Examining Family Structure and Half-Sibling Influence on Adolescent Well-Being

Abstract

Using data from a statewide education project for adolescents (N = 736), this study explored family structure of youth in two-parent families and their influence on coping, sexual activity delay, and substance use. Although research in this area has tended to define family structure by the parent–child relationship, this study emphasized variations in family structure based on sibling relatedness. Results indicate differences when examining “traditional” classifications, as well as classifications based on half-sibling presence, such that those in nuclear families and those without half-siblings are advantaged over other groups. Using groups based on a combination of family structure and half-sibling presence, differences were found between youth in nuclear families and step-nuclear hybrid families (biological and stepchildren). Biological and stepchildren in step-nuclear hybrid families did not significantly differ on any measure. A race by family structure interaction effect was seen for measures of coping. On average, differences were seen for European Americans but not for African Americans.

Keywords

adolescent well-being family structure half-siblings ethnicity

In the past several decades, children’s experiences in diverse family types have become increasingly common. Because of nonmarital births and parental relationship instability, estimates are that many children will spend at least a portion of their life in a single-parent home and as many as one half will be part of a stepfamily during their growing years (Bumpass, Kelly Raley, & Sweet, 1995; Cherlin, 2010; Parker, 2011). The likelihood of living in a “nontraditional” family structure is increased for African American children and those in low socioeconomic status (SES) families (Cherlin, 2010; Parker, 2011).

The instability of the family environment and changing family structure are strong focal points for research on adolescent development. Findings from decades of comparative studies consistently show that children growing up in stepfamilies or single-parent families have, on average, worse academic, behavioral, and psychological outcomes; lower educational and SES attainment; and greater incidence of delinquent behavior, teen pregnancy, and substance abuse than children in nuclear families (Cherlin, 2008; Coleman, Ganong, & Fine, 2000; Halpern-Meekin & Tach, 2008; Hetherington et al., 1999; Sweeney, 2010).

Heterogeneity in Family Structure Based on Sibling Relationships

The research on family structure and child outcomes, however, is limited due to broad measurement and definitions of family structure. The majority of family structure research defines family type by the relationship each child has with the adults in the household; it does not consider the variations within families based on sibling relationships. This has resulted in misclassifications, particularly of children living with both their biological parents who have a half-sibling. A “mutual child” in a stepfamily has typically been classified as living in a two-biological parent family (i.e., nuclear family), whereas his or her half-sibling is classified as living in a stepfamily (Apel & Kaukinen, 2008; Gennetian, 2005; Ginther & Pollak, 2004; Halpern-Meekin & Tach, 2008; Strow & Strow, 2008).

The consideration of sibling relationships and their effect on child and adolescent outcomes is a growing but still newer area of investigation in the broad literature on adolescent well-being. Recent studies find both the presence and the quality of sibling relationships to be associated with elements of emotional, cognitive, and behavioral development (e.g., Deater-Deckard, Dunn, & Lussier, 2002; Linares, MiMin, Shrout, Brody, & Pettit, 2007; Padilla-Walker, Harper, & Jensen, 2010). Importantly, most studies examining the link between sibling relationships and adolescent well-being have primarily utilized samples of biologically related “full” siblings. Although some recent research explores varying degrees of solidarity (functional, associational, normative, and affectual) among full, half, and step siblings (Bedford & Diderich, 2009), much remains to be known about sibling relationships within different family types and the role they play in child and adolescent outcomes (Deater-Deckard et al., 2002; Hetherington et al., 1999).

It seems particularly important that studies consider the unique family contexts of first nuclear families and subsequent nuclear families and distinguish between children raised by two biological parents in a first family (with their full siblings) and those in a “step-nuclear hybrid” family (with half-siblings from one or both parent’s previous relationships). Most likely, mutual children in a step-nuclear hybrid family experience a unique family structure in that they are part of a nuclear family within a stepfamily. Although mutual children have not experienced the divorce or separation of their parents, they differ from children in first families because they have complex sibling relationships and one or both of their parents are in a remarriage (Halpern-Meekin & Tach, 2008). Similarly, stepchildren are also commonly classified as living in a stepfamily, without distinguishing those with and without half-siblings.

Theoretical Framework for Current Study

Theories guiding the study of the effects of family transitions on child development tend to focus on the parent and the parent–child context and center on parent–partner transitions and events (e.g., bioevolutionary theory, resource theory, stress theory, and selection theory; Ginther & Pollak, 2004; Halpern-Meekin & Tach, 2008; Hetherington et al., 1999). However, sibling events can contribute to child/adolescent outcomes as well (Halpern-Meekin & Tach, 2008). One of the biggest family transitions can be the addition of a new child into the family. Boss (1980) found the birth of a baby to be a “normative stress event.” Although the introduction of a new child into any type of family will likely disrupt family routines and result in the reassignment of family roles and boundaries, the birth of a child can be particularly stressful for extant children in a stepfamily (Schlomer, Ellis, & Garber, 2010).

Recent work (Schlomer et al., 2010) suggests that “parent–offspring conflict theory” (POCT; Trivers, 1974) has utility as a means of understanding complex sibling relationships and their effects. POCT explicates a cognitive-behavioral process based on evolutionary biology assumptions that genetic asymmetries and divergence between expected investment and actual investment creates conflict between parents and children. A basic tenet of POCT is that parent–child conflict will be greater in multiple sibling families because of the disconnect between the amount of parental resources invested in each child and the amount of investment expected by the child. Furthermore, the suggestion that parental investment is based on biological relatedness informs the assumption that there will be less conflict among those who are biologically related than among those who are less biologically or nonbiologically related. Thus, it is expected that the introduction of a younger half-sibling will result in higher levels of conflict between extant children (i.e., older children in the family) and their biological parent than will the introduction of a younger full sibling. It is also expected that there will be greater conflict between the stepparent and stepchild (i.e., nonbiologically related parent–child dyad). It is therefore reasoned that extant children (i.e., the stepchild in a step-nuclear hybrid family) will experience more conflict with both residential parents and should fare worse on measures of well-being than stepchildren without a half-sibling and will fare worse than the younger half-sibling (i.e., the biological child in a step-nuclear hybrid family; Schlomer et al., 2010).

Family systems theory, however, suggests that what happens in one family subsystem affects every other subsystem (Minuchin & Fishman, 1981). The conflict experienced by the stepchild with his/her biological parent and with the stepparent may spill over and negatively affect the biological child in the hybrid family as well. A systems view suggests that biological children in step-nuclear hybrid families may be at risk for the same negative outcomes experienced by stepchildren in the same family (Dupuis, 2010; Minuchin & Fishman, 1981) and at greater risk than biological children in a first nuclear family. Compared with a child growing up in a first family, the mutual child in a step-nuclear hybrid family may be exposed to added tensions and stress, such as parental conflict with a former spouse, conflict between stepparent and stepchild, increased conflict between parent and older half-sibling, and stress related to sharing resources across households with former partners (i.e., economic strain; Strow & Strow, 2008). As selection theory suggests, it also may be that mutual children in stepfamilies may be more likely to have parents with the negative characteristics associated with divorce and relationship instability and may therefore be more likely to experience less positive parenting than the average child in a first nuclear family (Halpern-Meekin & Tach, 2008).

The study of mutual children in stepfamilies also benefits from concepts in social learning theory, which suggests that a key mechanism of behavioral development is modeling the behaviors and attributes of important others (Bandura, 1969; McHale, Bissell, & Kim, 2009). Although it is possible that more socially responsible adolescents are setting more positive examples for their younger half-siblings who then respond favorably through imitation (Hetherington et al., 1999), research indicates that family transitions may put older stepchildren at risk for delinquent and negative behavior (e.g., Halpern-Meekin & Tach, 2008; Sweeney, 2010); thus, there is an increased chance that a mutual child in a stepfamily may be exposed to negative role models, even if the relationship quality between siblings is high (Hetherington et al., 1999). Thus, multiple complementary theoretical viewpoints suggest that children with half-siblings may be at increased risk for negative outcomes than those without.

Empirical Studies of Children in Step-Nuclear Hybrid Families

In recent years, a handful of studies have emerged that examine the experiences of stepchildren and biological children with complex sibling relationships (Evenhouse & Reilly, 2004; Gennetian, 2005; Ginther & Pollak, 2004; Halpern-Meekin & Tach, 2008; Strow & Strow, 2008; Tillman, 2008). To date, evidence shows that children and adolescents raised by both of their biological parents in step-nuclear hybrid family households have, on average, worse outcomes in terms of educational achievement, behavior, and psychological well-being than children raised in nuclear family households (Gennetian, 2005; Ginther & Pollak, 2004; Halpern-Meekin & Tach, 2008; Strow & Strow, 2008). In addition, a few studies investigated and found no differences in outcomes between mutual children and step-children (Halpern-Meekin & Tach, 2008; Strow & Strow, 2008). Findings from one study also indicate differences between children in a stepfamily with and without half-siblings (Strow & Strow, 2008). Those with half-siblings experience greater conflict with their stepparent and their biological parent (Schlomer et al., 2010; Stewart, 2005) and demonstrate worse outcomes compared with stepchildren in a stepfamily without half-siblings (Strow & Strow, 2008).

Current Study

All recent studies of family structure that explore the presence or influence of complex sibling relationships suggest increased attention in this area to replicate and extend the findings, given the considerable number of children with hybrid family relationships and the still limited attention in the research and dearth of efforts to capture the full diversity of children’s family structure experiences. In particular, mutual children in step-nuclear hybrid families have been “lost” in most family structure studies; their experiences overshadowed by those of biological children in first families. Emerging evidence indicates that children in these hybrid families are at greater risk for negative outcomes. The current study builds on previous work, utilizing a community-based, ethnically balanced, recent sample of youth and focuses on a range of outcomes (i.e., coping, sexual activity delay, and substance use). The study focused on youth in two-parent families, both married and cohabiting. Comparisons that consider both “traditional” family structure categories and the presence of a half-sibling were conducted.

In addition, the role of race and gender was examined rather than controlled for as in most previous work. Only one study considered the effect of gender and found some distinctions in outcomes (Tillman, 2008); no previous study of family structure and complex sibling relationships has considered the possibility of racial differences. Scholars note that African American youth may be comparatively more accustomed to transitions and may be exposed to adaptive family processes that more readily incorporate nonbiologically related members into the family system based on sociohistorical experiences (Crosbie-Burnett & Lewis, 1993; McLanahan & Sandefur, 1994). Some very recent research finds no differences in well-being indicators in four large samples of African American youth in nuclear and stepfamilies (Adler-Baeder et al., 2010). It may be that having a half-sibling does not increase the risk of negative outcomes for African American youth. Comparisons that consider race are warranted.

Method

Sample and Procedure

This research was approved by the institutional review board and conducted in accordance with ethical standards for voluntary participation, informed consent, and confidentiality. Data were collected by family resource centers located in eight counties in a Southern state. Three of those counties are considered rural, whereas the others are considered urban. Each family resource center was responsible for recruiting youth to participate in the relationship education classes that were held in both school-based and nonschool-based settings. Data were collected from various sites including middle or high schools, churches, after school programs, and summer camps. The surveys were administered by educators trained in evaluation and data collection procedures. Informed consent letters, master code lists containing participant information, and completed surveys were mailed to the university and processed by the project research team.

The sample for this study was drawn from 1,417 adolescent youth who completed preprogram surveys in the fifth year of the project (2010-2011) since these surveys contained more detailed family structure questions. The analytic sample was restricted to those living in two-parent households (both married and cohabiting), leaving 817 adolescents. Furthermore, because of small numbers and because racial comparison analyses were planned, ethnicities other than European and African American were not used, resulting in a final study sample size of 736 adolescents.

Sixty-eight percent of adolescents in the sample identified as European American youth (n = 498) and 32% as African American youth (n = 238). The average age of participants was 15.67 (median = 15; range = 9, SD = 1.07). Gender composition was 56% female (n = 410) and 44% male (n = 321). Participants were asked about their parent’s education. For mothers, 13% did not complete high school (n = 92), 24% completed high school only (n =174), 22% completed some college (n = 158), 9% completed a 2-year college program (n = 67), 22% completed 4 years of college (n = 159), and 10% had postcollege education (n = 70). For fathers, 17% did not complete high school (n = 121), 33% completed high school only (n = 234), 18% completed some college (n = 126), 10% completed a 2-year college program (n = 74), 15% completed 4 years of college (n = 104), and 7% had postcollege education (n = 52).

Measures

Demographic Variables

Participants completed items on the survey indicating their age, gender, and ethnic background. Educational level of the participant’s mother was obtained and used as a proxy for SES. Mother’s educational attainment is used rather than father’s because of research indicating that children are more likely to live with their mother than father after a separation or divorce (Kelly, 2007). Adolescents were asked to circle one of the following with regard to their parents’ education level: less than high school, completed high school, some college, 2-year college/technical school degree, 4-year college degree, or postcollege degree (e.g., master’s, PhD, MD, EdD). Responses were coded as a continuous variable from 1 to 6; higher values indicate a higher level of education and, relatedly, higher SES. Age was coded as a continuous variable and represents the actual age in years reported. Gender was coded with 1 indicating girls and 0 indicating boys. Ethnic background was coded with 1 indicating European American and 0 indicating African American. Parent marital status was coded with 1 indicating that the two parents were married and 0 indicating that they were cohabitating. Gender, ethnicity, participant age, SES, and parent marital status were used as controls for all analyses.

Family Structure Groupings

For the first set of analyses, participants were classified into traditional family structure categories of two-parent nuclear family or stepfamily based on their reported relationship to their parents. In the sample, 379 (51%) are classified as living in a two-parent nuclear family and 357 (49%) as living in a stepfamily. For the second set of analyses, participants were classified as either having a half-sibling or not having a half-sibling (i.e., coded 0, 1 for half-sibling presence), based on their responses to family member items. Of those in the sample, 362 (49%) participants reported having a half-sibling.

For the final two sets of analyses, respondents were categorized into four complex family structure groups based on sibling relatedness and parent marital status. Adolescents were considered to be in a nuclear family type if they were currently living with both biological parents and were fully biologically related to all siblings in the household (n = 295 [40%]; 79% European American, 21% African American). Biological children in a step-nuclear hybrid family were currently living with both biological parents and had at least one half-sibling (n = 84 [12%]; 59% European American, 41% African American). Stepchildren in step-nuclear hybrid families were currently living with one biological parent and had at least one half-sibling (n = 238 [32%]; 61% European American, 39% African American). Finally, stepchildren in simple stepfamilies were currently living with one biological parent and had no half-siblings (n = 119 [16%]; 59% European American, 41% African American).

Coping

The Problem-Focused Style of Coping (Heppner, Cook, Wright, & Johnson, 1995) is an 18-item multidimensional measure of problem solving (i.e., an indicator of social skills) consisting of three subscales: Reflective Style, Reactive Style, and Suppressive Style. Only the Reflective Style (e.g., “I consider the short-term and long-term consequences of each possible solution to my problems”) and Reactive Style (i.e., “My old feelings get in the way of solving current problems”) were collected and used in this study. Respondents were asked to rate how often they engaged in each item across a 5-point Likert-type scale ranging from 1 (almost never) to 5 (almost all of the time). Evidence for the construct validity and utility of the Problem-Focused Style of Coping scales has been reported in Heppner et al. (1995; α = .77 for Reflective; α = .73 for Reactive). Chronbach’s alpha reliability for the current study was α = .67. Although reliability of .70 or higher is generally most desirable, values between .60 and .70 are not unacceptable (DeVellis, 1991). Reactive Style items were reverse coded and combined with the Reflective Style items; thus, higher scores on this scale indicated higher coping.

Sexual Activity Delay

Participants were asked six questions about their intent to wait to have sex until they are older (Gardner, Giese, & Parrott, 2004). Using a 5-point Likert-type scale, responses ranged from strongly disagree (1) to strongly agree (5). Questions included items such as “I intend to have sex while I am a teen (reverse coded)” and “I intend to wait to have sex until I can handle the things that may result from having sex.” Possible scores ranged from 5 to 25, with higher scores indicating greater intent to wait. Evidence for the construct validity and utility of the scale has been reported in Gardner et al. (2004; α = .81). Chronbach’s alpha reliability for the current study was α = .80.

Alcohol and Drug Use

Participants were asked to answer 10 questions about their substance use (adapted from Winters & Zenilman, 1994). Questions assessed current use, perspectives on use, and difficulties with discontinuing use and include items such as “Have you used alcohol (such as wine, beer, hard liquor)?” “Have you used drugs (such as pot, coke, heroin or other opiates, uppers, downers, hallucinogens, or inhalants)?” “Has anyone ever suggested that you might have an alcohol problem?” and “Have you tried to cut down or quit using drugs?” Responses are either yes (1) or no (0). Response values were added up to calculate a risk level for each participant. Scores ranged from 10 (highest risk) to 0 (no risk).

Results

“Traditional” Family Structure Categories

A one-way between-groups analysis of covariance (ANCOVA) was conducted to compare the differences in measures of coping, sexual activity delay, and alcohol and drug use between two groups of adolescents based on “traditional” family structure categories (Table 1). There was a significant difference between nuclear and stepfamily groups on Sexual Activity Delay, F(1, 702) = 4.25, p = .040, partial η² = .006, and Alcohol and Drug Use, F(1, 664) = 13.1, p = .000, partial η² = .019. On average, adolescents in stepfamilies were less likely to delay sexual activity and more likely to engage in alcohol and drug use than adolescents in nuclear families. There was no significant difference between nuclear and stepfamily groups on Coping, F(1, 714) = 2.4, p = .122, partial η² = .003.

Table 1.

Between-Groups Analysis of Covariance of Outcomes on All Categories of Family Structure^a.

	Coping				Sexual activity delay				Alcohol/drug use
Group	M	SD	F	Sig.	M	SD	F	Sig.	M	SD	F	Sig.
Traditional categories			2.40	1.22			4.25	.040*			13.11	.000***
Nuclear (n = 379)	3.35	0.79			3.58	.92			1.26	1.54
Stepfamily (n = 357)	3.15	0.79			3.35	.94			1.88	1.77
Half-sibling categories			4.49	.034*			5.12	.024*			8.42	.004**
Half-sibling (n = 362)	3.17	.79			3.36	.92			1.80	1.79
No half-sibling (n = 374)	3.34	.76			3.57	.95			1.34	1.55
Complex categories			2.14	.09^~			3.21	.023*			6.1	.000***
Nuclear (n = 295)	3.4	.77			3.61	.94			1.16	1.48
Bio in hybrid (n = 84)	3.19	.75			3.46	.82			1.62	1.67
Step in hybrid (n = 238)	3.12	.81			3.29	.93			1.91	1.84
Simple step (n = 119)	3.22	.77			3.46	.96			1.82	1.62

Gender, ethnicity, participant age, socioeconomic status, and parent marital status were used as controls for all analyses.

p < .10. *p < .05. **p < .01. ***p < .001.

Examining the Presence of a Half-Sibling, Regardless of Family Structure

ANCOVA was conducted to compare the differences in measures of coping, sexual activity delay, and alcohol and drug use between two groups of adolescents based on the presence of a half-sibling, regardless of family structure category (Table 1). There was a significant difference between those with half-siblings and those without half-siblings on all three outcomes assessed—Sexual Activity Delay, F(1, 700) = 5.12, p = .024, partial η² = .007, Alcohol and Drug Use, F(1, 664) = 8.42, p = .004, partial η² = .013, and Coping, F(1, 731) = 4.49, p = .034, partial η² = .006. On average, adolescents with half-siblings were less likely to delay sexual activity, more likely to engage in alcohol and drug use than adolescents without half-siblings, and reported poorer coping skills.

Differences in Indicators of Well-Being Based on Combinations of Half-Sibling Presence and Family Structure

ANCOVA was conducted to compare the differences in measures of sexual activity delay, alcohol and drug use, and coping between four groupings of family structure based on parents’ marital/relationship status and the presence of a half-sibling (Table 1). There was a significant difference based on family structure groups on Sexual Activity Delay, F(3, 698) = 3.21, p = .023, partial η² = .014. Least significant difference (LSD) post hoc tests indicated a difference between adolescents in a nuclear family (M = 3.61, SD = 0.95) and stepchildren in a step-nuclear hybrid family (M = 3.3, SD = 0.93). On average, adolescents in nuclear families were more likely to delay sexual activity than stepchildren in step-nuclear hybrid families (i.e., those with half-siblings). There was also a difference found between stepchildren in step-nuclear hybrid families (M = 3.3, SD = 0.93) and stepchildren in simple stepfamilies (M = 3.46, SD = 0.96; i.e., stepchildren in simple stepfamilies were more likely to delay sexual activity). No difference was found between biological and stepchildren in step-nuclear hybrid families or biological children in step-nuclear hybrid families and those in nuclear families. Additionally, there was no difference in sexual activity delay between adolescents in a nuclear family and those in a simple stepfamily.

Similarly, there was a significant difference based on family structure groups on Alcohol and Drug Use, F(3, 662) = 6.91, p = .000, partial η² = .03. LSD Post hoc tests indicated a difference between adolescents in a nuclear family (M = 1.16, SD = 1.48) and biological children in step-nuclear hybrid families (M = 1.62, SD = 1.67), stepchildren in step-nuclear hybrid families (M = 1.91, SD = 1.85), and stepchildren in simple stepfamilies (M = 1.82, SD = 1.62). On average, adolescents in nuclear families were less likely to participate in alcohol and drug use than adolescents in any other family type. There were no differences among the youth in non–first nuclear families.

The test for between-group differences on Coping approached significance, F(3,712) = 2.14, p = .09, partial η² = .009. LSD post hoc indicated a trend toward significant differences between adolescents in a nuclear family (M = 3.4, SD = 0.77) and stepchildren in a step-nuclear hybrid family (M = 3.12, SD = 0.8), such that on average, adolescents in a nuclear family indicated better coping skills than stepchildren in a step-nuclear hybrid family.

Interactions of Family Structure and Gender and Family Structure and Race

To test for interactions of family structure and gender and family structure and race (Research Question 4), 2 (gender) × 4 (family structure) and 2 (race) × 4 (family structure) full factorial models were conducted (Table 2). There were no significant race by family structure interaction effects on Sexual Activity Delay, F(3, 695) = .382, p = .773, partial η² = .004, or Alcohol and Drug Use, F(3, 658) = .942, p = .42, partial η² = .004. Additionally, there was no significant gender by family structure interaction effect on Sexual Activity Delay, F(3, 695) = .1.51, p = .916, partial η² = .001, or Alcohol and Drug Use, F(3, 659) = 1.56, p = .198, partial η² = .007.

Table 2.

The Interaction Effect of Family Structure by Ethnicity on Coping.

	Nuclear (n = 295)		Bio in blended (n = 84)		Step in blended (n = 238)		Simple step (n = 119)
	M	SD	M	SD	M	SD	M	SD	F	Sig.
Interaction									2.65	.048
European Americans	3.47*	0.70	3.10*	0.68	3.15*	0.74	3.22	0.76
African Americans	3.12	0.70	3.26	0.59	3.10	0.75	3.22	0.54

p < .10. *p < .05. **p < .01. ***p < .001.

A significant race by family structure interaction effect on Coping was found, F(3, 539) = 2.65, p = .048, partial η² = .015. Post hoc analyses indicated a difference between European American adolescents in a nuclear family (M = 3.47, SD = 0.70) and European American biological adolescents in a step-nuclear hybrid family (M = 3.09, SD = 0.68) as well as European American stepchildren in a step-nuclear hybrid family (M = 3.15, SD = 0.74). On average, European American adolescents in nuclear families scored significantly higher on measures of coping than European American biological and stepchildren in step-nuclear hybrid families. This same pattern was not evident for African American youth. There was no significant gender by family structure interaction effect for Coping, F(3, 705) = .63, p = .596, partial η² = .003.

Although there was no significant race by family structure interaction effects on Sexual Activity Delay or Alcohol and Drug Use, further exploratory analyses were conducted based on the previous empirical studies that suggest limited outcome differences based on family structure for African American youth in two-parent families (i.e., step and nuclear; Adler-Baeder et al., 2010). Differences were found when examining within subgroups. For European American participants, a significant between-groups difference was found for Sexual Activity Delay, F(3, 48) = 2.86, p = .036, partial η² = .018. LSD post hoc analyses indicated a difference between adolescents in nuclear families (M = 3.66, SD = 0.95) and stepchildren in step-nuclear hybrid families (M = 3.3, SD = 0.94). On average, European American adolescents in nuclear families were more likely to delay sexual activity than European American stepchildren in step-nuclear hybrid families. No differences were found between family groups among African American youth, F(3, 211) = .76, p = .519, partial η² = .01.

Similarly, when examining European American participants, a significant between-groups difference was found for Alcohol and Drug Use, F(3, 447) = 5.1, p = .002, partial η² = .033. LSD post hoc analyses indicated a difference between adolescents in nuclear families (M = 1.18, SD = 1.51) and stepchildren in step-nuclear hybrid families (M = 2.1, SD = 1.95). On average, European American adolescents in nuclear families were less likely to use alcohol and drugs than European American stepchildren in step-nuclear hybrid families. There was no difference found between family groups for African American youth, F(3, 207) = 1.44, p = .233, partial η² = .02.

Discussion

Recent studies of family structure that explore the presence or influence of complex sibling relationships indicate that children in hybrid families are at greater risk for negative outcomes. Traditional classifications of family structure that consider only the biological relationship between the parent(s) in the household and the children fail to capture the complexity of family structure for many adolescents, particularly shared biological children in step-nuclear hybrid families. Although children in this type of family are most commonly classified as living in a nuclear family, their experience is distinct from other children in nuclear families because of the presence of one or more half-siblings. Additionally, using such broad classifications does not allow for the distinction between adolescents in stepfamilies who have half-siblings and those who do not. Therefore, the purpose of this study was to utilize a more nuanced view of family structure by capturing parent–child relationships as well as sibling relationships and to extend previous work by utilizing a community-based, racially balanced, recent sample of youth and focus on a range of outcomes related to health and well-being. Additional unique contributions are the examinations of the potential role of race and gender on family structure comparisons. Another contribution of the study is the explication of theoretical assumptions to guide research related to adolescent outcomes in complex family structures and to frame results.

Traditional Family Structure Categories

Consistent with previous findings (e.g., Cherlin, 2008; Coleman et al., 2000; Halpern-Meekin & Tach, 2008; Hetherington et al., 1999; Sweeney, 2010), comparisons of “traditional” family structure groupings of nuclear and stepfamilies revealed expected differences for two of the three outcomes measured. On average, adolescents living in nuclear families were more likely to delay sexual activity and less likely to use alcohol and drugs than adolescents living in stepfamilies. This finding offers support for bioevolutionary, resource, stress, and selection theories, which all suggest that living in a stepfamily enhances the likelihood of risk-taking behaviors of adolescents (e.g., Cherlin, 2008; Coleman et al., 2000). Although these results were expected, traditional classification of family structure fails to capture the complexity and the variations within these two family structure groups, particularly those involving complicated sibling relationships.

The Effect of Having a Half-Sibling Regardless of Family Structure

To explore the effect of complicated sibling relationships, the study then examined groups based on the presence of a half-sibling, regardless of relationship to the parents. Findings indicate a difference between adolescents with a half-sibling and those without on all three outcome measures. On average, adolescents with no half-siblings were more likely to delay sexual activity, less likely to use alcohol and drugs, and had higher coping skills compared with adolescents with one or more half-siblings. These findings are consistent with previous work (Strow & Strow, 2008; Tillman, 2008) and suggest further attention to sibling relationships is needed since the experiences of those in nuclear families who also have half-siblings are not represented in the nuclear/stepfamily comparison findings.

Of the adolescents who had at least one half-sibling (n = 362), 23% were classified as living in a nuclear family (n = 84) in the first set of analyses, although these youth were actually living in step-nuclear hybrid families (i.e., a different “type” of nuclear family). Grouped one way, they appear to be part of an advantaged group; grouped another way, they appear to be part of a disadvantaged or at-risk group. Neither dichotomous split likely tells their story accurately. Clearly, there is a need to consider both parent−child relationships and sibling relationships when grouping youth based on family context to better understand distinct experiences.

Examining the Effect of Family Structure Using a Combination of Half-Sibling Presence and Relationship to Parents

When comparing the four family structure groups based on sibling composition and relationship to parents, the most advantaged group appeared to be youth in first nuclear families and the comparatively most disadvantaged group appeared to be stepchildren in step-nuclear hybrid families. On average, adolescents in nuclear families were more likely to delay sexual activity than stepchildren in step-nuclear hybrid families. Although not significant, there appeared to be a trend toward higher coping skills for youth in nuclear families than stepchildren in step-nuclear hybrid families. Interestingly, there was also a significant difference between stepchildren in step-nuclear hybrid families and stepchildren in simple stepfamilies (i.e., with no half-siblings) on sexual activity delay, but no difference between youth in simple stepfamilies and youth in nuclear families. This is consistent with results from Strow and Strow (2008) that also found no difference between those in simple stepfamilies and those in nuclear families. This is an important finding that warrants future work as it suggests that previous summarizing empirical “truths” that children in stepfamilies are disadvantaged (e.g., Cherlin, 2008) are inaccurate and do not capture the diversity within stepfamilies. Depending on the sibling constellation and the outcome examined, being in a stepfamily may not elevate the risk for some negative outcomes, such as sexual activity and coping, as much as previously thought.

In the current study, we were particularly interested in exploring outcome differences between youth in nuclear families and biological children in step-nuclear hybrid families. For the study sample, adolescents in nuclear families, on average, were less likely to use alcohol and drugs than biological adolescents in step-nuclear hybrid families (i.e., with half-siblings). These findings are consistent with those of Ginther and Pollak (2004), Gennetian (2005), Tillman (2008), Halpern-Meekin and Tach (2008), and Strow and Strow (2008), who also found significant differences between biological children in step-nuclear hybrid families and those in nuclear families. Not finding distinctions between youth in nuclear families and biological children in step-nuclear hybrid families on the other two outcome measures (i.e., coping and sexual activity delay) may have more to do with the greater diversity in this sample compared with samples in previous studies and is discussed further in the following section on interactions of family structure and race and gender.

Results lend some support for assumptions explicated a priori from bioevolutionary and POCT theories. Combined, the theories lead to the expectation for greater conflict between nonbiologically related family members (i.e., stepparent–stepchild). Thus, it is reasoned that extant children (i.e., the stepchild in a step-nuclear hybrid family) will experience more conflict with both residential parents and should fare worse on measures of well-being than the younger half-sibling (i.e., the biological child in a step-nuclear hybrid family) and stepchildren without younger half-siblings (Schlomer et al., 2010). This theory offers explanation for why, in our sample, the stepchildren in step-nuclear hybrid families fared worse than any other group, but stepchildren without half-siblings were no worse than those in nuclear families on two of the three outcome measures (i.e., coping and sexual activity delay).

The assumptions of family systems theory are only partially supported by the current findings. A family systems perspective suggests similar outcomes for biological children in step-nuclear hybrid families and stepchildren in step-nuclear hybrid families, and this was found for one of the three outcome measures (i.e., alcohol and drug use). Similarly, selection theory was only partially supported by the current findings. Based on this theory, we would expect to see no differences among youth in nonnuclear families (i.e., youth in stepfamilies with and without half-siblings and biological children in step-nuclear families). This was the case for only one of the three outcome measures (i.e., alcohol and drug use) and, in fact, as mentioned above, on measures of coping and sexual activity delay, adolescents in simple stepfamilies fared no worse than those in nuclear families.

Interaction Effects for Family Structure and Race and Family Structure and Gender

It appears that a better picture results when considering race. Patterns of differences among family structures were distinct within racial subgroups. For European American adolescents, differences in coping were found between those in a nuclear family and those in a step-nuclear hybrid family (both biological and stepchildren), such that those in nuclear families showed better coping skills. In contrast, there were no differences found between family structure groups on coping for African American adolescents. This finding for European American youth is consistent with previous studies that also found differences between biological adolescents in step-nuclear hybrid families and those in nuclear families. Because of the diversity in the current sample, this difference was not evident when utilizing the full sample. It is possible, and probable, that had the current sample been more predominantly European American, as in previous studies, the results would have indicated a more pronounced difference between biological children in first nuclear families and biological children in step-nuclear hybrid families.

Similarly examinations within racial groups revealed a pattern of differences between European Americans in nuclear families and European American stepchildren in step-nuclear hybrid families on measures of sexual activity delay and alcohol and drug use, but found no differences based on family structure on either measure among African American youth. Overall, these results highlight the value of examining outcomes for youth based on family structure within cultural context.

The racial differences are consistent with recent suggestions and research that show no differences in well-being indicators across several large samples of African American youth in nuclear and stepfamilies (Adler-Baeder et al., 2010). Suggestions are that African American youth may be more acculturated to inclusivity of both biologically related and nonbiologically related family members. It is also possible that African American youth are more accustomed to transitions and may be exposed to adaptive family processes that are more accepting of new (nonbiologically related) members in the family system (Adler-Baeder et al., 2010; Crosbie-Burnett & Lewis, 1993; McLanahan & Sandefur, 1994). It may be that having complicated sibling relationships does not necessarily increase the risk of negative outcomes for African American youth and that two-parent family structure, whether nuclear or stepfamily, is the critical feature for predicting more positive outcomes (Moore & Chase-Lansdale, 2001).

Limitations

The current study offers some important insights into the experiences of adolescents living in various family structures and with varying biological relatedness to family members. Limitations of the study, however, are acknowledged and some cautions in the interpretations of the findings are suggested. Although the overall sample size is substantial, the subsamples of family structure groups based on sibling composition and relationship to parent are relatively small, particularly when considering other distinguishing characteristics such as race and gender. Furthermore, there is limited variability in the risk-taking measures used, which questions validity, as is common with self-report measures. A possible reason for this is the nature of the measures themselves. When asking about sexual behavior and alcohol and drug use, some youth may have been reluctant to answer truthfully, even with confidentiality assurances. Previous research has found that with most risk behaviors, including sexual activity and substance use, there is often a problem with underreporting. It is assumed that social desirability and/or fear of reprimand may lead some youth to deny the behavior (Dolcini, Adler, & Ginsberg, 1996). Objective measures of outcomes would increase confidence in results. Finally, although this study did include both cohabiting and married two-parent families, no distinction was made between groups in analyses because of the predominance of married couples in our sample. Research indicates that cohabiting unions are often more unstable and shorter in duration than marriages and that this may have negative impact on youth outcomes (e.g., Manning & Lamb, 2003). Therefore, it is recommended that future research further distinguish these differing contexts and examine any between-group differences that may exist.

Conclusions and Future Directions

The current study offers additional information and a step forward for this small, but growing literature on youth outcomes based on parent relatedness and sibling relatedness. Novel contributions to the field are made in several ways. First, the study examines the effect of distinct family structure classifications on various outcomes, extending the work beyond that of the educational/cognitive realm, including the consideration of well-being and risk-taking behaviors. Additionally, unique to this study, the role of race and gender were examined and differences were found based on race, exposing an important and virtually untapped area for future research.

Efforts to expand this literature will inform both the adolescent development research literature as well as practice. Currently, practitioners are operating from an incomplete research base when working with children, youth, and families in complex family structures. Recently, there has been a push in the literature for development and implementation of sibling interventions (Feinberg, Sakuma, Hostetler, & McHale, 2012; Feinberg, Solmeyer, & McHale, 2012; Kramer, 2010) given that sibling relationships may influence adjustment. Although these studies have primarily focused on full-sibling relationships, the findings of this and previous studies suggest it may be important for practitioners to focus on sibling relationships of varying types. It is also important for practitioners to consider cultural context so the information can be used to shape appropriate intervention and education criteria.

Overall, the influence of sibling relationships on adolescent outcomes is an understudied area. In future research it would be beneficial to examine the dynamics and the influence of various sibling relationships within more complex family systems, given the growing prevalence of these family experiences. These families offer the opportunity to understand how relationships with siblings of varied biological relatedness affect indicators of adolescent well-being. We noted in coding for half-sibling presence in the current study that there may be differences in objective accounts of who is in the household and subjective reports of who is considered part of the family. Curiously, a significant portion of youth indicated a parent had a child from a previous relationship, yet did not list any half-siblings when detailing information in the “who is in your family” table. These “boundary ambiguity” issues (Boss, 1980) in step-nuclear hybrid families deserve further exploration for their meaningfulness in predicting outcomes.

Although these responses may indicate boundary ambiguity, it may also be an issue of measurement itself. Perhaps some of those who did not include the half-sibling in the “who is in your family” table have not lived with that half-sibling at any point in their life, or maybe they were not familiar with the term “half.” There are numerous possibilities, and given the potential influence of sibling relatedness, future research should focus on the development of more detailed and accurate family structure measures. Future measures should assess as much detail as is practically possible to better delineate and understand varying experiences and their impact on youth outcomes. This study, like much of the research, uses static measures of family structure at a given time point rather than measures that capture cumulative experience and instability (Wu & Martinson, 1993), which may limit our understanding of family structure as it is linked to adolescent outcomes (Manning & Bulanda, 2007). Researchers should focus on creating measures that will capture the full story of individuals’ current and cumulative experiences.

Additionally, future research is needed that examines stepfamilies in which there are no half-siblings, as adolescents in this type of family did not appear to be disadvantaged compared with youth in nuclear families on several important well-being indicators. As noted, this finding calls into question the common assertion that stepchildren face greater risks. Furthermore, broadening the scope of research to examine sibling composition in single-parent families is needed. It is also imperative that categorical comparison studies give way to within-group process studies to better understand the nuanced mechanisms involved in the development of adolescent well-being. And of critical importance, future studies should aim to involve more diverse samples and examine more nuanced differences within racial groups. Use of an ecocultural lens is important in family studies overall and appears to be highly relevant in studies of complex families.

Finally, the current study offers new insight into the theoretical assumptions used in this area of research. Interesting ethnic differences indicate that although some theories hold for European Americans, no a priori theoretical assumptions were supported for the African American sample. For European Americans, findings partially support POCT, family systems, and social learning theories.

Less support is offered for bioevolutionary, selection, resource, and stress theories, which expect clear advantages for children living with both of their biological parents. For African Americans, no differences were found between groups. For European Americans, no difference was found between biological and stepchildren in step-nuclear hybrid families, suggesting that even biological children in this family type may be at risk for more negative outcomes. Furthermore, there was also no difference between stepchildren in simple stepfamilies and youth in first nuclear families.

The current study offers an initial step forward in building a broader literature on family structure that considers and distinguishes complexity and the child’s unique family context and culture, highlighting the experiences of the previously “unseen.” Dilworth-Anderson, Burton, and Johnson (1993) remind us that nuanced elements and their impact on family life can and do exist “before the observer is aware of it and determines its relevance” (p. 638). Given that a large percentage of American children are growing up in more complex family forms (Cherlin, 2010; Evenhouse & Reilly, 2004), further attention to family structure and family context is warranted. Rather than continuing to use broad family structure categories based solely on the relationship of the parents to the child, efforts to further distinguish family types by considering multiple family characteristics can reveal important differences in children’s experiences and outcomes. Although more parsimonious categorizing of family types may be desirable for statistical and methodological reasons, researchers are encouraged to dive into the complexity of family structure and more finely distinguish children’s family contexts and family structure transactional patterns.

Footnotes

Authors’ Note

Any opinions, findings, conclusions, or recommendations expressed in this article are those of the author(s) and do not necessarily reflect the views of the U.S. Department of Health and Human Services, Administration for Children and Families (the funding agency).

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 was supported through a grant from the U.S. Department of Health and Human Services, Administration for Children and Families (Grant No. 90FM00006).

References

Adler-Baeder

Russell

Kerpelman

Pittman

Ketring

Smith

Stringer

(2010). Thriving in stepfamilies: Exploring competence and well-being among African American youth. Journal of Adolescent Health, 46, 396-398. doi:10.1016/j.jadohealth.2009.10.014

Apel

Kaukinen

(2008). On the relationship between family structure and antisocial behavior: Parental cohabitation and blended households. Criminology, 46(1), 35-70.

Bandura

(1969). Social-learning theory of identificatory processes. In Goslin

D. A.

(Ed.), Handbook of socialization theory and research (pp. 213-262). Chicago, IL: Rand McNally.

Bedford

Diderich

(2009). Adulthood, sibling relationships in. In Reis

Sprecher

(Eds.), Encyclopedia of human relationships (pp. 42-45). Thousand Oaks, CA: Sage.

Boss

P. G.

(1980). Normative family stress: Family boundary changes across the life-span. Family Relations, 29, 445-450.

Bumpass

Kelly Raley

Sweet

(1995). The changing character of stepfamilies: Implications of cohabitation and nonmarital childbearing. Demography, 32, 425-436.

Cherlin

A. J.

(2008). Multiple partnerships & children’s wellbeing. Family Matters, 80, 33-36.

Cherlin

A. J.

(2010). Demographic trends in the united states: A review of research in the 2000s. Journal of Marriage and Family, 72, 403-419. doi:10.1111/j.17413737.2010.00710.x

Coleman

Ganong

Fine

(2000). Reinvestigating remarriage: Another decade of progress. Journal of Marriage and the Family, 62, 1288-1307.

10.

Crosbie-Burnett

Lewis

E. A.

(1993). Use of African-American family structures and functioning to address the challenges of European-American postdivorce families. Family Relations, 42, 243-248.

11.

Deater-Deckard

Dunn

Lussier

(2002). Sibling relationships and social-emotional adjustment in different family contexts. Social Development, 11, 571-590.

12.

DeVellis

R. R.

(1991). Scale development: Theory and applications. Thousand Oaks, CA: Sage.

13.

Dilworth-Anderson

Burton

L. M.

Johnson

L. B.

(1993). Reframing theories for understanding race, ethnicity, and families. In Boss

Doherty

W. J.

LaRossa

Schumm

Steinmetz

(Eds.), Sourcebook of family theories and methods: A contextual approach (pp. 627-646). New York, NY: Plenum Press.

14.

Dolcini

Adler

N. E.

Ginsberg

(1996). Factors influencing agreement between self-reports and biological measures of smoking among adolescents. Journal of Research on Adolescence, 6, 515-542.

15.

Dupuis

(2010). Examining the blended family: The application of systems theory toward an understanding of the blended family system. Journal of Couple & Relationship Therapy, 9, 239-251. doi:10.1080/15332691.2010.491784

16.

Evenhouse

Reilly

(2004). A sibling study of stepchild well-being. Journal of Human Resources, 39, 248-276.

17.

Feinberg

M. E.

Sakuma

Hostetler

McHale

S. M.

(2012). Enhancing sibling relationships to prevent adolescent problem behaviors: Theory, design and feasibility of siblings are special. Evaluation & Program Planning, 36, 97-106. doi:10.1016/j.evalprogplan.2012.08.003

18.

Feinberg

M. E.

Solmeyer

A. R.

McHale

S. M.

(2012). The third rail of family systems: Sibling relationships, mental and behavioral health, and preventive intervention in childhood and adolescence. Clinical Child and Family Psychology Review, 15(1), 43-57.

19.

Gardner

S. P.

Giese

Parrott

S. M.

(2004). Evaluation of the connections: Relationships and marriage curriculum. Family Relations, 53, 521-527.

20.

Gennetian

(2005). One or two parents? Half or step siblings? The effect of family structure on young children’s achievement. Journal of Population Economics, 18, 415-436.

21.

Ginther

Pollak

(2004). Family structure and children’s educational outcomes: Blended families, stylized facts, and descriptive regressions. Demography, 41, 671-696.

22.

Halpern-Meekin

Tach

(2008). Heterogeneity in two-parent families and adolescent well-being. Journal of Marriage and Family, 70, 435-451. doi:10.1111/j.1741-3737.2008.00492.x

23.

Heppner

Cook

Wright

Johnson

(1995). Progress in resolving problems: A problem-focused style of coping. Journal of Counseling Psychology, 42, 279-293.

24.

Hetherington

Henderson

Reiss

Anderson

Bridges

Chan

Taylor

(1999). Adolescent siblings in stepfamilies: Family functioning and adolescent adjustment. Monographs of the Society for Research in Child Development, 64(4).

25.

Kelly

J. B.

(2007). Children’s living arrangements following separation and divorce: Insights from empirical and clinical research. Family Process, 46(1), 35-52. doi:10.1111/j.1545-5300.2006.00190.x

26.

Kramer

(2010). The essential ingredients of successful sibling relationships: An emerging framework for advancing theory and practice. Child Development Perspectives, 4(2), 80-86. doi:10.1111/j.1750-8606.2010.00122.x

27.

Linares

MiMin

Shrout

P. E.

Brody

G. H.

Pettit

G. S.

(2007). Placement shift, sibling relationship quality, and child outcomes in foster care: A controlled study. Journal of Family Psychology, 21, 736-743.

28.

Manning

W. D.

Bulanda

R. E.

(2007). Cohabitation and measurement of family trajectories. In Hofferth

Casper

(Eds.), Handbook of measurement issues in family research (pp.199-219). Mahwah, NJ: Lawrence Erlbaum.

29.

Manning

Lamb

(2003). Adolescent well-being in cohabiting, married, and single-parent families. Journal of Marriage and Family, 65, 876-893.

30.

McHale

S. M.

Bissell

Kim

(2009). Sibling relationship, family, and genetic factors in sibling similarity in sexual risk. Journal of Family Psychology, 23, 562-572.

31.

McLanahan

Sandefur

(1994). Growing up with a single parent: What hurts, what helps? Cambridge, MA: Harvard University Press.

32.

Minuchin

Fishman

(1981). Family therapy techniques. Cambridge, MA. Harvard University Press.

33.

Moore

M. R.

Chase-Lansdale

(2001). Sexual intercourse and pregnancy among African American girls in high-poverty neighborhoods: The role of family and perceived community environment. Journal of Marriage and Family, 63, 1146-1157.

34.

Padilla-Walker

L. M.

Harper

J. M.

Jensen

A. C.

(2010). Self-regulation as a mediator between sibling relationship quality and early adolescents’ positive and negative outcomes. Journal of Family Psychology, 24, 419-428. doi:10.1037/a0020387

35.

Parker

(2011). A portrait of stepfamilies. Retrieved from http://www.pewsocialtrends.org/2011/01/13/a-portrait-of-stepfamilies

36.

Schlomer

Ellis

Garber

(2010). Mother-child conflict and sibling relatedness: A test of hypotheses from parent-offspring conflict theory. Journal of Research on Adolescence, 20, 287-306.

37.

Stewart

S. D.

(2005). How the birth of a child affects involvement with stepchildren. Journal of Marriage and Family, 67, 461-473. doi:10.1111/j.0022-2445.2005.00128.x

38.

Strow

(2008). Evidence that the presence of a half-sibling negatively impacts a child’s personal development. American Journal of Economics & Sociology, 67, 177-206.

39.

Sweeney

(2010). Remarriage and stepfamilies: Strategic sites for family scholarship in the 21st century. Journal of Marriage and Family, 72, 667-684.

40.

Tillman

(2008). “Non-traditional” siblings and the academic outcomes of adolescents. Social Science Research, 37(1), 88-108. doi:10.1016/j.ssresearch.2007.06.007

41.

Trivers

R. L.

(1974). Parent–offspring conflict. American Zoologist, 14, 249-264.

42.

Winters

Zenilman

(1994). Simple screening instrument for outreach for alcohol and other drug abuse and infectious diseases (Vol. 11, Publication No. (SMA) 02-3683). Rockville, MD: Center for Substance Abuse Treatment: US Department of Health and Human Services (SAMHSA).

43.

L. L.

Martinson

B. C.

(1993). Family structure and the risk of a premarital birth. American Sociological Review, 58, 210-232.