Click for next page ( 96


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 95
4 Detenn~nants of Adolescent Sexual Behavior and Decision Making The research on adolescent pregnancy in the United States over the past 15 years has made enormous strides In enhancing the understanding of teenage sexual behavior and decision making. For teenagers of different ages, at different stages of cognitive and socioemotional development, and living under different social, economic, and cultural circumstances, choices concerning sexual behavior reflect very different degrees of ra- tional thinking and conscious decision making. For many, choices tO ~ . . . Satiate intercourse, to continue sexua ac~vlty, tO use contraception, or tO marry versus bearing a child and raising it as a single parent may in fact be nonchoices. A substantial body of research exists on the variety of ~ndivid- ual, family, and somal factors associated with adolescent sexual activity; Chapter 2 presented the trends of the past decade and a half In this activity. This chapter discusses the determinants of six components of adolescent sexual behavior: 3~tiai~on of sexual acidity, contraceptive use, abortion, marriage before childbearing (legitimation), adoption, and childbearing and rearing outside mamage. DETERMINANTS OF ADOLESCENT SEXUAL ACTIV1-IY Research suggests that a number of factors are strongly associated with the initiation of sexual aciinty before mamage. Among the most impor- tent of these are individual characteristics such as puberty and other development ad charactenstics, age, race, and socioeconomic status, reli- g~ousness, mtedigence and acadern~c achievement, and dating behavior; family charactenstics, such as family background and parental support and controls; and the influence of peer groups. 95

OCR for page 95
96 ADOLESCENt SEXUALITY PREGNANCY AND CHILDBEARING Individual Characteristics Pubertal Development There is almost universal agreement among the studies that have addressed the issue that early pubertal development (e.g., age of menarche for girls, body development and hormonal levels for boys) is strongly associated with the early initiation of sexual activity (Billy and U3ry, 1983; U6ry, 1979; Moms et al., 1982; Westney et al., 1983; ZeInik et al., 1981~. However, the importance of physical maturity varies by sex and race. Recent studies of pubertal development, sexual motivation, and sexual behavior among white adolescent boys and girls provide strong evidence for the hormonal basis of motivation and behavior in boys. Among girls, hormonal levels were shown to have strong effects on levels of sexual interest but only weak effects on sexual behavior (U3ry et al., 1985a, 198Sb). These researchers conclude that girls' actual behav- ior is influenced to a greater extent by their social environment than by physical maturation. There are no comparable data for black boys and girls. However, earlier work suggests that the association between puber- tal development and sexual behavior was stronger for white than black girls (Ze~ik et al., 1981) and for white boys than for black boys (Moms et al., 1982; Billy and U3ry, 1983~. Evidence that a sizable minority of black boys report initiation of intercourse prior to puberty (Westney et al., 1983; Clark et al., 1984) suggests a stronger effect of social environment for black boys than for white boys. As Ho~erth concludes (Vol. Il:Ch. I), although there appears to be a strong relationship between pubertal development, hormone levels, and sexual activity, social factors do intervene in determining when and how both boys and girls initiate sexual intercourse, given maturation. For girls especially, biological factors do not appear to operate independently of the individual's social context and concept of sexual readiness. Therefore, how social factors mediate maturational factors remains an important yet not Filly explored issue. Age at Initiatior: Available data suggest that more adolescents are be- coming sexually active at earlier ages. Nevertheless, the older the teenager, the more likely he or she is to have had intercourse (Zeluik et al., 1981; computations of cumulative sexual activity by single year of age US3mg data from the National Survey of Family Growth in Vol. Il). Apparently, only a minority of young people do not become sexually experienced while Sty in their teens: more than 80 percent of males and 70 percent of females

OCR for page 95
DETERMINANTS OF SEXUAL BEHAVIOR 97 report having had intercourse before their twentieth birthday. The pro- port~on of sexually active teenagers increases with age. lust under 17 percent of boys and 6 percent of g}rIs reported having had intercourse before the age of 15. Nearly 67 percent of boys and 44 percent of girls reported that they were sexually expenenced by age 18 (computations of cumulative sexual activity by single year of age using data from the National Longitudinal Survey in Vol. Il). Regardless of age, the first time adolescent girls have intercourse, they tend to have partners who are about three years older; boys' initial partners are approximately one year older (Zelnik and Shah, 1983~. Race an' Socioeconomic Status Black boys and gills become sexually expenenced at earlier ages than their white counterparts (on average about two years earlier) and, at every age, more black than white teenagers are having intercourse (ZeInik et al., 1981; Vol. Il.; Bauman and U3ry, 1981; Newcomer et al., 1980; Newcomer and Udry, 1983~. Sexually expen- enced blacks, however, appear to have intercourse slightly less frequently than whites (Ze~ik et al., 1981; Zabin and Clark, 19811. While there is some evidence that young black girls are slightly more likely to be physi- caDy mature than whites of comparable ages (Harlan et al., 1980; Devaney and Hubley, 1981), these differences between the races in physical matu- rity seem too small to explain the large race differences In early premantal sexual activity (Moore et al., 1985~. Disagreement exists over the source of racial differences In the propor- tion of teenagers who are sexually active and the age of sexual Ovation. Some researchers attribute the dispanty who3Iy or ~ large part to soclo- econom~c differences among blacks and whites. Others trace it to sigmfi- cant normative differences ~ the acceptability of early sexual behavior. These explanations are not quite so divergent as they might Erst appear, since many who believe that there are subcultural differences trace them to economic and social disadvantage. One hypothesis ~ this regard suggests that neighborhood env~romnents are very ~rnportant. Because of past histories of residential segregation, most blacks (even made - class blacks) live ~ neighborhoods that are substani3aDy poorer than their white coun- terpares, and their children are subjected to different pressures than are their white peers (St. John and Gras~ck, 1982; Hogan and Sagawa, 1983~. Simiiariy, the length of One an Lingual or his or her family has lived In poverty may affect sexual attitudes and behavior. In addition, many who account for racial differences by socioeconomic

OCR for page 95
98 ADOLESCENT SEXUALITY; PREGNANCY; AND CHILDBEARING explanations concede that chronic economic disadvantages may give rise to different outlooks on marriage and family formation, which in turn affect the acceptability of early sexual behavior (Moore et al., 1986; Abrahamse et al., 1985~. Blacks report a greater tolerance for sexual activity outside a mantal relationship than whites; they rate mamage as less ~rnportant than do whites; and they perceive a greater tolerance in their neighborhoods for childbearing outside marriage (Moore et al., 1985~. Williams (1977, cited in Moore et al., 1986), for example, reports that of the pregnant black teenagers in his Rochester, N.Y., sample, 70 percent expected a favorable reaction from peers and 65 percent antici- pated a positive reaction from the baby's father, compared with 40 and 43 percent, respectively, among the pregnant white girls he studied-. Simi- larly, black teenagers interviewed in the 1976 National Survey of Young Women were considerably less likely to perceive condemnation of an unmarried mother in their neighborhood than were white teenagers (Zeinik et al., 1981~. In some studies, blacks also indicate a preference for a younger age at first birth than age at List marriage, while whites report jUSt the opposite (Zabm et al., 1984; Peterson as reported in Moore et al., 1986~. Such attitudes do not cause premarital sexual activity and teenage pregnancy, but neither do they discourage it (Moore et al., 1986~. it may be that they simply reflect the prevalence of teenage pregnancy in the black community. As Hofferth powts Out (Vol. Il:Ch. I), it is Iffier to understand the role of attitudes because most studies have been unable to sufficiently control for them. The point at which values and attitudes about premantal sexual activity are usually measured is after sexual ID1tia- non, an] therefore it seems likely that sexual experience may have already influenced the respondent's news. A recent unpublished analysis of data Dom the 1981 National Survey of Chidden attempted to em - ine the sources of racial differences In levels of sexual acnnty and age of Vitiation of sexual activity (Furstenberg et al., 1985b). Using a vanety of Indicators of socioeconomic status and social disadvantage, it was possible to reduce only a small portion of the racial difference ~ the incidence of sexual expenence before age 16. However, much of the difference between blacks an] whites could be explained by talcing into account the racial composition of the schools attended by the survey participants. The investigators found high proportions of sexually experienced teenagers among blacks of the same somal background who attended racially isolated schools compared with those who attended

OCR for page 95
DETERMINANTS OF SEXUAL BEHAVIOR 99 schools with white students. Among blacks and whites who attended racially diverse schools, there were only modest differences in the proba- bility of sexual intercourse. In addition, there was some evidence that the racial composition of the schools was also assc~ated with differences in the acceptability of premantal childbearing. To some extent school segrega- tion is, undoubtedly, a surrogate measure for low socioeconomic status among many blacks and reflects the pervasive conditions of disadvantage that characterize the neighborhood environments in which they live. However, further research on this difficult issue is needed. In conclusion, the research on racial differences continues to show strong black-white differences in sexual intercourse at young ages, even controlling for socioeconomic differences. However, questions have been raised as to the adequacy of these controls, given the substantially unequal distribution of socioeconomic status by race, the inequality within categories of socioeconomic status, and the failure to control for the length of time an individual or his or her family has lived in poverty (Hofferth, Vol. Il:Ch. I). Siniilarly, repeated documentation of differ- ences in attitudes and behaviors does not definitively resolve the source of racial differences. Whether any such race differences represent sub- group values or more transient attitudinal adjustments to external c=- cumstances is not clear. The data do suggest, however, that there are differences in community standards and expectations that affect the acceptability of early sexual behavior in the peer group. If this finding is true, it has important implications for the kinds of strategies that are likely to be successful in lowering the rates of early pregnancy and childbearing among blacks as wed as whites. Current efforts by black community organizations to modify adolescent attitudes may further illuminate this issue. Further research on this rather neglected topic is needed. Religiousness Religiousness appears to be an important factor dist~n- guish~g early Tom later initiators of sexual activity. Delaney and Hub- ley (1981) found that women ages 15-19 were more likely to be sexually active if they were not regular church attenders and if they reported that religion was not veer important to them. These findings are supported by numerous other studies (Inazu and Fox, 1980; Ze~ik et al., 1981; lessor and lessor, 1975~. Most researchers who have addressed this issue have found that the tendency to be devout and observant of religious custom and teaching Is more important than any specific religious afElia

OCR for page 95
100 ADOLESCENT SEXUALITY; PREGNANCY AND CHILDBEARING lion. In particular, Catholicism, which was once regarded as a good index of conservatism on Tnorai issues, has more recently been found not to be a very accurate predictor of sexual experience. Devaney and Hubley (1981) found no difference in the likelihood of reporting sexual experi- ence between Catholics and those of other denominations. Protestant fundamentalism, however, which has gained visibility and followers in recent years, has frequently been associated with strong conservative positions on issues of sexual behavior. As Hofferth reports (Vol. Il:Ch. 1), teenage adherents to fundamentalist denominations have been found less likely to have had sexual intercourse outside marriage than members of other denominations (Thornton and Camburn, 1983~. Religious teenagers may also be those who are more traditional in general than other teenagers and therefore less likely to engage in behaviors that push toward adulthood (e.g., smoking, drinking). They may also have stronger social supports to enforce behavioral nones. Intettigence, Academic Aspirat20ns, a??4Achieveme?~t A number of stud- ies suggest a strong association between low intellectual ability, low academic achievement, a lack of educational goals, and early sexual experience among both blacks and whites. Adolescent girls who score low on intelligence tests and place little value on educational attainment are more likely to have intercourse at an early age than those who are educationally ambitious. Conversely, those who score high on intelli- gence tests, are academically motivated, and are doing wed in school are less likely to Agate sexual activity at a young age (Most, 1983; Devaney and Hubley, 1981; Furstenberg, 1976; Hogan and Kitagawa, 1983; Ugly et al., 1975; Moore et al., 1985; Jessor and lessor, 1975; lessor et al., 1983~. An earlier analysis by Mott (1983) indicated similar results for young men ages 17-20. The association between ability, educational aspirations, and perfor- mance and the lower likelihood of early sexual experience is undoubtedly tied to severs interacting social, economic, psychological, and situa- tionil variables. For example, parents' level of education and their aspirations for their children can significantly influence teenagers' own attitudes and expectations about academic achievement (Davies and Kandel, 1981; Spenner and Featherman, 1978~. Parents with more edu- cation are generally more affluent than parents with less education. Children in families with more educated parents tend to be more goal- onented, to place a higher value on achievement, and to be more on

OCR for page 95
DETERMINANTS OF SEXUAL BEHAVIOR 101 ented to work than play (Conger, 1973). Chilinan (1980b) suggests that these characteristics may make a teenager less likely to engage in premar- ital intercourse dunng the junior high or high school years. She also suggests that, especially for girls, involvement in educational achieve- ment (thereby pleasing parents and teachers) may inhibit interest in boys or may make girls less interesting to boys. The Nature of SexualBehavior Regardless of age, each adolescent boy and girl who enters a sexual relationship does so at a particular level of socioemotional and cognitive development and with whatever self- perceptions he or she has foe, as well as within a particular social and cuin~ral context. At least until recently, studies showed that adolescents, especially young white women, gradually advanced their level of sexual intimacy through a series of dating aIld "going steady" experiences (Vener and Stewart, 19741. There was generally a learning period dunng which a boy and a girl became better acquainted and developed art affectionate relationship. Research on the incidence of these behaviors suggests that this pattern of gradually developing sexual intimacy dur- ing adolescence is Still common among white youth, although it is beginning earlier and progressing more rapidly. it does not appear to hold true for young blacks, however. Existing research Endings concerning the nature of teenage sexual behavior, as McAnarney and Schreider (1984) suggest, derive from data on midge or older adolescents. Their applicability to the youngest teens is not clear. When young adolescent girls beam having intercourse, it is generally infrequent and unpredictable (Kantner and Zeinik, 1972~. They frequently report the need to be "spontaneous," and therefore the event does not really reflect rational and planned behavior. Especially among very young adolescents, their level of logical operational th~k- ing may not be sufficiently developed for them to recognize that having or not hanng intercourse is a choice and that without contraception it caI1 result in pregnancy (McAnarney, 1982~. As they become older, many researchers believe, teenagers become better able to make well-reasoned, conscious decisions about their sexual behavior. It is interesting to note, however, that research from other developed countries shows that even young sexually active teenagers can effectively avoid pregnancy (Alan Guttmacher Institute, In press). Self-esteem appears unrelated to the initiation of sexual activity for both boys and girls (Most, 1983; C,retLonch and Grote, 1980~. Nevertheless, there is a dearth of weD

OCR for page 95
102 ADOLESCENT SEX UALI7Y; PREGNANCY AND CHILDBEARING developed theoretical models to explain and predict sexual decision making by adolescents (Libby and CarIson, 1973~. Research findings also suggest that sexual activity is generally not an isolated behavior. Adolescents who are sexuaTiy active at an early age are also frequently involved in other behaviors that push toward indepen- dence and adulthood, often in conflict with adult norms for them. Among transition behaviors most often associated with early sexual acidity are smoking, drinking, and drug use Lessor and.lessor, 1975; lessor et al., 1983~. The extent to which there may be a causal link among these behaviors is not fully understood. There is ambivalence In our society about whether these behaviors represent a healthy assertion of independence or denant (if not delinquent) behavior (Ensm3mger, Vol. IT:Ch. 2~. In this regard, lessor et al. (1983) suggest that the decision by an adolescent to become sexually active may more accurately reflect the conscious or unconscious decision to assume a particular life-style rather than to adopt a single isolated behavior. Early dating appears to be associated with early sexual experience (Furstenberg, 1976; Spanier, 1975~. The more frequently teenagers (es- pemally garish date, the more likely they are to have intercourse (Simon and Gagnon, 1970; Presser, 1976b). Ire addition, while survey data show that more teenagers are sexually active and that there has been a relative decrease In the number of their partners (Zeluik, 1983), several studies suggest that the more committed the relationship between young peo- ple, the more likely they are to have intercourse (SpaIlier, 1975; Fursten- berg, 1976; Sorenson, 1973; Reiss, 19761. Family Characteristics P`2r~nm1 .C~unnortanl Controls or ~ A number of studies have found that the nature oftheir relationships tenth parents affects teenagers' sexual behav- for. Adolescent girls are more likely to have premantal intercourse if their mothers fait to combine affection with firm, mild discipline and to set clearly dewed limits 0D behavior. However, as Hofferth (Vol. Il:Ch. i) suggests, since adolescence is a time of testing one's independence and gradually growing away from parents, it is also possible that a decline in the closeness of a mother-daughter relationship follows the initiation of sexual activity rather than preceding or causing it. Alternatively, both decline in closeness and initiation of sexual intercourse could be the result of in creased independence.

OCR for page 95
DETERMINANTS OFSEXUALBEHAVIOR 103 Some studies also show that young people are more likely to be sexual'~y experienced if they perceive themselves to be in poor communi- cation with their parents and feel that they receive little parental support (Simon et al., 1972; lessor andlessor, 1975~. As Hofferth (Vol. IT:Ch. I) points out, however, parent-child relationships and parent-child com- munication, although important, seem to have an ambiguous associa- tion. As a result, there is no clear implication for program development. There is evidence that close relationships may be associated with less sexual activity among younger teenagers (tnazu and Fox, 1980~. lessor end Jessor (1975) simi'~arly found that the more consistent the values of teenagers and their parents, the greater sense of connectedness and supportiveness between them, and the closer the young people's ties to home, the less likely they were to become sexually active. Yet communication may be associated with a higher '~eve] of sexual activity among teenagers than a lower one, especially among older teenagers. First, in many cases, less parent-child communication takes place than is commonly assumed; second, such communication, whether so pronde information or to prescribe behavior, may not be fully heard by the child; and third, communication about sexual behav- ior frequently does not occur until after initiation of sexual activity (Newcomer and U3ry, 1983; Inazu and Fox, 1980~. Fox (1981) points out that parents' (espec~aTiy mothers') roles in sex education are reia- tively minor, and that the more traditionaTiy oriented mothers are on matters of sexual morality, sex roles, etc., the less likely they are to initiate discussions of these topics with their children. Unfortunately, however, as Hofferth (Vol. II:Ch. 1) points out, there is little research to specify the context of communication or to distinguish the effects of communication before and after initiation of sexual activity. One recent study that was able to make this distinction found no relationship between the frequency of communication about sexual topics (before initiation) with the mother or father and the sexual activity of the daughter (Kahn et al., 1984~. For boys, communication with the mother was associated with less subsequent sexual activity; communication with the father was associated with more sexual activity. Kahn et al. (1984) conclude thee perhaps fathers implicitly, if not explic- it~y, condone sexed activity among sons, without prodding the empha- sis on responsibility that mothers communicate. As with parental communication, the research on the relationship between parental supervision or control of adolescent behavior and the

OCR for page 95
104 ADOLESCENT SEXUALITY PREGNANCY AND CHILDBEARING initiation of sexual actiwty also suggests conflicting results. Hogan and Kitagawa (1983) found that more supervision was associated with less sexual activity among a sample of inner-city black Uris. In contrast, other researchers have found that more supervision was unrelated to the initiation of sexual activity (Inazu and Fox, 1980; Newcomer and UBry, 1983~. As Hofferth (Vol. Il:Ch. 1) points out, generalizations about the impact of parental supervision on sexual activity are not possible without further study. Other Fam~iy Characteristics There is a strong relationship between a mother's sexual and fertility experience as a teenager and that of her daughter (Newcomer and U3ry, 1983; Presser, 1976b). The earlier the mother's first sexual experience and first birth, the earlier the daughter's expenence. Other family [actors that appear to affect the level and quality of parental supports an] controls, an] perhaps in turn influence sexual behavior among teenagers, include family intactness, family composi- tion, and mother's age at marnage. Several studies have shown that girls in nonint act or female-headed families are more likely to become sexually expenenced at an early age than those in two-parent families (Zeluik et al., 1981; Newcomer and U6ry, 1983; Moore et al, 1985; Inazu and Fox, 1980~. Similarly, the larger the family (the more siblings present), the more likely that an older sibling will be sexually active and provide a mode] for younger siblings (Hogan and Kitagawa, 1983~. Although there is strong evidence of these associations, the mecha- nisms by which they affect adolescent sexual behavior are not fully understood. For example, some researchers hypothesize that the stress resulting from parental separation or divorce and from the presence of several siblings may cause teenage children (especially daughters) to perceive a lack of attention and affection from their mothers and may lead them to seek such attention in sexual relationships. Others suggest that the inevitable stress of such circumstances may make it more difE- c:uit for parents to adequately supervise their teenagers. Still others suggest that, particularly in families in which there has been a divorce or the mother became sexually active at an early age, parents (especially mothers) may indirectly communicate an attitude of permissiveness. All of these are plausible explanations, aIld they warrant further explora- tion.

OCR for page 95
DE TERMING FITS OF SEX US L BEHAVIOR 1 05 Peer Group InfZue??ce Although research on peer influences on early sexual activity is rela- tively limited, the attitudes arid behavior of peers is frequently cared as the single most important factor affecting the initiation of intercourse by adolescents. it appears, however, that peer influence may have been overrated, particuiariy among blacks and white males. Hofferth (Vol. Il:Ch. I) cites several significant problems with the research, including the facts that (~) the same individual typically reports on his or her own as wed] as his or her friends' attitudes and behavior without independent validation and (2) data have been gathered at only one point in time, thus preventing researchers from detecting delayed effects. Several studies suggest that same-sex peers are a major source of information about sex (Libby and Carison, 1973; Miller, 1976; Thorn- burg, 1978~. In addition, Cve~kovich and Grote (1980) report that the proportion of their same-sex peers that teenagers believe are sexually experienced and how sexually liberal they believe them to be are power- fill predictors of sexual experience among adolescent boys and girls. Newcomer et al. (1980), however, conclude that individual behavior and attitudes are more closely related to what teenagers think their friends do and believe than to what is actually going on. It appears that many teenagers act on perceptions of their friends' attitudes and behavior, whether or not their perceptions are correct. Peer pressure can take several forms (e.g., challenges and dares, coer- cion, somal acceptability), and its influence seems to vary among young people of different ages and genders. There is some evidence that white boys choose their friends on the basis of sexual activity. Blacks, however, appear neither to be influenced by friends' behavior nor to choose friends on that basis (Billy and U3ry, 1983~. Giris may be swayed to some extent by what they think or know their female friends are doing, but they are more strongly influenced by their best male friends and their sexual partners (Miller and Simon, 1974; Herold, 1980; Cverkonch and Grote, 1980; Billy and Ugly, 1983~. I~ewis and I~ewis (1984) found that among young adolescents (~14) peer pressure in the form of challenges and dares significantly influences sexual involvement at several levels (e.g., kissing, fondling, and intercourse), especiaTiy among girls. Billy and U3ry (1983, 1984), however, suggest that among black boys and girls, peer influence is relatively minor. In general, white girls appear to be most susceptible to peer influences in sexual decision making.

OCR for page 95
112 ADOLESCENT SEXUALITY PREGNANCy AND CHILDBEARING Unirltendedness Among the most important factors affecting the outcome of a preg- nancy is whether it was intended. Girls who report that they wanted the pregnancy are more likely to give birth, while those who report that they did not want the pregnancy are more likely to have an abortion (Zein~k et al., 1981~. The issue of intendedness is complicated, however, since reports of whether a pregnancy was wanted are generally collected after the pregnancy has been discovered. In many situations it seems likely that although the conception was unintended, a girl may deade in retrospect (either consciously or unconsciously) that she must have wanted to be pregnant once she learns that she is pregnant and has decided against obtaining an abortion. Academic Aspirations and Achievement Girls who are doing well in school before pregnancy and who have a strong future orientation are more likely to choose abortion to resolve an unintended pregnancy than those who are not good students and who lack high educational and vocational goals (Steinhoff, 1976; Evans et al., 1976; Eisen et al., 1983; Leibow~tz et al., 1980; Devaney and Hubley, 1981~. This hordes for both blacks and whites. As with contraception, parental education also appears to be a very significant factor in preg- nangy resolution. The higher the parents' level of education, the greater the likelihood that a teenager win have an abortion rather than carry an unintended pregnancy to term (Zeinik et al., 1981~. Contraceptive Use Concern that abortion may become a substitute for contraception is not supported by the available research. In 1979, teenagers who had terminated an unintended pregnancy by abortion were less likely to have expenenced a second pregnancy within two years than those girls who carried their first pregnancy to term (Koenig and Zeinik, 1982~. Data from the National Center for Health Statistics suggest that about 12 percent of abortions to 15- tO 17-year-olds =d 22 percent of abortions tO i8- tO 19-year-olds are repeat abortions (National Center for Health Statistics, 1984b). Furthe~ore, clinic studies show that, three weeks following an abortion, less than 10 percent of girls were not using any

OCR for page 95
DE7ERMIA7ANTS OF SEXUAL BEHAVIOR 113 method of contraception, while more than 80 percent were using the pill or JUD (Alan Guttmacher Institute, 19811. Although such studies using clinic patients undoubtedly reflect the behavior of a self-selected sample, they do suggest that contraceptive behavior following art abortion may be better rather than worse among some teenagers (Forrest and Hen- shaw, 19831. Family Characteristics Several studies of adolescent girls who choose abortion have found that family background factors are significant predictors. In particular, it appears that whites are more likely than blacks to terminate an unin- tended pregnancy, and that girls from families with higher somoeco- nomic status are more likely to abort than those from poverty back- grounds, especially from families on welfare (Zeinik et al., 1981~. A note of caution In the interpretation of socioeconomic data is important, however, because survey respondents tend to underreport pregnancy and abortion. Giris from less religious families have been found more likely to choose abortion than those from more devout families. Surpns- ~ngly, however, rates of abortion were found to be higher for white Catholic girls than for either white non-Catholics or Hispanic Catho- lies, suggesting that religious affiliation may not be an important deter- mination in the decision to abort (Eisen et al., 1983~. Parents', espec~aDy mothers', attitudes about abortion have also been shown to significantly influence the outcome of an unintended preg- nancy. Giris whose mothers are more favorably disposed to abortion are less likely to have a birth (Eisen et al., 1983~. Among very young teenagers, it appears that parents have a major influence on the decision to terminate a pregnancy (Steinhoff, 1976; Rosen, 19801. Peer Influences The attitudes of peers also seem to influence decisions concerning pregnancy resolution. The more positive a girlfriend's or boyEriend's opinion of abortion, the less likely an adolescent girl is to have a birth (Eisen et al., 19831. In contrast, girls who have friends or family mem- bers who are teenage single parents are more likely to carry their preg- nancies to term (Eisen et al., 1983~.

OCR for page 95
114 ADOLESCENT SEXUALITY; PREGNANCY; AND CHILDBEARING Access to A portion Services Teenagers are less likely than women in their twenties to obtain abortions dunng the safer, earlier weeks of gestation. The younger the teenager, the more likely she is to delay. As the Alan Guttmacher Institute (1981) reports, only 34 percent of abortions to girls age 15 and younger are performed dunng the first eight weeks of gestation, com- pared with 41 percent of abortions among girls ages IS-19, and 51 percent of those among women ages 2~24. At the other extreme, 14 percent of abortions to girls I: and younger are performed at 16 weeks and later (Alan Gut~macher Institute, 1981~. Such delays increase the health risks associated with pregnancy termination. The Alan Gut~macher Institute (1981) cites several likely reasons for the delay. First, many teenagers, particularly the very young, fail to recognize the signs of pregnancy early. Many adolescent girls ordinarily experience menstrual *regulanties and therefore do not distinguish them from early signs of pregnancy. Many others simply deny the unpleasant reality of an unintended pregnancy until it becomes unavoid- able. In addition, access to abortion sernces appears to be limited for many teenagers. Though most school-age adolescents, especially very young teenagers, consult their parents in deciding tO obtain an abortion, parental consent requirements in many states (or the perception of such requirements) are thought to inhibit some teenagers from seeking and obtaining abortions. Geographical distance from clinics or hospitals that perfonn abortions, as well as costs, have also been shown to limit teenagers' access to abortion, especially school-age girls and those from poor families (Alan Guttmacher Institute, 1981~. DETERMINANTS OF MARRIAGE BEFORE CHILDBEARING (LEGll IMATION) Although there are more conceptions to unmarred adolescents now than a generation ago, they are less likely today to resolve a pregnancy by marrying. The proportion of unmarred adolescents conceiving who married before the birth decreased from approx~nately 31 percent in 1970 to approximately 23 percent in 1981 (O'Connell and Rogers, 1984). The ex~st~g research on determinants of marriage to legitimate a birth is limited. Nevertheless, race, age at initiation, and the availability of financial assistance appear to be significant factors.

OCR for page 95
DETERMINANTS OF SEX UAL BEHAVIOR 1 1 ' Race and Socioeconomic Status White teenagers and those from families of higher socioeconomic status are more likely, if they are pregnant, to marry before bearing a child (ZeInik et al., 1981; O'Connell and Moore, 1980~. In 198~198l, approximately 28 percent of white unmarred teenagers ages 15-19 who conceived were marned before the birth. In contrast, only about 9 percent of all black teenagers married to legitimate a birth (O'Connell and Rogers, 1984~. Indeed, In all age groups, black women are now more likely to be unmarred than to be mamed (Moore et al., 1986~. Among teenagers, however, the primary reason is never having mamed rather than divorce, separation, or the death of a spouse. Ale Older teenagers are more likely than younger ones to marry to leg~ti- mate a birth. In 1981 only 11 percent of 14-year-olds ~named, while 63 percent of 19-year-olds marned before beanug a child (O'Connell and Rogers, 1984~. The highest proportion of marriages occurs among older white teenagers. Availability of Financial Assistance Two recent studies have found that two factors are strongly associated with decisions to carry a premantal pregnancy to term: (~) the availabil- ity of financial aid from the family of ong~n and (2) the availability of public financial assistance (Eisen et al., 1983; Leibow~tz et al., 1980~. They found that a major factor dist~guish~ng those who married from those who gave birth without marriage was the source of support. GirIs whose families had been receiving financial aid from the state during their pregnancies were less likely to marry than those who had not received such assistance (Eisen et al., 1983~. In contrast, Moore and Caldwell (1977) found no significant association between Aid to Fami- lies With Dependent Children benefit levels and acceptance rates and whether a pregnant teenager married before the birth. DETERMINANTS OF ADOPTION Because there are I1O systematically collected national data on adop- tion, it iS impossible tO denve precise estimates of the number and proportion of teenagers who choose this means of pregnancy resolution.

OCR for page 95
116 Al)OLESCENT SEXUALITY PREGNANCY A ~D CHILDBEARING There is also very limited research comparing teenagers who make a~optiorl plans and those who keep their babies and raise them as single mothers. As Hofferth (Vol. Il:Ch. 4) reports, the results suggest that teenagers who make adoption plans are similar to those who have abortions but differ from those who take on parenting responsibilities. They tend to be older and to come from families of higher socioeconomic status. They tend to have stronger academic ambitions and to be per- forming better in school. They tend to hold more traditional attitudes about abortion and family life. In contrast, parenting teenagers tend to have less schooling, to have dropped out of school, to have less well formulated educational and occupational goals, and to come from single- parent families (Hofferth, Vol. Il:Ch. 4~. There is virtually no research on factors affecting the decision of unmarred adolescents to place their children for adoption. However, two studies are now under way that may shed some light on this important topic (Kailen, 1984; Resnick, 1984~. DETERIvlINANTS OF NONMARITAL CHLDBEARING The existing research suggests that several factors are strongly assoa- ate] with nonmantal childbearing: race, attitudes, poverty and unem- ployment, and the availability of financial assistance. Race Although the rate of childbearing has increased dramatically among unmarred white adolescents since 1970, black adolescents have always been more likely to give birth outside marriage. Black teenagers account for 14 percent of the adolescent population and 46 percent of all births to unmarred 15- to 19-year-olds (National Center for Health Statistics, 1984b). In 1982, over 98 percent of births to black teenagers under age 15 occurred outside marriage, and 87 percent of births to black teenagers ages iS-19 occurred outside marriage. The comparable figures for white teenagers in these age groups were 78 and 36 percent, respectively (Vol. Il:appendix tables, section on births). The underlying causes of childbearing among unmamed young black adolescents are complex and ~i~CU]t t0 disentangle. Most researchers agree that the rising proportion of births tO unmarred black teenagers over the past generation is attributable to the declining rate of mamage tO legitimate a birth. In part it is also at~nbutable to changing patterns of

OCR for page 95
DETERMINANTS OFSEXUALBEHAVIOR 117 sexual activity and contraception among blacks. Black girls are twice as likely as white girls to have premantal intercourse, and they generally become sexually active at younger ages than whites. The higher propor- tion of blacks who are very young when they first have intercourse is associated with a higher incidence of sexual activity without contracep- tion. As one might expect, blacks are more likely to become pregnant and to become pregnant at younger ages. They are also disproportion- ate~y more likely to resort to abortion. However, since the higher abortion rate among blacks does not erase the dramatic race difference in rates of unintended pregnancy, births are much more prevalent among black than white teenagers (Moore et al., 1986~. Attitudes Toward Nonmantal Childbearing There are many accepted family forms in the black community, in- cluding the nuclear family (father, mother, and children), the attenuated family (father or mother and children), the extended, multigeneratior~al family (some combination of grandparents, aunts, uncles, parents, and children), and augmented families (one of the above types that also includes unrelated friends, boarders, or guests) (Billingsiey, 1970~. As several scholars have commented, this diversity of family forms makes single parenthood less unusual and provides more socially acceptable opportunities for accommodating young black single mothers and their children than are available in the white community (Williams, 1977, cite] in Moore et al., 1986; Miller, 1983~. Stack (1974) describes a support network among low-income urban blacks that helps cushion individuals from the conditions of life in poverty. She concludes that many young mothers End greater security in this network than in marnage. Black families seem to be more supportive of young mothers, and it may be that this supportiveness ameliorates a teenager's fear of becoming a mother (Moore et al., 19861. Still, many researchers and advocates argue that childbearing by unmarred adolescents is not highly valued in the black community. Although there is a greater tolerance of unmarried parenthood, it is generally viewed as unfortunate (Moore et al., 1986~. Furstenberg (1976) found that among a low-income urban clime population, only 20 percent of teenagers were pleased to learn they were pregnant; another 20 percent had mixed feelings; and the remain- der were disappointed or upset. Very few of these teenagers reported that their mothers were pleased; most mothers were reportedly hurt and depressed or angry (Furstenberg, 1976~.

OCR for page 95
118 ADOL=CENT SEXUALITY PREGNA ACT AND CHILDBEARING As discussed earlier, differences in attitudes about childbearing by very young unmarried teenagers appear to be closely related to community and neighborhood standards of tolerance and acceptability. Most blacks live in poorer neighborhoods than whites. Therefore, as many research- ers argue, race differences ~ adolescent pregnancy and childbeanng may be sigmficantly linked to social and economic disadvantage (Hogan and Kitagawa, 1983~. Recent research also suggests that an important factor influencing attitudes about nonmantal childbearing are perceptions of opportunities that are unrelated to reproductive behavior. In short, willingness to bear a child outside marriage is closely related to the implied costs of doing so (Abrahamse et al., 19851. These researchers infer that among low- mcome girls with Tow academic and occupational expectations Ming In single-parent families, the perceived opportunity costs of early, nonman- tal childbearing were very low, since their willingness to risk such parenthood was quite high. They infer that young girls whose lives and perceived opportunities are not currently gratifying may be more open to motherhood, which they may perceive as a potential source of gratifi- cation. Because blacks are more likely than whites to be poor, to live in non~tact families, and to demonstrate low academic ability and have low expectations, they may also be more willing to bear a child while unmarried (Abrah~mse et al., 1985~. Natha~son and Becker (1983), however, were unable to find any relationship between perceived oppor- tuniiies arid the contraceptive behavior of adolescents. Abrahamse et al. (1985) also showed a strong relationship between rebelliousness (i.e., disciplinary problems ~ school, Cutting classes, and absenteeism) among adolescent girls and willingness to become an un- maIried mother Similarly, girls who reported that they rarely talked to their parents about their plans and activities expressed greater willing- ness to have a child outside marriage than those who reported that they talked to they parents often. And girts who reported that their parents were less likely to mo~tor md keep track of their activities were also at greater risk of unmarried motherhood, although this pattem was found to be much stronger for whites than for blacks (Abrahamse et al., 19851. Poverty and Unemployment As previously discussed, bleak social and economic prospects for many black girls from low-income families may be associated with their early

OCR for page 95
DETERMINANTS OF SEXUAL BEHAVIOR 119 initiation of sexual activity and lack of effective contraceptive practice. Poverty and poor employment opportunities are closely associated with nonmantal childbearing (Presser, 1974; Ross and Sawhili, 197~; Fur- stenberg, 1976~. High rates of youth unemployment and a lack of economic resources, especially among black teenagers, frequently make marriage unmanageable for an adolescent couple, despite the impending birth of a baby. Availability of Financial Assistance Concern over the high rates of welfare dependence in the United States have led many critics to question whether the availability of Aid to Families With Dependent Children (AFDC) and other, noncash bene- fits is an unfortunate incentive for young women to give birth outside marriage. The existing body of research suggests that there is no en- dence to support this assumption, although the relationship between welfare and unmamed adolescent childbearing is complex (Ross and SawhiD, 1975; Presser, 1974; Moore and Caldwe3D, 1977; Ellwood and Bane, 1984~. Presser (1974) found that there were no significant differ- ences in fertility attitudes or behaviors among the welfare recipients and noureapients in her New York City study. Similarly, Furstenberg (1976) found that unmarried teenage girls do not get pregnant in order to receive public assistance, but that girls from low-income, female-headed families (many of whom are receiving welfare benefits) are more likely to become single mothers themselves. Moore and CaldweD (1977) suggest that because welfare assistance is available, a young woman faced with a premarital pregnancy may be more likely to choose single parenthood over abortion or adoption or maniage, especially if the father is a poor prospect for support. They found little empirical evidence, however, that welfare benefit levels affect decisions to become sexually active, to become pregnant, or to marry or have an abortion, or to relinquish a child for adoption. As they note, the vast majority of adolescent preg- naDcies are untended, and welfare is only one of a number of factors that influence teenagers' decisions regarding pregnancy resolution. Fi- nary, EDwood and Bane (1984) conclude that largely unmeasurable differences in culture, attitudes, and expectations, rather than differ- ences In levels of welfare support, explain differences ~ birth rates to unmarried teenagers across the country.

OCR for page 95
120 ADOLESCENT SEXUALITY PREGNA.~CY AND CHILDBEARING CONCLUSION As this chapter has described, a variety of individual, family, and social factors are associated with sexual behavior and decision making. Some of these factors directly affect decisions to initiate sexual activity, to contra- cept, to abort, to marry, or to have a child while unmarried. Others affect decisions indirectly by influencing other relevant factors. Among the most important factors are adolescents' attitudes about sexual behavior, contraception, abortion, mamage, and single parent- hood. Attitudes are inevitably tied to the specific social, economic, and cultural circumstances of a person's life, as weD as to a person's overall development as a masculine or feminine human being. Attitudes are relatedin complex ways to the development of aspirations, interests, and abilities, the capability to form intimate interpersonal relationships, and the transition from dependence on families of origin to independence, marnage, and parenthood. Several studies of social and psychological factors associated with adolescents' sexual behavior conclude that self-perception (not self- esteem)-that is, the sense of what and who one is, can be, and wants to be is at the heart of teenagers' sexual decision making. The perception (rather than the reality) of peer attitudes and behaviors also appears to be central and, as McAnarney and Schreider (1984) suggest, applies to both boys and girls. It is what governs one's internal response to external influences and events, and it is the basis for assessing the risks and consequences of sexual behavior. We have seen that an important aspect of self-perception among teenagers is their educational, occupational, and family formation expec- tations Expectations, in turn, are significantly influenced by percep- tions of opportunities, regardless of whether these perceptions reflect reality. Teenagers, especially girls, with a strong achievement onenta- tion and clear future goals are less likely to become sexually involved at an early age, more likely to be regular and effective contraceptors if they are sexually active, and less likely to bear a child if they experience an unintended pregnancy. In contrast, girls who lack a strong achievement orientation and who have low educational expectations are more likely to become sexually involved at a young age, to be less regular and effective contraceptors, and to carry an unintended pregnancy to term. These findings suggest that for adolescents with clearly formulated expectations and high aspirations, their perceptions of the risks of preg

OCR for page 95
DETERMI^~A~TS OF SEXUAL BEHAVIOR 121 nancy, when measured against their perceptions of Inure potential, are quite high. Many other teenagers, however, do not perceive the risks as great enough to deter sexual activity without contraception. They are the ones at highest risk of pregnancy and chil~lbeanng. Research underscores the variety of family background characteris- tics, psychological factors, and environmental conditions that influence teenagers' self-perceptions and, in turn, influence their perceptions of the risks of pregnancy and childbearing. Race, socioeconomic status, family structure, family size, and parents' education are strongly associ- ate] with attitudes about sexual and fertility behavior. Yet, as several researchers point out, not all adolescent girls from poor black inner-c~ty backgrounds or rural white poverty an] not all girls from single-parent households or from large families are at higher risk of early pregnancy and childbearing (Furstenberg and Brooks-Gunn, 1985b; McAnarney and Schre:der, 1984) What makes the difference? It remains for Inure research to answer this essential question about the factors affecting sexual decision making among adolescents and the mechanisms by which they work.

OCR for page 95