Journal of Youth and Adolescence

, Volume 47, Issue 4, pp 793–806 | Cite as

Growth in Adolescent Self-Regulation and Impact on Sexual Risk-Taking: A Curve-of-Factors Analysis

  • AliceAnn Crandall
  • Brianna M. Magnusson
  • M. Lelinneth B. Novilla
Empirical Research


Adolescent self-regulation is increasingly seen as an important predictor of sexual risk-taking behaviors, but little is understood about how changes in self-regulation affect later sexual risk-taking. Family financial stress may affect the development of self-regulation and later engagement in sexual risk-taking. We examined whether family financial stress influences self-regulation in early adolescence (age 13) and growth in self-regulation throughout adolescence (from age 13–17 years). We then assessed the effects of family financial stress, baseline self-regulation, and the development of self-regulation on adolescent sexual risk-taking behaviors at age 18 years. Using a curve-of-factors model, we examined these relationships in a 6-year longitudinal study of 470 adolescents (52% female) and their parents from a large northwestern city in the United States. Results indicated that family financial stress was negatively associated with baseline self-regulation but not with growth in self-regulation throughout adolescence. Both baseline self-regulation and growth in self-regulation were predictive of decreased likelihood of engaging in sexual risk-taking. Family financial stress was not predictive of later sexual risk-taking. Intervening to support the development of self-regulation in adolescence may be especially protective against later sexual risk-taking.


Self-regulation Sexual risk-taking Family stress Growth curve analysis Structural equation modeling 


Sexual risk-taking behaviors are associated with increased risk for sexually transmitted infections (STIs) (Eaton et al. 2012) and unplanned and teen pregnancies (Centers for Disease Control & Prevention 2016). STIs and unplanned pregnancies are costly to nations and to individuals, leading to billions of dollars of treatment costs (Centers for Disease Control & Prevention 2016) and a loss of social and economic capital at the individual level (Kavanaugh et al. 2017).

Adolescent self-regulation, defined as the ability to regulate emotions, cognitions, and behaviors in service of reaching one’s goals (Murray et al. 2015), is emerging as an important predictor of lower engagement in sexual risk-taking in adolescence and young adulthood (Griffin et al. 2012). However, despite increasing evidence of the importance of self-regulation on sexual risk-taking behaviors, prior research has focused on self-regulation at one time point and has not assessed how the development of self-regulation throughout adolescence affects later sexual risk-taking behaviors.

This study builds on a previous study that showed that adolescent self-regulation mediated the relationship between family financial stress and risky sexual behaviors in later adolescence (Crandall et al. 2017). In the prior study, using data from the Flourishing Families Project, greater family financial distress at age 13–16 years was predictive of lower adolescent self-regulation 1 year later, and lower self-regulation was associated with higher report of sexual risk-taking in young adulthood. The prior study assessed self-regulation at one time point, controlling for self-regulation in the prior year. For the current study, the sample also came from the Flourishing Families Project, but we used a developmental approach with 6 years of data to assess the relationships between family financial stress, self-regulation, and sexual risk-taking. We first examined how family financial stress in early adolescence affected both baseline self-regulation (at age 13 years) and growth in self-regulation from ages 13 to 17 years. We then assessed the effects of self-regulation at age 13 years and the growth in self-regulation throughout the teenage years on sexual risk-taking at age 18 years. We used a curve-of-factors model in a structural equation modeling framework to assess the data. This allowed us to account for bias due to measurement error and to analyze multiple relationships simultaneously.

Sexual Risk-Taking and Consequences

Adolescence is a developmental period in which persons may engage in experimentation across a variety of behaviors including involvement in risk-taking behaviors (Steinberg 2008). This period of physical, mental, and emotional development is further characterized by sexual maturation. Sexual maturation, increased participation in risk-taking and coincident changes in the social acceptance of sexual exploration are associated with increased risk-taking among some youth during this life stage. Sexual risk-taking is typically defined as participation in sexual activities that increase the chances of contracting an STI or becoming involved in an unintended or mistimed pregnancy. Specific behaviors include unprotected intercourse outside of a long-term monogamous relationship, sex with high-risk partners, and concurrent or overlapping partnerships.

Youth (15–24 years of age) contract an estimated 10 million new cases of STIs each year, accounting for approximately 50% of the total cases in the United States (Centers for Disease Control & Prevention 2016). In addition to their immediate health consequences, STIs are a leading cause of tubal factor infertility in women throughout the world (Tsevat et al. 2017). The United States spends an estimated $16 billion on the direct treatment of STIs annually (Centers for Disease Control & Prevention 2016).

Despite recent reports of declining unintended pregnancy, nearly one-half of U.S. pregnancies are reported by their mother to be unintended at conception (Finer and Zolna 2016). As with STIs, adolescents and young adults are disproportionately likely to experience an unintended pregnancy. In 2011, approximately 75% of pregnancies in females 15–19 years and 59% of those in females 20–24 years were unintended (Finer and Zolna 2016). Unintended pregnancy in adolescence or emerging adulthood may reduce the social mobility of those involved, particularly for young women. A 2017 qualitative study identified that disruption of education trajectories (especially for young women), reduction in earnings due to job loss or ability to work fewer hours, relationship difficulties or dissolution, and negative effects on physical and mental health were all commonly experienced by both men and women who had experienced an unintended pregnancy (Kavanaugh et al. 2017).

Developmental Factors Affecting Sexual Risk-Taking

Sexual risk-taking and its consequences are associated with social disadvantage. Exposure to harsh environments including low neighborhood quality, unsafe communities, and poverty are associated with adolescent involvement in risk-taking behaviors including sexual risk-taking in both White (Hampson et al. 2016) and African American adolescents (Gibbons et al. 2012). Among urban adolescents aged 11–16 years, concentrated poverty and other measures of social disadvantage were associated with multiple sexual partners during adolescence (Browning et al. 2008). In the United States and United Kingdom, HIV (El-Sadr et al. 2010) and chlamydia (Crichton et al. 2014) prevalence are concentrated in marginalized communities.

Aspects of low self-regulation correlate with risky sexual behavior in adolescents and young adults. Specifically, self-regulation is associated with later age at initiation of oral and coital sex behaviors, fewer lifetime partners, and increased condom and contraceptive use (Moilanen 2015); impulsivity (an aspect of low self-regulation) is associated with engaging in risky sex acts and impulsive sexual behaviors (Birthrong and Latzman 2014).

Development of Self-Regulation

The adolescent period, defined as the stage of growth between 10–19 years of age (World Health Organization 2017), is a crucial phase of physical growth and mental and emotional development. During this period, children, who are transitioning into young adult responsibilities, are continually forming their identity, asserting their independence, and developing their ability to rein in their impulses while attempting to behave in a way consistent with their deepest-held values. Such behavioral, emotional, and cognitive coherence toward purposeful action is known as self-regulation (Murray et al. 2015). Self-regulation is foundational to the development of emotional well-being and the adoption of healthy behaviors (Murray et al. 2015).

During adolescence, higher order reasoning, planning, and problem-solving skills develop, and this development continues throughout young adulthood. These executive functions combined with adolescents’ added life experiences, increased synaptic pruning, and better neuroconnectivity, lead to an enhanced ability to integrate thoughts, emotions, and behaviors in the service of attaining their goals (Gestsdottir and Lerner 2008).

Aspects of self-regulation are housed in the pre-frontal cortex, which develops over the first two to three decades of life (Niendam et al. 2012). Recent neurological research indicates that adolescence is a sensitive period for the development of the prefrontal cortex (Fuhrmann et al. 2015). Although there is evidence of brain plasticity and changes in self-regulation throughout the life course (Dahlin et al. 2008), sensitive periods differ from general neuroplasticity. During sensitive periods, the human organism expects exposure to triggers that will enhance development of self-regulation between early and later adolescence (Fuhrmann et al. 2015). However, if those triggers do not occur or if adolescents are exposed to environmental toxins such as substance abuse (Volkow et al. 2014) or adverse social environmental conditions like excessive stress (Lupien et al. 2009), development may be delayed.

Family Financial Distress and Self-Regulation

The development of self-regulation throughout childhood and adolescence is not yet well understood, though both individual experiences and ecological influences, such as family poverty and stress, can impact self-regulation. For instance, the ability to actively cope with stress, also known as effortful coping, is a skill that requires the integration of emotional and cognitive self-regulation and is still being developed among children and adolescents (Murray et al. 2015; Woltering and Lewis 2009). How severe a stressor is, how early it is introduced, and how persistently it permeates into childhood and adolescence can influence an individual’s ability to control personal actions and behaviors. Murray et al. (2015) explained that manageable stress may actually build self-regulatory skills, but chronic or repeated stress can tax one’s self-regulatory skills and have lasting consequences on one’s neurobiological development. King and colleagues (2012), using a diverse sample, found that although pre-adolescents who experienced greater than average amounts of contextual stress had lower baseline self-regulation (defined as effortful control and impulsivity), they caught up with their peers by early adolescence, while those with low stress showed the least growth in self-regulation over time. Adolescent self-regulation in this study was reported by the mother and the child (King et al. 2012). Given that self-regulation was reported using scale measures, it is possible that the lower growth in self-regulation over time among adolescents with the lowest levels of contextual stress was due to a ceiling effect. All self-regulation scale measures were based off of 5-point and 8-point rating systems, and children with lower stress at baseline would have had, on average, self-regulation ratings closer to the maximum than their peers with lower self-regulation, thus limiting the potential for growth among children with higher baseline scores. However, the results are encouraging in that they show that children with more stress may catch up with their peers over time.

Persistent stressors, such as traumatic experiences and family financial challenges, can disrupt the stress response system (Shonkoff et al. 2012) to the point of overwhelming an adolescent’s ability to self-regulate. The more severe the financial stress at the family level, the higher is its potential to negatively impact the parent–child relationship and quality of communication, which in turn, can affect self-regulatory skills. Since parents serve as co-regulators with their children, when the “warm” and “responsive” interactions between parent(s) and child (Murray et al. 2015) are lost or significantly reduced, adolescents may resort to expressing vs. restraining their frustrations through negative behavior. Financial stress, such as poverty, also contributes to poor self-regulation through the “psychology of scarcity” (Mullainathan and Shafir 2013). Economic deprivation brings with it the paucity of not only money, food, and material resources but also the reduction in one’s mental flexibility and capacity to focus and think through problems and the ability to control one’s behavior (Barbarin 2013).

The timing of stress and poverty may also be important when considering their effect on self-regulation. Higher income-to-needs ratio predicts the baseline of self-regulation in early childhood (Li et al. 2017) and in middle childhood (Evans and Rosenbaum 2008), but not the later rates of developmental change in self-regulation (Hackman et al. 2015). Similarly, greater stress (King et al. 2012) and family poverty (Evans and Rosenbaum 2008) have been associated with lower baseline self-regulation in early adolescence, but there is limited research on its development throughout adolescence. Nascent research on stress in the context of poverty, indicates that poverty in childhood has negative structural effects on areas of the brain (specifically the prefrontal cortex) that are involved in self-regulation and stress response, particularly for children living in less supportive home environments (Blair 2010). During adolescence, stress in the home environment has been linked with greater stress in the adolescent (Blair 2010), and this stress appears to negatively impair prefrontal cortex functioning associated with self-regulation (Lupien et al. 2009).

The above review of literature suggests that family stress and poverty likely affect the development of self-regulation more acutely at certain periods of the life course than at other times, as evidenced by lower levels of baseline self-regulation in children and adolescents who had been exposed to family stress and/or poverty. Results are conflicting as to whether stress and poverty affected the rate of growth of self-regulation. Family stress and poverty may particularly affect the growth of self-regulation during adolescence because it is a sensitive period for brain development. Family poverty and stress generally appear to be toxic to the growth of self-regulation (Lupien et al. 2009), but at manageable levels, stress may actually aid in catch-up growth (King et al. 2012).

Current Study

Prior research indicates that family poverty (Hackman et al. 2015; Li et al. 2017), family stress (King et al. 2012), and family stress in the context of family poverty (Blair 2010) affect baseline levels of self-regulation in childhood and early adolescence. Low self-regulation is associated with engaging in risky and impulsive sexual behaviors (Birthrong and Latzman 2014). The majority of research has been conducted in early and middle childhood samples and the development of self-regulation throughout adolescence is less well understood. Longitudinal growth models are needed to assess the relationship between family financial stress with growth in adolescent self-regulation and later sexual risk-taking behaviors.

The purpose of this study was to assess the role of family financial stress on baseline levels of adolescent self-regulation at age 13 and in the growth of self-regulation from age 13 to 17 years, and then to assess the effects of baseline and growth of self-regulation on young adult risky sexual behaviors. This study addressed four primary aims. (1) The first aim was to assess whether adolescent self-regulation improved over time. Given that adolescence is a sensitive period for brain development relating to self-regulation (Fuhrmann et al. 2015), we hypothesized that parent and adolescent self-report of self-regulation would increase between the ages of 13 to 17 years. (2) Next, we assessed whether family financial stress in early adolescence impaired baseline self-regulation. Previous work demonstrated that family financial stress when adolescents were 13–16 years impaired self-regulation 2 years later (Crandall et al. 2017). We hypothesized that, similar to these results, higher family financial stress at baseline (when the adolescent was 13 years old) would be associated with lower adolescent baseline self-regulation. (3) Relatedly, we aimed to test whether family financial distress affected the rate of growth in self-regulation over time. King and colleagues (2012), found that adolescents exposed to stress had lower baseline self-regulation but developed self-regulation more quickly over time. We hypothesized that adolescents exposed to family financial stress would have greater growth in self-regulation over time than adolescents who were exposed to less family financial stress at baseline. (4) The final aim of this study was to assess the overall effects of adolescent self-regulation (baseline and growth) on risky sexual behaviors at age 18 years. We hypothesized that higher baseline self-regulation and increased growth in self-regulation would both predict lower report of adolescent sexual risk-taking at age 18.


The sample included 470 adolescents and their parents who were from a large northwestern city in the United States and participated in the Flourishing Families Project. The Flourishing Families Project was initiated in 2007 and is an ongoing longitudinal study assessing family life and adolescent well-being throughout adolescence and the transition to adulthood. In this sample, slightly more than half (52%) of adolescent respondents were female, and 67% of families were of European American ethnicity. The majority (66%) of parent respondents reported that they were married at baseline, 60% of the primary responding parents reported that they had a bachelor’s degree or higher, 10% of the sample had an annual income less than $20,000, and 35% had an annual income of at least $75,000 (see Table 1 for full participant demographics). This study included six waves of data, beginning when adolescent respondents were age 13 years and with annual follow-ups through age 18 years. Retention was high between waves, with >90% retention rates in most waves. Attrition was unrelated to income, and in many cases, participants who did not participate in one wave returned in a future wave.
Table 1

Demographic characteristics of the sample (N = 470)


Percent (%)





Responding parent married at baseline


Responding parent highest level of education

 <High school


 High school


 Some college




 Graduate degree


Family annual income



 $20,000 to $49,999


 $50,000 to 74,999


 > or equal to $75,000



The majority of participants (n = 410) were randomly recruited into the survey using Polk Directories/InfoUSA, a national survey database with household information based on landline telephones, magazine subscriptions, open voting records, land ownership records, and personal home contacts. These families came from census tracts that were socioeconomically representative of the local school districts. Because the initial sample underrepresented lower SES and minority families, the remaining 60 families were recruited into the study through referrals and flyers to increase the diversity of families. This approach increased the socioeconomic and racial diversity of the sample. In the earlier waves, two interviewers visited each family’s home to interview the adolescent and both of their parents (or one parent in one-parent households). Later waves of the survey were placed online as adolescents began to transition out of their parent’s household. Only the responses of the adolescent and the primary responding parent, usually the mother, were included in this study. The procedures of the Flourishing Families Project have been further described previously (Crandall et al. 2017).


Risky sexual behaviors

Adolescent risky sexual behaviors were measured when the respondent was 18 years old. Adolescents responded to five items from the Sexual Risk Taking with Uncommitted Partners subscale of the Sexual Risk Survey1 (Turchik and Garske 2008). This scale had high reliability in college-age samples (α = .88) (Turchik and Garske 2008). The items included number of sexual partners, untrustworthy partners, having sex with a stranger, relationship commitment, and failure to discuss sexual histories before having sex over the past 6 months. Responses for all items were positively skewed because just under half (46.44%) of respondents reported that they had not been sexually active in the past 6 months. To account for the skewed data, we created categorical variables based on the response distribution for each item. Number of sexual partners was coded into quartiles (0 = no sexual partners, 1 = 1 sexual partner, 2 = 2–3 sexual partners, 4 = ≥4 sexual partners in the past 6 months). Having sex with a stranger and having sex with a partner you did not trust and were both dichotomized (0 = did not have sex with a stranger/partner you did not trust; 1 = had sex with at least one partner who was a stranger/you did not trust). Having sex with a partner that you were not involved in a relationship with and having sex with a new partner before discussing sexual histories were both categorized into tertiles (0 = none, 1 = one partner, 2 = ≥2 partners in the past 6 months). Internal reliability for the current sample was good (α = .82).

Family financial stress

Family financial stress was measured when the adolescent was 13 years old. One parent, usually the mother, reported on seven items that measured parents’ stress about family finances in the preceding 12 months. Five of the items came from the Chronic Stress scale (Umberson et al. 2005) and two items came from the Family Transitions Project (Spilman and Burzette 2006). Exploratory factor analysis previously suggested that the seven items fit a one-factor structure well (Crandall et al. 2017). The items included worry about their (the parents’) ability to pay bills, make ends meet, debt, and having enough money for housing and healthcare. Parents were asked to rate their concerns using a 4-or 5-point Likert Scale. Higher scores indicated more stress. Cronbach’s alpha for the seven items in this sample was high, α = .89.

Adolescent self-regulation

Adolescent self-regulation was measured over five time points using parent and adolescent self-report (from adolescent age 13 years to 17 years). We used a 12-item revised scale of self-regulation (Novak and Clayton 2001) that included items measuring the adolescent’s ability to regulate emotions, cognitions, and behaviors. The original instrument was designed only for adolescent self-report. To allow for parent report of adolescent self-regulation, items were reworded using the stem “How true is this statement of your child?” Adolescents self-reported on items with the following stem “How true is this statement about you?” Response categories ranged from 1 (Never True) to 4 (Always True). Higher scores indicated higher self-regulation. Consistent with a previous study using this self-regulation data (Crandall et al. 2017), exploratory factor analysis revealed six factors in each wave including parent and adolescent report of emotional self-regulation, cognitive self-regulation, and behavioral self-regulation. Items from each of the six subscales were summed and averaged in each wave. Also similar to the previous study, adolescent self-report of cognitive self-regulation did not fit the self-regulation factor well during exploratory factor analysis and was dropped from further analysis, leaving us with measures from five subscales.


Child gender (1 = female, 0 = male), parent marital status (1 = married, 0 = not married), and sampling technique used to recruit participants (0 = referrals/flyers, 1 = Polk Directories) were included in the model as control variables. Race was not included as a covariate because it was not associated with risky sexual behaviors in this sample.

Statistical Analysis

Descriptive statistics, including item means and distributions, were conducted in Stata 14 (StataCorporation 2014). To set up the measurement model, we conducted separate confirmatory factor analysis on family financial distress (at baseline), adolescent self-regulation at each of the five time points, and risky sexual behaviors (at age 18 years). Confirmatory factor analysis was conducted using a structural equation modeling framework in Mplus Version 7 (Muthén and Muthén 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012) using the robust weighted least squares maximum likelihood estimator, which is appropriate for categorical data. We used the following fit indices and cut-offs to assess model fit: Comparative Fit Index (CFI) > 0.90 indicated adequate fit; Root Mean Square Error of Approximation (RMSEA) < 0.08 indicated adequate fit (Little 2013).

To assess change in adolescent self-regulation over time, we first fit a curve-of-factors longitudinal growth model with the five waves of self-regulation and no covariates. A curve-of-factors model fits a growth curve to the factor scores for each wave of adolescent self-regulation. In a curve-of-factors model, the intercept and slope are fit as second-order factors (Duncan et al. 2006). To do this, we first conducted a factor analysis of the individual observed items (parent and adolescent self-report of emotional self-regulation, cognitive self-regulation, and behavioral self-regulation) to produce the self-regulation factors scores at each time point. We then fit a growth curve to these factor scores. After assessing whether there was significant change (development) in self-regulation over time, we added baseline family financial stress and risky sexual behaviors at age 18 years to the model by regressing the self-regulation intercept and slope on family financial stress and regressing risky sexual behaviors on family financial stress and the self-regulation intercept and slope. For the final model, we added the control variables (child gender, parent marital status, and sampling method). Figure 1 contains the analytic model. For an example of the Mplus code used for this curve-of-factors model and for more information on constraints used in a curve-of-factors models, see Appendix A.
Fig. 1

Analytic model

All models were estimated using a structural equation modeling framework in Mplus Version 7. Family income was modestly correlated to missing data for some of the family financial stress items (Rs = ~.10), with higher income modestly correlated with more missing data for some items and lower income related to more missing data on other items. Family income was not associated with missing data for items relating to self-regulation. Given the small and contradicting correlations between income and missing responses on family financial stress and no relationship between family income and report of adolescent self-regulation, we assumed the data to be missing at random. We used full information maximum likelihood (FIML) to account for missing data. The same model fit indices and cut-offs used for conducting the CFA were applied in the curve-of-factors final model.


Confirmatory Factor Analysis

We conducted confirmatory factor analysis (CFA) on self-regulation, including all five waves of self-regulation in one CFA model. Model fit was good (RMSEA = 0.049, CFI = 0.968) and factor loadings ranged from 0.38 to 0.75. Model fit for the CFA for family financial stress when the respondent was 13 years old was adequate (RMSEA = 0.065, CFI = 0.967), and factor loadings ranged from 0.59 to 0.90. For risky sexual behaviors, model fit was good (RMSEA = 0.007, CFI = 1.000), and factor loadings ranged from 0.88 to 0.94.

Invariance Testing of Self-Regulation over Time

To ensure that the self-regulation measure was invariant across time, we tested for configural, weak, and strong measurement invariance to examine whether the factor loadings and intercepts were equivalent across all five waves of data. If strong equivalence (both factor loadings and intercepts were equivalent over time) was met, then it would be appropriate to examine mean differences in a construct over time. When testing for measurement invariance, Cheung and Rensvold (2002), recommended examining changes in the CFI between models. Changes in CFI of more than .01 would be an indication of measurement non-equivalence. In this study, changes in CFI between configural, weak, and strong invariance models were all less than .01, indicating measurement equivalence in adolescent self-regulation across all five waves of data and that it was appropriate to fit a curve-of-factors model.

Curve-of-Factors Model Results

We next assessed whether adolescents experienced growth in self-regulation over time. The mean slope was 0.03 (p < 0.001), indicating that self-regulation increased between ages 13 and 17 years. There was evidence that adolescents significantly varied in their rate of growth in self-regulation (p < .001). We examined self-regulation growth using both linear and quadratic models. The linear growth model fit the data best and was retained for subsequent analyses.

We then added the parent report of family financial stress when the adolescent was 13 years old; the adolescent report of risky sexual behaviors at age 18 years, and baseline control variables to the growth curve model (see Table 2). Adolescent gender, parental marital status, and how the adolescent was recruited into the study (sampling method) had minimal effect on these model paths. Therefore, we trimmed non-significant control variable paths for a more parsimonious model (final model fit: RMSEA = 0.034, CFI = 0.924). Model results did not vary by gender (χ 2 difference test = 19.18, df = 14, p = 0.16).
Table 2

The longitudinal (adolescents 13–18 years) intersection of family financial distress and adolescent self-regulation on young adult sexual risk-taking, n = 470


Self-regulation intercept

Self-regulation slope (13–18 years)

Sexual risk-taking behaviors—age 18

Baseline family financial stress

−0.23 (0.06)***

0.00 (0.11)

0.09 (0.08)

Self-regulation intercept


−0.35 (0.08)***

Self-regulation slope


−0.23 (0.10)*

Model controls for adolescent gender, parent marital status at baseline, and how the family was recruited into the study (sampling method)

Model fit: RMSEA = 0.034, CFI = 0.924

*p < .05; **p < .10; ***p < .001

Results indicated that although greater family financial stress at baseline was associated with lower baseline self-regulation (−0.23, p < .001), it was not predictive of growth in self-regulation during the teen years, nor was it directly predictive of risky sexual behaviors. Higher baseline self-regulation was correlated with decreased growth in self-regulation between ages 13–17 years (−0.31, p< .01). Baseline higher adolescent self-regulation predicted lower adolescent report of risky sex at age 18 (−0.35, p < .001). Increased growth in self-regulation was also predictive of lower report of adolescent risky sexual behaviors (−0.23, p < .05).

Sensitivity Analysis

In the main model (Table 2), we tested baseline family financial stress as a continuous variable. We conducted a sensitivity analysis using a dichotomous measure of family financial stress to assess if the slope of self-regulation varied in families with high vs. low financial stress at baseline. We examined the intercept and slope of self-regulation comparing families with the mean or higher family financial stress vs. below the mean family financial distress. We also examined families in the upper quartile of family financial distress vs. the rest of the sample. There were no significant group differences in the self-regulation intercept or slope between dichotomous family financial stress groups, and thus we retained the continuous measure of family financial stress.

We also tested a model including the intercept and slope of family financial stress (from adolescent age 13–17 years). The results showed that family financial stress decreased as adolescents aged, however, including the slope of family financial stress did not change the main results of the model paths and worsened model fit.


Prior research has demonstrated that higher adolescent self-regulation (Crandall et al. 2017; Griffin et al. 2012) and lower family financial stress (Ponnet et al. 2015) decrease the likelihood of participating in risky sexual behaviors in older adolescence and young adulthood. It is less clear from extant literature how changes in self-regulation throughout adolescence affect later sexual risk-taking behaviors. Family financial distress has been associated with lower self-regulation in early adolescence (Shonkoff et al. 2012), though the impact of family financial stress on the development of self-regulation throughout adolescence may be more complex (King et al. 2012). In this study, we built on prior literature by assessing the impact of family financial stress on both early adolescent self-regulation and the growth in self-regulation throughout adolescence. Further, we looked at the longitudinal effects of baseline self-regulation and growth in self-regulation on sexual risk-taking behaviors in young adults. We used a curve-of-factors model to account for measurement error and to assess growth in self-regulation over time.

The results indicated that self-regulation modestly improved throughout adolescence, consistent with our aim 1 hypothesis. Family financial stress impaired baseline self-regulation (aim 2), but contrary to our hypothesis for aim 3, family financial stress was not associated with growth in self-regulation. The results indicated that higher baseline self-regulation (at age 13 years) and greater growth in self-regulation from ages 13–17 years predicted lower self-report of sexual risk-taking behaviors in the past 6 months at age 18 years, which was consistent with our hypothesis for aim 4.

Self-Regulation and Risky Sexual Behaviors

Based on these results, lower self-regulation in early adolescence may place young people on a trajectory for greater likelihood of engaging in unhealthy sexual behaviors in young adulthood. However, lower self-regulation in early adolescence does not necessarily doom an adolescent to life-long poor ability to manage their emotions, cognitions, and behaviors. We found that adolescents who catch-up to their peers in self-regulation, or in other words, those adolescents who have greater growth in self-regulation from early to late adolescence, are also less likely to engage in risky sexual behaviors in young adulthood, regardless of their baseline levels of self-regulation. Thus, both higher early self-regulation and greater improved self-regulation are associated with less sexual risk-taking at age 18 years. Likewise, lower early self-regulation and less development of self-regulation throughout adolescence increases the likelihood of self-report of risky sexual behaviors in young adulthood.

Family Financial Stress and Self-Regulation

Family financial stress was associated with lower self-regulation at age 13 years, but fortunately for adolescents coming from stressful, impoverished backgrounds, these social determinants may not negatively affect their development of self-regulation throughout adolescence. These results are consistent with previous studies that indicated that poverty and stress appear to influence baseline levels of self-regulation (Evans and Rosenbaum 2008; King et al. 2012), but not the development of self-regulation in later childhood and adolescence (Hackman et al. 2015). However, extant literature suggests that persistent and chronic stress may overwhelm individual neurobiological development and the associated development of self-regulation (Murray et al. 2015). The current study included a subset of adolescents coming from low-income households, but was largely a lower- to moderate-risk sample. To assess if the results would be similar in more vulnerable populations, this study should be repeated in samples experiencing chronic, elevated stress and poverty.

Implications for Interventions

Our study has important implications for adolescent intervention work to reduce STIs and teenage pregnancies and the associated consequences. Whereas much attention has been paid to improving self-regulation in early childhood (e.g., Tools of the Mind and other educational initiatives that focus on building non-cognitive skills in early childhood and elementary school; see Diamond et al. 2007; Kautz et al. 2014), based on our results, a focus on improving self-regulation among adolescents may also be worthwhile. Consistent with the findings of Murray and colleagues (2015), an individual’s self-regulatory skills are not hard-wired at adolescence. Extensive neurobiological development occurs beyond childhood (Murray et al. 2015), and young people have experiences that provide opportunities to practice self-regulatory skills. This allows for further development of self-regulation and presents opportunities for interventions.

The Harvard Center on the Developing Child (2014) developed a set of strategies to help adolescents improve their self-regulation. Key strategies included encouraging adolescents to identify simple, meaningful goals, talk through the steps of a difficult activity, and break down projects into smaller pieces (Harvard Center on the Developing Child 2014). There are some examples of successful interventions using these strategies, particularly interventions aimed at improving self-regulation in the service of academic achievement. Duckworth and colleagues (2011) developed an intervention designed to help adolescents develop mental contrasting (a process that facilitates goal commitment by helping adolescents to visualize their future goal and contrasting that vision with current potential obstacles that will have to be overcome to achieve the goal) and form implementation intentions (helping adolescents to develop a plan for action to achieve a goal that includes determining the when, where, and how). Adolescents in the intervention group were able to complete 60% more practice test questions for a high-stakes test compared to their peers in a control group (Duckworth et al. 2011). Yeager and colleagues (2014) found that an intervention that helped students to develop self-transcendent purpose for learning (e.g., a meaning behind their learning) improved their academic self-regulation, resulting in better academic performance (Yeager et al. 2014).

Among adolescents coming from the most vulnerable environments, understanding when in the developmental process a stressor occurs and how stressors facilitate or limit the development of self-regulation is critical to introducing interventions that can strengthen self-regulation skills. Such interventions should aim to potentially attenuate and/or reverse the disruptive effects of stressors on one’s neurobiological development, particularly those stressors that are severe and prolonged (Murray et al. 2015).

Strengths and Limitations

A key strength to this study was that we had five waves of parent and adolescent self-report of self-regulation, and thus we were able to assess changes in self-regulation from early adolescence (age 13 years) to later adolescence (age 17 years). However, a limitation of this study was that we only had a questionnaire measure of self-regulation rather than task measures. Task measures are considered the gold standard when measuring volitional aspects of self-regulation like cognitive, emotional, and behavioral control. Questionnaire measures show parent and self-perceptions of adolescent self-regulation, but may not be as effective at assessing some developmental changes associated with brain development that would be expected. In our study, higher baseline self-regulation was associated with less growth in self-regulation over time. This might be partly because adolescents who were already performing well at age 13 years did not have as much room to improve based on the rating scale. Thus, future research using task measures of self-regulation is warranted to verify the results of the current study. Despite the lack of a task to assess self-regulation, having two reporters of self-regulation, measuring multiple aspects of self-regulation (cognitive, emotional, and behavioral), and using latent variable measures to account for bias due to measurement error were strengths of this study.

We were only able to measure risky sexual behaviors at age 18 years because these items were not included in the survey until the 6th wave of the study. This was a limitation of the study given that adolescents coming from disadvantaged communities may be at particular risk for early sexual debut (Cavazos-Rehg et al. 2010). Further, in our sample, race was not shown to be correlated with risky sexual behaviors. Given that previous studies have shown a link between race and risky sexual behaviors regardless of socioeconomic status (Dariotis et al. 2011), future longitudinal studies are needed that assess the longitudinal relationship between the development of self-regulation and risky sexual behaviors throughout adolescence in multiple racial and ethnic groups.


Adolescence is a developmental stage characterized by periods of rapid brain development including the development of higher order cognitive processes such as reasoning and problem-solving (Gestsdottir and Lerner 2008) and also increased exploration that can lead to risk-taking behaviors such as unprotected sex or sex with multiple partners (Steinberg 2008). Sexual risk-taking in adolescence and young adulthood can have long-lasting consequences including infertility due to STIs (Tsevat et al. 2017) or early childbearing and unintended pregnancy that can lead to lower earning power, greater relationship instability, and worse physical and mental health (Kavanaugh et al. 2017). The results of the current study suggest that higher self-regulation in early adolescence and greater growth in self-regulation throughout adolescence decreases the likelihood of engaging in sexual risk-taking behaviors in young adulthood. Greater family financial distress is associated with lower self-regulation in early adolescence, but does not appear to impair the development of self-regulation throughout the adolescent period. These results have important implications for interventions as they suggest that improving adolescent self-regulation is important to decreasing risk-taking behaviors that can have life-long serious consequences. Adolescents coming from families experiencing financial distress may be at increased risk for engaging in later risk-taking behaviors due to lower self-regulation in early adolescence compared to their peers coming from families with less financial distress. However, the results suggest that improving adolescent self-regulation is not only important but also possible and may help to facilitate growth in self-regulation leading to decreased likelihood of engaging in risky sexual behaviors in young adulthood.


  1. 1.

    The current study included these five items from the Sexual Risk Survey, items 8, 16, 17, 19 and 22.



We thank the College of Family, Home, and Social Science, and the many donors and supporters of the Family Studies Center at Brigham Young University who provided generous financial assistance for the Flourishing Families Project for many years. We also thank Jeremy B. Yorgason for his suggestions and careful review of our curve-of-factors Mplus code.

Author Contributions

A.C. conceived of the study and the analytical design, performed statistical analyses and interpretation, and drafted the manuscript. B.M.M. and M.L.B.N. helped with the conceptualization and design of the study and interpretation of results. All authors helped draft the manuscript and read and approved the final manuscript.


The Flourishing Families Project was funded by Brigham Young University (U.S.) College of Family, Home, and Social Science (Principal Investigator: Randal D. Day).

Compliance with Ethical Standards

Conflict of Interests

The authors declare that they have no competing interests.

Ethical Approval

The Brigham Young University Institutional Review Board (IRB) approved the Flourishing Families Project. The Flourishing Families Project involved human participants who provided informed consent in accordance with the procedures established with the institutional ethics committee. This current study was a secondary data analysis using the Flourishing Families Project data.


  1. Barbarin, O. (2013). A longitudinal examination of socioemotional learning in African American and Latino boys across the transition from pre‐K to kindergarten. American Journal of Orthopsychiatry, 83(2–3), 156.CrossRefPubMedGoogle Scholar
  2. Birthrong, A., & Latzman, R. D. (2014). Aspects of impulsivity are differentially associated with risky sexual behaviors. Personality and Individual Differences, 57, 8–13.CrossRefGoogle Scholar
  3. Blair, C. (2010). Stress and the development of self-regulation in context. Child Development Perspectives, 4(3), 181–188. doi: 10.1111/j.1750-8606.2010.00145.x.CrossRefPubMedPubMedCentralGoogle Scholar
  4. Browning, C. R., Burrington, L. A., Leventhal, T., & Brooks-Gunn, J. (2008). Neighborhood structural inequality, collective efficacy, and sexual risk behavior among urban youth. Journal of Health and Social Behavior, 49(3), 269–285.CrossRefPubMedPubMedCentralGoogle Scholar
  5. Cavazos-Rehg, P. A., Spitznagel, E. L., Bucholz, K. K., Nurnberger, J., Edenberg, H. J., Kramer, J. R., & Bierut, L. J. (2010). Predictors of sexual debut at age 16 or younger. Archives of Sexual Behavior, 39(3), 664–673.CrossRefPubMedGoogle Scholar
  6. Centers for Disease Control and Prevention. (2016). Sexually transmitted disease surveillance 2015. Atlanta: U.S. Department of Health and Human Services.Google Scholar
  7. Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling, 9(2), 233–255.CrossRefGoogle Scholar
  8. Crandall, A., Magnusson, B. M., Novilla, M. L. B., Novilla, L. K. B., & Dyer, W. J. (2017). Family financial stress and adolescent sexual risk-taking: The role of self-regulation. Journal of Youth and Adolescence, 46(1), 45–62. doi: 10.1007/s10964-016-0543-x.CrossRefPubMedGoogle Scholar
  9. Crichton, J., Hickman, M., Campbell, R., Heron, J., Horner, P., & Macleod, J. (2014). Prevalence of chlamydia in young adulthood and association with life course socioeconomic position: Birth cohort study. Plos One, 9(8), e104943.CrossRefPubMedPubMedCentralGoogle Scholar
  10. Dahlin, E., Nyberg, L., Bäckman, L., & Neely, A. S. (2008). Plasticity of executive functioning in young and older adults: Immediate training gains, transfer, and long-term maintenance. Psychology and Aging, 23(4), 720–730.CrossRefPubMedGoogle Scholar
  11. Dariotis, J. K., Sifakis, F., Pleck, J. H., Astone, N. M., & Sonenstein, F. L. (2011). Racial and ethnic disparities in sexual risk behaviors and STDs during young men’s transition to adulthood. Perspectives on Sexual and Reproductive Health, 43(1), 51–59.CrossRefPubMedPubMedCentralGoogle Scholar
  12. Diamond, A., Barnett, W. S., Thomas, J., & Munro, S. (2007). Preschool program improves cognitive control. Science (New York, N.Y.), 318(5855), 1387.CrossRefGoogle Scholar
  13. Duckworth, A. L., Grant, H., Loew, B., Oettingen, G., & Gollwitzer, P. M. (2011). Self‐regulation strategies improve self‐discipline in adolescents: Benefits of mental contrasting and implementation intentions. Educational Psychology, 31(1), 17–26.CrossRefGoogle Scholar
  14. Duncan, T. E., Duncan, S. C., & Strycker, L. A. (2006). An introduction to latent variable growth curve modeling: Concepts, issues, and application. Mahway, New Jersey: Lawrence Erlbaum Associates, Inc.Google Scholar
  15. Eaton, D. K., Kann, L., Kinchen, S., Shanklin, S., Flint, K. H., Hawkins, J., & Chyen, D. (2012). Youth risk behavior surveillance—United States, 2011. Morbidity and Mortality Weekly Report: Surveillance Summaries, 61(4), 1–162.Google Scholar
  16. El-Sadr, W. M., Mayer, K. H., & Hodder, S. L. (2010). AIDS in America—forgotten but not gone. The New England Journal of Medicine, 362(11), 967.CrossRefPubMedPubMedCentralGoogle Scholar
  17. Evans, G. W., & Rosenbaum, J. (2008). Self-regulation and the income-achievement gap. Early Childhood Research Quarterly, 23(4), 504–514. doi: 10.1016/j.ecresq.2008.07.002.CrossRefGoogle Scholar
  18. Finer, L. B., & Zolna, M. R. (2016). Declines in unintended pregnancy in the United States, 2008–2011. New England Journal of Medicine, 374(9), 843–852.CrossRefPubMedPubMedCentralGoogle Scholar
  19. Fuhrmann, D., Knoll, L. J., & Blakemore, S.-J. (2015). Adolescence as a sensitive period of brain development. Trends in Cognitive Sciences, 19(10), 558–566. doi: 10.1016/j.tics.2015.07.008.CrossRefPubMedGoogle Scholar
  20. Gestsdottir, S., & Lerner, R. M. (2008). Positive development in adolescence: The development and role of intentional self-regulation. Human Development, 51(3), 202–224.CrossRefGoogle Scholar
  21. Gibbons, F. X., Roberts, M. E., Gerrard, M., Li, Z., Beach, S. R., Simons, R. L., & Philibert, R. A. (2012). The impact of stress on the life history strategies of African American adolescents: Cognitions, genetic moderation, and the role of discrimination. Developmental Psychology, 48(3), 722.CrossRefPubMedPubMedCentralGoogle Scholar
  22. Griffin, K. W., Scheier, L. M., Acevedo, B., Grenard, J. L., & Botvin, G. J. (2012). Long-term effects of self-control on alcohol use and sexual behavior among urban minority young women. International Journal of Environmental Research and Public Health, 9(1), 1–23.PubMedGoogle Scholar
  23. Hackman, D. A., Gallop, R., Evans, G. W., & Farah, M. J. (2015). Socioeconomic status and executive function: Developmental trajectories and mediation. Developmental Science, 18(5), 686–702. doi: 10.1111/desc.12246.CrossRefPubMedGoogle Scholar
  24. Hampson, S. E., Andrews, J. A., Barckley, M., Gerrard, M., & Gibbons, F. X. (2016). Harsh environments, life history strategies, and adjustment: A longitudinal study of oregon youth. Personality and Individual Differences, 88, 120–124.CrossRefPubMedPubMedCentralGoogle Scholar
  25. Harvard Center on the Developing Child (2014). Enhancing and practicing executive function skills with children from infancy to adolescence,
  26. Kautz, T., Heckman, J.J., Diris, R., Ter Weel, B., & Borghans, L. (2014). Fostering and measuring skills: Improving cognitive and non-cognitive skills to promote lifetime success (No. w20749). National Bureau of Economic Research. Paris, France: Organization for Economic Co-operation and Development.Google Scholar
  27. Kavanaugh, M. L., Kost, K., Frohwirth, L., Maddow-Zimet, I., & Gor, V. (2017). Parents’ experience of unintended childbearing: A qualitative study of factors that mitigate or exacerbate effects. Social Science & Medicine, 174, 133–141.CrossRefGoogle Scholar
  28. King, K. M., Lengua, L. J., & Monahan, K. C. (2012). Individual differences in the development of self-regulation during pre-adolescence: Connections to context and adjustment. Journal of Abnormal Child Psychology, 41(1), 57–69. doi: 10.1007/s10802-012-9665-0.CrossRefGoogle Scholar
  29. Li, M., Riis, J. L., Ghazarian, S. R., & Johnson, S. B. (2017). Income, family context, and self-regulation in 5-year-old children. Journal of Developmental & Behavioral Pediatrics,. doi: 10.1097/dbp.0000000000000380. Publish Ahead of Print.Google Scholar
  30. Little, T. D. (2013). Longitudinal structural equation modeling. New York, NY: The Guilford Press.Google Scholar
  31. Lupien, S. J., McEwen, B. S., Gunnar, M. R., & Heim, C. (2009). Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nature Reviews Neuroscience, 10(6), 434–445.CrossRefPubMedGoogle Scholar
  32. Moilanen, K. L. (2015). Short-and long-term self-regulation and sexual risk-taking behaviors in unmarried heterosexual young adults. The Journal of Sex Research, 52(7), 758–769.CrossRefPubMedGoogle Scholar
  33. Mullainathan, S., & Shafir, E. (2013). Scarcity: Why having too little means so much. Macmillan. New York, NY: Times BooksGoogle Scholar
  34. Murray, D. W., Rosanbalm, K., Christopoulos, C., & Hamoudi, A. (2015). Self-regulation and toxic stress: Foundations for understanding self-regulation from an applied developmental perspective. Washington, DC: Office of Planning, Research and Evaluation, Administration for Children and Families, U.S. Department of Health and Human Services. OPRE Report #2015-21.Google Scholar
  35. Muthén, L., & Muthén, B. (1998–2012). Mplus user’s guide. Los Angeles: Muthen & Muthen.Google Scholar
  36. Niendam, T. A., Laird, A. R., Ray, K. L., Dean, Y. M., Glahn, D. C., & Carter, C. S. (2012). Meta-analytic evidence for a superordinate cognitive control network subserving diverse executive functions. Cognitive, Affective & Behavioral Neuroscience, 12(2), 241–268. doi: 10.3758/s13415-011-0083-5.CrossRefGoogle Scholar
  37. Novak, S. P., & Clayton, R. R. (2001). The influence of school environment and self-regulation on transitions between stages of cigarette smoking: A multilevel analysis. Health Psychology, 20(3), 196–207. doi: 10.1037/0278-6133.20.3.196.CrossRefPubMedGoogle Scholar
  38. Ponnet, K., Van Leeuwen, K., Wouters, E., & Mortelmans, D. (2015). A family system approach to investigate family-based pathways between financial stress and adolescent problem behavior. Journal of Research on Adolescence, 25(4), 765–780. doi: 10.1111/jora.12171.CrossRefGoogle Scholar
  39. Shonkoff, J. P., Garner, A. S., Siegel, B. S., Dobbins, M. I., Earls, M. F., McGuinn, L., & Care, D. (2012). The lifelong effects of early childhood adversity and toxic stress. Pediatrics, 129(1), e232–e246.CrossRefPubMedGoogle Scholar
  40. Spilman, S. K., & Burzette, R. G. (2006). Critical Transitions Project Technical Reports, F10 2003. Ames, IA: Family Transitions Project, Iowa State University.Google Scholar
  41. StataCorporation. (2014). Stata statistical software. College Station, TX.: StataCorp LP. Release 14.Google Scholar
  42. Steinberg, L. (2008). A social neuroscience perspective on adolescent risk-taking. Developmental Review, 28(1), 78–106.CrossRefPubMedPubMedCentralGoogle Scholar
  43. Tsevat, D. G., Wiesenfeld, H. C., Parks, C., & Peipert, J. F. (2017). Sexually transmitted diseases and infertility. American Journal of Obstetrics and Gynecology, 216(1), 1–9.CrossRefPubMedPubMedCentralGoogle Scholar
  44. Turchik, J. A., & Garske, J. P. (2008). Measurement of sexual risk taking among college students. Archives of Sexual Behavior, 38(6), 936–948. doi: 10.1007/s10508-008-9388-z.CrossRefPubMedGoogle Scholar
  45. Umberson, D., Williams, K., Powers, D. A., Liu, H., & Needham, B. (2005). Stress in childhood and adulthood: Effects on marital quality over time. Journal of Marriage and Family, 67(5), 1332–1347. doi: 10.1111/j.1741-3737.2005.00220.x.CrossRefPubMedPubMedCentralGoogle Scholar
  46. Volkow, N. D., Baler, R. D., Compton, W. M., & Weiss, S. R. B. (2014). Adverse health effects of marijuana use. New England Journal of Medicine, 370(23), 2219–2227. doi: 10.1056/NEJMra1402309.CrossRefPubMedPubMedCentralGoogle Scholar
  47. Woltering, S., & Lewis, M. D. (2009). Developmental pathways of emotion regulation in childhood: A neuropsychological perspective. Mind, Brain, and Education, 3(3), 160–169.CrossRefGoogle Scholar
  48. World Health Organization (2017). Maternal, newborn, child and adolescent health: Adolescent development,
  49. Yeager, D. S., Henderson, M. D., Paunesku, D., Walton, G. M., D’Mello, S., Spitzer, B. J., & Duckworth, A. L. (2014). Boring but important: A self-transcendent purpose for learning fosters academic self-regulation. Journal of Personality and Social Psychology, 107(4), 559.CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2017

Authors and Affiliations

  1. 1.Department of Health ScienceBrigham Young UniversityProvoUSA

Personalised recommendations