Pubertal Development and Sexual Intercourse Among Adolescent Girls

Abstract

There are strong reasons to assume that early onset of puberty accelerates coital debut among adolescent girls. Although many studies support this assumption, evidence regarding the putative causal processes is limited and inconclusive. In this research, longitudinal data from the 1986 Northern Finland Birth Cohort Study (N = 2,596) were used to address three theoretical explanations: (a) a direct effect premised on biological processes, (b) a mediated path based on social psychological processes, and (c) a spurious effect derived from the evolutionary theory of socialization. In support of the social psychological pathway, the negative association between age at menarche and coital status at age 15 was almost fully mediated by differential social exposure—an empirical construct measuring involvement in high-risk social contexts.

Keywords

delinquency peers sexual behavior/risk

Early onset of pubertal maturation is regarded as an important antecedent of sexual activity among adolescent girls (De Genna, Larkby, & Cornelius, 2011; Negriff, Susman, & Trickett, 2011). The literature on pubertal timing suggests three plausible explanations for this association. First, early maturing girls may be more predisposed to sexual behavior as a direct function of physiological development. Hormonal changes associated with pubertal development can be considered as the foundation of libido (Smith, Udry, & Morris, 1985). In support of the direct effect, a pubertal rise in testosterone has been shown to predict the timing of coital debut among adolescent females (Halpern, Udry, & Suchindran, 1997). Thus, it is possible that biological processes suffice to explain the influence of pubertal development on sexual intercourse.

Second, social-psychological processes have been proposed to mediate the association between physiological maturity and first intercourse (Cavanagh, 2004; Deardorff, Gonzales, Christopher, Roosa, & Millsap, 2005). Early maturation has been found to increase girls’ exposure to social settings consisting of older peers, including males (Halpern, Kaestle, & Hallfors, 2007). Adolescent boys tend to associate bigger breasts and other signs of biological maturity with sexual availability (Martin, 1996; Waylen & Wolke, 2004), and the girls may find older males more appealing compared to boys their own age (Marín, Kirby, Hudes, Coyle, & Gómez, 2006). Associations with older and mixed gender peer groups present girls with increased opportunities and pressures to participate in illicit substance use, unsupervised house parties, and other activities conducive to sexual encounters (Costello, Sung, Worthman, & Angold, 2007; Ge, Brody, Conger, & Simons, 2006; Haynie, 2003). According to this hypothesis, early maturing girls are vulnerable to adjustment problems because they are cognitively and emotionally unprepared to cope with the social consequences of advanced physiological development (Moffitt, Caspi, Rutter, & Silva, 2001; Stattin & Magnusson, 1990).

Third, the evolutionary theory of socialization argues that human reproductive strategies are sensitive to environmental cues (Belsky, Steinberg, & Draper, 1991). Specifically, it has been proposed that adverse family circumstances accelerate the onset of menarche—presumably because they signal uncertainty with respect to long-term reproductive success. In support of this theory, father absence, family conflict, maternal depression, and harsh parenting have been found to predict the onset of menarche (Belsky et al., 2007; Belsky, Houts, & Fearon, 2010; Deardorff et al., 2011; Ellis & Garber, 2000). On the other hand, research on adolescent development has identified family adversity and its potential consequences as influential determinants of adolescent sexual behavior independently of pubertal maturation (Caminis, Henrich, Ruchkin, Stone, & Martin, 2007; Price & Hyde, 2009; Schofield, Bierman, Heinrichs, & Nix, 2008). Thus, to understand the association between early pubertal maturation and coital debut, it is prudent to attend to the potentially confounding influence of early childhood risk factors, including not only indicators of family adversity but also child characteristics, such as difficult temperament, that are predictive of maltreatment or negative parenting outcomes (Belsky et al., 2007; Jaffee et al., 2004).

In sum, as depicted in Figure 1, current literature suggests three plausible pathways linking early pubertal development and sexual debut among adolescent girls. The biological perspective (thick solid arrow) is consistent with a direct link between reproductive maturity and coital status. The social-psychological perspective (thin solid arrows) argues that early pubertal maturation increases exposure to peer groups and lifestyles conducive to sexual encounters. Finally, research on determinants of pubertal timing draws attention to the potentially spurious influence of adverse childhood dynamics and life circumstances.

Figure 1.

Etiological pathways in the association between pubertal timing and adolescent female sexual intercourse.

Evidence From Prior Research

The social-psychological theory of puberty effects has received considerable attention in the literature. As a particularly creative way to examine the role of social context, Caspi, Lynam, Moffitt, and Silva (1993) compared girls from mixed-gender and all-girl schools. In this research, the association between early puberty and problem behavior was limited to girls in mixed settings; early timing had no effect among girls attending all-girl schools. This finding is consistent with the assumption that exposure to male peers is a necessary link in the path from early puberty to female delinquency. Focusing on social contexts and peers associations outside the school, a more recent study, by Stattin, Kerr, and Skoog (2011), finds further support for the theory that access to older friends and opposite-sex relationships is a major factor in the causal path linking early maturation to problem behavior among adolescent girls.

Research by Haynie (2003) showed that early puberty is particularly strongly related to “party delinquency,” that is, such victimless offending as drinking, smoking, truancy, and disorderly conduct, and that delinquent peer associations and dating behavior are the principal mechanisms accounting for this effect. In general, there is persuasive evidence of peer mediation processes in studies of pubertal development and various types of adverse outcomes, including depression, school failure, and substance abuse (Cavanagh, Riegle-Crumb, & Crosnoe, 2007; Halpern et al., 2007; Marklein, Negriff, & Dorn, 2009; Tanner-Smith, 2012).

By contrast, evidence of social-psychological mediation is more limited and less conclusive in studies of sexual behavior (Crockett, Bingham, Chopak, & Vicary, 1996; Negriff et al., 2011; Waylen & Wolke, 2004). Using data from the United States, Cavanagh (2004) discovered significant racial/ethnic differences in the association between early maturation and sexual debut: It was strongest among Latinas, less strong but substantial among White girls, and entirely absent in the African American sample. Moreover, relevant indicators of the peer group context—presence of older peers, male peers, and delinquent friends—were able to account for a substantial portion (one third) of the total puberty effect among White girls but none of the very strong bivariate associations observed in the Hispanic sample.

According to Cavanagh (2004), the observed pattern of findings may reflect cultural differences in adolescent female sexuality: “Because Whites tend to enter adolescence in a social context with stronger norms against early sex, more powerful motivations or motivators may be required for them to engage in early sexual behaviors” (p. 290). Generalizing beyond the United States, this insight suggests that the effect of biological maturation on coital debut may exhibit substantial variation across sociocultural context.

The evidence in support of the evolutionary theory of socialization remains mixed (Moffitt, Caspi, Belsky, & Silva, 1992; Posner, 2006). However, recent research based on rigorous designs suggests that early exposure to erratic parenting, family disruption, socioeconomic disadvantage, and residential instability accelerate the onset of menarche (Arım, Tramonte, Shapka, Dahinten, & Willms, 2011; Belsky et al., 2010; Tither & Ellis, 2008). To our knowledge, our study is the first to integrate the mechanisms suggested by the evolutionary theory of socialization as an explicit part of a causal model testing the social psychological mediation hypothesis.

The Present Study

As noted in the literature (Negriff et al., 2011; Waylen & Wolke, 2004), there remains a shortage of studies addressing theoretical explanations of the association between pubertal maturation and adolescent sexual behavior. Existing research is limited to relatively small samples and dominated by evidence from a single cultural context (the United States). In particular, there is less etiological research focusing specifically on sexual intercourse as opposed to more age-appropriate forms of sexual and romantic activity, such as holding hands, heavy petting, and kissing. We suspect this has to do with the fact that small samples do not permit detailed investigations of rare events. The evidence reported in this research is based on a large-scale birth cohort study from Finland.

The Finnish climate surrounding female sexuality can be characterized as more liberal and gender equal than the climates in many other nations, including the United States (Suvivuo, Tossavainen, & Kontula, 2010; Tolman, 2002). Finnish schools offer comprehensive sex education at ages 13 to 14, with a primary focus on teaching responsibility and natural attitudes about sexual behavior; sexual abstinence is a low priority (Kontula, 2007). Cross-national research suggests that the prevalence of sexual double-standards is comparatively low in Finland (Haavio-Mannila & Kontula, 2003), which may explain why Finnish women report having had as many sexual partners before marriage as men—an average of 7, which is high even by European standards (Kontula, 2003). It is possible that growing up in a culture that encourages young girls to embrace and take control of their sexuality affords adolescent girls in Finland superior cognitive and emotional resources to manage the social risks associated with early puberty (Suvivuo et al., 2010). The significantly lower rates of teenage pregnancy, abortion, and fertility among Finnish girls, compared to their American counterparts, are consistent with this assumption (Singh & Darroch, 2000).

On the other hand, it is conceivable that a more gender-equal and permissive normative order presents early maturing girls with increased opportunities to engage in activities—such as heavy drinking and spending time in unsupervised settings—that are particularly conducive to adolescent sexual encounters (Browning, Leventhal, & Brooks-Gunn, 2005; Stattin et al., 2011). Flammer, Alsaker, and Noack (1999) found that Finnish adolescents spend time “hanging around with their friends” twice as frequently as American adolescents. The rate of heavy drinking among Finnish adolescents—an activity that typically occurs in unsupervised settings (Felson, Savolainen, Bjarnason, Anderson, & Zohra, 2011)—is substantially higher than the rate among American adolescents and higher than the average rate reported for other European countries (Hibell et al., 2004). Thus, the Finnish context appears to be particularly fertile ground for observing peer and behavioral mediation effects proposed by the social-psychological explanation.

Method

Participants

We use data from the 1986 Northern Finland Birth Cohort Study (NFBCS) of individuals born between July 1, 1985, and June 30, 1986, in the two northernmost provinces of Finland (Hurtig et al., 2007; Järvelin, Hartikainen-Sorri, & Rantakallio, 1993). The major data collection waves included a prenatal survey of mothers, hospital records at birth, a health study at age 7 (survey of parents), a learning disability study at age 8 (survey of teachers), and an adolescent study of health and well-being at age 15 (self-report survey followed by a clinical exam). To capture information about pubertal timing and sexual behavior, the present study is limited to females who participated in the adolescence self-report survey and the clinical examination, both of which were administered during the last year of compulsory education (ages 15-16, N = 2,596).

Because of the joint influence of residential mobility away from the region and survey nonparticipation, the adolescent wave has an attrition rate of about 16%. Examination of attrition bias across study variables revealed statistically significant differences in only two of the items used in measures of family adversity (maternal smoking and births outside marital or cohabiting union). These differences are not very large (results available from authors), however, and no significant differences were observed in measures of child characteristics (hyperactivity and antisocial behavior) among participant and nonparticipant groups. Moreover, the final estimation sample is large and exhibits sufficient variation in the analytic variables of interest.

Measures

Coital status

The experience of sexual intercourse is measured with a question from the adolescent self-report survey (age 15). This dichotomous variable identifies adolescents who have had sexual intercourse at least once in their lifetime (no = 0, yes = 1).

Timing of puberty

Self-reported information about the onset of menarche was used to measure differences in pubertal timing. The relevant question asked, “How old were you when you had your first menstruation?” We coded this information into four categories: 11 or younger, 12, 13, and 14 or older. Although age at menarche captures only one (albeit major) stage of female pubertal development, it is recognized as a valid and reliable tool for identifying early maturing girls (Cavanagh, 2004; Deardorff et al., 2005; Dorn & Biro, 2011).

Social exposure

In this research, “social exposure” is defined as involvement in (or access to) unsupervised social settings consisting of older and mixed-gender peers engaged in delinquent behaviors. This construct was measured as a latent variable with three indicators: peer delinquency, hanging out with older peers, and early alcohol use. Peer delinquency is a scale (α = .61) created by standardizing and calculating the mean of three questionnaire items: (a) “Do you spend time with friends who get into trouble a lot?” (b) “Has anyone ever offered you drugs (other an alcohol)?” and (c) “Do you have friends who use drugs?” (These items were standardized first because they had different response formats.) Hanging out with older peers was measured with a single item: “I prefer spending time with friends who are older than me” (never = 0, sometimes = 1, and frequently = 2). Early alcohol use was a scale (α = .84) calculated as the mean of four items: (a) age at first time drinking beer, (b) age at first time drinking wine, (c) age at first time drinking spirits, and (d) age at first time drinking to the point of intoxication.

Family adversity index

In the absence of items suitable for one-dimensional scaling, family adversity was measured in terms of a cumulative risk score covering seven loosely related domains: (a) maternal smoking, (b) maternal drinking during pregnancy, (c) born to a single mother (neither married nor cohabiting), (d) living without one biological parent at age 7, (e) residential instability (at least three nonlocal moves by age 15), (f) neither parent gainfully employed (at child age 7), and (g) poor economic situation of the household as reported and perceived by the respondent at age 15. Each item is a dichotomy indicating the presence of an adversity.

Child characteristics

Finally, we attend to differences in child characteristics to capture likely precursors of harsh discipline, family conflict, and other indicators of family adversity. Our measure of difficult temperament is a latent construct based on the Antisocial Behavior (α = .80) and Hyperactivity (α = .71) subscales from the Rutter Scale for Teachers (McGee et al., 1985; Rutter, 1967; Taanila, Ebeling, Kotimaa, Moilanen, & Järvelin, 2004). The questionnaires were filled out by first grade teachers when participants were around 8 years old. This approach is justified by two assumptions. First, because of the heritability of these traits, antisocial and hyperactive children are more likely to have parents with similar characteristics (Rutter, 2006). Second, research on child effects shows that difficult temperament in children increases the likelihood of harsh parenting styles (Jaffee et al., 2004). In particular, research by Belsky et al. (2007) suggests a developmental path in which infant negative emotionality influences parenting, which, in turn, affects the timing of puberty.

Modeling Approach

Structural equation modeling analyses were conducted in Mplus 6.1 (Muthén & Muthén, 2010) using the weighted least squares means-variance (WLSMV) estimator, which is appropriate for categorical indicators and outcomes (e.g., coital status). WLSMV also incorporates pairwise missing data procedures to maximize the use of available data, resulting in a full analysis sample of 2,596 cases.

In addition to the chi-square statistic, two alternative fit indices were used to evaluate model fit: the comparative fit index (CFI) and the root mean square error of approximation (RMSEA). Current guidelines suggest that CFI values close to .95 or greater and RMSEA values less than .06 indicate acceptable model fit (Hu & Bentler, 1999). The statistical significance of the indirect effect was calculated via the delta method using the Mplus model indirect option (Sobel, 1982).

Findings

Descriptive Statistics

Table 1 presents bivariate correlations and descriptive statistics for study variables. Although all of the correlations are in the expected direction, it is notable that the intervening variables derived from the evolutionary theory of socialization (antisocial behavior, hyperactivity, and family adversity) are not statistically significantly related to age at menarche. Among other observations, the univariate statistics show that 34% of the girls in the sample had experienced coital debut at the time of the survey and that the distribution of the age at menarche approximates the normal curve.

Table 1.

Descriptive Statistics and Correlations.

	1	2	3	4	5	6	7	M	SD	Range
1. Coital status								0.34	0.47	0–1
2. Age at menarche	−.19*							12.51	0.98	11–14
3. Peer delinquency	.38*	−.13*						−0.06	2.17	−1.7–9.4
4. Older peers	.19*	−.13*	.26*					0.86	0.64	0–2
5. Early onset drinking	.35*	−.08*	.22*	.08*				2.40	1.57	0–6
6. Antisocial behavior	.11*	−.04	.13*	.09*	.05*			0.06	0.17	0–2
7. Hyperactivity	.12*	−.01	.10*	.09*	.04*	.64*		0.14	0.30	0–2
8. Family adversity	.16*	−.03	.13*	.08*	.14*	.11*	.11*	0.73	0.97	0–5

Valid N varies from 2,596 to 2,421 because of missing data.

p < .05.

To offer a more detailed picture of the negative association between age at menarche and coital debut, Figure 2 presents the prevalence of sexual intercourse in the four categories of pubertal timing. As can be seen, the association is quite linear, with a 48% prevalence rate among girls experiencing the earliest onset of menarche (younger than 12) compared to a 19% prevalence rate among girls with late onset (age 14 or older).

Figure 2.

Coital status at age 15 by age at menarche (%).

Structural Equation Model

As an initial step, a confirmatory factor analysis (CFA) was conducted in which the measurement model and all covariances among the study constructs were estimated. Results showed that the fit between the data and the CFA was acceptable, χ²(13, N = 2596) = 92.10, p < .05, CFI = .97, RMSEA = .05. Each factor loading was statistically significant (p < .05), with standardized values ranging from .40 to .86. Estimated correlations among the constructs are reported in Table 2. Coital status had statistically significant positive associations with difficult temperament, family adversity, and social exposure and a statistically significant negative association with age at menarche. Additional associations among the constructs were statistically significant in expected directions, although neither difficult temperament nor family adversity was statistically significantly associated with age at menarche.

Table 2.

Correlations Among Study Constructs From the Confirmatory Factor Analysis.

Construct	1	2	3	4
1. Age at menarche
2. Difficult temperament	−.03
3. Family adversity	−.04	.14*
4. Social exposure	−.26*	.25*	.25*
6. Coital status	−.26*	.17*	.20*	.83*

N = 2,596.

p < .05.

The structural equation model corresponding to processes described in Figure 1 was estimated next. This model was saturated, estimating all possible paths; therefore, model fit was the same as for the CFA. Standardized factor loadings, standardized path estimates, and correlations are reported in Figure 3. Results showed that age at menarche was a statistically significant positive predictor of social exposure, which, in turn, was a statistically significant positive predictor of coital status. Consistent with the social-psychological perspective, age at menarche no longer was significantly related to the outcome after taking into account the social exposure construct. The indirect effect of age at menarche on coital status through social exposure was statistically significant (β = –.20, p < .05), accounting for 81% of the total effect of pubertal timing. Although difficult temperament and family adversity were unrelated to both age at menarche and coital status, they were statistically significant positive predictors of differential social exposure. Taken together, the study constructs accounted for an estimated 69% of variance in the outcome.

Figure 3.

Structural equation model of hypothesized pathways in the association between age at menarche and coital status.

Discussion

Research on adolescent female development finds that early onset of puberty is related to adverse outcomes in multiple domains. Although this issue has received considerable attention in the literature, the evidence concerning causal mechanisms remains limited. The purpose of this research was to examine etiological pathways in the association between pubertal timing and sexual intercourse in a Finnish sample of adolescent girls.

Our theoretical model recognized three potential explanations for the association: a direct effect premised on biological processes, a mediated path based on social-psychological processes, and a spurious effect derived from the evolutionary theory of socialization. A structural equation model estimating the hypothesized pathways provided strong evidence of social-psychological mediation. Social exposure—an empirical construct measuring behaviors and peer associations consistent with the mediation model—accounted for most (81%) of the strong bivariate association between age at menarche and coital status. In the presence of this mediator, the direct effect of age at menarche was no longer statistically significant. We did not find support for the spurious pathways suggested by the evolutionary theory of socialization, as neither family adversity nor difficult temperament was predictive of age at menarche.

Prior research finds strong support for the theory that peer associations mediate the influence of early maturation on delinquency and other high-risk outcomes among adolescent girls. Our research adds to this literature by demonstrating the salience of these processes for midadolescent coital status. The amount of mediation observed in our research is large compared to prior studies of early puberty and sexual behavior, which tend to report substantial main effects of pubertal timing net of social-psychological mediating or moderating processes (Cavanagh, 2004; Marín et al., 2006; Negriff et al., 2011; Smith et al., 1985). These differences may reflect the influence of cultural context: It is possible that the macro-social environment in Finland provides exceptional opportunities for unstructured socializing among adolescents across boundaries of age and gender.

As noted previously, Finnish adolescents spend time “hanging around with their friends” twice as frequently as American adolescents (Flammer et al., 1999), and they are significantly more likely to engage in binge drinking (Hibell et al., 2004), which is an activity that typically requires contact with individuals old enough to purchase alcohol and the absence of adult supervision. In addition, there are important differences in the ecological context of adolescent time use between Finland and United States. Finns are more likely to live in what Felson and Boba (2010) have described as a convergent city, whereas Americans are more likely to live in a divergent metropolis. Consequently, Finnish (and other European) teens are more likely to walk and take public transportation to go shopping, to see movies, or to “hang out” in city centers. By contrast, American youth younger than 16 are more likely to be driven by parents to shopping malls and other highly monitored semipublic areas (see Anderson & Hughes, 2009). More research from different national contexts is needed to determine whether these macro-level differences matter to processes that link early pubertal timing to sexual intercourse.

Based on secondary data analysis, this study is not without limitations. The fact that the girls included in this study were 15 years old in 2001 may limit the generalizability of the findings to more contemporary populations. It is possible that the proliferation of the Internet and social networking sites has altered the relevant process of social exposure. On the other hand, for physical intimacy and sexual intercourse to happen, adolescents still have to meet and spend time in person. Thus, although technological innovations likely facilitate these encounters, we suspect the proximate contexts of social exposure have remained the same. Ultimately, it is up to future research to examine these assumptions.

Second, the fact that we found no support for the evolutionary theory of socialization might be viewed by its proponents as an indictment of the measures used in the analysis. In particular, the measure of family adversity is dominated by structural indicators of disadvantage. We attempted to remedy this limitation by considering difficult temperament (a child characteristic) as a proxy measure of exposure to harsh discipline. It is possible that results would have been different had we had access to more direct indicators of parenting behavior. We encourage future research to reexamine these processes with improved measurement of relevant constructs.

Second, the indicators of social exposure were measured concurrently with coital status, which likely increases the size of the mediation effect and makes it difficult to establish the causal order between the mediator and the outcome. Unfortunately, there is only one wave of adolescent data collection available in the NFBCS. Thus, our research cannot rule out the possibility that changes in coital status precede problem behavior and relevant peer associations. Consistent with this argument, Armour and Haynie (2007) report that early sexual debut increases involvement in delinquency. However, Harden, Mendle, Hill, Turkheimer, and Emery (2008) failed to replicate this finding using a more rigorous (twin-study) design to address the influence of confounding factors. In fact, their research suggests that earlier age of first sex predicts slightly lower levels of delinquency.

Further complicating this issue, Negriff et al. (2011) found that early pubertal timing at Time 1 predicted increased sexual activity at Time 2, which in turn was positively related to delinquency at Time 3; and none of these associations was mediated by changes in peer context. Note, however, that the measure of sexual activity used in their research included relatively innocuous behaviors, such as holding hands and kissing. It is entirely consistent with the social-psychological mediation hypothesis to assume that more age-appropriate acts of intimacy are part of the initial process in which early maturing girls become involved in peers groups and lifestyles conducive to sexual intercourse. Romantic activities of a less advanced nature can be understood as signals of availability and interest rather than consequences of peer exposure. By contrast, it seems implausible to treat coital debut as the first step in the causal chain linking early maturation to delinquent peer associations.

In conclusion, this study supports the hypothesis that increased exposure to prodelinquent social contexts accounts for the association between early pubertal maturation and sexual intercourse. In light of this finding, efforts to delay coital debut among early maturing girls should focus on reducing opportunities for unstructured socializing with older peers in mixed-gender settings. For example, there may be specific settings—times and places—that are particularly likely to facilitate these interactions. It also would be helpful to know the extent to which these encounters can be characterized as consensual/agentic as opposed to predatory (Tolman, 2002). If it is common for male teens with delinquent tendencies to actively target younger girls of advanced physiological development, intervention efforts that focus solely on early maturing girls will be less effective than those that also address the behavior of males.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is a part of a research project funded by the Academy of Finland and the Thule Institute at the University of Oulu.

Author Biographies

Jukka Savolainen is Associate Professor in the School of Criminology and Criminal Justice at the University of Nebraska. His research is focused on longitudinal and cross-national studies of crime, violence, delinquency, and adolescent risk behavior.

W. Alex Mason is Associate Director of the Boys Town National Research Institute for Child and Family Studies. His major research interests include the developmental etiology and family-based prevention of adolescent and young adult substance misuse and co-occurring problems.

Lorine A. Hughes is Associate Professor in the School of Criminology and Criminal Justice at the University of Nebraska. Her current interests include youth street gangs, quantitative methods, and social network analysis.

Hanna Ebeling is Professor of Child Psychiatry at the University Hospital of Oulu, Finland. Her research is focused on the epidemiology of child and adolescent mental health.

Tuula M. Hurtig is a postdoctoral scholar at the Institute of Health Sciences, University of Oulu, and at the Clinic of Child Psychiatry, the University Hospital of Oulu.

Anja M. Taanila is Professor of Public Health at the University of Oulu and the Managing Director of the 1986 Northern Finland Birth Cohort Study.

References

Anderson

A. L.

Hughes

L. A.

(2009). Exposure to situations conducive to delinquent behavior. Journal of Research in Crime and Delinquency, 46(1), 5-34.

Arım

R. G.

Tramonte

Shapka

J. D.

Dahinten

S. V.

Willms

D. J.

(2011). The family antecedents and the subsequent outcomes of early puberty. Journal of Youth and Adolescence, 40(11), 1423-1435.

Armour

Haynie

(2007). Adolescent sexual debut and later delinquency. Journal of Youth and Adolescence, 36(2), 141-152.

Belsky

Houts

R. M.

Fearon

R. M. P.

(2010). Infant attachment security and the timing of puberty. Psychological Science, 21(9), 1195-1201.

Belsky

Steinberg

Draper

(1991). Childhood experience, interpersonal development, and reproductive strategy: An evolutionary theory of socialization. Child Development, 62(4), 647-670.

Belsky

Steinberg

L. D.

Houts

R. M.

Friedman

S. L.

DeHart

Cauffman

NICHD Early Child Care Research Network. (2007). Family rearing antecedents of pubertal timing. Child Development, 78(4), 1302-1321.

Browning

C. R.

Leventhal

Brooks-Gunn

(2005). Sexual initiation in early adolescence: The nexus of parental and community control. American Sociological Review, 70(5), 758-778.

Caminis

Henrich

Ruchkin

Stone

M. S.

Martin

(2007). Psychosocial predictors of sexual initiation and high-risk sexual behaviors in early adolescence. Child and Adolescent Psychiatry and Mental Health, 1(1), 14.

Caspi

Lynam

Moffitt

T. E.

Silva

P. A.

(1993). Unraveling girls’ delinquency: Biological, dispositional, and contextual contributions to adolescent misbehavior. Developmental Psychology, 29(1), 19-30.

10.

Cavanagh

S. E.

(2004). The sexual debut of girls in early adolescence: The intersection of race, pubertal timing, and friendship group characteristics. Journal of Research on Adolescence, 14(3), 285-312.

11.

Cavanagh

S. E.

Riegle-Crumb

Crosnoe

(2007). Puberty and the education of girls. Social Psychology Quarterly, 70(2), 186-198.

12.

Costello

E. J.

Sung

Worthman

Angold

(2007). Pubertal maturation and the development of alcohol use and abuse. Drug and Alcohol Dependence, 88(Suppl. 1), 50-59.

13.

Crockett

L. J.

Bingham

C. R.

Chopak

J. S.

Vicary

J. R.

(1996). Timing of first sexual intercourse: The role of social control, social learning, and problem behavior. Journal of Youth and Adolescence, 25, 89-111.

14.

De Genna

N. M.

Larkby

Cornelius

M. D.

(2011). Pubertal timing and early sexual intercourse in the offspring of teenage mothers. Journal of Youth and Adolescence, 40(10), 1315-1328.

15.

Deardorff

Ekwaru

J. P.

Kushi

L. H.

Ellis

B. J.

Greenspan

L. C.

Mirabedi

Hiatt

R. A.

(2011). Father absence, body mass index, and pubertal timing in girls: Differential effects by family income and ethnicity. Journal of Adolescent Health, 48(5), 441-447.

16.

Deardorff

Gonzales

N. A.

Christopher

F. S.

Roosa

M. W.

Millsap

R. E.

(2005). Early puberty and adolescent pregnancy: The influence of alcohol use. Pediatrics, 116(6), 1451-1456.

17.

Dorn

L. D.

Biro

F. M.

(2011). Puberty and its measurement: A decade in review. Journal of Research on Adolescence, 21(1), 180-195.

18.

Ellis

B. J.

Garber

(2000). Psychosocial antecedents of variation in girls’ pubertal timing: Maternal depression, stepfather presence, and marital and family stress. Child Development, 71(2), 485-501.

19.

Felson

Boba

(2010). Crime and everyday life (4th ed.). Thousand Oaks, CA: Sage.

20.

Felson

R. B.

Savolainen

Bjarnason

Anderson

A. L.

Zohra

I. T.

(2011). The cultural context of adolescent drinking and violence in 30 European countries. Criminology, 49, 699-728.

21.

Flammer

Alsaker

F. D.

Noack

(1999). Time use by adolescents in an international perspective: The case of leisure activities. In Flammer

Alsaker

F. D.

(Eds.), The adolescent experience: European and American adolescents in the 1990s (pp. 33-60). Mahwah, NJ: Lawrence Erlbaum.

22.

Brody

Conger

Simons

(2006). Pubertal maturation and African American children’s internalizing and externalizing symptoms. Journal of Youth and Adolescence, 35(4), 528-537.

23.

Haavio-Mannila

Kontula

(2003). Single and double sexual standards in Finland, Estonia, and St. Petersburg. Journal of Sex Research, 40(1), 36-49.

24.

Halpern

Kaestle

Hallfors

(2007). Perceived physical maturity, age of romantic partner, and adolescent risk behavior. Prevention Science, 8(1), 1-10.

25.

Halpern

C. T.

Udry

J. R.

Suchindran

(1997). Testosterone predicts initiation of coitus in adolescent females. Psychosomatic Medicine, 59(2), 161-171.

26.

Harden

Mendle

Hill

Turkheimer

Emery

(2008). Rethinking timing of first sex and delinquency. Journal of Youth and Adolescence, 37(4), 373-385.

27.

Haynie

D. L.

(2003). Contexts of risk? Explaining the link between girls’ pubertal development and their delinquency involvement. Social Forces, 82(1), 355-397.

28.

Hibell

Andersson

Bjarnason

Ahlström

Balakireva

Kokkevi

Morgan

(2004). The ESPAD report 2003. Alcohol and other drug use among students in 35 European countries. Stockholm: Swedish Council for Information on Alcohol and Other Drugs.

29.

Bentler

P. M.

(1999). Covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1-55.

30.

Hurtig

Ebeling

Taanila

Miettunen

Smalley

S. L.

McGough

J. J.

Moilanen

I. K.

(2007). ADHD symptoms and subtypes: Relationship between childhood and adolescent symptoms. Journal of the American Academy of Child and Adolescent Psychiatry, 46, 1605-1613.

31.

Jaffee

S. R.

Caspi

Moffitt

T. E.

Polo-Tomas

Price

T. S.

Taylor

(2004). The limits of child effects: Evidence for genetically mediated child effects on corporal punishment but not on physical maltreatment. Developmental Psychology, 40(6), 1047-1058.

32.

Järvelin

Hartikainen-Sorri

Rantakallio

(1993). Labour induction policy in hospitals of different levels of specialisation. British Journal of Obstetrics & Gynaecology, 100(4), 310-315.

33.

Kontula

(2003). Trends in teenage sexual behaviour: Pregnancies, sexually transmitted infections and HIV infections in Europe. In Bajos

Guillaume

Kontula

(Eds.), Reproductive health behaviour of young Europeans (Vol. 1, Population Studies No. 42, pp. 77-127). Strasbourg, France: Council of Europe Publishing.

34.

Kontula

(2007, April). National Survey on Sex Education and Sexual Knowledge among Adolescents in Finland. Paper presented in the 18th World Congress of Sexology, First World Congress on Sexual Health, Sydney, Australia.

35.

Marín

B. V.

Kirby

D. B.

Hudes

E. S.

Coyle

K. K.

Gómez

C. A.

(2006). Boyfriends, girlfriends and teenagers’ risk of sexual involvement. Perspectives on Sexual and Reproductive Health, 38(2), 76-83.

36.

Marklein

Negriff

Dorn

(2009). Pubertal timing, friend smoking, and substance use in adolescent girls. Prevention Science, 10(2), 141-150.

37.

Martin

K. A.

(1996). Puberty, sexuality, and the self: Boys and girls at adolescence. Florence, KY: Taylor & Francis/Routledge.

38.

McGee

Williams

Bradshaw

Chapel

J. L.

Robins

A. A.

Silva

P. A.

(1985). The Rutter Scale for completion by teachers: Factor structure and relationships with cognitive abilities and family adversity for a sample of New Zealand children. Journal of Child Psychology and Psychiatry, 26(5), 727-739.

39.

Moffitt

T. E.

Caspi

Belsky

Silva

P. A.

(1992). Childhood experience and the onset of menarche: A test of a sociobiological model. Child Development, 63(1), 47-58.

40.

Moffitt

T. E.

Caspi

Rutter

Silva

P. A.

(2001). Sex differences in antisocial behaviour: Conduct disorder, delinquency, and violence in the Dunedin Longitudinal Study. New York, NY: Cambridge University Press.

41.

Muthén

L. K.

Muthén

B. O.

(2010). Mplus user’s guide (6th ed.). Los Angeles, CA: Muthén & Muthén.

42.

Negriff

Susman

E. J.

Trickett

P. K.

(2011). The developmental pathway from pubertal timing to delinquency and sexual activity from early to late adolescence. Journal of Youth and Adolescence, 40(10), 1343-1356.

43.

Posner

R. B.

(2006). Early menarche: A review of research on trends in timing, racial differences, etiology and psychosocial consequences. Sex Roles, 54(5), 315-322.

44.

Price

Hyde

(2009). When two isn’t better than one: Predictors of early sexual activity in adolescence using a cumulative risk model. Journal of Youth and Adolescence, 38(8), 1059-1071.

45.

Rutter

(1967). A children’s behaviour questionnaire for completion by teachers: Preliminary findings. Journal of Child Psychology and Psychiatry, 8, 1-11.

46.

Rutter

(2006). Genes and behavior: Nature-nurture interplay explained. Malden, MA: Blackwell.

47.

Schofield

H.-L. T.

Bierman

K. L.

Heinrichs

Nix

R. L.

(2008). Predicting early sexual activity with behavior problems exhibited at school entry and in early adolescence. Journal of Abnormal Child Psychology, 36(8), 1175-1188.

48.

Singh

Darroch

J. E.

(2000). Adolescent pregnancy and childbearing: Levels and trends in developed countries. Family Planning Perspectives, 32(1), 14-23.

49.

Smith

E. A.

Udry

J. R.

Morris

N. M.

(1985). Pubertal development and friends: A biosocial explanation of adolescent sexual behavior. Journal of Health and Social Behavior, 26(3), 183-192.

50.

Sobel

M. E.

(1982). Asymptotic confidence intervals for indirect effects in structural equation models. In Leinhardt

(Ed.), Sociological methodology (pp. 290-312). Washington, DC: American Sociological Association.

51.

Stattin

Kerr

Skoog

(2011). Early pubertal timing and girls’ problem behavior: Integrating two hypotheses. Journal of Youth and Adolescence, 40(10), 1271-1287.

52.

Stattin

Magnusson

(1990). Pubertal maturation in female development: Paths through life (Vol. 2). Hillsdale, NJ: Lawrence Erlbaum.

53.

Suvivuo

Tossavainen

Kontula

(2010). “Can there be such a delightful feeling as this?” Variations of sexual scripts in Finnish girls’ narratives. Journal of Adolescent Research, 25(5), 669-689.

54.

Taanila

Ebeling

Kotimaa

Moilanen

Järvelin

(2004). Is a large family a protective factor against behavioural and emotional problems at the age of 8 years? Acta Paediatrica, 93, 508-517.

55.

Tanner-Smith

E. E.

(2012). Pubertal development and adolescent girls’ substance use. Journal of Early Adolescence, 32(5), 621-649.

56.

Tither

J. M.

Ellis

B. J.

(2008). Impact of fathers on daughters’ age at menarche: A genetically and environmentally controlled sibling study. Developmental Psychology, 44(5), 1409-1420.

57.

Tolman

D. L.

(2002). Dilemmas of desire: Teenage girls talk about sexuality. Cambridge, MA: Harvard University Press.

58.

Waylen

Wolke

(2004). Sex “n” drugs “n” rock “n” roll: The meaning and social consequences of pubertal timing. European Journal of Endocrinology, 151(Suppl. 3), 151-159.