Introduction

The family exerts a powerful influence on the developing child, and this impact may continue into adolescence to an even greater degree than previously believed (Joronen and Astedt-Kurki 2005; Resnick et al. 1997; Schor and American Academy of Pediatrics Task Force on the Family 2003). While the influence of the family on youth development is complex, serving as both an important risk and protective factor, most research has focused on elements of the family environment that increase the likelihood of negative adolescent functioning (Akse et al. 2004; Cuffe et al. 2005; Ge et al. 1996; Hale et al. 2005; Reinherz et al. 1993b, 1995). Fewer studies have critically examined how families may promote healthier outcomes for adolescents. To help fill this gap in the literature, we examined the relationship between potentially modifiable health-promoting family factors and a wide array of developmentally-relevant areas of late adolescent functioning. Gaining a better understanding of the specific health-promoting features of the family will aid in the development of science-based parenting programs to help strengthen families.

Adolescence as a Period of Risk

The adolescent period is marked by an increase in risk behaviors (Arnett 1992, 1999). Many adolescents engage in smoking, alcohol and drug use, physical aggression and violence, and suicidal ideation (Eaton et al. 2006; Johnson et al. 2006; Reinherz et al. 1995). While some consider a degree of problem behavior as age-appropriate as adolescents develop autonomy from their parents, these risky behaviors may nonetheless compromise long-term functioning (D’Amico et al. 2005; Rohde et al. 2001). Furthermore, adolescence is a peak period of onset for many psychiatric disorders, including depression, that may compromise an adolescent’s ability to acquire the life skills needed to make a successful transition to adulthood (Giaconia et al. 2001). Given that the family remains an important social context for adolescents (Kumpfer and Alvarado 2003), it is critical to understand how promoting positive familial relationships during this developmental period may both decrease the risk of engaging in problem behaviors and increase the likelihood of positive functioning.

Linking Adolescent Family Functioning to Healthy Outcomes

Much of the research on the protective aspects of the family has been conducted in samples of high-risk youth, such as those exposed to stressful or traumatic life events (e.g., Bal et al. 2003) or with risk factors for mental disorders (e.g., Carbonell et al. 1998, 2002). Findings generally show that supportive, cohesive families help decrease the likelihood of negative outcomes among these high-risk subgroups (Bal et al. 2003; Carbonell et al. 1998, 2002; Cederblad et al. 1994; Hall-Lande et al. 2007; Tremblay et al. 1999).

There is also growing evidence that strong family connections foster positive development among the larger population of youth without high-risk backgrounds who may nevertheless experience later difficulties. Adolescents who believe they hold an important position as a member of the family and who matter to their family members tend to have higher self-esteem, healthier self-concepts, less depression, and greater overall well-being (Marshall 2001; Rosenberg and McCullough 1981; Taylor and Turner 2001; Youngblade et al. 2007). Moreover, family closeness, communication, and engagement have all been associated with both a reduced risk of negative outcomes among adolescents and an increased likelihood of positive adaptation (Ford-Gilboe 1997; Youngblade et al. 2007). A high level of family cohesion can also protect against the occurrence of serious behavior problems in children (Lucia and Breslau 2006) and has been shown to promote an adolescent’s sense of control over his or her health (Zdanowicz et al. 2004).

Current Study

This study builds on prior research by prospectively examining whether hypothesized health-promoting family factors during mid-adolescence resulted in subsequent adaptive functioning during late adolescence among a community population as a whole, rather than among a select group at risk for poor psychological and social functioning (e.g., abused or depressed youth). We investigated the extent to which the family may promote positive behavior (e.g., good academic functioning) and reduce the likelihood of negative outcomes (e.g., mental and behavior problems). We were also interested in the number and types of adaptive outcomes associated with each family factor. Our focus on potentially modifiable factors was intended to yield practical information that can be used by social workers and other mental health practitioners and educators to develop programs designed to strengthen those aspects of the family environment that promote adolescent well-being.

Method

We examined the relationship between potentially modifiable health-promoting family factors during mid-adolescence (age 15) and indicators of age-appropriate functioning in late adolescence (age 18). The family factors selected for study encompass three major areas: (1) feeling valued in the family, (2) family cohesion, and (3) family social support. These factors reflect characteristics of “healthy families” hypothesized by researchers and practitioners to promote positive development in children and adolescents (Schor and American Academy of Pediatrics Task Force on the Family 2003). Functioning in late adolescence was assessed with a comprehensive array of measures indicating how well adolescents achieved the developmental tasks that characterize a successful transition to adulthood (Schulenberg et al. 2004). These indices of functioning covered four domains: (1) academic functioning; (2) current mental health status; (3) suicidal behavior; and (4) social, psychological, and behavioral functioning. These measures reflect areas of current functioning that may continue to have an impact on quality of life well beyond adolescence into adulthood, such as dropping out of school and mental disorders (Paradis et al. 2006).

Sample

The current analyses draw on data from a community-based study that has traced the life course of a single-aged cohort from childhood to adulthood. These analyses focus on data collected from ages 15 (1987) and 18 (1990) and include the 370 participants, 184 males and 186 females, who participated in both data waves. At age 18 most participants were seniors in high school. Almost all participants were white (98%) and the socioeconomic status of their families was predominately working- or lower-middle class (Reinherz et al. 1993a).

The original 1977 sample included all youth (50% female) entering kindergarten within one public school system participating in state-mandated preschool testing of developmental, academic, and behavioral factors (N = 763). The school district was located in a predominately white, working-class community in the northeastern U.S., and the original sample reflected the composition of the community at that time. Between 1977 and 1990 attrition occurred primarily at the end of kindergarten when students transferred from the public school system to parochial and private schools. Because data collection occurred within the public schools, those transferring out of this school system after kindergarten were purposely excluded from follow-up. Analyses verified that participants who remained in the study at grade 12 did not differ from those lost to follow-up on any key demographic, behavioral, or emotional factors (Giaconia et al. 1994; Reinherz et al. 1993a).

Measures

Information on hypothesized health-promoting family factors and indices of late adolescent functioning was assessed using school records and self-report measures with demonstrated reliability, validity, and/or utility.

Health-Promoting Family Factors at Age 15

Feeling Valued in the Family

Participants’ perceptions of being valued in the family were evaluated by seven true-false items from the Piers-Harris Children’s Self-Concept Scale (Piers 1984). The index included items such as “I am an important member of my family” and “I am a disappointment to my family” (α = 0.47).

Family Cohesion

Self-reported cohesion in the immediate family (i.e., parents and siblings) was measured using the Family Adaptability and Cohesion Evaluation Scales (FACES III) (Olson et al. 1986). The cohesion subscale of the FACES III included 10 items (rated on a 5-point scale from “almost never” to “almost always”) such as “Family members feel very close to each other” and “Family togetherness is very important” (α = 0.82).

Family Social Support

The Arizona Social Support Interview Schedule (Barrera 1980) assessed whether adolescents believed that immediate family members were available to provide advice. The family was considered to be available for support if participants identified at least one parent or sibling in response to the open-ended question: “Who would you go to if a situation came up when you needed some advice?”.

Functioning in Late Adolescence (Age 18)

Academic Functioning

Five dichotomous indicators of academic functioning were based on self-reports and school records. These indicators included both positive aspects of academic functioning (e.g., earning good grades [A or B average] and receiving honors) and academic and school-related difficulties (e.g., failing one or more courses, being suspended or expelled from school in the past year, and dropping out of high school).

Current (1-Year) Mental Disorders

The Diagnostic Interview Schedule, version III-R (DIS-III-R) (Robins et al. 1989), provided diagnoses of current (1-year) disorders for three serious problems among adolescents: (1) major depression, (2) alcohol abuse/dependence, and (3) drug abuse/dependence. The DIS-III-R, which provides diagnoses according to DSM-III-R criteria (American Psychiatric Association 1987), is a structured clinical interview administered by extensively trained interviewers with research or clinical experience (Reinherz et al. 1993a).

Suicidal Behavior

We assessed current thoughts of suicide as well as lifetime suicide attempts. Current suicidal ideation at age 18 was identified using items from the Children’s Depression Inventory (CDI) (Kovacs 1992) and the Youth Self-Report (YSR) (Achenbach 1991). Participants were coded as having suicidal ideation if they endorsed the statement “I think about killing myself but I would not do it” or the statement “I want to kill myself” from the CDI, or if they provided a “somewhat” or “very true” response to the YSR statement “I think about killing myself”. Lifetime suicide attempts by age 18 were determined from a positive response to the DIS-III-R item “Have you ever attempted suicide?”.

Social, Psychological, and Behavioral Functioning

Participant reports of self-esteem were collected using the 10-item Rosenberg Self-Esteem Scale (Rosenberg 1986). Adolescents rated (on a 4-point scale from “strongly disagree” to “strongly agree”) the extent to which they agreed with statements such as “I feel that I have a number of good qualities” and “On the whole, I am satisfied with myself” (α = 0.85). Higher scores reflected increased levels of self-esteem. Interpersonal problems were assessed using a 6-item scale (Reinherz et al. 1993a) that asked adolescents how often during the previous 6 months they experienced problems such as “not having enough close friends” or “having problems communicating with others” (α = 0.75). Higher values indicated more interpersonal problems. Internalizing and externalizing behavior problems were assessed by self-reports on the YSR (Achenbach 1991). The 31 items in the internalizing behavior scale (α = 0.86) encompassed three areas: withdrawn behavior, somatic complaints, and anxious-depressed behavior. The 30 items in the externalizing behavior problems scale (α = 0.83) assessed delinquent and aggressive behavior. Higher values on both scales indicated increased levels of problem behavior.

Statistical Analyses

Because previous work has shown that family factors and psychosocial functioning may differ for males and females (Frost et al. 1999; Ge et al. 1996; Hale et al. 2005; Taylor and Turner 2001), a series of bivariate analyses was conducted to test whether the associations between hypothesized health-promoting family factors and indices of late adolescence adjustment differed by gender. Gender differences were evaluated by examining the significance of the gender-by-family factor interaction term in logistic regression models (for dichotomous functioning outcomes) and linear regression models (for continuous functioning outcomes). Only three significant gender differences were noted: for males only, feeling valued in the family was related to alcohol abuse/dependence (Wald χ2 for interaction term = 4.66, p = 0.03) and externalizing behavior problems (t for interaction term = 2.57, p = 0.01), and again for males only family cohesion was associated with school failure (Wald χ2 for interaction term = 5.61, p = 0.02). In these instances, results were reported separately for each gender. In all other analyses gender was included as a covariate in analyses of covariance, logistic regression models, and linear regression models. An alpha level of 0.05 was used and all significance tests were two-tailed.

Results

Feeling Valued in the Family

As shown in Table 1, feeling that one was a valued family member at age 15 played a significant role both in reducing the likelihood of an array of negative late adolescent outcomes and in promoting healthy psychosocial functioning at age 18. More specifically, those with higher levels of perceived value as a family member had a lower risk of all current psychiatric disorders at age 18 (depression, alcohol abuse/dependence [males only], and drug abuse/dependence), as well as both types of suicidal behavior (current suicidal ideation and lifetime suicide attempts). Adolescents with increased levels of feeling valued in the family subsequently reported significantly higher self-esteem at age 18, fewer interpersonal problems, and less internalizing behavior. For males only, feeling valued by the family at age 15 was also linked to less delinquent and aggressive (externalizing) behavior three years later. This family factor, however, was not significantly associated with any aspect of age 18 academic functioning for all respondents.

Table 1 The relationship between feeling valued in the family and family cohesion and late adolescent functioning
Full size table

Family Cohesion

Self-ratings of family cohesion during mid-adolescence were significantly related to later functioning in three of the four domains examined: academic functioning; current mental disorders; and social, psychological, and behavioral functioning (Table 2). Adolescents who reported higher levels of family cohesion at age 15 were significantly more likely to have maintained an A or B grade average and to have received school-related honors or awards. For males only, those with higher levels of family cohesion were also less likely to have failed one or more courses. Equally striking, higher levels of family cohesion substantially reduced the risk for current alcohol and drug abuse/dependence and were associated with fewer externalizing behaviors (i.e., aggressive and delinquent behavior).

Table 2 The relationship between family social support and late adolescent functioning
Full size table

Family Social Support

Adolescents who believed they could rely on family members for advice fared significantly better than their peers without this type of social support across all four domains of late adolescent functioning (Table 2). More specifically, the availability of parents and/or siblings for advice at age 15 increased the likelihood of receiving school-related honors at age 18 and reduced the risk of receiving a failing grade, suspension or expulsion, and dropping out of high school. The availability of family support at age 15 also substantially reduced the risk for alcohol and drug abuse/dependence. Importantly, having family members to turn to for advice was related to a decreased likelihood of suicidal ideation and lifetime suicide attempts. This health-promoting family factor was also significantly associated with fewer externalizing behavior problems for both genders.

Discussion

Our prospective study illustrates that the family remains an important and vital social context during mid-adolescence (Joronen and Astedt-Kurki 2005). Findings add to an accumulating body of literature underscoring the critical role healthy family relationships in adolescence play in enhancing strong positive functioning and reducing the risk of negative outcomes that may have implications well into adulthood. While there were general influences of the family factors studied here across multiple domains of adolescent functioning, the patterns of association differed by type of health-promoting factor. For instance, feeling valued in the family was the only factor found to significantly reduce the risk of major depression at age 18. In contrast, while both feeling valued in the family and having a family member available to provide advice reduced the risk of suicidal behavior, no association was observed between family cohesion and suicidal behavior. Such specificity has both theoretical and applied applications for designing targeted family programs; it illustrates the need for social workers and other mental health professionals to design interventions aimed at strengthening multiple aspects of the family environment.

Feeling like an important family member during mid-adolescence was linked to multiple aspects of well-being three years later. In comparison to the other elements of the family environment studied here, feeling valued by family members was particularly influential in promoting positive social, psychological, and behavioral functioning, such as increased self-esteem and decreased interpersonal problems. Perceptions of being a valued family member impacted both internalizing (e.g., major depression and withdrawn behavior, somatic complaints, and anxious-depressed behavior as reported on the YSR) and externalizing difficulties (e.g., substance problems) in late adolescence. There was also some indication that this family factor may influence the functioning of males and females somewhat differently. For instance, higher levels of feeling valued in the family decreased the risk of alcohol abuse/dependence and externalizing behavior problems, but only for males. This is in accord with studies that have shown that parental acceptance is more closely related to externalizing behaviors in males than in females (Ge et al. 1996).

As found in our prior work and the work of others (Carbonell et al. 2002; Ford-Gilboe 1997; Lucia and Breslau 2006), a high level of family cohesion promotes a number of positive outcomes in late adolescence and mitigates against negative outcomes. In addition to promoting academic achievement (e.g., good grades), family cohesion is particularly effective in reducing the occurrence of externalizing mental and behavioral problems, including substance disorders and aggressive and delinquent behavior. Unlike feeling valued in the family, cohesion does not impact internalizing outcomes. This again suggests that although key elements of the family environment may be related, they impact different areas of child and adolescent well-being.

Of the three hypothesized health-promoting factors, social support was found to be significantly related to the largest range of late adolescent outcomes, across all four domains. These findings are not surprising in light of previous research (Barrera 1980; Carbonell et al. 2002; Cuffe et al. 2005; Lucia and Breslau 2006). The greatest influence of family support was in the area of academic functioning. Having a family member available for advice promoted both academic success and reduced the risk of poor outcomes (such as dropping out of school) that may have negative long-term repercussions. It is also noteworthy that having a supportive family substantially reduced the risk of current alcohol and drug problems. Researchers have noted the importance of identifying factors that reduce the development of these disorders during adolescence, since early onset of these problems is linked to compromised functioning well into adulthood (D’Amico et al. 2005; Rohde et al. 2001). The important link between the availability of family members for advice and adaptive late adolescent functioning speaks to the need adolescents have for a supportive and nurturing home environment, despite the expected adolescent developmental push for autonomy from parents and family (Steinberg 1998).

Limitations

Several limitations should be noted in interpreting and applying findings from the current analyses. First, because the sample was predominately white and from working- or lower-middle class backgrounds, results may not be generalizable to more racially and economically diverse groups. Second, although health-promoting family factors were measured prospectively at age 15 and clearly predated the assessment of functioning in late adolescence (at age 18), other pre-existing factors may have influenced both the family factors and observed level of functioning in late adolescence. Third, a large number of bivariate statistical tests were conducted in identifying family factors that promoted healthy functioning in late adolescence, increasing the risk for Type I errors. In spite of these limitations, the variables used in these analyses were chosen from the family literature and have practical utility in program planning. The focus on promoting positive outcomes is compatible with the recent thinking in the field (Hawkins et al. 2005; Kumpfer and Alvarado 2003)

Implications and Conclusions

The influence of effective parenting in promoting positive adolescent behavior has been found to be even greater than that of peers and the media in promoting problem adolescent behavior (Joronen and Astedt-Kurki 2005; Kumpfer and Alvarado 2003). Parents must be empowered to understand that a strong family has the ability to overcome the impact of negative influences. As policy and service program advocates, it is critical that we provide accurate, research-based information on what defines a strong family and how to be an effective parent. A lack of this information, coupled with a lack of evidence-based programs offered by community agencies, has resulted in limited opportunities for parents to become more effective in their child-rearing practices and to develop stronger families (Kumpfer and Alvarado 2003). Integrated and comprehensive programs aimed at both preventing negative behavior and fostering positive well-being among children and adolescents should seek to enhance social support, a sense of worth in the family, and cohesion. Resources and adequate information needs to be readily available for social workers and other mental health workers.

It is widely agreed upon that comprehensive, integrated prevention programs produce the most substantial and long-lasting results in enhancing the well-being of children and adolescents (Durlak et al. 2007; Flay and Allred 2003; Weissberg et al. 2003). While such programs may incorporate the school and the community, the role of the family in them is critical as well (Caspe and Lopez 2006). Recent research has also suggested that negative adolescent behavior can be avoided through earlier childhood intervention by strengthening family relationships and dynamics (Kumpfer and Alvarado 2003). It is encouraging that a number of current prevention programs are teaching parents the skills needed to create a cohesive, supportive family environment (Flay and Allred 2003; Hawkins et al. 2005; Kumpfer and Alvarado 2003; Redmond et al. 1999).

A common thread throughout the differing approaches to strengthening families is the overarching goal of enhancing an adolescent’s self-concept (Flay and Allred 2003). A sense of feeling valued by one’s family is a critical component of self-concept. Indeed, our findings show that adolescents who feel valued by their families have higher overall self-esteem and fewer behavior problems than those who do not feel like a valued family member. These results underscore the necessity of communicating to parents and families that it is important for their children to feel that they are important members of their families.

In summary, the current study revealed strong and compelling relationships between health-promoting family factors and late adolescent functioning. Yet the continuing task remains for those who work with youth to translate the findings of studies such as this into viable and on-going programs of health promotion, prevention, and treatment.