Table of Contents | Previous | Next |
Chapter Two: Describing Adolescents' Beliefs, Attitudes, and Behaviors with Respect to Romantic Relationships
The legislation that reauthorized the TANF program allocates money for eight different activities to promote healthy marriages (Deficit Reduction Act, 2006). Of these, two activities directly affect adolescents: public service campaigns to promote the value of marriage, and high school curricula that teach the value of marriage and healthy relationship skills. The decision to specify these activities presumably rests on an implicit theory about the needs of adolescents. Specifically, these activities assume that the beliefs, attitudes, and behaviors of adolescents need changing and that potentially dysfunctional patterns among adolescents need to be fixed. Is this true? How much do we know about what adolescent romantic relationships are like? What do adolescents believe about marriage, divorce, and cohabitation and what do they value about their own current and future relationships? How do adolescents’ beliefs and values translate into behaviors with respect to romantic relationships?
Programs that target adolescent romantic relationships are likely to be most effective to the extent that they are informed by accurate answers to these questions. Thus, the goal of this chapter is to assemble and review the existing research that describes adolescent romantic relationships. In particular, the research reviewed in this section examines what adolescents believe is true about romantic relationships, what they value in their own relationships and what they hope for in their future relationships, and how they actually conduct their relationships during adolescence. Unless otherwise noted, all between-group differences described here are statistically significant.
Data Sources
Accurately and comprehensively describing current patterns and trends in adolescent relationships is challenging for several reasons. First, as noted in Chapter One, there simply has not been much research that directly addresses issues related to adolescent romantic relationships. Second, results from large-scale, nationally representative surveys of youth have focused primarily on sexual behavior and other health risk behaviors, often to the exclusion of the romantic relationships in which those behaviors take place. Third, studies that provide a broader window into adolescent romantic relationships, including research that links adolescent dating to psychological and behavioral outcomes, tend to have limited generalizability because of small sample sizes or unique study settings (Carver, Joyner, and Udry, 2003).
For nationally representative data on romantic relationships among adolescents and young adults, research to date has relied almost exclusively on just five studies: the National Longitudinal Study of Adolescent Health (Add Health), the Youth Risk Behavior Survey (YRBS), the National Survey of Family Growth (NSFG), the National Survey of Adolescent Males (NSAM), and the National Longitudinal Survey of Youth (NLSY). Each of these surveys covers a range of topics, with some covering a wide range of health risk and health-related behaviors, and others focusing somewhat more narrowly on sexual behavior and romantic relationships. All sampled through multistage, stratified, clustered sampling, usually including over-samples of certain underrepresented racial or ethnic groups. Because these five studies provide the basis for so much of what is known about adolescent relationships, each study is worth describing in more detail.
National Longitudinal Study of Adolescent Health (Add Health)
Designed primarily to examine the precursors of health-related behaviors among adolescents, Add Health has collected three waves of data to date (Harris et al., 1997). Wave I assembled a school-based sample of youth in grades 7–12 during the 1994–1995 school year. Wave II assessed the same respondents one year later, and Wave III contacted them approximately seven years later, when respondents were in young adulthood. The survey was conducted through written self-administered questionnaires, and data on sexual behavior specifically were collected through in-home interviews using audio-enhanced, computer-assisted, self-interviewing methods. With respect to understanding adolescent behaviors in romantic relationships, a unique feature of Add Health is that respondents were asked to identify up to three romantic relationships that they had experienced in the previous 18 months and to answer questions about those relationships. The survey also includes comprehensive information about sexual activity and sexual risk taking, although it includes very few items assessing relationship attitudes. A single chapter describing the relationship data from Add Health (Carver, Joyner, and Udry, 2003) is currently one of the most frequently cited sources of data on adolescent experiences in romantic relationships.
Youth Risk Behavior Survey (YRBS)
The YRBS refers to a national survey conducted by the Centers for Disease Control and Prevention (CDC) involving a school-based nationally representative sample of adolescents in grades 9–12. The YRBS has been conducted every two years, beginning in 1991, and data collection occurs through written self-administered questionnaires. The YRBSS (note the additional S) refers to the national school-based survey plus state and local school-based surveys conducted by state and local education and health agencies. The YRBS monitors six categories of priority health-risk behaviors among youth and young adults: (1) behaviors that contribute to unintentional injuries and violence; (2) tobacco use; (3) alcohol and other drug use; (4) sexual behaviors that contribute to unintended pregnancy and sexually transmitted diseases (STDs); including human immunodeficiency virus (HIV) infections; (5) unhealthy dietary behaviors; and (6) physical inactivity. In addition, the YRBSS monitors general health status and the prevalence of obesity and asthma. Our report primarily describes data from the national YRBS surveys (e.g., Garofalo et al., 1998; Garofalo et al., 1999; Silverman et al., 2001).
National Survey of Family Growth (NSFG)
Designed primarily to assess fertility and fertility-related behaviors, the NSFG collects data through in-person interviews and self-administered questionnaires of people ages 15–44 living in U.S. households. The survey has been conducted multiple times, on a different cohort of respondents each time. Cycles 1–5 were conducted in 1973, 1976, 1982, 1988, and 1995. They included only women and focused on marriage, divorce, contraception, infertility, and the health of women and infants in the United States. Cycle 6 was conducted in 2002 and included both men and women. Of the five national data sets emphasized here, the NSFG is the only one that does not limit its sample to teenagers.
National Survey of Adolescent Males (NSAM)
Designed as a counterpart to the NSFG, the NSAM includes only male teenagers ages 15–21 from nationally representative U.S. households. There are two cohorts within the NSAM. The original, longitudinal cohort was surveyed in 1988 (males ages 15–19), 1990–1991 (males ages 16–21), and 1995 (males ages 21–27). The new cohort was surveyed in 1995 and included males ages 15–19. The study was conducted through face-to-face interviews and self- administered questionnaires.
National Longitudinal Survey of Youth (NLSY)
The NLSY is one of the oldest survey initiatives. The survey was first administered in 1979 (NLSY79) and included a nationally representative sample of 12,686 young men and women who were 14–22 years old when they were first surveyed in 1979. These individuals were interviewed annually through 1994 and are currently interviewed on a biennial basis. NLSY97 was launched in 1997 and contains a nationally representative sample of about 9,000 youths who were 12 to 16 years old as of December 31, 1996. Data collection has been conducted via interviews and questionnaires. The NLSY surveys are focused more on educational and workforce-related issues, but they do contain some data on adolescent relationships and romantic and sexual behavior.
In addition to these five data sets, other data on specific aspects of adolescent relationships are available. For example, Monitoring the Future (MTF), a survey focusing on drug use that includes high school student respondents from across the country, asks about marital expectations and values (Johnston et al., 1995). The National Survey of Families and Households (NSFH) measures marital attitudes and beliefs, focusing on the views of young adults (ages 19–34). We include some references to the NSFH where no data on younger individuals are available. Data on birth statistics come from the National Center for Health Statistics, based on birth certificate data for every child born in the United States. Abortion rates are from the CDC’s National Center for Chronic Disease Prevention and Health Promotion, primarily based on data on women having abortions from most state health departments. Abortion data are also available from the Alan Guttmacher Institute’s annual survey of abortion providers. Teenage marriage and divorce rates are based on samples of records of states that participate in the marriage-registration area (MRA) and divorce-registration area (DRA).
Limitations of the Available Data
The data sources described above provide most of the information reviewed in the rest of this chapter. Although they are nationally representative, these studies also tend to be limited in several ways. First, the data in the surveys are generated entirely by the self-reports of the respondents. As has been well documented (Singer and Presser, 1989), self-reports, especially about sensitive topics like romantic relationships, are highly vulnerable to memory biases and distortions. For example, partners are often poor at recalling their behaviors in romantic relationships (e.g., Jacobson and Moore, 1981) and rely instead on their general feelings about the relationship to guess what their behaviors were likely to have been (e.g., Weiss, 1984). Respondents may also distort their responses deliberately to create a favorable impression for the researcher, e.g., by neglecting to admit to illegal or undesirable attitudes and behaviors. Because self-reports are susceptible to these biases, data generated by self-reports are likely to underestimate the prevalence of negative behaviors and attitudes.
Second, most of the national data sets include only a small set of items relevant for understanding intimate relationships. Thus, only a narrow range of attitudes and behaviors have been addressed, even within the studies that have examined these issues directly.
Third, for surveys with school-based samples, results may not generalize to out-of-school youth. This is a particularly noteworthy limitation when considering low-income youth. These youth may be at greater risk for engaging in high-risk behaviors and their potentially negative consequences, and yet may be most likely to be underrepresented in surveys drawing upon school-based samples.
Fourth, although some of these surveys are longitudinal (i.e., the same individuals provided data at several assessments over time), analyses of most of these data have tended to be cross-sectional (i.e., examining associations among variables assessed at a single assessment). Cross-sectional analyses can estimate associations among variables but do not support conclusions about causal relationships between variables.
Fifth, because the questions asked and the samples addressed vary across studies, the existing literature offers little basis for comparing results across different studies. Because most of this research has been cross-sectional, this means that there are scant data that allow descriptions of trends in adolescent romantic relationships over time, and all inferences about trends over time should be considered tentative.
Sixth, although analyses of data from these surveys usually contain breakdowns by race and ethnicity, they very rarely contain breakdowns by household income or by variables—such as poverty status, neighborhood quality, or urban versus rural—that may serve as proxies for household income. Because race and ethnicity in themselves are poor proxies for income (e.g., most blacks and Hispanics are not low-income; see Bachrach, 1998), this limits what existing analyses of these data sets can say about the population currently being targeted by healthy marriage initiatives.
The studies that have addressed youth relationships in low-income or urban populations have tended to be far smaller in scale (e.g., Connolly et al., 2004; Hogan, Astone, and Kitagawa, 1985; Hogan and Kitagawa, 1985; Miller, Forehand, and Kotchick, 2000; O’Donnell, O’Donnell, and Stueve, 2001; Silver and Bauman, 2006; VanOss Marín et al., 2006; Whitaker, Miller, and Clark, 2000). Many of the studies are qualitative and thus provide a richer picture of youth perceptions than might be possible with prevalence estimates and average responses to quantitative measures. Yet, whereas small-scale studies provide valuable insight into adolescent subgroups of particular interest to the current project, they suffer from their own limitations. First, they also tend to be cross-sectional, limiting their ability to support causal inferences. Second, they are also based primarily on self-report data. Third, the studies’ smaller sample sizes limit their statistical power to identify significant relationships among variables. Finally, the characteristics of the respondents and the settings in which the studies take place impose limits on the generalizability of the study results, even when comparing across studies involving similar subgroups of adolescents. Despite these limitations, we describe results from several small-scale studies, especially when these studies address gaps in the results from national surveys.
Adolescents’ Attitudes and Beliefs Regarding Dating and Romance
The fact that adolescents’ experiences in romantic relationships are significantly associated with such outcomes as violence and depression (e.g., Capaldi and Gorman-Smith, 2003; Monroe et al., 1999; Wekerle and Avgoustis, 2003) suggests that these relationships play an important role in adolescents’ lives. Small-scale studies suggest that even young children are aware of (Connolly et al., 1999) and generally preoccupied with (Simon, Elder, and Evans, 1992) romantic issues. By age 9, children express clearly distinct conceptions of romantic relationships versus opposite-sex friendships, with their descriptions of romance uniquely characterized by longing, physical attraction, and commitment (Connolly et al., 1999). Opposite-sex friendships, in contrast, are perceived to be less intimate (i.e., involving less self-disclosure and less closeness) than same-sex friendships (Lempers and Clark-Lempers, 1993; O’Donnell, 1979), suggesting a continuum of intimacy from opposite-sex friendships at the low end, to romantic relationships at the high end.
In general, adolescents value romantic relationships highly and seek them out for themselves. One study of sixth through eighth graders observed high levels of interest in having a boyfriend or girlfriend across all grade levels, indicating that interest in a romantic relationship is already established by grade 6 (Darling et al., 1999). However, in a large ethnically diverse sample of Canadian youth in grades 5 through 8, interest in having a boyfriend or girlfriend in the near future was higher for those in grades 7 and 8 than for those in grades 5 and 6 (Connolly et al., 2004). The difference in findings may have to do with differences in the ethnic makeup or dating experience of the populations represented by these two samples. In the Canadian study, it was also shown that interest in having a boy- or girlfriend was greater among youth from European or Caribbean backgrounds than among Asian-Canadians and among youth who had recent experience with dating. Interest was not higher among youth who had participated in mixed-gender group activities but had not been out on a date, and there were no differences in romantic interest among boys versus girls (Connolly et al., 2004). Another factor that may play a role in dating interest, and that is closely linked to age, is pubertal development. Youth become more interested in romantic relationships during puberty (Harris et al., 1997; Miller and Benson, 1999).
Beliefs about the qualities that are important for successful romantic relationships evolve as youth move from middle school to high school and on to college, and this evolution in part reflects increasing clarity about the concept of “boy(girl) friend” across these years. In grade 6, youth are more likely to ask what “counts” as a date or as a girlfriend or boyfriend; but by grade 8, youth understand questions about romantic relationships immediately (Darling et al., 1999). Descriptions of romantic relationships at age 15 commonly include self-disclosure and support, in addition to physical attraction and commitment (Feiring, 1999b). Physical attraction is most likely to be part of boys’ descriptions, whereas self-disclosure and support are more prevalent in girls’ descriptions (Feiring, 1996, 1999a). Older teenagers’ conceptions of what it means to have a boyfriend or girlfriend also include both self-disclosure and companionship but are less likely to include physical attractiveness or passion (Connolly and Johnson, 1996; Furman and Buhrmester, 1992; Furman and Wehner, 1994). When asked to supply reasons for dating particular individuals, middle school youth place greater emphasis on superficial characteristics of potential partners, whereas college students focus more on partners’ future plans and mutuality of feelings (Galotti, Kozberg, and Appleman, 1990; Roscoe, Diana, and Brooks, 1987). In general, Connolly and Goldberg (1999) argue that younger adolescents’ romantic relationships are primarily characterized by physical attraction (e.g., infatuations or “crushes” predominate), and that this emphasis becomes moderated in the later teen years as teens grow to value intimacy and commitment. Waldinger and colleagues (2002), in research on ideas about relationships generally (not just romantic relationships) found that the same themes appear in the narratives of youth at ages 14 to 16 as at age 25, but that the young adults’ narratives were more complex than the adolescents’ narratives, possibly as a result of more experience with relationships. We speculate that, as youth achieve more independence from their parents, the future implications of a romantic relationship become more salient, accounting for the some of the change in relationship beliefs and values as they get older.
When describing dating and romance, young teens in a non-random sample of high-school-age youth or younger focused on positive aspects of relationships (Feiring, 1996). During early adolescence, a majority of females believe that romantic relationships should be important but not everything in life, that romance should be heterosexual, and that a girl should not have feelings for a boy who is already attached to another girl (Simon, Elder, and Evans, 1992). By age 18, in contrast, descriptions of romantic relationships frequently describe jealousy and concerns about compatibility as part of romantic relationships, with girls more likely than boys to mention compatibility, reciprocity, and jealousy in their descriptions (Feiring, 1999a). Qualitative research has identified jealousy and infidelity as especially salient themes in African-American adolescents’ descriptions of committed sexual relationships (Eyre et al., 1998).
Several studies have directly examined adolescents’ beliefs about the role of sexual behavior within various types of relationships. As noted earlier, romantic relationships are the primary contexts for sexual behavior among adolescents (Collins and Sroufe, 1999; Sprecher, Barbee, and Schwartz, 1995). Consistent with this observation, within a multiethnic sample of 452 18- to 25-year-olds, initiation of new sexual activities is considered more appropriate when it occurs within a serious romantic relationship (Feldman, Turner, and Araujo, 1999). According to a national survey of 12- to 19-year-olds, approximately two-thirds “strongly agree” that sex should occur only within such a relationship (Albert, 2004). Indeed, the majority of adolescents have their first sexual intercourse with someone they are going steady with or know well and like a lot (Abma et al., 2004). Consistent with this fact, the modal reason reported for first intercourse is “to have the partner love them more” (Rodgers, 1996). Although adolescents are most accepting of sexual behavior within the context of a romantic relationship, it is worth noting that the majority of respondents also endorse the view that intercourse is not appropriate for high school teens, with rates of disapproval at 63 percent among boys, and 76 percent among girls (Albert, 2004). Sexual activity may be less strongly linked to relationship commitment and emotional intimacy among gay, lesbian, or bisexual youth (Herdt and Boxer, 1993), for whom a romance or dating may be too public while sex is more private (Diamond, Savin-Williams, and Dubé, 1999). These youth, especially lesbians, may also develop passionate, asexual same-sex friendships (Diamond, Savin-Williams, and Dubé, 1999).
Adolescents’ Expectations and Attitudes About Marriage, Cohabitation, and Divorce
A decade ago, most high school students in the MTF study expected to marry, and reported that having a happy marriage was important to them (Johnston et al., 1995). In these data, 72 percent of boys and 82 percent of girls expected to get married at some point. According to a recent analysis of the 2002 NSFG (Flanigan, Huffman, and Smith, 2005), most teens continue to feel favorably about marriage, although there are differences in the strength of support for marriage across genders and racial and ethnic groups. With respect to gender differences, among boys ages 15–19, a strong majority (69 percent) agree that “it is better for a person to get married than to go through life being single.” Among girls ages 15–19, a majority (54 percent) feel this way, but many girls (42 percent) disagree.
According to the 2002 NSFG data, girls’ attitudes toward marriage do not differ by race/ ethnicity, but boys’ attitudes do (Flanigan, Huffman, and Smith, 2005). Although a majority across all ethnic groups favor marriage, endorsement of marriage over remaining single ranges from 60 percent among black males to 75 percent among Hispanic males, with white males falling in between. Closer examination of these patterns reveals that boys have more favorable marriage attitudes than girls among Hispanics and whites, but there are no similar gender differences among black teens—black teens of both genders are less favorable toward marriage than white teens. It is not clear whether or not these differences persist into adulthood. For example, among the 19- to 34-year-old respondents in the NSFH, blacks, Hispanics, and whites were all equally likely to say that they planned to marry (Brown, 2000; Bulcroft and Bulcroft, 1993; Bumpass, Sweet, and Cherlin, 1991). The differences between the studies may be the result of developmental changes in attitudes toward marriage, or they may result from the fact that the NSFG measured attitudes toward marriage whereas the NSFH tapped plans to marry. Although adolescent attitudes toward marriage differ by gender and race/ethnicity, they do not appear to differ by age. For example, attitudes toward marriage are similarly positive among younger (ages 15–17) and older (ages 18–19) teens (Flanigan, Huffman, and Smith, 2005).
Of particular interest to policymakers seeking to promote marriage are the marital attitudes of young unwed parents. Four percent of girls ages 15–19 gave birth in 2004, and most teens who give birth are unmarried (Hamilton et al., 2005). Studies focusing on unwed teen parents suggest that they may view marriage differently from other youth, if not less favorably. For example, two small studies of teenage mothers indicate that they believe men are unreliable and likely to be unfaithful, suggesting that their expectations for a stable marriage are low (Furstenberg, 1995; Maynard, 1995). Although many young mothers are in a relationship with the father of their child (Furstenberg, Brooks-Gunn, and Morgan, 1987), in one sample of more than 100 low-income black teen mothers and their partners, most of the respondents nonetheless believed they would not get married (Burton, 1995). These young mothers believed that the fathers of their children would be unable to support the child, suggesting that they remained unmarried in part for this reason. An ethnographic study of 292 low-income single mothers also identified mistrust for men as a salient obstacle to marriage for this subgroup (Edin, 2000). Perhaps as a result, one study of a population with high rates of teen pregnancy—black low-income girls—found that fertility and marriage are only loosely linked in the minds of this group (East, 1998).
Just as adolescents generally feel positively toward marriage and expect to get married themselves, they generally disapprove of divorce. Just under half of teens from the 2002 NSFG sample (48 percent of girls and 42 percent of boys) agree “divorce is usually the best solution when a couple can’t seem to work out their marriage problems.” There are no substantial differences in these attitudes based on gender, sexual experience, or age. As with attitudes toward marriage, however, there are differences by race/ethnicity among boys. Hispanic (51 percent) and black (53 percent) males are more likely then white males (38 percent) to endorse divorce in the face of marital problems (Flanigan, Huffman, and Smith, 2005). It is worth noting that, even though youth generally do not approve of divorce, many nevertheless expect to experience divorce themselves. In the MTF, where 72 percent of boys and 82 percent of girls expected to get married, only 59 percent of respondents believed that they would stay married (Johnston et al., 1995). This finding points out the important differences between adolescents’ attitudes toward marriage and relationships in the abstract, and their potentially quite different expectations for their own experiences (Cherlin, 1988).
Most young people are accepting of cohabitation, according to the same NSFG analysis. Only about a third of boys (32 percent) and girls (36 percent) agree, “A young couple should not live together unless they are married.” There are no differences by race/ethnicity among boys, but white (34 percent) girls are less likely to disapprove of cohabitation than Hispanic (40 percent) or black (39 percent) girls. Younger boys are more likely to disapprove of cohabitation than older boys, but there are no age differences among girls (Flanigan, Huffman, and Smith, 2005).
As noted earlier, only limited data suggest how adolescents’ attitudes toward marriage, divorce, and cohabitation have changed over time. The best of these data come from the MTF survey, which has been conducted annually for three decades (Schulenberg et al., 1995). Data from the years 1976 to 1992 indicate that adolescents’ attitudes toward marriage became slightly less positive during that time, with fewer indicating that they agree or mostly agree that people have fuller and happier lives if they marry. During the same period, however, the percentage indicating that it is extremely or quite important to them to have a good marriage and family life remained stable, again pointing out the different results obtained when questions are asked about relationships in the abstract or about the individual’s own experiences. In 1992, more youth agreed or mostly agreed that it was a good idea for a couple to live together before marriage than they had in 1976—this increase was relatively dramatic, a shift from 40 percent to 54 percent. As the authors note, this constitutes a change toward normative acceptance of cohabitation. During the same period, adolescents’ predictions about their age at marriage increased by approximately four years, especially among girls. This may reflect the increasing importance of college education for women during the time frame covered by the assessments. Overall, these data suggest shifts in how adolescents understand the normative path toward marriage but do not indicate changes in their expectations of marrying at some time in the future.
Correlates of Adolescents’ Attitudes and Beliefs Regarding Romantic Relationships
The average trends described thus far in this chapter mask considerable diversity in the attitudes of adolescents toward intimate relationships and marriage. Research in this area is only beginning to examine variables that may account for this diversity. This section reviews several variables that have been examined in more than one study.
Sexual Experience
One survey of first-year college students (approximately age 19) indicates that virgins feel more favorable toward marriage than those with multiple sex partners (those with one partner fell in between; see Salts et al., 1994). However, the more recent and generalizable NSFG data do not fully corroborate this finding. Among female respondents in the 2002 NSFG, there were no differences in marriage attitudes depending on sexual experience; among boys the difference fell just short of statistical significance (71.5 percent of virgin males were pro-marriage compared to 66 percent of nonvirgins; see Flanigan, Huffman, and Smith, 2005). An association between marital attitudes and adolescent sexual activity could underlie the association between teen childbearing and attitudes toward marriage, to the extent that teens who are more sexually active are also more likely to become pregnant. However, there are no published studies providing evidence of such a link.
Parents and Peers
Adolescents may develop their ideas about romantic relationships through their observation of and interaction with parents and peers (Rose and Frieze, 1989). In support of this idea, Simon and colleagues (Simon, Elder, and Evans, 1992) observed that adolescents’ normative beliefs about how one should feel and behave in a romantic relationship develop through repeated discussions with their peers. Other studies also suggest that peers reinforce the importance of romantic relationships, especially among girls (Douvan and Adelson, 1966).
Media
Adolescents are major consumers of mass media, on average spending about eight hours each day with television, music, movies, video games, magazines, and the Internet (Roberts, Foehr, and Rideout, 2005). Connolly and Goldberg (1999) speculate that the media influence adolescents’ attitudes and beliefs about romantic relationships. Most research that has examined media effects on youth has focused on television, since it is the medium that adolescents spend the most time using (Roberts, Foehr, and Rideout, 2005).
Television often portrays dating relationships as adversarial. These depictions may affect young viewers’ beliefs about what is typical in relations between real men and women (Bandura, 1994; Gerbner et al., 1994). Indeed, Ward (1995) has argued that television distorts adolescents’ beliefs about intimate relationships by promoting a view of dating in which dating is a competition—women are objects that men aggressively strive to obtain, and both genders play a strategic and manipulative game to obtain the partner they desire. In various laboratory studies with high school students from diverse ethnic and racial backgrounds, Ward has found support for this view (e.g., Ward and Friedman, 2006; Ward, Hansbrough, and Walker, 2005). In a recent study of undergraduate women, more frequent viewing of reality dating television programs was correlated with the belief that dating is a game, and that men and women are adversaries (Zubriggen and Moregan, 2006). Taking the notion of conflictual relationships a step further, Johnson et al. (1995) described an experimental study showing that black adolescent girls exposed to rap videos expressed greater acceptance of teen dating violence than those not exposed.
A notable exception to television’s portrayal of intimate relationships as adversarial is the depiction of married couples. For example, few television programs portray serious conflict between spouses (Heintz, 1992), and when conflict is depicted, it is typically resolved easily and with appropriate communication (Comstock and Strzyzewski, 1990). This may promote positive views of marital relationships among adolescents. In fact, television viewing is positively associated with high school students’ reports that they want to get married, stay married to the same person for life, and have children (Signorielli, 1982). However, the same study found that television viewing was correlated with the opinion that people see so few good or happy marriages in the real world that they question marriage as a way of life. The latter finding is consistent with the concern expressed by some theorists (e.g., Cantor, 1990) that exposure to media portraying marriage in an overly positive light might lead to idealized beliefs about what marriage and relationships entail, leaving youth unprepared for the hard work of relationship maintenance. It may also lead them to judge the real-life marriages that they encounter more harshly.
Television may also influence adolescents’ beliefs about the role of sex in intimate relationships. Sexual behavior is common in a variety of television programming, and when it is depicted, it is typically between unmarried individuals (Greenberg and Busselle, 1996) and often involves individuals who have little or no prior relationship (Kunkel et al., 2005). Exposure to television portrayals of nonmarital sex has been shown in laboratory studies to produce greater acceptance of sexual indiscretions among teen viewers (Greenberg, 1994). Correlational research suggests that regular exposure to sex on TV may also lead to teens’ dissatisfaction with their own relationships, particularly the sexual aspects of them (Greenberg, 1994). Similar work also indicates that exposure to sexual content on television and in other media may lead to initiation of intercourse earlier in adolescence (Brown et al., 2006; Collins et al., 2004; Martino et al., 2006).
Of course, sexual content is most prevalent in pornographic material, which is not typically shown on television but is now readily available in many homes via the Internet. Rates of exposure to pornography through magazines, films, and the Internet range from 48 percent to 92 percent among teens, depending on the medium and the sample studied (Bryant, 1985; Malamuth and Impett, 2001). Zillman (1994) argues that the values instilled by pornography are antithetical to those necessary to forming and maintaining traditional families, undermining commitment and moral behavior. Kubey and Donovan (2001) suggest that spouses and monogamy can be devalued as a result of exposure. There has been very little empirical research examining such effects, and nothing addressing young viewers. Studies of young adults, however, provide preliminary support for their contentions. Men who regularly view pornography are more dissatisfied with their spouses (Zillman and Weaver, 1989) and see their girlfriends as less attractive (Weaver, Masland, and Zillman, 1984). Men and women exposed to pornography in an experimental setting report less interest in having children than do controls (Zillman, 1994). To the extent that adolescents are still developing their conceptions of romantic relationships, the effects of exposure to pornography may be stronger and longer lasting for adolescents, although no experimental research has addressed this possibility directly.
Adolescents’ Behaviors in Romantic Relationships
Do adolescents’ beliefs and attitudes regarding romantic relationships correspond to how they actually conduct their relationships? This section describes current trends in adolescent relationship behaviors and discusses changes over time if such data are available. Included in this review is research on timing of entry into romantic and sexual relationships, the number and duration of relationships, degree of involvement between partners, characteristics of partners (i.e., observed preferences in partner selection), the use of contraception, and dating violence.
Through the course of adolescence, adolescents’ peer relationships generally progress from same-sex friendships, to mixed-gender friendships, to romantic and sexual relationships with the opposite sex. Dating activities over this stage of life tend to progress from outings in mixed-sex groups, to opposite-sex friendships, to dating (Feiring, 1996).
Data from several studies support this progression. For example, among children ages 11–12 in the NLSY79, the vast majority (92 percent) had never or rarely dated; if they had, they had done so sporadically. In contrast, fewer adolescents ages 13–14 and 15–16 reported never or almost never dating (71 percent and 46 percent, respectively) (Cooksey, Mott, and Neubauer, 2002). In a noteworthy racial difference, blacks ages 13–14 and 15–16 in this sample were more likely than their nonblack peers to report never or almost never having dated (79 percent versus 69 percent and 59 percent versus 43 percent, respectively); (Cooksey, Mott, and Neubauer, 2002). Using data from the Add Health study, Carver, Joyner, and Udry (2003) found a similar pattern, showing that, in the 18 months prior to the survey, 26 percent of 12-year-olds had been in a romantic relationship, whereas 68 percent of 18-year-olds had been in a romantic relationship. A Canadian study paints the same picture, noting that, once in a romantic relationship, adolescents were less likely to return to a previous stage of dating or affiliative relationships (Connolly et al., 2004).
One reason that the age at which adolescents enter into romantic relationships is important is that the timing of initiation of dating is closely associated with sexual initiation. For example, in the aforementioned study of the teenagers of the NLSY79 sample, 50 percent of those who reported having sex for the first time between ages 11–12 and 15–16 indicated that they had begun dating first. This general tendency masks significant variance by gender and race; females and blacks were less likely than males and nonblacks to have initiated dating prior to initiating sex (Cooksey, Mott, and Neubauer, 2002).
In sum, nationally representative samples indicate that most people experience their first romantic relationships during adolescence and these romantic relationships usually precede adolescents’ first sexual experiences. The specific timing of entry into these experiences is associated with gender and race/ethnicity. Although the data exist to compare patterns in low-income and upper-income groups, such comparisons have not yet been conducted.
Duration of Adolescent Romantic Relationships
Reporting on the timing of when adolescents typically enter adolescent romantic relationships is challenging because adolescents do not always know exactly when their relationships began, particularly if partners were friends before they were romantic partners. Younger boys and black adolescent males have the most difficulty identifying when their relationship began, and older girls and white adolescent females have the least difficulty (Carver, Joyner, and Udry, 2003). Without reliable reports of when relationships begin, estimating the duration of adolescent relationships is also a challenge, even though adolescents are very aware of the date that their relationships end because breakups are usually associated with a specific event.
Estimating the duration of adolescent romantic relationships is further complicated by the methods used to study the relationships. For example, the Add Health study asked adolescents who had been in a relationship in the last 18 months to provide information about up to three of their romantic relationships. Data analyses only included data on the relationship that was listed first, which was likely to be the most important and/or the longest relationship, thereby underreporting the frequency and duration of short-term relationships (Carver, Joyner, and Udry, 2003). Keeping this bias in mind, the Add Health study observed a median duration of adolescent romantic relationships of 16 months for girls and 12 months for boys. Older adolescents in the Add Health study reported having longer relationships. The median duration was 5 months for adolescents under age 14, 8 months for adolescents between 15 and 16, and 21 months for adolescents over age 16. Substantial differences in median relationship duration were observed across racial and ethnic groups—black adolescents reported a median duration of more than 24 months (Giordano, Manning, and Longmore, 2005), whereas white, Hispanic, and Asian adolescents reported median durations of 12 months, 15 months, and 11 months, respectively.
In sum, whereas data on the duration of adolescent relationships remain incomplete, the data that exist suggest that the average romantic relationship during adolescence is not transient but rather lasts up to a year or more. It is not clear whether the same patterns would be observed within low-income populations specifically.
Degree of Involvement Between Partners
Involvement in romantic relationships during adolescence can be defined in terms of how adolescents describe their own relationships and feelings about their partners, as well as the type, amount, and frequency of activities or behaviors that occur within the relationship.
The Add Health study is a valuable source of information about these issues. To assess the level of social connectedness in adolescent romantic relationships, the Add Health study asked adolescents a variety of questions about one of their recent relationships. Within the Add Health Wave II data (Carver, Joyner, and Udry, 2003), 81 percent of adolescents in romantic relationships considered themselves to be a “couple,” 67 percent went out together with no one else present, 68 percent told their partners that they loved them, 62 percent gave each other gifts, and 49 percent spent less time with their friends in order to spend more time with their partner. Overall, boys and girls tended to provide similar reports of spending less time with friends in order to spend time with their partner; however, girls tended more often than boys to report engaging in other acts expressing intimacy or commitment, such as thinking of themselves as a couple (86 percent versus 76 percent), giving each other gifts (66 percent versus 57 percent), saying that they love the other person (71 percent versus 75 percent) and going out together with their partner alone (71 percent versus 67 percent). Older adolescents reported engaging in these behaviors more frequently than younger adolescents, and, in separate analyses of the same data set, white adolescents report these behaviors more frequently than adolescents of other races and ethnicities (Giordano, Manning, and Longmore, 2005).
Because Add Health allows the respondent to choose a relationship to describe, it may be biased toward assessing more serious relationships. Yet the NLSY study also asked adolescents about dating habits and going steady, and found, consistent with the results from Add Heath, that as adolescents got older, they described their relationships as more involved. Among the youth in the NLSY79 sample, 27 percent of those ages 15–16 went on more than one date per week, compared with 10 percent of those ages 13–14 (Cooksey, Mott, and Neubauer, 2002). A convenience sample of 493 youth ages 12 to 24 recruited from youth organizations, public schools, and private schools also found that, compared with younger adolescents, older adolescents had more dating experiences; had been in love more times; scored higher on measures of passion, intimacy, and commitment-related romantic beliefs; and scored lower on measures of romantic idealization (Montgomery, 2005). Consistent with the results from Add Health, females scored higher than males on measures of intimacy and self-consciousness, although they reported being in love fewer times than males and were less likely to believe in love at first sight. An independent small-scale study of adolescents describes their typical dating activities as including going out to dinner and hanging out at school, the mall, or each other’s homes (Feiring et al., 2002).
Multivariate analyses of the NLSY data described above suggest that the degree of involvement between partners is associated with sexual activity among adolescents. Specifically, these data indicate that sexual activity is more closely associated with how adolescents define their relationships generally (e.g., “going steady” versus casual dating) than with the self-reported frequency of specific behaviors like going on dates (Cooksey, Mott, and Neubauer, 2002). Other studies also find that, although sexual activity is prevalent across a range of types of relationships, it most frequently occurs in relationships described as close and committed. For example, among males 15–19 years old in the 2002 NSFG who had had sexual intercourse in the past three months, 62 percent describe themselves as “going steady,” 14 percent considered themselves to be “just friends” with their female partner, 10 percent are cohabiting, 9 percent reported that they go out once in a while with the sexual partner, and 1 percent are married (Martinez et al., 2006). For females under age 20 in the 2002 NSFG study who had ever had sexual intercourse, 75 percent described themselves as “going steady,” 14 percent described their first sexual partner as someone they had “just met or were just friends with,” 7 percent went out with the partner once in a while, and 4 percent were cohabiting or married (Abma et al., 2004). The younger a female’s age at sexual initiation, the more likely she is to describe the partner as someone whom she had just met or was just friends with. Thirty-three percent of female teenagers who first had sexual intercourse at age 14 or younger fit in this category, compared with only 11 percent of female teenagers who were ages 17–19 at the time they first had sexual intercourse (Abma et al., 2004). It is worth noting that these analyses examine the proportion of sexual activity among adolescents that takes place within romantic relationships. There have been no comparable analyses of the proportion of romantic relationships that include sexual activity, an independent and relevant question.
In sum, although data on the degree of involvement between partners in adolescent romantic relationships remain sparse, the best existing data suggest that romantic relationships among adolescents are most frequently characterized by a relatively well-defined bond between the members of a couple, and the likelihood of experiencing this bond increases throughout adolescence. It is not known how the intensity of romantic relationships varies across income groups.
Number of Romantic and Sexual Relationships in Adolescence
Over the past few decades, some evidence indicates a downward trend in the prevalence of adolescent dating. Based on data from the Monitoring the Future study from 1975 to 2004, dating appears to be becoming less common. Since the early 1990s, data for students in eighth, tenth, and twelfth grade show that the percentage of students who do not date at all is increasing. From 1992 to 2004, the percentage of eighth graders who did not date at all increased from 48 to 51 percent; the increase for tenth graders was 28 to 37 percent; and for twelfth graders, the increase was from 14 to 27 percent. Using the same data, from 1991 to 2004, the percentage of twelfth graders who date frequently, defined as going on one or more dates per week, declined from 34 percent to 27 percent. In 2004, eighth grade females were more likely to report never dating than male eighth graders (57 percent compared with 46 percent). The differences between genders for having never dated shrink as they age: 38 percent of tenth grade girls and 36 percent of tenth grade boys never date and the difference between twelfth grade girls and boys was not statistically significant (Child Trends, 2006). The problem with interpreting these data is that the term dating is defined only vaguely across these studies. It is not clear how declines in dating may be associated with trends in other kinds of involvements among adolescents.
In a cross-sectional look at Add Health Wave I data, 55 percent of all adolescents in the sample (ages 12–18) experienced a romantic relationship in the 18 months preceding the study. Another 10 percent of adolescents were involved in a “liked” relationship, where the adolescents did not report that they were in a romantic relationship but reported involvement in behavior indicative of romantic relationships, such as holding hands, kissing, or telling someone they loved or liked them. Thirty-five percent of adolescents in the sample had not been involved in any romantic relationship in the past 18 months. The prevalence of involvement in romantic relationships followed a consistent increasing pattern in older ages compared to younger. With the exception of Asian adolescents, who had consistently low rates of romantic relationships, white, black, Hispanic, and Native American adolescents all had similar rates of relationship experience (Carver, Joyner, and Udry, 2003).
The NSFG study provides nationally representative estimates of the number of sexual partners adolescents have had. Although reports of sexual partners do not map perfectly onto number of relationship partners, the fact that most sexual activity among adolescents takes place within romantic relationships suggests that these estimates may be close approximations of the number of romantic relationships that adolescents experience. Indeed, consistent with the data from other studies, the 2002 NSFG data (Abma et al., 2004) indicate that 61 percent of sexually active female teens and 47 percent of sexually active male teens had only one sexual partner in the last 12 months. Sexually active male teens are more likely than their female counterparts to report having had two to three partners in the previous year (32 percent versus 22 percent, respectively), and also more likely to have had no partners in the previous year (14 percent versus 9 percent, respectively). Gender differences do not emerge for teens reporting four or more partners in the previous year (8 percent for both genders).
Some evidence suggests that low-income adolescents may have a greater than average number of sexual partners. For example, one study of 1,287 urban, minority youth living in economically disadvantaged neighborhoods showed that, by tenth grade, 54 percent of males and 18 percent of females had had four or more lifetime sexual partners, with no significant differences for ethnicity within gender (O’Donnell, O’Donnell, and Stueve, 2001). It is worth exploring whether this pattern holds true among the low-income segments of national data sets.
In sum, consistent with the finding that most romantic relationships among adolescents last more than one year, most adolescents are not experiencing romantic and sexual relationships with multiple partners. At the same time, there is substantial variability across adolescents. The fact that low-income adolescents may report higher numbers of sexual partners at earlier ages may indicate higher numbers of romantic relationships as well, but this possibility remains to be explored directly.
Characteristics of Romantic Partners in Adolescence
Research on adolescent romantic relationships tends to focus on the individual respondent, rarely asking questions about the partner. The data that do exist generally suggest that adolescents tend to be in relationships with partners who are similar to themselves in age, race, socioeconomic background, academic interests, and likelihood of engaging in risk-taking behavior.
With respect to age, although adolescents in the Add Health study did have relationships with older and younger partners, most adolescents had partners whose age was within a few years of their own (Carver, Joyner, and Udry, 2003). Adolescent boys tend to have older romantic partners until they reach late adolescence (17–18), and then they tend to select partners who are younger than themselves. Across ages and across racial and ethnic groups, more than two-thirds of adolescent boys in the Add Health study described relationships with partners who were within a year of their own age. In contrast, most female respondents in the Add Health study described relationships with partners older than themselves, and this holds true for all ages. Among females of all ages, the mean age of partners was oldest for blacks and His-panics (two years older in both groups), and youngest for whites (1.68 years older).
The Add Health data also reveal that adolescents tend to enter romantic relationships with others of the same race and ethnicity. However, the prevalence of interracial relationships appears to be negatively affected by the racial composition of the school environment (i.e., the higher the proportion of students of the same race, the fewer interracial relationships reported) (Carver, Joyner, and Udry, 2003).
Regarding other characteristics of partners, a study involving a subset of Add Health data collected within one midsize, close-knit, predominantly working-class Midwestern U.S. town showed that adolescents tended to select partners similar with respect to: socioeconomic status, grade point average, college plans, attachment to school, trouble in school, drinking behavior, IQ, and grade (Bearman, Moody, and Stovel, 2004). The same adolescents also tended to select partners similar with respect to sexual experience, suspension from school, smoking, and age, though females showed a slight preference for male partners in a higher grade (Bearman, Moody, and Stovel, 2004). Within the range of denominations available in this sample, a partner’s religious denomination did not appear to be an important aspect of partners’ characteristics.
Characteristics of adolescents’ partners are also related to the likelihood of relationships being sexual in nature. Data from Wave II of the Add Health study found that, for both genders, being friends with a partner before the relationship became romantic decreased the odds that the couple was having sexual intercourse, compared with being acquaintances prior to the romantic relationship (Kaestle and Halpern, 2005a). The likelihood of a couple in a romantic relationship being sexually active increased with relationships that were longer in duration, involved older partners, and involved partners who did not live in a family with two biological parents (Kaestle and Halpern, 2005a). A different study found that adolescents are likely to have their first sexual experiences with partners who have already been sexually active (Rodgers, 1996).
Another noteworthy aspect of partner preference is the preference for partners of the same or opposite sex. Some researchers have estimated that 10 percent of U.S. youth identify as gay, lesbian, or bisexual at some point in their lives (D’Augelli, 1988). Wave I Add Health data found that, for respondents who reported having a relationship in the last 18 months, 2 percent of boys and 4 percent of girls reported a romantic relationship with someone of the same sex (Carver, Joyner, and Udry, 2003). Although individuals in same-sex romantic relationships during adolescence may or may not end up getting married to someone of the opposite sex as adults, studies have found that, among this group of teens, romantic difficulties are significantly associated with risk for depression, substance abuse, and truancy (e.g., Anderson, 1993; Diamond, 2003; Savin-Williams, 1994).
In sum, adolescent romantic relationships are generally characterized by similarity between partners, but the degree of similarity, at least in terms of race and ethnicity, is affected by the composition of the pool of available partners.
Intimate Partner Violence in Adolescence
A final element relevant to describing romantic relationships among adolescents is the prevalence of dating violence. In Wave I of the Add Health study (conducted during 1994–1995), 12 percent of respondents had been the victim of physical dating violence and 20 percent had been the victim of psychological dating violence at some point in the previous 18 months (Halpern et al., 2001). Within the window between Wave I and II of the Add Health study, 12 percent of male adolescents and 14 percent of female adolescents reported being abused by an intimate partner (Roberts, Klein, and Fisher, 2003). In Wave II data from the Add Health study, a history of any abuse by an intimate partner was reported by 21 percent of male respondents and 22 percent of females. Specifically, 19 percent of adolescents reported that their partner had sworn at them, 13 percent reported that their partner had insulted them in front of others, 8 percent reported that their partner had pushed or shoved them, 3 percent had been threatened with violence by their partners, and 3 percent reported that their partner threw something at them. Prevalence of abusive behavior was similar for both genders. Abusive behavior appears to increase as adolescents’ age, but begins to decline at around age 18 (Carver et al., 2003).
In the YRBS studies, the prevalence of dating violence has been quite stable over time. Of boys in the 1999 YRBS study, the prevalence of physical dating violence was 9 percent over the past 12 months (Howard and Wang, 2003). Of students in the 2003 YRBS study, again 9 percent reported being victims of physical dating violence in the past 12 months (Black et al., 2006). Of students included in the 2005 YRBS study, another 9 percent reported that they had been “hit, slapped, or physically hurt on purpose by their boyfriend or girlfriend (i.e., dating violence)” at some point in the 12 months preceding the survey. Dating violence was more prevalent among black and Hispanic students compared with white students (12 percent, 10 percent, and 8 percent, respectively), and more prevalent among older students compared with younger students (11 percent of twelfth graders, 10 percent of eleventh graders, 9 percent of tenth graders, and 7 percent of ninth graders) (Eaton et al., 2006).
To examine conflict within adolescent romantic relationships in more detail, a study of 869 high school students in the San Francisco Bay area asked those who had been involved in romantic relationships to rate the frequency with which they used specific conflict resolution tactics to resolve disagreements with their partners. In decreasing order of use, respondents indicated that they used compromise, distraction, avoidance, overt anger, seeking social support, and violence. Gender, age, and racial/ethnic differences in the use of the six tactics included the following: Girls reported using compromise and overt anger more frequently than boys, whereas boys described using distraction more frequently than girls; older teens described using compromise more frequently than younger ones; blacks described using violence more frequently than whites, whereas whites described using compromise more frequently than blacks (Feldman and Gowen, 1998).
Summary and Conclusions
The data on adolescents’ attitudes and values with respect to romantic relationships, while sparse, come together to form a consistent picture. From an early age, adolescents understand that romantic relationships are distinct from other kinds of relationships they experience (e.g., family relationships or friendships). Moreover, by middle school, most adolescents desire a satisfying romantic relationship for themselves. How youth define romantic relationships appears to develop over the course of adolescence—from an emphasis on the positive aspects of romance, such as physical attraction and companionship, during the middle-school years, to an acknowledgement of more complex issues like jealousy and infidelity among older adolescents. They express disapproval of sexual intercourse among teens, but also believe that sex, when it occurs, is most appropriate within the context of a committed romantic relationship. The apparent tension between these two views may reflect the limits of the self-report method (i.e., respondents may be attempting to respond as they believe researchers would like them to respond) as opposed to any real conflict in adolescents’ views about sex and relationships. In general, when asked to describe their attitudes and values regarding romantic relationships, most adolescents endorse values consistent with prevailing societal standards.
With respect to the future, data from a number of different sources converge to suggest that most adolescents view a healthy and happy marriage as an important life goal and expect to marry eventually. The value of marriage for adolescents does not vary much by age, but males are more likely than females, and Hispanic males more likely than white and black males, to endorse being married over being single. Consistent with their positive attitudes toward marriage, most adolescents disapprove of divorce. Attitudes toward divorce do not vary by age or gender, but do vary by race/ethnicity, such that Hispanic and black males view divorce as more acceptable than white males do. The limited data on historical trends suggest that these attitudes have been relatively stable over the past few decades.
Yet, despite their favorable views toward marriage as an institution, most adolescents also believe that marriage is a risky proposition. This is especially true among samples of adolescent unwed parents, who value marriage but nevertheless perceive obstacles to achieving a satisfying marriage for themselves. Even within the general population, significant proportions of adolescents who expect to marry also expect to experience divorce. Thus, adolescents appear to recognize that situational constraints may prevent them from achieving the life-long marriages that they desire. Perhaps as a consequence, recent decades have seen a substantial rise in adolescents’ acceptance of cohabitation.
The gap between adolescents’ aspirations and their expectations suggests a potentially important avenue for relationship education directed toward adolescents’ beliefs and values regarding marriage. Specifically, curriculum developers may wish to emphasize healthy adult marriage as not only desirable, but possible and achievable as well. In conjunction with programs to address real barriers to marriage (e.g., employability, substance abuse, lack of financial stability), such a curriculum could benefit adolescents by teaching specific steps that adolescents can take to maximize their chances of reaching their goals.
Adolescents’ romantic and sexual behaviors generally correspond with their self-reported attitudes and values. For example, adolescents’ strong desires for romantic relationships are consistent with the observation that most young people experience their first romantic relationships during adolescence. Just as adolescents generally endorse the idea that sexual behaviors are most appropriate in the context of a close, romantic relationship, so do the majority of adolescents engage in sexual behaviors within the context of exclusive romantic relationships. Moreover, these relationships are described, on average, as relatively long-lasting (i.e., enduring for more than a year) and characterized by a relatively high degree of intimacy, with couples exchanging expressions of love, expecting each other to be sexually exclusive, and appearing as a couple in public.
Yet these mean trends clearly mask considerable variability associated with gender, ethnicity, and level of income, and this variability remains to be explored in detail. In particular, data from small samples suggest that low-income adolescents, especially low-income males, may be experiencing a higher number of shorter and less intimate relationships than members of other groups. More generally, the existing literature suggests that the majority of adolescents already endorses the goals of healthy marriage initiatives and seeks to behave in ways that are consistent with those goals.
The research reviewed in this chapter describes a picture in broad strokes, but that picture can be refined and elaborated in important ways. For example, only a handful of smaller studies explicitly explored the romantic representations and experiences of gay and lesbian youth. Whereas there is some suggestion that income influences perceived marital prospects, the existing literature provides insufficient evidence to determine whether the relationship attitudes and values of low-income youth are contributors to the low likelihood of successful marriage in this group. Finally, because measurement of these variables has been inconsistent across studies, it is difficult to assess trends in adolescents’ attitudes and behaviors over time. As Florsheim (2003a, p. 372) has observed, “ The changing face of adolescent romance only underscores the importance (and the urgency) of collecting data that will allow us to track trends and fluctuations and determine their meaning and direction.”
| Table of Contents | Previous | Next |