Click for next page ( 34


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 33
/' Trends in Adolescent Sexuality and Fertility The incidence of adolescent childbeanng is the result of severs social and demographic processes. The size of the teenage population, the pro- ponion of teenagers who are marred, the incidence of sexual activity among the unmarried, the consistency of contraceptive use, and the effectiveness of methods used are aD factors that affect the probability of pregnancy. Among those who become pregnant, a number of resolutions are possible, including miscarriage, abortion, mamage, adoption, and motherhood without mamage. The frequency of these behaviors (except possibly miscarriage) has been changing over the past two decades, and in many cases the trends are moving in different directions. Moreover, some of these changes offset others, or one change affects the size of the population at risk of another behavior. Consequently, it iS necessary tO explore the entire range of behaviors In order to understand the phenom- ena of teenage pregnancy and childbearing. Data from several sources are available. Vital statistics data report marriages, births, abortions, and deaths ~ the U.S. population. Census data provide wformanon concerning the size and composition of the teenage population, the characteristics of their families, their living ar- rangements, their school es~rolIment, their employment status, and their economic well-be~g. Other national survey data provide ~nfo'~ation about sexual activity. They also provide infonna~aon about pregnancy and abortion, although these behaviors appear to be less reliably reported. While census data and vita] stansucs data perinit us to trace trends in teenage childbearing over several decades, survey data coring informa- tion about adolescent sexual behavior and pregnancy have been available 33

OCR for page 33
34 ADOLESCENT SEXUALITY PREGNANCY AND CHILDBEARING only since the early 1970s. Data concerning racial subgroups have been available from the census and the vital statistics system, as weD as from surveys containing info' citation on adolescent sexual and fertility behav- ior. Data descnbing ethnicity, however, began to be available only in the 1970s as some subgroups, especially Hispanics, became more prominent minorities in the United States. The census and other federal data sources based on census samples began to distinguish Hispanic ongiI1 in 1970; in 1978 selected states began to report inflation concerning ethnic~ty through the Vital statistics system. Although data on Hispanics is included In recent surveys of sexual and fertility behavior, the samples are fre- quently tOO small to permit statistically meaningful national estimates according tO the relevant payables (e.g., age, marital status). In addition, data concerning Hispanic ong~n are frequently not mutually exclusive of data concerning race. Hispanics may be of either race, and data concerning blacks and whites may also include Hispanics. Accordingly, information concerning trends ~ the sexual and fertility behavior of Hispanic teenag- ers is incomplete and not comparable to that for racial subgroups. This chapter presents an overview of the sonal and demographic charac- tensiics of the teenagers in the United States and trends in their sexual and fertility behavior. It also presents a critique of the strengths and weak- nesses of the available data. More detailed inflation is presented In the statistical appendix tO tips repon, which is part of Vogue Il. ADOLESCENTS IN THE UNIT= STATES The post-World War I! baby boom extended through the 1950s, peak- mg ~ 1957. As a reset, the population of U.S. teenagers ages I~19 grew until i976, when it peaked at 21.4 minion. Since then it has declined steadily to 18.4 miLion in 1985, consisting of approximately 9.0 million girts and 9.4 minion boys. Recent populanon projections medicate that the number of teenagers will continue to decline to approximately 16.9 million in i990; it Will rise again to approximately 18.9 million by 2000 (Table 2-~. Despite the increase and subsequent decline In the adolescent population during the past three decades, the proportion of the tots U.S. population 15-19 years of age has remained between 7 and SO percept since 1940 (Bureau of the Census, 1980~. Over the past generation, the number and proportion of racial and ethnic minorities in the United States has increased. In 1984 blacks com

OCR for page 33
TRENDS IN AL)OLESCE~T SEXUALITY AND FERTILITY 35 TABLE 2-1 Total U.S. Populatior~ Ages 1~24 Years Old by Sex, 1960-2000 (in thousands) Males 1~14 1~-19 2~24 1~14 15-19 Females 2~24 1960 8,:24 6,634 :,272 8,249 6,586 ~,528 1960 9,636 8,656 6,884 9,323 8,395 6,794 1970 10,622 9,714 8,0~4 10,230 9,Sl7 8,;44 1975 10,534 10,757 9,640 10,112 10,465 9,677 1976 10~251 10,896 9,893 9,837 10,;82 9,901 1980 9,316 10,726 10,697 8,92: 10,376 10,678 1981 9,374 1D,419 10,813 8,964 10,074 10,779 1982 9,318 10,104 10,793 8,899 9,767 10,765 1985 8,590 9,398 10,820 8,207 9,019 10,481 1990 8,586 8,670 9,443 8,207 8,299 9,137 2000 9,986 9,681 8,723 9,332 9,262 8,422 SOURCES: Bureau of the Census, Detailed Population Ch~actenstics, U.S. Sum- mary, 1960, 1970, 1980; "Population Estimates and Projections," Current Population Reports, Senes P-23, No. 960, 198~. pnsed approx~nately 14 percent of the adolescent population, compared with 13 percent in 1970 and I] percent in 1960. In 1984 Hispamcs comprised approximately 7 percent, compared with 5 percent in 1970 (earlier data for Hispanics are not available). Overall the proportion of adolescents who are nonwhite has increased by 47 percent since 1960 (Bureau of the Census, 19801. Mamage Over this same period, early marriage has become less prevalent. Fewer young women and even fewer young men marry while stiD ~ they teens than did one or two decades ago. In 1984, both mates and females ages 15- 19 were less likely to have ever been married than their counterparts in the early 1970s. The propomon of Al females under age 20 who remained single maeased from 84 percent in 1960 to 91 percent in 1976 and to 93 percent in 1984. Among males the increase was from 96 percent in 1960 to 97 percent in 1976 to 98.S percent in 1984. Although there was only a relai~,rely small decline over the decade ~ the percentage of white teenag- ers who married, there was a much sharper decline among blacks, and it

OCR for page 33
36 ADOLESCENT SEXUALITY PREGNANCY AND CHILDBEARING began a decade earlier. Between 1960 and 1984 the percentage of black women ages iS-19 who were ever married dropped from 16.2 percent to I.6 percent. Similarly, the decline for black males was from 3.4 to I.8 percent (Table 2-2~. Hispanic teenagers, both male and female, were more likely to have married in 1984 than in 1970, and they were more likely to have mamed than either whites or blacks. Older teenagers (18-19) have always been significantly more likely to be married than those who are under 18. In 1984, females 1~19 years old were four and a half tunes more likely to be married than those 15-17 years old, and older males were nearly seven times more likely to be marned than those of school age. Because of the greater proportion of 18- and 19- year-olds who are mamed, it is the decline among these older teens that primarily accounts for the overall decline in teenage mamage. Among younger teenagers the percentage of males and females who are mamed has remained fairly stable at a very Tow level over the past two decades (see Table 2-2). SchooZ:ng Most teenagers under age 18 are enrolled in school, while fewer 18- and 19-year-olds are students. In 1984, 98 percent of It and 15-year-olds and 92 percent of 16- and 17-year-olds were ~ school, compared with appro~n- mately 50 percent of 18- and 19-year-olds School enrollment has re- ma~ned fairly constant among those under 18 over the past two decades and has maeased slightly among those 18 and over. The proportion of males and females who are iI1 school is virtually equivalent for all age groups (Bureau of the Census, 1985c). Patterns of school enrollment vary by race and ethnic group. Approm- mately equal proportions of white end black teenagers ages 1015 and 16- 17 are enrobed in school, while fewer Hispanics ages I~17 are students. Among teenagers I~19 years old, whites are more likely to be enrobed than either blacks or Hispanics. Among older teenagers not In school, Hispanics are substantially more likely than whites or blacks to lack a high school diploma. Regardless of race or ethnic group, virtually all young people ages 1~15 aIld 1~17 who are not enrolled in school have dropped out before graduation. In 1984 approximately 610,000 teenagers ages 10 17 and 1.1 million teenagers ages 1~19 were out of school but had not graduated (Table 2-3).

OCR for page 33
TRENDS IN ADOLESCENT SEXUALITY AND FERTILITY 37 TABLE 2-2 Percentage of Never-Mamed Boys and Girls Ages 15-19 by Race, 196~1984 Boys 1~-17 Girls 18-19 tS-19 15-17 18-19 lS-19 1960 Total 99.1 91.1 96.3 93.2 67.8 83.9 White 99.1 91.0 96 2 93.3 67.6 83.9 Nonwhite 99.2 91.9 96.6 92.3 69.3 83.8 1970 Total 98.6 91.3 95.9 9;.3 76.6 88.1 White 98.7 91.3 9~.9 95.4 76.4 88.0 Black 98.0 91.0 9~.5 95.0 77.7 88.6 Hispanic 97.7 87.4 94.0 93.1 70.6 84.7 1973 Total 99.2 90.4 96 ha 96.2 75.8 89 6c White 99.1 89.: 96.2a 96.2 74.4 89.1 Nonwhite 99.: 95.6 98 4a 96.1 83.9 92.2 Hispanic N/A N/A N/A N/A N/A N/A 1916 Total 99.4 91.9 97 oa 97. ~78.3 90.~ White 99.3 91.2 96 1a 96.8a 77.2 90.2 Black 99.6 95.9 Mesa 98 la 85.0 93.8~ Hispanic 99.; 92.7 97 7a 94.6 ~74.9 87.1a 1980 Total 99.4 94.2 97.3 97.0 82.8 91.1 White 99.4 93.6 97.0 96.7 81.5 90.4 Black 99.4 97.7 98.8 98.3 90.9 95.4 Hispanic 98.0 92.2 95.8 94.6 79.2 88.2 1981 Total 99.2 95.7 97.8 97.2 84.7 92.0 White 99.2 9~.4 97.7 96.9 83.4 91.3 Black 99.6 97.0 98.6 98.8 92.7 96.4 Hispanic 99.2 91.8 96.3 95.3 74.0 86.7 1984 Total 99.7 96.8 98.S 98.0 87.1 93.4 White 99.6 96.5 98.3 97.7 85.2 92.4 Black 100.0 98.2 99.3 99.3 97.2 98.4 Hispanic 99.0 93.S 96.8 95.7 79.1 88.8 NOTE: Hispanic persons may be of any race and blacl; and white totals may include Hispanics. Precludes boys and girls 14 years of age. SOURCE: Bureau of the Census, "Manral Stands end Living Arrangements," Current Po~uLaizon Reports, Series P-20, 1960, 1970, 1913, 197:, 1976, 1981, 1984, 1985.

OCR for page 33
38 a, c) Cal C5 c) no a: Cal - ~: Cal Cal o o Cal As et _, Cot _, - C) - o - - o o - Cot ~ Cal - ~ o <: ~ ~ _ _ CO Cal_ ~ ~ ~ o . . .. . .. . . con ~ uscon ~ ~con ~ r __I_ ~ Us Us~ ~ ~ Us Us 0 ~ Cal _Cal Us ~ ~ ~ _ ~ ~ ~ ~_._ ~ us crycot of or_. ~ t- . . .. . .. . .. . . Cal ~ ~Cal oo oo~ ~ ~Us ~ 0 ~en _ _ ~ ~ oo ~ 0 0~ - 0 _ ~ t-~ Us 0 em Cal 0 ~rat ~ rat Cal 0 0 Cal 0 ~ ~0 Cal -~ ~ ~ ~ ~ . . .. . .. . . . . . I_ ~ 0~ _ ~ Cal ~ Us LO~ ~ ~ Cat ~0` 0 o~ ~ ~ _ ~n~ _ ~c~ ~ ~ ~ ~ c~C~ 0` ~u~ 0 ~- ~D C~ c~ Lr, r~cr~ e~ _ _ _~ _ _~ ~ ~ ~U~ L~ ~4 - ._ - ~C _0 ~ ~ ~ ~ ~ C ~o~ 0 ~ C~ ~ C~U~ O ~ ~ 0` ~ ~ ~D 0~ ~ ~0 ~ ~0 ~-~- =~= ~__ X q_ C~ 2 O O .. _ C) _ _ C~, C~ '_ O ~e~ ~ .- _ ~ C) C) 3 cc ~C~ ,_ _ C ~ - _ CO o o o ~ c:;.a .- ~ _ ,_ .= Ct O ~1 _ _ ~- ~S _ C~ = - O ~ _ :: O ~ 4) ' O ~s _ 4~ ~ ~- +0 . .} O ~ & ~o ~ C) ~ _ , . C~ ~ ~ ": C, ._ ._ ~ ~ C~ .. o' U; E~ ~ o O 0 Z C~ ~ ~0

OCR for page 33
TRENDS IN ADOLESCENT SEXUALITY AND FERTILITY 39 Employment Approximately 7.8 million teenagers ages 16-19 reported that they were in the civilian labor force in July 1985. Approximately 6.3 million were employed either part or full time, despite the fact that many of them were also enrolled in school. Another I.; million were unem- ployed, because they either could not find a job or were not looking. Forty-eight percent of white teenagers ages 16-19 were employed com- pared with 25 percent of blacks. Unemployment (seasonally adjusted) was significantly greater among black teenagers than among white teenagers, 41 percent compared with 16 percent, and it was slightly higher for males of both races than for females (Table 2-41. Data on Hispanic employment and unemployment status are not published by age category. TABLE 2 - Employment Status by Race, Sex, and Age, 198~198: (seasonally adjusted, in thousands) 1984 198; - Whitesages16-19 Civilian labor force 6,952 6,852 participation rate 57.; 57.7 Employed 5,893 ;,733 Employment-popula~ion ratio 48.7 48.3 Unemployed 1,059 1,119 Unemployment rate 15.2 16.3 Men 15.2 17.5 Women 14.3 15.0 Blacks Ages 16-l) Civilian labor force 849 915 participation rate 39.4 42.4 Employed 490 :3? Employment-population ratio 22.7 24.9 Unemployed 3;9 378 Unemployment rare 42.3 41.3 Men 42.3 43.3 Women 42.2 39.0 NOTE: Hispanic breal;downs no: available for the years presented. SOURCE: U.S. Department of Labor, "The Employment Situation: July 1985," News, U.S. Department of Labor 8;-304, August 2, 1985.

OCR for page 33
40 ADOLESCENT SEXUALITY; PREGNANCY AND CHILDBEARING THE POPULATION AT RISK OF PREGNANCY Estimates of the adolescent population at risk of pregnancy depend on information concerning fecundity among adolescent females and sexual activity and contraceptive use among both males and females. Fecundity The Second National Health and Nutrition Examination Survey, 1976-1980, shows that more than 75 percent of adolescent females have started menstruation by age 13 and 96 percent have begun by age 15. The mean age of menarche differs slightly by race: for blacks the mean age is 12.5 years, for whites it is 12.7 years; the difference is not statistically significant (see Vol. Il:append~x tables, section on changing contexts). Sexual Activity Less than half of all unmarried teenage girls in the United States are sexually active (i.e., have experienced coitus at least once); between 1971 and 1979, however, there was a dramatic increase in nonmarital sexual activity among 15- to 19-year-old females. As Figure 2-! shows, during this period the proportion of unmarried girls in this age group who had ever had intercourse increased from 28 to 46 percent. Between 1979 and 1982 there was a slight decline, to 42 percent, in the proportion of those girts who were sexually active. (This may not represent a statistically significant reduction, and it iS based on comparisons of data from differ- ent surveys.) Despite the decline (or leveling off), rates of sexual activity among unmamed teenage girls appear tO be substantially higher than they were in 1971 (Table 2-~. Throughout this period, blacks have had significantly higher levels of sexual intercourse outside marriage than whites. Ire 1982, nearly 53 percent of black girls ages 15-19 had had intercourse, compared with about 40 percent of whites. However, the proportion of sexually active white teenage gills increased steadily between 1971 and 1979 and then declined very slightly in the early 1980s. In contrast, levels of sexual activity among black girls rose sharply between 1971 and 1976, re- mained virtually unchanged between 1976 and 1979, and then declined substantially between 1979 and 1982. In 1982, the 13 percent difference between blacks and whites in levels of nonmarital sexual intercourse was

OCR for page 33
TRENDS IN ADOLESCENT SEX UALI7 Y AND FERTILI7 Y 41 100 l`' 80 c: 'IS ~ 60 x '0 40 LO er ~ 20 1971 .1 ~I ~i I 1976 YEAR 1979 1982 ~ Blacks All Races Whites FIGURE 2-1 Sexual activity of adolescent girls ages 1;-19, 1971-1982. Sources: M. Zeloil; and I.F. Kantner (1980~; W. F. Pratt (1984~; unpublished tabulations, 1984. the lowest it has ever been since such data were made available, and significantly less than the 31 percent difference reported ~ 1976. The apparent decline in sexual activity among blacks is primarily responsible for the overall decline dunug the early 1980s in the proportion of u$~marned adolescent girls who had ever had intercourse (Table 2-61. Data concerning trends in sexual activity among unmarried Hispanic adolescent girls are unavailable. Nevertheless, on the basis of estimates fiom the 1982 National Survey of Family Growth, it appears that levels of premantal sexual activity among Hispanic teenagers are closer to the level for whites than for blacks (see Vol. Il:appendix tables, section on adolescent sewal activity). By age 20, most unmarried young men and women are sexually active: over 80 percent of males and over 70 percent of females report that they have had intercourse at least once. With each successive year of age, a greater proportion of adolescents are sexually expenenced. While only 5 percent of teenage girls and 17 percent of teenage boys report having had intercourse by their fifteenth birthday, 44 percent of girls and 64 percent of boys report that they were sexually active by their eigh

OCR for page 33
42 ADOLESCENT SEXUALITY PREGNANCY AND CHILDBEARING TABLE 2-: Percentage Never-Marned Giris Ages 1;-19 Lining in Metropolitan Areas Expenencing Sexual Intercourse, 1971-1982 _ Race and Percent Change Age 1982 1979 1976 19111971-1982 All racesa 42.2 46.0 39.2 27.652.9 15 17.8 22.5 18.6 14.423.6 16 28.1 37.8 28.9 20.934.4 17 41.0 48.S 42.9 26.157.0 18 52.7 56.9 ;1.4 39.732.7 19 61.7 69.0 59.5 46.433.0 White 40.3 42.3 33.6 23.273.7 lS 17.3 18.3 13.8 11.353.1 16 26.9 35.4 23.7 17.058.0 17 39.5 44.1 36.1 20.29~.: 18 48.6 :2.6 46.0 35.636. j 19 59.3 64.9 :3.6 40.745.7 Black 02.9 64.8 64.3 52.41.0 15 23.2 41.1 38.9 31.2- 1~.6 16 36.3 50.4 So.1 44.4- 18.2 17 46.7 73.3 71.0 58.9- 20.7 18 7~.7 76.3 76.2 60.22~.7 19 ?8.0 88.5 83.9 78.3- 0 4 reincludes blacks, whites, and other races. SOURCES: M. Zel:iik andJ.F. Kantner, 1980, "Sexual Aciinty, Contraceptive Use and Pregnancy Among Metropolitan-Area Teenagers: 1971-1979," Family Plan'2ing Perspectives 12~), September/October: W. F. Pratt, 1984, NCHS Nanonal Surrey of Family Growth, 1982, Cycle III; unpublished tabulations, 1984. teenth birthday. Boys appear to initiate sexual activity earlier than girls but, by the later teenage years (~19), the proportion of girls `~ho report having had intercourse more closely approaches that of boys. Blacks (especially males) appear to initiate sexual activity earlier than whites, and at every age blacks are more likely than whites or Hispanics to be sexually active. Race differences in the proportion who are sexually active are especially pronounced among younger teenagers (Table 2-6~. Despite the increased proportion of teenagers who are sexually active, many have intercourse infrequently. Data from the National Survey of Young Women (Vol. Il.: appendix tables, section on adolescent sexual activity) indicate that in 1979 nearly a quarter of sexually active gigs ages 15-19 had had intercourse only once or twice during the previous month. Over 40 percent reported that they had not had intercourse at all

OCR for page 33
TRENDS IN ADOLESCENT SEXUALITY AND FERTILITY 43 TABLE 2-6 Cumulative Sexual Activity by Age of Initiation and Sex for the National l~ongitud~nal Sunrey of Youth . Cumulative Percentage Sexually Active Age Boys Girls Total Sample (A = 4,637 boys, 4~648 girls) Is 16.6 5.4 16 28.7 12.6 11 47.9 27.1 18 64.0 44.0 19 77.6 62.9 20 83.0 73.6 White (N = 2,828 boys, 2,788 girls) 15 12.1 4.7 16 23.3 11.3 17 42.8 25.2 18 60.1 41.6 19 75.0 60.8 20 81.] 72.0 Black (N = 1,146 boys, 1,1;7 girls) 15 42.4 9.7 16 59.6 20.1 17 77.3 39.; 18 85.6 09.4 19 92.2 71.0 20 93.9 84.7 Hispanic (N = 683 boys, 703 girls) lS 19.3 4.3 16 32.0 11.2 17 49.7 23.7 18 67.1 40.2 19 78.; 58.6 20 84.2 69.5 NOTES: Sample is limited tO respondents age 20 and over at 1983 survey date. Percentages reference birthday for specified ages, e.g., 15 mews by fifteenth birthday or end of age 14. Hispanics may be of any race, and black and white lotus may include Hispanics. SOURCE: Special tabulations from the National Longitudinal Survey of Youth, 1983, Center for Human Resource Research, Ohio State University.

OCR for page 33
64 - - ._ _} ._ _, ._ _ - c) & - cr~ I_ 1 it_ Cal C) Cal C) - o C) ._ :D o - _, ._ Cat _, C~ ._ ., - o ._ ._ C~ o C~ ._ ~_ C~ = - o^ ~ Z ~ 11 o ~o om o ~o o ~ C~ o o~ 7 o m^ ~ t1 o ~Z o c~ ~ 11 ~ ~ Z O ___ C) - ._ - 3 o ~ o ~ . .. . o _ t-o .. . . o o C~ .. . . C~ oo ~_. C^' O I ~ 11 o ~~) ~ 0 c, ~Z O O~ -00 _ 0^\ . ._ ~_. _ 11 o ~ - ~ ~ ~ O U ~ t_ C~ V, ~Z O 0 :- o0 ~ 11 o, ~ o=_ ~ Z Oo C~ ~ C~ ~ = ~C~ ~) 0` c~ 11 o ~ ~ ~ c, ~Z 00 ~ ~ -0 , O ~et ~ O ~ e~ _ ~ ~ ,> .> _ ,_ L) _ o O _ e' 5 e o OC ~C~ O ~ . .. . . O r~c~ O .. C~C~ C~ - o ._ ._ C~ - _~ ._ . ~ .. ~ ~ CC o C~ C~

OCR for page 33
TRENDS IN AI)OLESCENT SEXUALITY AND FERTILITY 6: Nonmantal Chillbeanng In the early 1980s, more than half of all births to adolescents occurred outside marriage, compared with only about one-third in 1970. In 1984, births to unmarred mothers under age 20 numbered more tears 270,000; approximately 46 percent of them were to mothers under age 18 (Vol. Il:appendix tables, section on births). Despite a decline in the number and rate of births to women in all adolescent age groups, rates of nonmantal childbearing increased steadily dunng the 1970s as rates of mantal childbeanug among teenagers decreased. However, nonmantal childbearing has also increased among adult women over the past decade and a half. As a result, births to unmarried teenagers now account for a smaller proportion of aD nonmantal births, approximately one-third in 1984 compared with approximately one half in 1970 (Vol. Il:appendix tables, section on births). Trends In nonmantal childbeanng among adolescents vary dramati- caDy by race. Consistently, blacks have been significantly more likely to give birth outside marriage than whites. In 1984, the rate of nonmarital childbearing among blacks ages 15-19 was 4.5 times greater than the rate among whites- 87. ~ per 1,000 unmamed black women, compared with 19.0 per i,000 unmarred white women. However, since 1970 there has been a sharp increase in the rate of unmarried parenthood among white teenagers and in the number of births to white girls: nonmantal birth rates for whites ages 15-19 rose by 74 percent between 1970 and 1984. It is this rise that explains the overall increase ~ births to unmarried teenagers, since the rate of black nonmantal childbeanng among women under age 20 actually declined by 10 percent dunng this period (Table 2-16~. Hispanic teenagers are more likely than non-Hisp~c whites but less likely than non-Hispanic blacks to give birth outside marriage. Appro=- mately 45 percent of births to Hispanic women ages I~19 were non- mantal births in 1984, compared with 34 percent of non-Hispan~c white births =d 87 percent of non-Hispanic black births (unpublished tabula- tions by the Division of Vital Statistics, National Center for Health Statistics, 19861. Despite the substantial increase in unmarred parenthood among whites during the 1970s and the early 1980s, the proportion of all nonmarital births to teenage mothers (ages 15-19) remains significantly higher for blacks than for whites. In 1984, this proportion was more than twice as high for blacks as for whites 89 percent of live births Lo

OCR for page 33
66. Cal CC es o C' C, AS Cal ~ o o -= 3 es LO o =O Cal _ _ 3 ~ _ ~ ~= .~ _ ~ _ l ~ ~ ~ ~ ~ ~ ~ ~ o ~ ~ ~ ~ 0 ~ . . . . . . . . . . . . . . .. ~ ~ ~ ~ ~ o ~ ~ ~ ~ US US ~ 0 Cal Cal ~ ~ Cal Cal ~ ~ ~ ~ ~ ~ ~ o ~ ~ ~ ~ ~ ~ Cal ~ ~ Go ~ ~ ~oo Cal ~ o o 0 ~ ~ Cal ~ o _ As ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ cut ~ ~ ~ ~ ~ ~ 0 0 ~ cryof - . . . . . . . . . . . . . . . ~ ~ r~ ~ ~ ~ 0 ~ ~ ~ L~ ~ ~ ~ ~D 0` 0 oo oo ~ co ~ c~ o~ ~ cr~ ~ ~ ~ c~ o ~-~ 1 0 u~ ~ _ c~ ~ 0 ~ ~ ~ ~ I~ ~n ~ ~- 0 r~ ~ ~ ~ ~ ~ C~ ~ ~ O O O O _ ~- ~ - C~ ~ C~ C~ ~ C~ ~ ~ O ~ ~ C~ C . . . . . . . . . . . . . . . O 0` 0 0 ~ ~ U~ U~ ~ ~ ~ ~ C~ ~ C~ C~ C~ C~ C~ C~J C~ C~ C~ ~ C~ C~ C~ C~ C~ O X ~ ~ ~ ~ ~ ~ 0 ~ O ~ ~ ~C~ e2- w) r- 0 -) ~ ~ ~ ~ ~ ~ o~ ~ ~o 0 _ ~ ~ _ ~ C~ C~ C~ ~ C~ ~ ~ ~ ~ C ~C~ _ ~ ~ ~ ~ - ~ ~ - ~ ~ ~ - ~ ~1 ~ ~ ~ ~ ~ C~ U~ CX) O ~ ~ O ~ ~ ~D . . . . . . . . - O ~ ~ ~- ~ ~ ~- ~ L~ ~ ~ LO ~- ~4 ~ ~ ~4 o o c~ ~ o ~ ~ o u~ ~ ~ ~ o ~ ~ ~ ~ o c~ ~ ~ ~ ~ ~ ~ ~ o ~ ~ c~ ~ ~ ~ ~ c~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 0 _ ~ ~ ~ u~ ~ ~ ~ r~ ~ c~ o~ o~ r~ ~ ~D ~ ~ O ~ ~ r~ a u~ LO 0\ ~ 00 ~ u~ ~ ~ r~ ~ L~ ~ er u~ ~ ~ c~ c~ - - o o ~ o~ 0 0 ~ ~ ~ - - - - ~ - - - ~ - ~ 0 o ~ ~c~ - ~ - ~ ~- 1 1 o ~> ~ ~ ~ ~ ~ e~ u~ ~ ~ ~ o o~ S - C~ - O O O ~ O ~ ~ ~ 0 0 1- t- c~ ~ c~ c~ c~ c~ - ~ ~ - ~ ~ bRo ~o ~ ~ c: c~ - o ~ o~ c, 5 4, O - {C C~ C~ ~4 0 ~ ~ ~ ~ U~ ~ ~ ~ c~ ~ 0 ~ 0 a~ r~ ~ ~ t_ t- t_ t- Ct~ C~ ~ C~ 0` ~ C~ 0` ~ 0\ ~ C~ _ _' ~ ~4 ~ _ _ ~ _ _4 _ _ _ _4 C~ CO C~ _ - io t~ c-4 I_ C} ~ {,, _ 0` ~ - _' C) C ~_ ,_ C)~ ~_ - __ ~ C~_ C~ C~ C,O ~ C~ C ~O C) ~ CL 3 ~ C) C~ C) _ _ .- C~ _ ._ C~ o ." - + q_ C~ o ~._ X ~e~ ~ _ CC ~ ~Z + _ {C =. _ O ~ _ L- ~ O & C) ~ Ct C) _ ~ C~ _ _ e~ _ :~ ., ~S ~S ~ _ _ D" 0 a: ~; Ct _4 C~ ~ ._ C) ~ ._ - _ ~e _ ~0 O ~ _ . ~ e~s O _ C) -O i C~ ._ ~ _ C~ _ ~ ~ cr ~c ~ C ~_ C) ~5 ~_ _ _ _ ee Ct ~ ~ ,0 X e' Z _ ~_ ,0 ~ 3: O X O Z 4) C~ CO

OCR for page 33
TRENDS IN ADOLESCENT SEXUALITY ACID FERTILITY 61 black women compared with 41.5 percent for white women (National Center for Health Statistics, 1986~. Among Hispanic adolescents, the proportion of nonmantal births was ;0.l percent (unpublished tabula- tions by the Division of Vital Statistics, National Center for Health Statistics, 1986~. In short, although white teenagers were more likely in 1984 to give birth outside marriage than they were in 1970, black teenagers were still at significantly greater risk than either whites or Hispanics of having a nonmantal birth. Future Projections What do these trends mean for the absolute numbers of children born to teenage mothers since 1970, and what do they suggest for adolescent childbearing into the 1990s? Over the past decade and a half, the annual number of children born to teenage mothers (regardless of marital sta- tus) dropped from 64;,000 in 1970 to 479,600 in 1984. This reflects a decline both in the birth rate and, since 1976, in the size of the adolescent population. Congressional Budget Office analysts project that the total number of births to teenagers is likely to continue to decline somewhat for the next several years, as the number of adolescent girls continues to decline. Even if the adolescent birth rate remains at the 1984 level, the total number of births to teenage mothers would drop to 422,000 by 1992 (tonne and Adams, 1985~. Future trends in nonmantal childbear- ing depend on individual decisions concerning marriage, which are more difficult to predict. DATA ISSUES Crucial to our understanding of adolescent pregnancy and childbear- ing is the availability of accurate data on the relevant issues. Without data, research cannot be conducted; with poor data, research conclusions may be misleading. Data used in most studies of teenage childbeanng come from one of three sources: surveys of individuals, government record systems, and "formation Tom service programs. Each type of data plays a different role aIld informs discussion from its particular perspective. When results obtained from several different types of data pout toward the same conclusion, one can have particular confidence in that conclusion. Thus, though each type of data has a different contribution to make, aD three types are unportant. Each Is discussed ~ this section in An.

OCR for page 33
68 ADOLESCENT SEXUALITY; PREGNANCY; AND CHILDBEARING The most complete data available on adolescent parenthood are on the total number of adolescents in the U.S. population and the number of births to adolescents. As indicated throughout this chapter, however, data on teenage sexuality, contraceptive use, pregnancy, abortion, and adoption are available though often incomplete, and the extent of misre- porting and underreporting is often not accurately known. These data come from several sources, each source providing infor~a- tion concerning specific aspects of adolescent sexuality and fertility and relevant associated factors. However, there are significant mconsisten- cies among these data sets-for example, ir3 the use of age, race, and income categories, in the time frames for reporting, and ~ the defini- tions of core concepts, such as "at risk of pregnancy." Consequently, it iS frequently difficult for researchers tO integrate data from different sources in order to better understand causal relationships that may exist between the observed characteristics and the behaviors of teenagers. In addition, it is difficult to trace the relative importance of various factors at different points in the decision-making sequence leading to teenage childbearing. Of special significance are the problems of explaining race differences in adolescent sexual activity, pregnancy, abortion, and nonmantal child- bearing. Although available data contribute to understanding the associ- ations between an indindual's characteristics and behaviors for exarn- ple, living ~ a single-parent family, early sexual initiation, and early nonmantal childbearing they do not lend themselves tO forming COI1- clusions concerning the chain of causality and how this may have changed over time. A special data issue that has been highlighted throughout this chapter concerns the lack of complete and consistent information on ethnicity. In particular, we have pointed out the difficulties in presenting accurate estimates of the sexual and fertility behavior of Hispanic adolescents. The problems are even more pronounced for other ethnic groups, such as Nanve Americans. problems of omission, small samples, and inconsis- tent and noncomparable categorization have hindered knowledge of the behavior of signiEcant population subgroups. Sun~eys Much of the information on adolescent sexual acnnty and pregnancy discussed in this report is derived from household surveys in which individual teenagers are interviewed, typically in person in the* homes.

OCR for page 33
TRENDS IN ADOLESCENT SEXUALITY AND FERTILITY 69 Respondents are chosen for the interviews using carefully developed procedures designed to create a sample of individuals who accurately represent the larger population. Difficulties inevitably arise because some of the respondents selected refuse to be interviewed or the inter- - viewer never manages to find them at home. When conducting a study among adolescents on a sensitive topic for which parental permission is required, there is an additional step in the process, and additional refusals may result. If such refusals occur randomly that is, if the people who refuse or could not be located are just like the people who actually participate then the substantive results would not be affected; how- ever, this is rarely the case. Those persons who refuse differ in ways that cannot always be predicted but that may affect the conclusions of the research. If, for example, parents who hold very conservative and stnct views about adolescent sexual activity are more likely to refuse permis- sion for their child to be interviewed, then children reared in conserva- tive homes will be underrepresented in the analyses conducted with the data. Another source of ~if~cuity arises when respondents do not accurately report their attitudes or experiences. This seems to be an acute problem for some respondents who have been pregnant and who have had an abortion or relinquished their children for adoption. Data on sexual activity appear to be more reliably reported by teenagers (although it is quite possible that some males overreport sexual activity). In addition, over time there have been changes in the public perception of particular behaviors, which may make them easier to discuss. For example, part of the increase ~ sexual activity during the 1970s may simply reflect a greater openness ~ reporting nonmantal sexual activity. It is very diffi- cult to ascertain the accuracy of data coming from confidential ~nter- news unless it can be verified by other sources. One such source is the nsing incidence of reported sexually transmitted disease and abortion dunng the 1970s, which suggests that an increase in sexual activity has indeed occurred. In addition, data from several types of surveys concur in their est~rnated levels of sexual activity; thus data on sexual activity appear to be reliable in general. However, data Dom providers of abor- tion sentences indicate that there is considerable underreporting of abor- tion by respondents to surveys, particularly among unmarried black girls an] boys and white girts. Thus estimates of pregnancy and of abortion denved from survey data must be Mewed with caution. In the absence of nonsurvey data on adoption, it is impossible to validate the incidence of adoption reported in surveys except by companug the number of

OCR for page 33
70 ADOI ASCENT SEXUALITY; PREGNANCY; AND CHILDBEARING women who report having adopted an American child to the number who report relinquishing a child. There are other shortcomings of existing survey data In particular, data from boys and from very young teenagers have been Biscuit to obtain. Teenage boys are less likely to agree to be interviewed about sexual activity and pregnancy, it appears, and they are less likely to provide accurate inforTnation. Hence, nonresponse and misreporting are more likely among boys. Parents frequently deny requests for interviews of very young adolescents. In addition, since sexual activity among teenagers 14 and younger is fairly uncommon, a very large sample is necessary in order to obtain enough cases to support statistical analysis AL of these problems interview refusals, discrepancies in data from different sources, large samples that include few individuals with the relevant characteristics increase the cost of data collection. Although there is good reason to have samples large enough so that blacks, whites, and Hispanics can be studied separately, the costs of data collection can be so large that periodic surveys may be infrequent or corners may be cut, and data quality supers. And yet the substantive concerns are valid It is important to have more and better data on young adolescents and on males, and it is very important to be able to study socioeconomic differ- ences within as wed as between racial and ethnic subgroups, as a basis for designing more sensitive policies and more effective interventions. Record Systems Although U.S. data are not as complete as data from the population registration systems of many European nations, several of the U.S. rental record systems are of high quality. In particular, most births are recorded on birth certificates and most deaths are similarly recorded on death certificates. Thus, the number of births to teenagers can be tracked accurately across time, and infant mortality can be measured with con- siderable certainty. By contrast, the reporting of miscarriages is far Tom complete. Similarly, in abortion data collected by the Centers for Disease Control (CDC) from state health departments, the number of abortions is significantly incomplete, primanly because some states do not have abortion reporting systems, and in many states that do have systems, not all service prodders are covered. Estimates based on data obtained di- rectly from providers of abortion sernces by the Alan Gut~macher Institute are consistently higher; however, these data do not provide

OCR for page 33
TRENDS IN ADOLESCE.~IT SEX UALI7Y AND FERTILITY 71 information regarding the age, race, and marital status of women who obtain abortions. Information collected by both organizations must be integrated in order to obtain more accurate estimates of the incidence and rates of abortions among age, race, and marital subgroups. Ad~opnon is a potentially significant type of pregnancy resolution about which very little is known. Because the national adoption report- ing system was `discontinued in 1975, it is currently impossible to obtain a complete enumeration of adoptions nationwide and to assess trends in adoptions. Currently, the only system that gathers annual information is the Voluntary Cooperative Information System, managed by the Amen- can Public Welfare System. This system collects data only on children placed for adoption by public child welfare agencies and therefore does not count private placements. In addition, while it collects some infor- mation on the characteristics of adopter! children and adoptive families, it contains little or no information on the characteristics of birth parents, the adoption process, and the subsequent fertility, mantal status, and economic status of the birth mother. Data From Service Programs While household surveys are conducted for research purposes, and recor~keeping agencies are funded by the government to maintain basic public stanstical data about the population, infonnation collected by service programs is typically gathered for either fiscal or management pw poses and thus is only occasionally suited for research purposes. For example, data may be collected on insurance or Medicaid coverage but not on family income or socioeconomic status. Clients may be given a different identification number every time they appear for service, mak- ing it unpossible to track they use of the service across time. In addition, samples that are obtained from clinic populations tend to be small and are almost never representative. Thus, conclusions based on such samples cannot be extended to individuals who do not appear tO receive the service. For example, contraceptive use among adolescents attending birth control clinics is likely to differ from use by those who see private physicians, or who use drugstore methods, or who use no method at all. Since control groups composed of similar individuals who do not receive the service are rarely obtained, it iS usually difficult tO isolate the effect of receiving the service from the selection process involved in requesting the service In the first place.

OCR for page 33
72 ADOLESCENT SEX US LITY; PREGNANCY; AND CHILDBEARING CONCLUSIONS This chapter has descnbed trends and differentials in adolescent sexual and fertility behavior, including the size and charactenstics of the popu- lation at risk of early pregnancy and childbearing. The available data suggest a number of conclusions: Adolescent sexual activity increased sharply during the 1970s but appears to have declined slightly or leveled off since then. Among sexually active teenagers, contraceptive use has increased; however, many teenagers are inconsistent users, and many do not begin to use contraception for a year or more after they initiate intercourse. The number of pregnancies increased somewhat during the 1970s; subsequently the number of pregnancies has declined slightly. The pregnancy rate calculated for all adolescents has increased stead- ily since 1970. When calculated for those who are sexually active, however, the pregnancy rate has remained stable during the past decade as a result of increased contraceptive use. The rate of births to adolescent mothers declined in the 1970s, largely because of the rise in abortion and increased contraceptive use. Marriage among teenagers has declined since the early 1970s, and as a result, an increasing proportion and an increasing absolute number of births to adolescent mothers have occurred outside marriage. The sequence of choices presented in Chapter ~ provides a context for understanding the social and demographic processes that lead to adoles- cent childbearing; choices at each step in the sequence affect the size of the population at risk of subsequent behaviors. Figure 2-; adds to the contextual framework presented in Figure 1-! our best estimates of the size of the population involved at each step in the sequence. Inconsisten- cies among the relevant data sets make it difficult to develop accurate estimates of the numbers of young women at each step in the sequence. Recognized inaccuracies (e.g., in reported pregnancies and abortions) and gaps in the data (e.g., on adoption placements) preclude precise estimates at any given point in time or for a particular cohort of adoles- cents. Nevertheless, rough estimates based on calculations using data from several sources are interesting for two reasons. First, they help demon- strate the relative weight of these events by placing the number of in~in~uais at each step of the sequence In the context of the total

OCR for page 33
TRENDS IN ADOLESCENT SEXUALITY AND FERTILITY 73 No initiation of l intercourse 5,124 Contraceptive use 1.863 PJI adolescent girls ages 15-19 9,772 Initiation of intercourse 3,866 Effective contraception 1,565 | Mscamages |/ / 1 'I ~ / Abortions 34~ Adoption Placements _ 12(~) _ Remain in parental home (I Premarital Marital pregnancies pregnanc es 8S7 180 Prernantal births 261 Unmamed mothers 249(~-) Marriage before intercourse 147 No contraceptive use 2,003 I No conception 1,402 1 Marriage before birth tleg~mation) 184 Marriage subsequent to birth (I) l Establish own household ( A) FIGURE 2-o Sequence of decisions and estimates of population at each step, 1982 (in thousands).

OCR for page 33
74 ADOLESCENT SEXUALITY; PREGNANCy AND CHILDBEARING population of adolescent girls. Second, they have significant implica- tions for intervention strategies and for assessing the incremental effects of alternative approaches. As Figure 2-5 shows, in 1982 among the tote] of 9 77 million adoles- cent girls ages 15-19, less than half initiated sexual intercourse. Of those who did, just less than half 1.86 million were using some form of contraception. Of the estimated 2 million who were not using contra- ception, about 30 percent experienced a pregnancy in 1982; of the estimated 1.86 million contraceptive users, about 16 percent experienced contraceptive failure and became pregnant: of the estimated 897,000 premarital pregnancies that occurred, roughly one-third were to unsuc- cessful contraceptors and two-thirds were to nonusers. Among those who became pregnant, approximately 39 percent or about 348,000 obtained an abortion. Another estimated 12 percent or about 104,000 experienced ~ miscarnage. Approximately 445,000, just under half, carried their pregnancies to term: among this group, approximately 184,000 married to ieg~mate the birth, and roughly 261,000 gave birth without marriage. A very small proportion of these young women placed their children for adoption. The remainder, an estimated 249,000, became single parents. These estimates have important implications for identifying target populations and designing strategies for intervention. A significant pro- portion of ah teenagers initiate sexual activity each year, and a significant proportion of these experience an unintended pregnancy, either because they are not using contraception at all or are not using it effectively. A very small proportion of the total population of girls ages 15-19 become unmamed mothers in any given year, roughly 2.5 percent. Although they are a small target population, the cumulative proportion is much higher, and these young single mothers are at high risk of experiencing serious heath, social, educational, and economic problems, and the public costs for their support are substantial. Increases in the number who delay sexual initiation and who effectively use contraception if they are sexually active could significantly reduce the number of adolescent girls at risk of unintended pregnancy, abortion, and childbearing outside mamage. The next three chapters explore what is known about the changing societal context an] relevant factors affecting adolescent sexual activity, contraception, abortion, and childbearing outside marriage.