1958 and 1967 the number of Statesin the 13RA increased to 22; Alaska and Hawaiiwere included among the States in 1959 and 1960,respectively, and the following States were ad-mitted: Kansas 1959 , Maryland 1959 , Michigan1961 , Missouri 1961 , Ohio 1962 , and RhodeIsland 1963 .

Since 1960 detailed data have been compiledfrom samples of marriage and divorce recordsfrom the respective registration areas. The firstsample collection, that of 1960, included selectedcounties outside the two registration areas, mak-ing it possible to obtain estimates for the UnitedStates and the four regions. Thus 1960 is the onlydata year since 1932 for which national data otherthan totals are available. In addition to the sample,State totals by county and, for marriages, alsoby month were obtained from every State; fromthese data national totals were compiled. Marriageand divorce statistics for 1960 and subsequentyears were published in Vital Statistics of theUnited States Volume HI, and analytical reports,in Vitul and Health Statistics Series 21. Marriageand divorce statistics for the outlying areas ofAlaska before 1959, Hawaii before 1960, and the


9/68

Virgin Islands and marriage statistics for PuertoRico were published in Vital tatistics of theUnited tates The publication of data for outlyingareas was temporarily discontinued in 1966.

Provisional figures on marriages or mar-riage licenses have been published since 1944;since 1949 they have been published on a monthlybasis. Provisional monthly divorce totals havebeen published since 1950 for a growing numberof reporting States; 40 States and the District ofColumbia reported in 1967.Type of Information ollected

Nationwide data for 1932 and earlier years arebased on information gathered routinely by localclerks and courts for the purpose of issuingmarriage licenses and in court proceedings ofdivorce cases; no special information was col-lected. Divorce data were limited to variablesimportant from the legal viewpoint. In manyStates procedures for the issuance of marriagelicenses and the return of marriage certificateswere unsatisfactory, detailed statistical informa-tion could not be collected, and marriage statis-tics were limited to State and county totals.Marriage totals were intended to represent thenumber of marriages performed in the arearather than the number of marriage licensesissued, but for the years 1867-86 it is notalways clear what they do represent.

The situation improved for the 1887-1906period. The intent ion was to find both the numberof marriages and, whenever possible, also thatof marriage licenses and to use the latter tocheck the former. The authors of the report forthat period felt that: The effort to secure thenumber of marriages celebrated has met withsuccess in almost all the states and terri-tories... he same was true for 1916 and tie1922-32 period. Comparatively complete mar-riage totals made it possible to compute marriagerates for individual States and territories. ForStates with many counties not reporting, popula-tion totals were computed for the reportingcounties combined, and thus a rate could beobtained which served as an estimate for thewhole State. This method was used to analyzethe 1867-86 data in the report of 1887-1906statistics.

Another type of rate was the marriage rateper 1,000 unmarried population. This was basedon population data by marital status, first gath edin the 1890 census. Unfortunately, this rate wascomputed for the unmarried population withoutdistinguishing sex, making it virtually valueless,especially for the newly settled States of thefar and middle West, where the number ofunmarried men often was more than double thatof unmarried women. For the purpose of thepresent report rates for the unmarried wererecomputed by sex.

Detailed marriage data were first publishedfor the marriage-collection area, 25 Statesand the District of Columbia for 1939 and27 States and the District of Columbia for 1940.These data concerned the characteristics,of newlymarried persons. Special issues were devotedto a given characteristic of brides and groomsmarried in the collection area: age, maritalstatus, or age and race. (The latter study islimited to brides.) Data in the three publicationswere given ,by State of residence rather than byState where marriage was performed, and in thisthey differ from all other nationally publishedmarriage statistics. A special publication in-dicated the, relationship between place of marriageand place of residence of bride.

By 1954, the Standard Record of Marriagewas developed and recommended for adoption bythe States. A State marriage record confirmingclosely to the Standard Record became one of thecriteria for admission to the marriage-registra-tion area in 1957. The following characteristics ofbride and groom were included in the StandardRecord: age, race or color, marital status attime of marriage, number of this marriage,residence, and. birthplace. Two further charac-teristics, occupation and type of business orindustry, were on the Standard Record but thisinformation was not required from MRA Statesand not used to compile statistics. Three charac-teristics of the marriage ceremony could alsobe obtained from the Standard Record: month,day of the week, and whether the ceremony wascivil or religious. Not all variables listed abovewere tabulated every year, but all of them weretabulated for one year or more between 1957,when thetypes of

MRA was established, and 1967. Variousrates, percentage distributions, median


10/68

ages, and seasonally adjusted marriage data havebeen prepared for each year.State and county totals of divorces werereported much more completely than the com-parable marriage figures during the data col-lections prior to 1933. In addition, many tablesof detailed statistical information on divorceshave been published, while there were none formarriages. Consequently, publications of mar-riage and divorce statistics 1867-86 through 1932included aImost exclusively divorce data. It isdifficult to say if data on limited decrees, whichdo not give the parties the right to remarry,were included in the published figures. As earlyas 1887 In collecting the statistics of divorce itwas sought to separate, so far as designation isconcerned, the absolute from the limited di-vorces. ITSStatistics on annulments were publishedseparately for the years 1926-32.

In recent times divorce data have referredto absolute divorces and amulments combined,excluding various types of limited divorces. Inmost States where interlocutory decrees aregranted, they have been included in the data forthe year when the decree was rendered; thesedecrees take effect not at the time they aregranted but at a given time in the future, unlessthe parties to divorce become reconciled duringthe intervening period. The completeness ofreporting for divorces in recent years has beenbelow that for marriages, and often there hasbeen at least one State with such poor reportingthat a rate could not be computed. Estimates forsuch States were prepared to obtain nationaland regional rates and totals.In the detailed divorce statistics prior to1933 great attention was paid to legal aspects ofthe divorce case, with particular emphasis on theparty to whom the decree was granted, occasion-ally also referred to as the Iibellant. It seemsthat the terms libellant and party to whomgranted were used interchangeably. Most statis-tical information was given separately for di-vorces granted to husband and those granted towife, often even without corresponding totals;e.g., State divorce data by duration of marriagewere given separately for husbands and for wives,so when median duration for the State is computedtwo figures have to be taken into considerationfor each year of duration.

Eight questions were asked about everydivorce during the 1887 data collection:

1

2*3.4.5.6.7.8.

State and county where divorce was ren-dered.State or county where parties were mar-ried.Year when parties were married.Year when parties were divorced.Number of years married.The cause for which divorced.The number of children by the mar-riage.The party who was Iibellant.

Supplementary questions were asked in 45 countiesconsidered to be representative: 1 date ofmarriage, 2 date of separation, 3 date offiling petition, 4 date of decree, 5 whethernotice was served by publication, 6 whetherintemperance was a direct or indirect cause,and 7 whether alimony was granted.The information sought on the schedule usedin 1907 was much more extensive. In additionto State and county of occurrence, 18 questionswere asked:

12.3.4.5.6.7.8.9.10.

11.12.13.14.15.16.17.18.

State and country in which married.Date of marriage.Date of separation.Date of filing petition.Who was libellant?How was notice served?Was case contested?Was decree granted?Date of decree or judgment.Number of years married.Cause for which divorced.If not direct, was intemperance an in-direct cause?Kind of divorce.Number of children.Was alimony asked?Was alimony granted?Occupation of parties.Residence of libellee.

The 1916 report was limited to legal ques-tions; even data on duration of marriage andchildren were not given. The 1922-32 publicationsalso showed mostly legal data, though data onduration and children were given. No information


11/68

on personal characteristics was included in anypublication. The incomplete information by oc-cupation included in the 1887-1906 statistics wasan exception, but the plan to obtain data by racefor the same period proved to be impracticable.

The first publication that included personalcharacteristics of parties to divorce was thereport Statistics on Divorces and Annulments:Reporting Areas, 1951 Vital Stutistics4peciulRepoYts Volume ,38, No. 3, March 19, 1954 .This report included a table on age of thedivorced husband and wife and marriage orderwhether married once or more than once ,but data from only four States--Connecticut, Idaho,Missouri, and Tennesseewere presented. Sincethat date the importance of personal character-istics has increased considerably. In 1958, whenthe present divorce-registration area was orga-nized, the reporting of age became a prerequisitefor participation in the DR4. Race and the numberof this marriage, though considered highly desir-able, have been waived in individual cases. Asa result of this requirement, many States thatreported duration of marriage, number of chil-dren, or legal variables were not included in theDRA. The statistical items on the StandardRecord of Divorce or Annulment, which wasrecommended to States, were:

2.3.4.5.6.7.8.9.

10.11.12.13.14.15.

State and county of occurrence.Date of birth or age of husband andwife.Usual residence of husband and wife.Place of birth of husband and wife.Number of this marriage for husband andwife.Race or color of husbandUsual occupation of husbandKind of business or industryand wife.Place of this marriage.Date of marriage.Number of children under 18.Plaintiff.To whom decree was granted.Legal grounds.Date of decree.

and wife.and wife.

of husband

Of these items, the reporting of place of birth,usual occupation, and kind of business or in-dustry was never required. For reasons of

6

economy, the coding and punching of all legalitems were temporarily suspended in 1966.Detailed data on divorce characteristics are

available for varying numbers of reporting Statesfor the years 1948-59 and for the DRA for 1960and subsequent years. In addition, national andregional data are available for 1960 for durationof marriage, children of divorced couples, andState of marriage. Annual estimates of thenumber of children involved in divorce havebeen prepared for all years since 1953. Nationaland regional statistics for 1960 and DRA statisticsfor 1960 and subsequent years were preparedfrom probability samples of divorce records.

TIME TREN S INTOT LS N R TES

National Totals ofarriages and ivorces

The total number of marriages and divorcesincreased considerably during the 100 yearsunder study table 1 due to both increases inpopulation and changes in rates. Increases inregional and divisional totals were also in-fluenced by the distribution, of the unmarriedor married population in various parts ofthe country as well as the location of marriageand divorce mills places where many out-of-State people are married or divorced . Theincrease was more pronounced for divorces thanfor marriages: In 1967 there were more than50 times as many divorces in the United Statesas in 1867 10,000 to 523,000 but only 5.4times as many marriages 357,000 to 1,927,000figures 1 and 2 . During the 100 years annualmarriage totals increased in 70 of the years,declined in 29, and once remained the samefor two consecutive years. For divorces, thecomparable figures were 70, 20, and 10. Thelowest observed marriage total was 345,000 in1868, the highest 2,291,000 in 1946-a ratio of1 to 6.6. Divorce totals ranged from 10,000 to610,000.

Data on annulments of marriage, wheneveravailable, are included in the divorce totals. Inrecent years all States have reported annulmentscombined with divorces, though usually one Stateor more has failed to give the annual totals


12/68

2 4r 12 4002 200

2 C03I

t

1 2 CCQI i2 200/LJ/

1 8C0

1 6CQA

200 t J 2@J I I I I I I I 1 I o1660 1870 1690 1890 leoo 1910 1923 1930 le40 19ao 1960 1970

YEAR

Figure 1. Number of marriages: United States 1867 1967.

for the two types of decrees separately. Theoret-ically, annulments and divorces differ in thatmarriages are annulled for legal grounds thatarose before marriage and made the marriagevoid from the beginning, while divorces aregranted on legal grounds that arose after alegally valid marriage was performed. In manyStates, however, statutes do not follow this dis-t inction, and often divorces are granted on groundsthat arose before marriage or annulments ongrounds which arose after marriage. The numberof annulments is insignificant except in Californiaand New York table 2 . In 1964, the only recentyear when all States reported annulment totals,annulments comprised only 3.1 percent of divorcesand annulments combined. This percentage variedfrom 2.1 to 2.4 in 1927-32, another period whenall States reported.

ational arriage RatesThe net change in the national marriage rate

during the 100-year period was almost nil: 9.6per 1,000 population in 1867 and 9.7 in 1967.During the intervening years the rate dipped aslow as 7.9 in 1932 and rose to 16.4 in 1946, but90 annual rates fell within the interval 8.5through 11.4 per 1,000 table 1 .

Marriage rates were comparatively stableduring the late decades of the 19th century, with30 of the 34 annual rates varying from 8.7 to9.2 per 1,000 without showing a noticeable trend.The rate started to increase after the turnof the century, and this trend was strengthenedby the effects of World War I. The high point ofthis increase was a rate of 12.0 per 1,000 in1920. After 1920 the trend reversed itself as theeffects of war wore off, and soon afterward

7


13/68

700

600 600

500 Boo

400 400

300 300

200 200

10 0 100

0 I I .01860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970

YEAR

Figure 2. Number of divorces: United States 1867 1967.

the Great Depression intensified this decline.The rate reached the lowest point of 7.9 in1932.

A second cycle of increase and declinefollowed 1932. The rate increased to 12.1 by1940 and sustained by the effects of World WarII reached 16.4 in 1946. This was followed by adecline that brought the rate down to the secondlowest figure ever observed .4 in 1958. Aftera short period of stable rates at the 8.5 levelin 1959-62 the rate began to increase and roseslowly to 9.7 in 1967. h is still impossible totell whether this increase is the beginning ofa third cycle of increase and decline comparableto those of 1900-32 and 1932-58.

The number of marriages depends amongother things on the maximum number of coupleswho could possibly be married. When the numbersof unmarried men and women including thesingle widowed and divorced are approximatelyequal either figure can be considered a goodapproximation of potential couples or of themaximum number of possible marriages. In

areas where the number of unmarried womensignificantly cliff ers f rom the number of unmarriedmen the figure for the less numerous of thetwo sexes represents the maximum number ofpossible marriages; even if all unmarried personsof one sex were to get married there wouldremain a group of unmarried persons of themore numerous sex who could not possibly marry.

During the 1OO-year period under considera-tion the sex ratio of the unmarried populationchanged drastically. In 1890 the first year whenmarital status information was collected therewere 112.5 unmarried men aged 15 years and overper 100.0 unmarried women of that age. As manyunmarried male immigrants entered the UnitedStates in years after 1890 that ratio increased to115.3 in 1910. With immigration interrupted byWorld War I subsequent legislation limitingimmigration and a number of young men killedduring the wars the sex ratio of the unmarrieddeclined to 99.3 in 1940 and 83.4 in 1960.

National marriage rates per 1 000 unmarriedwomen aged 15 and over are shown in table 3 for


14/68

1940-67 and Census years since 1890. Rates per1,000 unmarried men and women aged 15 andoverfor census years are showh in table A. The higherof these two rates for a given year is the rateper 1,000 possible marriages; for years 1890-1930it was the rate for women, while for 1940-60 itwas that for men. This rate increased from 67.4 in1890 to 88.1 in 1960, with a maximum of 100.1 in1950.Table A. Marriage rates for unmarried menand women aged 15 and over: UnitedStates, census years 1890-1960

[R a t e p er 1 p op ul a t ion ]l 1

1960 --------------1950 --------------1940--------------1930--------------1920--------------1910--------------1900--------------1890--------------

88.1100.183.464.585.266.961.359.9

Women

73.590.282.867.892.377.168.267.4

National Divorce Rates

Thetrend of the divorceratediffers fromthatof the marriage rate by showing a consistent in-crease for aperiodof80years, 1867-1946 table 1and figure 3 . Although thisincreasewasacceler-atedby two World Warsandtemporarily reversedby the Depression, the trend wasresumed assoonas the situation returnedtonormal. Duringthe80-year period an annual rate was smaller than thatfor the preceding year in only seven cases 1913,1918, 1921,1922, 1930, 1931, and 1932 .

The time trend became completely differentafter 1946. It could be expectedthattheextremelyinflated post-World War 11rates would revert toa more normal size and that the increase wouldthen be resumed. This, however, did not happen.The inflated rate declined from 4.3 in 1946 to2.3in1955 and fluctuated from 2.1 t02.3 between1955 and 1963. A new wave of increases began

in 1963, andtherate reached 2.6in1967+ growthof 13 percent in 4 years. The rate was 9 per1,000 for 1968, and it increased to the provisionalrate of 4.0 for 1972, so this increase continuedafter 1967 and lasted much longer than that whichoccurred during and after World War H.During the first 60 years under considerationthe divorce rate increased on the average about75 percent every 20years, and if this increase hadcontinued, rates of 2.8 in 1947 and 4.9 in 1967 couldhave been expected. The year 1947 fell into theperiod of post-World War II increase, so theobserved rate was much higher than expected,but the following year, 1948, it was 2.8. In spiteof rapid growth in the divorce rate since 1963,the observed rate for 1967 was still much lowerthan that which could be expected on the basis ofthe 1867-1927 experience.The population exposed to the risk of divorceare all married couples, including those who livetogether and those who live apart S the numkrof married couples, particularly those livingapart, is not known, that of all women, or all men,reporting themselves as married can be used asan approximation. There are differences betweenthe two numbers, but these differences arecomparatively small and their effects on the rateare not pronounced. Therefore rates were com-puted only for women table 4 .

These rates indicate that in 1890, the firstyear with data by marital status, 3.0 couples per1,000 were divorced; this rate increased to 8.0in 1920, 8.8 in 1940, and 17.9 in 1946. It declinedto 8.9 in 1958 and then resumed the increase,reaching 11.2 in 1967. This increase is lesspronounced than that of the rate per 1,000 totalpopulation: The 1967 rate per 1,000 marriedwomen was 3.7 times as large as the 1890 rate,while a comparable ratio was 5.2 for the cruderate and 15.8 for the number of divorces granted.

Because of a rapid and consistent increase inpopulation, the number of divorces grew muchmore rapidly than the rate. The 1967 crude ratewas 8.7 times as large as that for 1867, but thenumber was over 50 times as large; from 1963 to1967 the rate increased 13 percent, but thenumber, 22 percent. Rates higher than in 1967were found during the six years 1944-49, but atotal higher than that for 1967 was found onlyonceAI0,000 in 1946.


15/68

11

16

15

14

13

12

11

10

9

a

7

6

5

4

3

2

1

Marriage

17

16

15

14

13

12

11

10

9

8

7

6

5

4 3 :{e

41 M ti , ,, , sf @, , , I

W 2+s -o, , , d+ a, , , , \

, 0, , + , , W, , , , , , , , , , f i um . . l ml l nm s n9nt 1 91F Io I I I I I I I o1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1980 1970

YEAR

Figure 3 Marriage and divorce rates per 1 000 population: United States 1867-1967.

rea Totals of Marriages and ivorces

Though both marriages and divorces in-creased throughout the United States during the100-year period, the increase varied considerablyamong regions and geographic divisions tables 5and 6 . For divorces the highest rate of increasebetween 1870 and 1967 was found in the West,followed by the South, North Central, and North-east Regions in that order. The increase in theNew England Division was the lowest; that inthe Pacific Division was the highest. Marriagetotals for 1870 could be estimated only for tworegions, the North Central and the West, and theratio between the 1967 and 1870 totals was 4.1and 46.4, respectively. From 1870 to 1967 the rateof increase in marriages in the Pacific Division

was 14 times the rate of increase in New Englandfor divorces this figure for the Pacific was 23times that for New England. Ratios between the1870 and 1967 totals for geographic divisions areshown in table B, though 1870 marriage totals arenot available for five of the nine divisions. If 1870figures are equal to 1.0, the 1967 ratios are asshown in table B.

Due to differences in the rate of increase,the distribution of marriages and divorces amongregions and divisions changed considerably duringthe period under study tables 5 and 6 . In 1870,77 percent of all divorces were granted in thetwo northern regions. This percentage declinecl to63 in 1890, 51 in 1930, and 37 in 1967. Simul-taneously,increased

the percentage granted in the Southfrom 17 to 27, 32, and 37, and the


16/68

Table B. Ratio of 1967 to 1870 marriageand divorce totals: United States andeach division[187 t ot a ls eq ua l 1 ]

DivLsion

United States-- w EnglandMiddle Atlantic ------East North Central---West North Central---South Atlantic -------East South Central---west South Centxal---Mountain -------------Pacific --------------

arriage5.52.7---4.7------2.7------37.0

Di.vorce

47.5 10.722.924.728.81:;.;

193:5160.0246.4

percentage granted in the West, from 6 to 11,i8, and 26.Formarriages thedistributionchangedmuch less, but the direction of the changes wasthe same, as can be seen in table5. One of theprincipal causes of these changes is the shiftinpopulation from East to West. Hence changesinthe rates are less pronounced than those of thetotals.rea arriage Rates

In the early years of marriage and divorcestatistics marriage data weremuchless completethan those on divorce. This applies particularlyto the 1887 collection, when no satisfactorymarriage figures were received from a largenumber of States and territories table 7 . Crudemarriage rates for 1870 could be computed foronly 10 States and the District of Columbia, rateswere limited to reporting counties in 15 States,and they could not be obtained for 25 States andterritories. For 1880 the comparable numberswere 17 and the District of Columbia, 12, and21. More complete totals began with the 1907collection. In 1890, when rates were availablefor all States and territories except Maine andSouth Carolina, the magnitude of State ratesvaried much less than at present, and due toless developed means of transportation, few ifany marriage mills existed. Marriage rates in

newly and sparcely settled areas were low due tothe scarcity of unmarried women.

Before 1940 there was little difference inmarriage rates among geographic divisions corn-pared to more recent years, when the MountainDivision often had rates almost twice those of thedivision with the next highest rate table 7 . Thegrowth of the marriage rate in the MountainDivision was due to development of marriagemills, particularly in Nevada, established sincethe late 1920s and early 1930s. Variation inrates for the remaining divisions depended inpart on the general level of the national marriagerate; the higher this rate, the wider the rangeof divisional rates, which widened from 1890 to1940 and narrowed afterward. In recent decades,the lowest marriage rates were found in the twodivisions of the Northeast Region New Englandand Middle Atlantic and the Pacific Division. Theproximity of the Mountain Division marriagemills may explain the comparatively low rates inthe Pacific Division. The two divisions in theNorth Central Region had rates higher than thosein the Northeast but, as a rule, lower than thenational average. The highest rates outside of theMountain Division were found in the South from1890 to 1916, in 1950 in the West South Central,in 1930 and 1940 in the East South CentraI, andin 1960 and 1967 in the South Atlantic Division.

The number of marriages depends in part onthe size and the sex ratio of the unmarriedpopulation of an area. The sex ratio for the un-married population has varied considerably amongthe States and has often differed from that for theNation. Therefore State rates per 1,000 unmarriedmales and per 1,000 unmarried females aged 15and over were computed for years in whichpopulation figures by marital status were availabletable 8 .

Since 1890 the number of unmarried womenhas been larger than the number of unmarried menin the Northeast and parts of the South. Occasion-ally the two figures were practically identical, asin New Jersey in 1930, where the marriage ratewas 49.0 for men and 48.8 for women. In mostareas west of the Mississippi River numbers ofunmarried women were comparatively very smallin the late 19th century and beginning of the 20th,often resulting in marriage rates for women thatwere three times or more the rates for men. One


17/68

of the largest differences was found in Montana in1890: 27.4 per 1,000 men but 141.5 for women, aratio of 1 to 5. Wherever unmarried women werescarce, the likelihood for men to get married wasmuch lower than elsewhere, but the likelihoodfor women was extremely high. Often well over10 percent of all unmarried women over 15were married during a single year. As thesettlement of the Western States progressed andtransportation improved, gaps between the tworates narrowed and in some cases the directionwas reversed. The only area where the oldpattern still prevailed in 1960 was Alaska, wherethe marriage rate was 54.3 per 1,000 unmarriedmen but 152.5 per 1,000 unmarried women.

Area Divorce Rates

of 1870. Nationally the rates were 2.6 and 0.3table 9), and their ratio was 8.7; this ratio waslowest for the Northeast only 3.7), approximatelyequal for the North Central and the West 6.5and 6.8, respectively), and very high 15.5) forthe South.

Divorce rates per 1,000 married womenpresent a geographic distribution very similar tothat of the rate per 1,000 total population table10). It should be noted that in all regions theincrease of the rate per 1,000 married coupleswas less pronounced than that of the rate per1,000 total population. AS information aboutmarital status was first collected during the 1890census and rates have been computed only forcensus years, 1890 and 1960 rates are compared.If the 1890 rates are equal to 1, the 1960 rateshave the values shown in table C.

In recent decades the geographic distribution Table C. Ratio of 1960 to 1890 divorceof divorce rates by magnitude has been invariablythe same table 9). The Northeast Region, partic-ularly the Middle Atlantic Division of that region,had the lowest rates, followed by the NorthCentral, South, and West Regions, in that order;the highest rates were found in the MountainDivision of the West. This pattern represents aconsiderable change from the early years of theperiod under study. Originally the South AtlanticDivision had rates lower than or equal to theMiddle Atlantic. In 1870, 1880, and 1890 the ratefor the South Region was similar to that for theNortheast, and as late as 1930 the rate for theNorth Central Region was higher than that forthe South. This was due mostly to high rates intwo North Central States, Indiana and Missouriwhich still have higher rates than other Statesin that region). Divorce rates of the threedivisions of the South Region have increased morerapidly than those of other areas. The onlyfeature that remained unchanged throughout thewhole period was that the rate for the WestRegion was always 2 to 4 times the lowestregional rate for the same year. On the otherhand, the great difference between the rates forthe Mountain and the Pacific Divisions of theWest is of comparatively recent origin; as lateas 1930 the two rates were approximately equal.

The changes in divorce rates can be sum-marized by using ratios of the 1967 rate to that

rates per 1,000 population and per 1,,000married couples: United States and eachregion[189 r a t es e qu a l 1 ]

~

UnitedStates -----uNortheast --------- 2.3 ,North Central ----- :::south ------------- ;:;West -------------- ::: .

in Miwr age n i vo rce 1867-1906, divorcerates for the census years 1870, 1880, 1890, and1900 were shown for counties that include 42major cities. These rates were brought up todate by adding values for 1930, 1950, 1960, and1965 county divorce totals were not published for1940) without including any new areas table 11).About three-fourths of the city rates were higherthan the corresponding State rate, for almost 20percent of the cities both rates were equal, andin a few instances the State rates were higher.In the 19th century, when the level of divorce


18/68

rates was generally low, only about 60 percent ofthe city rates were higher than the State rates;in the first half of the 20th century this proportionincreased to 80 percent, while in the 1960s itdeclined to 70.

The rates for all city areas increased con-siderably from 1870 to 1965. The largest in-crease was found in Richmond, Virginia, wherethe 1965 rate was 32 times the 1870 rate; thesmallest was in New Haven and Waterbury,Connecticut, where it grew by about one-fourth.

H R TERISTI SOF M RRI GES

Detailed marriage and divorce statisticsrefer either to the events themselves or to thepersons involved. Data are collected on charac-teristics of marriages or of divorces as well ason characteristics of brides and grooms and di-vorcing husbands and wives. Each marriage ordivorce is represented by one statistic for eachcharacteristic of the event e.g., the month whenit occurred but by two statistics when charac-teristics of spouses are given e.g., one for theage of the wife, another for that of the husband .

During this period less data on character-istics of the event were collected for marriagesthan for divorces. For many decades informationabout marriages was limited to area totals. It wasnot until 1941 that data by month became availableand not until 1960 that they covered the entireUnited States. The first monthly statistics werepublished by the Bureau of the Census in Popu.hZon Series PM-1, No. 4, under the titleMarriage Licenses Issued in Cities of 100,000Inhabitants or More, 1939 to 1944, With Statisticsby Month, 1941 to 1944. The title points to twolimitations of the data: they represent marriagelicenses instead of actual marriages, and theycover only 34 major cities and the 57 countiesthat included the remaining major citiesa totalof 91 local areas where about one-third of allmarriage licenses were issued. These earlymarriage license data are of use for comparativepurposes. They show the still-familiar seasonaldistribution, with a peak in June and lowestfigures for the first 3 months of the year.Beginning with 1960 statistics on two othercharacteristics of marriages have been published:

Table Percent distribution of marriagesby day of week on which marriage tookplace: United States, 1960

Day of week Percentdistribution

Total ---------b

Sunday -------------- 11.3Monday --------------Tuesday ------------- :::Wednesday ----------- 5.9Thursday ------------Friday -------------- 1;::Saturday ------------ 46.7

day of the week when they were performed andtype of marriage ceremony, whether civil orreligious.

The variation in numbers ofmarriages bydayof the week is even more pronounced than that bymonth, with almost one-half of all marriagestaking place on Saturdays. National data by day ofthe week are available for 1960, and their distri-bution is shown in table D. MRA figures for 1960and subsequent years show a very similar pattern.Due to this distribution, marriage totals for anymonth are affected by the number of Saturdays inthe month, and this source of variability has to betaken into consideration when monthly figures aremathematically adjusted for seasonal variation.

Marriages by type of ceremony show thenumbers of religious and civil marriages, and forsome years the former are classified by whetherthe clergyman was Protestant, Catholic, Jewish,or of another religion. The publication of sta-tistics by type of ceremony was temporarilydiscontinued in 1966 during program cutbacks re-sulting from budgetary restrictions. For all yearsfor which data are available the overwhelmingmajority of marriages were religious, and theproportion of religious marriages was higher forbrides and grooms, who were marrying for thefirst time than for those who had been previouslymarried. The respective percentages are shown intable E.

13


19/68

Table E. Percent distribution of marriages by whether ceremony was religious or civil,according to marriage order: United States, 965 and 1961965 196

Marriage orderReligious Civil Religious Civil

BrideFirst marriages -------------------------- 8 6 19 4 83 6 16 4Remarriages ------------------------------ 57 5 42 5 61 6 38 4

GroomFirst marriages -------------------------- 80.0 2 82 7 17 3Remarriages ------------------------------ 59.9 4 1 63 9 36 1

H R TERISTI SOF DIVOR ES

Duration of Marriage Prior to DecreeData on characteristics ofdivorcesare avail-

able for some of the very earlyyears; interestingcomparisons and contrasts can be made betweenthen and recent times. While the different setsof data may not be strictly comparable, they arereasonably acceptable as general indicators fora few basic comparisons.

The duration of marriage at time of divorcedecree is one of the few variables collectedsince1867. During the 1867-1906 period the durationofmarriage was computedby subtractingtheyearofmarriage from the year of decree; since then ithas been computed using the year and monthofeach event. The difference in method may affectthe percentage of eventswithaveryshortdurationsuch as less than 1 year or 1 year, but bothmethods should yield similar data for longerdurations.

The median duration of marriage was 7.4years in 1867 for the United States and7.1 yearsin 1967 for the DRA. The difference between thetwo figures seems small. During the interveningyears the median varied from aminimumof 5.8years for the 16 reporting States in1950t08.3in the four years 1889, 1891, 1900,and1901. Theratio between these two figures is only 1.4, muchlower than the comparable ratio for the crudedivorce rate. In fact, even disregarding years

for which the median duration is not availableand thus omitting the high post-World War IIrates, the ratio between the highest and lowestcrude rate is 9.3.

The time trend of the median duration table 12and figure 4 shows a decline during the first fewyears of the period under study, from 7.4 yearsin 1867 to 6.7 in 1871, followed by more than 10years of increasing duration, reaching 8.0 years in1887. In the subsequent years 1887-1905 themediins were high and varied little 8.0 to 8.3years , with what appears to be an incipient de-cline in 1905 and 1906. When data again becameavailable, the medians were much lower from 7.9years in 1906 to 6.6 in 1922 ; however, they in-creased to 7.1 in 1931 and 1932. During thesubsequent break in divorce statistics the mediansdeclined again to 6.4 years in 1948 and 5.8 in1950. From 1950 to 1963 there was another pe:riodof increase, followed by a slight decline 7.5 in1963 to 7.1 in 1967 .

These changes seem rather erratic and it isdifficult to explain their relation to the nationaldivorce rate. The median duration has increasedwhile the rate was increasing 1871-90 and1922-29 and while it was declining 1929-32 and 1950- 57 ; the median has also declined while therate was increasing 1963-67 . The duration ofmarriage declined during the two periods for whichdata are not available 1907- 21 and 1933-47 , asmedians for the years preceding the break in datawere considerably higher than those when datacollection was resumed. The medians declined


20/68

9 9

2 8>

8gz0gK 7 -2

/]

7z~~z 6 6

5 - 50 1 I I I I I I 1 I I 01860 1870 1880 1890 1900 1910 1920 1930 1940 1950 Iwo 1970

YEAR

Figure k Median duration of marriage prior to divorce: United States 1867-1906 1922-32 n I Jaw; reportingStates 19W-69; and d i vorce-reg i strat ion areaj 196 I -67.

from 7.9 years in 1906 to 6.6 in 1922 and from7.1 in 1932 to 6.4 in 1948. As each period forwhich data are not available included a World War,it seems likely that the disorganization causedby major upheavals is conducive to shorterdurations of marriage.

Duration medians for individual States inselected years table 13 indicate that the rangeof State medians has varied between about 4 and12 years. In general medians were high in theNortheast Region, particularly in Massachusettsand hew Jersey. They were low in the WestSouth Central Division and showed pronounceddeclines in the West Region.The percent distribution of divorces by dura-tion of marriage has changed during the periodunder study table 14 . The most important change,the increase in the number of couples with a veryshort duration of marriage, less than 2 years,may be partly due to changed methods of com-puting duration. Still, the change in methodscannot explain the fact that the proportion ofthose divorces tripledfrom 5 percent in the 19thcentury to 15 percent in the 1960s. Percentagesof divorces with duration of 5-9 years and, to alesser degree, those with 10-14 years declinedduring the same period.

Duration of Separation

Marriages cease to function normally whenthe spouses separate, and separation often, thoughnot always, is a preliminary to divorce. The dura-t ion of marriage prior to divorce divides into twomain periods: 1 duration of marriage prior toseparation and 2 duration of separation prior todecree. Before 1967, statistics on separationwere collected only once, in 1907, and publishedfor the entire 20- year period 1887-1906 for theUnited States and every State. These data referonly to couples who divorced during that period,and no data exist on the incidence of separationthat did not end in divorce.

The modal duration of marriage to separationwas 1 year or less, and 27 percent of couplesdivorced during the 1887-1906 period had livedtogether less than 2 years table 15 and figure 5 .Couples for whom the duration of marriage todivorce was less than 2 years comprised only 5percent. As a couple who divorced after a givenduration of marriage had to be separated withinthe same period, it can be seen that 5 percentwere both separated and divorced less than 2years after marriage, while 22 percent wereseparated less than 2 years after marriage but

15


21/68

n

30 I

I

: 1

~15

201

Marrwe to dworcemmmmtee Man mge to separation-- Separation to dwarce

1[A\ ,,, 5 10 15 20

DURATION IN YEARS

Figure 5 Percent distribution of divorces by duration of mar-riage and of separation: United States, 1867-1906.

divorced at a later date. The same generalizationapplies to any cumulative percentage of duration.Cumulative percentages of couples who separated,separated and divorced, and separated but di-vorced later are shown in table F. The lastcolumn was obtained by subtraction.

The time elapsing between separation anddivorce depends not only on the decision of the

spouses but also on statutory rules of the States.It takes a certain period of time for some legalgrounds to arise, and there are varying residencerequirements and procedural rules. The durationof separation was usually short: for 35 percentof the couples it lasted less than 2 years; for 44percent, 2-4 years; and for 17 percent, 5-9 years.Only a small proportion of divorces, less than 5percent, was granted after 10 years or more ofseparation.

In the years 1887-1906 the median durationof marriage to divorce in the United States was8.1 years, and State medians ranged from 6.3years in Arkansas to 11.4 in Massachusetts. Themedian duration of marriage to separation was4.7 years, ranging from 3.4 in Arkansas to 6.7 inNew York. The median duration of separation todivorce was 2.8 years in the United States; itvaried from 1.8 years in Indiana to 5.7 inMassachusetts table 16 .

Median duration of marriage ending in divorceduring the 20-year period depended also on theparty to whom the decree was granted. There waspractically no difference in the median durationof marriages at decree for the United States itwas 8.1 years when granted to the husband and8.2 when granted to the wife . However, the simi-larity disappears when the whole duration isdivided by the separation. The median duration ofmarriage to separation was 4.5 years when thedecree was granted to the husband and 4.9,, whento the wife, while the median duration of theseparation was 3.1 and 2.7 years, respectively.Thus when decrees were granted to the wife, asusually happened, the marriage had lasted longerbefore separation but not as long after separation.

Table F. Cumulative percentages of couples who were separated, separated and divorced,and separated but divorced later, by duration of marriage prior to divorce: UnitedStates, 1887-1906Separated SeparatedDuration of marriage Separated and but divorceddivorced later

Less than 2 years ----------------------------- 27.0 21.8Less than 5 years ----------------------------- 51.9 2;:; 23,.6Less than 10 years ---------------------------- 74.6 59.7 14..9Less than 15 years ---------------------------- 86.5 77.7 8.8Less than 20 years ---------------------------- 93.1 87.9 51.2

16


22/68

Place of MarriageInformation about the place where divorcing

couples were married was obtained for the UnitedStates and each State or territory for every yearof the 1867-1906 period and for 1916. National,regional, and DRA figures were obtained for 1960,and there are data for limited numbers of Statesfor the years 1952-65. This information was nottabulated for 1966 and 1967. Classification of theplace of marriage has varied considerably, butin all cases it is possible to compute the propor-tion of divorced couples who were married in thesame State in which they were divorced. For theUnited States a pronounced decline in this propor-tion took place during the period under studyfrom 79.2 percent in 1880 to 57.5 in 1960 table17 . The percentage has declined since 1870 in theNortheast Region, since 1880 in the North Centraland South, and since 1900 in the West. Thesechanges are due to two factors influencing thetrend in opposite directions: improved means oftransportation and settlement of new areas.Better transportation permitted more spatialmobility, which lead to 1 a higher likelihood ofmarriages performed outside of the State wherethe couple intended to settle, e.g., in a marriagemill or in the parental home of the bride, 2easier migration of married couples, 3 easiermigration of separated persons before divorce,and 4 more migratory divorces in divorce-millStates. All these factors contributed to an in-creased percentage of couples who were marriedoutside the State where divorced. On the otherhand, a high proportion of the married personsin areas that were still sparcely settled in the19th century had been married before movinginto the area. This percentage declined with timeand became more similar to that for the rest ofthe country, as in the case of the Dakotas,Oklahoma, and some States in the West Regiontable 17 .hildren of the ivorced

Information about children of divorcedcouples goes back to 1867 but is rather sketchyand incomplete. For the years 1867-1906 onlythree categories were given: couples with chil-dren, couples without children, and couples who

gave no information as to children, the lattercategory often being the most numerous. The totalnumber of children for couples reporting childrenwas published without information on how it wascomputed. Thus the only measure that could beused was the mean number of children per divorcewith children. This measure for the whole periodunder study is shown in table 18 and figure 6.

The means for the first collection period,1867-86, are quite different from those for thesecond collection period, 1887-1906. During thefirst 20 years the number of children per divorcewith children varied from 2.02 to 2.10; during thesecond period, from 1.84 to 1.92. The mean numberfell from 2.10 in 1886 to 1.91 in 1887. It wouldseem that this difference is due to the methodused for compiling the data. As information fromthe second collection is, in general, more accuratethan that from the first, reliable time trend databegin with 1887.

During the years 1887-1906 the mean declinedfrom 1.91 to 1.84. Exactly the same mean, 1.84,was found when publication of data on children wasresumed in 1922. At that point a decIine set inagain, and the low point of 1.74 was reached in1932. This decline was not due to the Depressionbecause it had already begun when the Depressionstarted, but the Depression may have acceleratedthe process. In the 1950s and 1960s the meannumber of children per divorce with children in-creased in the reporting areas from 1.76 in 1950to 2.18 in 1967.The available figures for regions, divisions,and States in selected years table 19 indicatethat the mean number of children per decree withchildren was lowest in the Northeast Region inall years for which regional ratios were com-puted. However, in 1967 this pattern seems tohave changed, as all available ratios for thereporting southern States are lower than thosefor any reporting State outside of the SouthRegion. Highest ratios were found in the Southin 1890, 1900, and 1922, and in the North CentralRegion in 1930 and 1960.

Legal Grounds

When divorce statistics were collected di-rectly from court files, the only available infor-

17


23/68

zz~

2.1

2.0

1.9

1.8

1 7

OL1860

I I1870 1880 1890 190Q

I I I1910 1920 1930

YEAR

I I1940 1950 1960

2.2

:?.1

:?.0

1.9

1.8

i1870

Fiaure 6. Mean number of chi 1drenper di vorce w th chi l dren reported: United States 1867 1906 1922 32 and 1860;report i ng States 1946 59; and di vorce regi strat i on area 1961 67.

mat ion was that routinely obtained duringlitigation. The majority of data collected in1887, 1907, 1917, and 1922 -32 were legal. Whenthe collection of divorce statistics was resumedafter the Depression, emphasis had shifted fromlegal to demographic variables. States that re-ported complete legal data but had no informationon personal characteristics of husband and wifewere not admitted to the DRA. In the years 1940-59only two legal variables were tabulated: legalgrounds for the decree and the party to whom thedecree was granted. In 1960 data identifying theplaintiff were added. The time series on legaldivorce statistics was broken in 1966, when it wasdecided not to tabulate the three legal items.Legal grounds for divorce are listed in Statestatutes, and the true cause of estrangementbetween the spouses can be used as a legal groundin court proceedings only if it coincides with oneof the listed grounds. These grounds vary fromState to State and can be changed by acts of Statelegislature. Still, there are types of behavior thatconstitute legal grounds for divorce in all or mostStates.the

18

Over 95 percent of all divorces granted duringperiod under study for which information is

available, were granted on the 10 grounds listedin table 20. In many cases similar grounds calledby different names in various States are com-bined in the table. Throughout the period understudy the importance of individual grounds haschanged. This may have been due to severalreasons: 1 new grounds were added or old onesreformulated, 2 the time period necessary fora ground to arise has changed, and 3 the attitudeof the parties has changed. The wish to have thelegal grounds reflect, as far as possible, theactual situation may have been replaced by thewish to use the most anodyne ground. The percent-age of decrees granted for adultery declinedthroughout the whole period, from 25 percent in1867-86 to 1. 4 in1965, orataratio of 18 to 1. Thepercents of decrees granted for desertion declinedin a ratio of 3 to 1; drunkenness declined also. Itmust be remembered that these declines do notnecessarily mean that there is now less adultery,desertion, or drunkenness than there was 100years ago. On the other hand, cruelty as a legalground increased 160 percent, and this is no doubtconnected with the broad interpretation given tothis term, especially to mental cruelty, in recentdecades; neglect and nonsupport increased also.


24/68

Such causes as incompatibility, separation, andindignities are used now more often than in thepast. In the years 1867-86 over two-thirds of alldivorces were granted for adultery and desertion.In 1965 roughly the same proportion were grantedfor cruelty, neglect or nonsupport, and indignities.

Party to Whom DecreeWas ranted

Information about the party to whom thedivorce was granted was considered particularlyimportant for div,orce statistics in the years1867-1932. Most published data were given sep-arately for divorces granted to the husband andthose granted to the wife, with even the totalfigures occasionally omitted. Information abouttwo aspects of this variable can be obtained: 1the distribution of divorces by party to whomgranted and 2 the relationship between theparty to whom the decree was granted and variousother variables. The first aspect is shown intable 21.

During the whole period under study the over-whelming majority of divorces were granted to thewife, and this majority increased slightly through-out the period. There is a definite territorialpattern: The proportion of decrees granted towomen in the South, particularly the South AtlanticDivision, was always lower than in other areas.During the early years of divorce statistics theoverwhelming majority of decrees in severalsouthern States were granted to husbands, butthis majority disappeared about the turn of thecentury. On the other extreme, wives have ob-tained about three-fourths of all decrees in theWest and, since 1916, in the North Central Region.

The classification of most items by party towhom decree was granted, particularly in yearspreceding 1933, permits some insights into the im-portance of this variable. In the section on sep-aration there was a short discussion of the dif-ferences in the duration of marriage to separationand in that of separation to decree between casesgranted to the husband and those granted to thewife during the 20- year period 1887-1906. In 1870and 1880 the median duration of marriage to de-cree was higher when the decree was granted tothe wife table 22 . Both medians became identical

in 1890 and 1900, and later the relationship wasreversed. In 1930, 1960, and 1965 higher medianswere observed for divorces granted to the husbandthan for those granted to the wife.In all years when information was available,the mean number of children per decree withchildren was higher when the decree was grantedto the husband than when it was granted to thewife; these means were 1.95 and 1.89 in 1890, 1.96and 1.84 in 1900, and 1.81 and 1.77 in 1930. Thedefendant husband was more likely than the de-fendant wife to reside in the State where thedecree was granted; 82.9 percent of defendanthusbands and 73.7 of defendant wives resided inthe divorce State in 1960, while for 1965 thecomparable percentages were 86.1 and 75.4.Other egal Variables

Information about several other legal vari-ables was collected at various times before 1933.In most cases the period of data collection wastoo short to yield any information of interest, andtime trend data are available for only two of thesevariableswhether the case was contested andwhether alimony was granted.

The proportion of contested divorce caseshas always been comparatively low: 15.4 percentfor the 1887-1906 period, 14.8 in 1922, decliningto 11.9 in 1929, and increasing to 14.1 after theDepression began. This increase seems to con-tradict the theory that divorces declined duringthe Depression because of costs, as a higherpercentage of defendants spent money in contest-ing the case. In absolute numbers, contesteddivorces declined 10.1 percent from 1929 to 1932,but they declined much less than uncontested di-vorces 20.9 percent .Information about alimony is found for the1887-1906 period, for 1916, and for six States in1939. The percentage of cases with alimony askedincreased from 13.2 in 1887-1906 to 20.3 in 1916for 1939 this figure is not available , and thoseof alimony granted from 9.2 to 15.3, and to 29.0in 1939. This left the proportion of cases withalimony refused almost unchanged: 4.0 percentin 1887-1906 and 4.8 percent in 1916. In 1916alimony was awarded to the wife in 4.7 percent ofdivorces granted to the husband and in 20.1 percentof those granted to the wife.

19


25/68

PERSON LCH R CTERISTICS

Collection of information on demographiccharacteristics of bride and groom and of divorc-ing spouses did not begin until the publication ofmarriage and divorce statistics was resumedafter the Depression. For divorces these dataare still very incomplete. For marriages the best-covered items are age at marriage and marriageorder, i.e., whether a person was married for thefirst time or remarried. From these data medianage of bride and groom was prepared for mostStates in selected years since 1940 table 23 andmedian age by marriage order for a group of 22States for the years 1954-67 table 24 .

The overwhelming majority of reportingStates showed declines in median age for bridesfrom 1940 to 1950, from 1950 to 1955, and from1955 to 1960 but increases from 1960 to 1967.For grooms the declines observed in earlieryears continued through the 1960-67 period. Re-cent increases in the median age of brides ac-

companied by declines of that for grooms indicatea narrowing of the difference in age betweennewly married spouses. Median age for groomsdeclined from 1960, to 1967 in all four regions;for brides the median increased slightly in theSouth and the North Central these two regionscombined accounting for 61 percent of all mar-riages in 1967 and did not change in the North-east. The median for the West depends in part onassumptions made about age distribution in fiveWestern States where information on age is notavailable Arizona, Colorado, Nevada, NewMexico, and Washington , as about 45 percent ofmarriages performed in the West Region wereperformed in these five States.Declines in the 1950?s and subsequent in-creases in the 1960s can be observed in medianages for all brides and first-married brides in agroup of 22 States, while the comparable mediansfor grooms show declines that have been levelingout in recent years. No clear-cut pattern can bedetected of median ages for remarried brides andgrooms table 24 .

REFERENCESCarroll D. Wright, Commissioner of Labor: A Report on

Marriage and Divorce in the united States 1867 to 1886Washington. U.S. Government Printing Office, 1889.21bid., p. 18.

Slbid., p. 19.4U.S. Bureau of the Census: Mamage and Divorce, 1867-

1906, Part 1, Summary, Laws, Foreign Statistics Washington,U.S. Government Printing Office, 1909; Part 2, General TablesWashington, U.S. Government Printing Office, 1908.

5 U.S. Bureau of the Census: Marriage and Divorce 1916Washington. U.S. Government Printing Office, 1919.

6Stouffer, S. A., and Spencer, L. M.: Recent increases inmarriage and divorce. Arm ]. Sociol 44:551-554, Jan. 1939.

7Nationai Center for Health Statistics: Demographic char-acteristics of persons married between January 1955 and June1958. Vital and Health Statistics PHS Pub. No. 1000-Series21-No. 2. Public Health Service. Washington. U.S. GovernmentPrinting Office, Apr. 1965.

8 U.S. Bureau of the Census: Jfarriage and Divorce1867-1906, Part 1, SUmmary, Laws Foreign Statistics Wash-ington, U.S. Government Printing Office, 1909. p. 5.

Wright, op. cit., p. 132.


26/68

LIST OF DET ILED T LESPage

22able 1.2.

Marriages divorces and rates: United States 1867-1967------------------------Regr;;d annulments: all reporting States California New York and all others------------------------------------------------------- -------------- ---- 23

243. Marriage rates for unmarried women aged 15 and over: United States 1940-67 and

selected census years-----------------------------------------------------------Divorce rates for married women aged 15 and over: United States 1920-67 and se-lected census years-------------------------------------------------------------4. 24

25Marriages and percent distribution by area of occurrence: United States eachregion and division selected years 1870-1967-----------------------------------5.

6. Divorces and percent distribution by area of occurrence: United States eachregion and division selected years 1870-1967----------------------------------- 277. Marriage rates: United States each region division and State or territoryselected years l87O-l967-------------------------------------------------------- 28

38. Marriage rates for unmarried men and women aged 15 and over: United States eachregion division and State or territory selected census years 1890-1960-------9.

1

Divorce rates: United States each region division and State or territory se-lected years l87O-l967---------------------------------------------------------- 34

36Divorce rates for married women: United States each region division and Stateor territory selected census years 1890-1960-----------------------------------

12.13.

14.

Divorce rates: counties containing selected ajor cities selected years 1870-1965-----------J---------------------------------------------------------------- 3839edian duration of marriage prior to divorce: United States 1867-1967----------

Median duration of arriage prior to divorce: United States and each State orterritory selected years l87O-l967--------------------------------------------- 4Percent distribution of divorces by duration of marriage prior to divorce: UnitedStates selected years l87O-l967------------------------------------------------ 41

4115. Divorces and percent distribution by duration of marriage and of separation:United States l887-l9O6--------------------------------------------------------16. Median and quartile duration of marriage and of separation: United States andeach State and territory l887-l9O6--------------------------------------------- 42

Percent of divorced couples married in the State where divorce was grante UnitedStates each region division and State or territory selected years 1870-1965--17. 43

4518.

19.

Mean number of children per divorce with children reported: United States se-lected years l867-l967----------------------------------------------------------Mean number of children per divorce with children reported: United States eachregion division and State or territory selected years 1870-1967-------------- 46Divorces and percent distribution by legal grounds:United States selected yearsl867-l965-----------------------------------------------------------------------

2

2148

Percent of divorces granted to wife: United States each region division andState or territory selected years 1870-1965------------------------------------ 522. Median and quartile duration of marriage prior to divorce by party to whom de-cree granted: United States selected years 1870-1965--------------------------- 5223. Median age of brides and grooms: United States each region division and Stateselected years l94O-67----------------------------------------------------------- 53

5524. Median age of brides and grooms for all marriages first marriages and remar-riages: 22 States l954-67------------------------------------------------------

21


27/68

Table 1. Marriages, divorces, and rates: United States, 1867-1967[All t ot a ls r ou nd ed a n dt in clu de es tim a tes . M a rr ia g e f ig ur es in clu de m a rr ia g e licen ses f rom s om e S t a tes ; d iv or ce f ig ur es in clu de r gp or t ed a n nu l

m en ts . F or popu la t ion ba ses s ee a ppen dix I I. Ra t es per 1 000 popu la t ion ]

YearMarriage

Number

1,927,0001,857,0001,800,0001,725,0001,654,0001,577,0001,548,0001,523,0001,494,0001,451,0001,518,0001,585,0001,531,0001,490,0001,546,0001,539,0001,595,0001,667,0001,580,0001,811,0001,992,0002,291,0001,613,0001,452,0001,577,0001,772,0001,696,0001,596,0001,404,0001,331,0001,451,0001,369,0001,327,0001,302,0001,098,000982,0001,061,0001,127,0001,233,0001,182,000

1,201,0001,203,0001,188,0001,185,0001,230,0001,134,0001,164,0001,274,0001,150,0001,000,0001,144,000

ate

9.79.59.39.08.88.58.58.58.58.48.99.59.39.29.81;:?11.110.612.413.916.412.210.911.713.212.712.110.710.311.310.710.410.38.77.98.61:::9.810.110.210.310.411.010.310.712.011.01?:1

Divorce

Number

523,000499,000479,00050,000428,000413,000414,000393,000395,000368,000381,000382,000377,000379,000390,000392,000381,000385,000397,000408,000483,000610,000485,000400,000359,000321,000293,000264,000;:;,;())249:000236,000218,000204,000165,000164,000188,000196,000206,000200,000196,000185,000175,000171,000165,000149,000160,000171,000142,000116,000122,000

Rate

2.62.52.52.42.3;:;2.22.22.12.22.32.32.42.52.52.52.62.72.8u;:;2.62.4M1.91.91.91.81.71.61.31.31.5H1.71.61.61.51.51.51.41.51.61.31.11.2

Year

1916----------1915----------1914----------1913----------1912----------L911----------1910----------1909----------1908----------1907----------1906----------1905----------1904----------1903----------1902----------1901----------1900----------1899----------1898----------1897----------1896----------1895----------1894----------1893----------1892----------1891----------1890----------1889----------1888----------1887----------1886----------1885----------1884----------1883----------1882----------1881----------1880----------1879----------1878----------1877----------1876----------1875----------1874----------1873----------1872----------1871----------1870----------1869----------1868----------1867----------

Marriage

Number

1,076,000L,008,000L,025,000L,021,000L,005,000955,000948,000897,000857,000937,000895,000842,000815,000818,000776,000742,000709,000673,000647,000643,000635,000620,000588,000601,000601,000592,000570,000563,000535,000513,000534,000507,000485,000501,000484,000464,000453,000438,000423,000411,000405,000409,000385,000386,000378,000359,000352,000348,000345,000357,000

late

NOTE: Alaska included beginning in 1959; Hawaii included beginning in 1960.

LO.6LO.OLO.3L0,5LO.5LO.2LO.39.9L i10.5LO.OL~:?9.89.69.3:8.9R8.6;.;.9.29.0::;8.7

u8.89.39.29.0.?::::;8.8;9.09.08.88.88.99.09.6

Divorce

Wmb er

.14,000.04,000.01,00091,00094,00089,00083,00080,00077,00077,00072,00068,00066,00065,00061,00061,00056,00051,00048,00045,00043,00040,00038,00037,00037,00036,00033,00032,00029,00028,00026,00023,00023,00023,00022,00021,00020,00017,00016,00016,00015,00014>00014,00013,00012,00012,00011,00011,00010,00010,000

late

1k;1:: :0.90.90.90.80.8M0.80.80.70.70.70.60.6M0.60.60.60.50.50.50.50.40.4:::0.40.40.40.30.30.30,30.30.30.30.3030.3:0.3


28/68

Table 2. Reported annulments: all reporting States. California. ew York and allothhrs 1~26-67 -

Year

1967------------------------------1966------------------------------1965------------------------------1964------------------------------1963------------------------------1962------------------------------1961------------------------------1960------------------------------1959------------------------------1958------------------------------1957------------------------------1956------------------------------1955------------------------------:;;;------------------------------------------------------------1952------------------------------1951------------------------------1950------------------------------1949------------------------------1948------------------------------;;:;------------------------------------------------------------1945------------------------------1944------------------------------1943------------------------------1942------------------------------1941------------------------------1933-40---------------------------1932------------------------------1931----------------.-------------1930------------------------------1929------------------------------1928------------------------------1927------------------------------1926------------------------------

Total

15 13214 66915 41813 94012 70112 69211 52011 18112 60312 5418 2317 7707 6877 8657 8987 7447 84310 8486 8957 2087 3249 8981 972963891725627---

3 9034 3394 3704 4084 237; ;;;Y

Califor-nia

6 8616 4806 9356 5876 1345 9845 6435 46635 4605 1075 0635 0104 7054 8204 8074 7164 8054 4714 7954 937

J: ------------------

1 2671 4991 4761 4761 4411 6291 282

lCountingthe District of Columbia as a State.~Estimated.incomplete.

New York

2 5232 8323 0352;;;::32 33132310;2:5692 9053 417

---------------------33 609------------------------------

1 0251 0491 0301 029990872964

Al1otherreport-ingStates

5 7485 3575 4484 7564 283; y;3;1464 2384 0173 1682 7602 9823 0453 0913 0283 0382 7692 1002 2712 1002 5981 ;;:891725627---1 6111 791

1 8641 9031 806: ;;;3

Number ofStatesreport-ingI


23


29/68

able 3. Marriage rates for unmarried women aged 15 and over: United States 1940-67 and se-lected census years

[Theunmarriedincludesingle idowed and divorcedpersons.or populationbases eeappendix11.Ratesper1 000unmarriedwomen]

Year

......... .....

Rate

76.475.675.074.673.471.272.273.573.672.078.082.480.979.883.783.286.690.286.7

Year

19481947194619451944194319421941194019301920191019001890

Rate

98.5106.2118.183.676.583.093.088.582.867.892.377.168.267.4


able 4. Divorce rates for married women aged 15 and over: United States 1920-67 and selectedcensus yearsForpopulationases seeappendix11.Rateslrer1 000marriedwomenl

Year Rate

11.210.910.610.09.69.49.69.29.38 .99.29 49.39.5 ::? ;:;10.611.213.617.914.412.011.01 19.4

Year

;;;:-----------------------.----- -------- -----1938193719361935193419331932193119301929192819271926192519241923192219211920191019001890

8.88.58.48.78.37.8::?6.17 17.58.07.87.87.57.27.27.16.6:::4.74.13.0



30/68

Table 5. Marriages and percent distributionby area of occurrence:United States, each region anddivision, selectedyears 1870-1967[National totafsfor years prior to1950were reestimatedizr 1950. Hence tiere~ondt ot donotn ecesstiIya dduptot henationdtot .Percentages are based on sums of regional figures]

Area

United Statesl---------------------Region

Northeast-------.-----.-----.----.-------Nomr;;Central--------------------------------------------------------------West--------.---------------------------

DivisionNew England------------------------------Middle AtlanticEast North Central----------------------West North Central-----------------------South Atlantic---------------------------East South Central-----------------------West South Central-----------------------Mountain--------------------------------Pacific---------------------------------

united states

Northeast-------------------------------North Central---------------------------south------------------------------------

DivisionNew England.........-----------------.-.-Mtddle AtlanticEast North Central-----------------------West North Central-----------------------South Atlantic---------------------------East South Central-----------------------West South Central-----------------------Mountain--------------------------------Pacific---------------------------.-----

1967 1960 1950 1940 1930

Number1,927,000

92,855277,913355,871141,843333,827135,8241;;::;::212,842

1

19.225.835.319.7

4.814.418.51;::l~:j.11

1,523.000

311,481403,5951534,2011274,104

76,206235,275283,285120,3101261,885111O,3O -J162,0161122,818151,286

1,667,000

365,538454,6291;;;: ;;

88.503277;035317,344137,285265,0611:;;;;


31/68

Table 5. Marriages and percent distribution by area of occurrence: United States each region anddivision selected years 1870-1967-Con.[National totals for years prior to1950were reestimated in1950 Hence there@ond totalsdo notnecesstiIyadd uptothe nationdtotals Percentages are based on sums of regional figures]

Area

United States -------------..------

RegionNortheast ----------------........--------North Central----------------------------south------------------------------------West-------------------------------------

DivisionNew England------------------------------Middle Atlantic--------------------------East North Central-----------------------West North Central-----------------------South Atlantic---------------------------East South Central-----------------------West South Central-----------------------Mountain ---------------------------------Pacific.............................--.--

United States----------------------

RegionNortheast ................................North Central...........................-South---------------.-----.----------....West-------------------------------------

DivisionNew England------------------------------Middle Atlantic --------------------------East North Central-----------------------West North Central-----------------------South Atlantic---------------------------East South Central-----------------------West South Central-----------------------Mountain---------------------------------Pacific----------------------------------

1916 1900 1890 1880 I 1870

1 076 000

272 275336 9922373 82285 761

71 599200 676213 097123 895:;; ;;;3116:778:: ;;;

a

1

25.531.535.08.0

6.718.819.911.614.71::;::?

709 000

174 394240 709252 16535 957

47 639126 755148 17292 537101 29678 33372 53615 478220 479

Number570 000

147 022205 290183 71427 126

242 789104 233u; ;::275:56260 75847 394U 21O215 916

453 000

2158 85;1168 84813 952

33 918298 49;259 39:139 18028 85;

Percent distribution1

24.834.235.95.1

6.818.021:113.214.411.11; ;2:9

1

26.136.532.64.8

22.713.813.410.8

100 0

it

352 000

2120 76;8 162

234 410---75 617

50 63:5 76;

1

lEstimated.Incomplete.31nclu~ing estimates for nonreporting areas.

26


32/68

Table 6. Divorcesand percent distributionby area of occurrence: United States, each regionand division,selectedyears 1870-1967[Nat ional totak for years prior to 1950 were reest imated in 1950 He

100 yrs marriage div usa

Documents