RESEARCH DOSSIER
Sibling Relationships and Family Subsystems
Abstract. Sibling relationships are enduring and changing relational systems embedded within the wider organization of family life. This critical narrative monograph examines sibling warmth, conflict, rivalry, power, social learning, identity comparison, caregiving, and differential parental treatment as measurable processes rather than consequences of birth order alone. It traces the historical shift from family constellation studies to multidimensional relationship measurement, reviews family-systems and developmental theories, and evaluates evidence from longitudinal sibling cohorts, observational studies, meta-analyses, evidence maps, and randomized preventive interventions. Major family transitions include the birth of a sibling, adolescence, adult independence, divorce, disability, foster care separation, and bereavement. Research designs are assessed for clustering, informant disagreement, temporal change, family confounding, cultural generalizability, and causal identification. The synthesis finds that close ties and hostility can coexist, parental differential treatment is associated with adjustment but operates within reciprocal family processes, and targeted interventions can improve some sibling and child outcomes. Important gaps remain in representative sampling, adult developmental trajectories, violence differentiation, rigorous longitudinal measurement, and generalizable causal tests. This monograph reports no newly collected data or original meta-analysis.
Keywords: sibling relationships; family systems; sibling subsystems; sibling warmth; sibling conflict; sibling rivalry; parental differential treatment; birth order; sibling socialization; sibling bullying; family transitions; developmental psychology; dyadic analysis; family intervention; longitudinal family research
Research Object, Scope, and Questions
Sibling relationships are enduring family ties produced through shared or intersecting family histories, recurring interaction, and the social meanings attached to kinship. Their psychological organization comprises warmth, affiliation, cooperation, conflict, status, rivalry, perceived fairness, disclosure, and negotiated autonomy. A family subsystem is a recurring configuration of relationships whose members develop patterns of communication and mutual regulation within the larger family. The sibling subsystem has distinctive opportunities for everyday interaction, although its operation depends on caregivers, household arrangements, cultural expectations, and transitions across the life course. Treating relationship quality as a measured process rather than a fixed consequence of family composition is foundational to this research object Furman and Buhrmester (1985) McHale et al. (2012).
This monograph asks how brothers and sisters influence one another, how a sibling dyad is embedded in the other relationships of a family, and which empirical designs distinguish association from causal influence. The primary objects of analysis are the sibling relationship, sibling-pair perceptions, reciprocal developmental processes, and contextual changes after births, parental separation, disability, caregiving demands, and family loss. Peer friendships, parent-child attachment, and clinical diagnoses are examined where they modify sibling processes, while their comprehensive treatment belongs to other scientific dossiers. The discussion includes biological, adoptive, step-, and other socially recognized sibling relationships where evidence permits precise comparison Whiteman et al. (2011) Holmes et al. (2024).
The method is a critical narrative synthesis of identified original studies, longitudinal family cohorts, intervention trials, meta-analyses, methodological reviews, and current scholarship. No original participants were recruited; no pooled effect estimate was calculated, and this work makes no claim to a PRISMA-systematic search. Every interpretive conclusion is tied to the population, informant, and design from which it arose. An observational association between sibling conflict and adolescent symptoms cannot by itself distinguish reciprocal influence, genetic similarities, parental processes, and contextual selection. The aim is a coherent scientific account that keeps these levels of explanation explicit while identifying testable questions for future work Richmond et al. (2005) Jensen and Thomsen (2024).
Historical Emergence of Sibling Research
Early family psychology often examined birth order, number of children, and family constellation as structural characteristics. These variables were useful demographic descriptors but could tell little about the emotional organization of any particular pair. The empirical turn toward relationship quality became clear in Furman and Buhrmester's 1985 investigation, which interviewed children about salient experiences and then developed a questionnaire. Four dimensions emerged in the reported analysis: warmth or closeness, relative status or power, conflict, and rivalry. That distinction changed the unit of study from membership in a family category to an actual relationship whose quality can vary even among structurally similar families Furman and Buhrmester (1985).
Research in the following decades traced developmental change, social learning, family organization, and adjustment. Buhrmester and Furman examined children's changing perceptions across middle childhood and adolescence, opening questions about shifting power and companionship. Later work situated siblings inside ecological and family-systems models, rather than treating their interactions as isolated social encounters. Whiteman, McHale, and Soli organized theoretical perspectives including social learning, attachment-related processes, comparison, and systems thinking; McHale, Updegraff, and Whiteman reviewed two decades of empirical findings and established the sibling relationship as a serious developmental context Buhrmester and Furman (1990) Whiteman et al. (2011) McHale et al. (2012).
This history also contains a methodological warning. Older popular claims about birth-order personalities often traveled further than their supporting evidence. An analysis of large national panels in the United States, Great Britain, and Germany found little enduring birth-order association with broad personality traits, while observing small effects in intelligence-related measures. This result does not eliminate developmental differences associated with an older or younger role in a particular household; it narrows what can be asserted about adult personality from birth position alone. The scientific record thus favors measurement of interaction patterns and context over inference from ordinal labels Rohrer et al. (2015) Jensen et al. (2023).
Family Systems and Subsystem Organization
A family systems perspective treats the family as an interdependent organization of relationships. Parent-child, coparenting, couple, and sibling ties form partially overlapping subsystems; a change in one dyad can alter patterns elsewhere. In a sibling conflict, for example, a caregiver's intervention can create or relieve an asymmetry of perceived support. A parent's work strain can alter availability, affect, and routines, thereby changing sibling opportunity for cooperation or dispute. The family-system claim is an explanatory proposition about reciprocal dependencies and feedback, and its value depends on observing temporal sequences across more than one relationship Whiteman et al. (2011) Feinberg et al. (2012).
Subsystem boundaries organize who participates in particular interactions, which decisions are shared, and whether children are assigned age-appropriate roles. Siblings may coordinate privately, disclose information selectively, ally in negotiations with caregivers, or draw caregivers into disputes. Concepts such as triangulation and spillover describe possible patterns: tension between adults may reappear in child interactions, and a sibling's response can elicit a parental reaction that reshapes the original conflict. These terms should identify observable sequences, not diagnose an entire family from a single incident. Research on sibling influences repeatedly shows the importance of placing dyadic behavior within a wider family constellation McHale et al. (2012) Eckert et al. (2026).
The subsystem model requires temporal and multilevel evidence. A cross-sectional association between unequal parental responses and poor sibling relationship quality is compatible with parental differences affecting sibling conflict, sibling conflict altering parenting, and common family characteristics influencing both. Repeated measurement permits separation of stable differences among families from changes within a family, though unmeasured time-varying causes can remain. A 2026 random-intercept cross-lagged analysis of parental differential treatment and relationship quality yielded a more complicated pattern than a single-direction deterioration hypothesis. Its findings reinforce the importance of feedback, domain specificity, and careful attention to the level at which an estimate is identified Eckert et al. (2026) Richmond et al. (2005).
What Relationship Quality Measures
Sibling relationship quality is multidimensional. Warmth may include expressed affection, shared activities, support, and trust; conflict may include disagreement, hostile behavior, or coercive confrontation; power concerns asymmetries in influence; rivalry concerns competition for valued attention or resources. The dimensions need not move together. Some pairs combine intense closeness with frequent arguments, and children can interpret the same episode differently. The original Sibling Relationship Questionnaire established why compositional variables and qualities of interaction should be measured separately. A rich assessment should therefore permit simultaneous positive and negative qualities rather than forcing a single favorable-to-unfavorable scale Furman and Buhrmester (1985) Buhrmester and Furman (1990).
Measures operate through informants and observation. One sibling may rate the relationship as supportive while the other experiences a burden of caretaking or repeated unfairness. Parents may underestimate private aggression or overestimate cooperation based on visible household routines. Directly coded activities can detect turn-taking and escalation, but observations in a short laboratory task may not represent unobserved behavior across ordinary weeks. Daily diaries, interviews, questionnaires, teacher reports, and repeated home observations each identify different aspects of relational experience. Evidence mapping of 277 studies found substantial heterogeneity in outcome domains and measurement, underscoring why apparent disagreements among studies need examination at the construct level Holmes et al. (2024) Updegraff et al. (2005).
Temporal comparability is an additional challenge. A preschooler's closeness may be expressed through proximity and imitation; an adolescent's through confidential conversation; an adult's through contact, care, and practical assistance. The same item can change meaning across these stages, making invariance tests and age-appropriate operational definitions essential. Furthermore, a pair's average quality and the difference between the siblings' own reports contain distinct information. Treating these differences as mere measurement error can erase the very asymmetries that organize sibling power and perceived fairness. Appropriate analysis distinguishes individual, dyadic, household, and developmental levels McHale et al. (2012) Çineli and Buyukkececi (2026).
Theoretical Explanations and Their Predictions
Social learning accounts propose that children acquire behavioral repertoires through observation, reinforcement, practice, and modeling during sibling interaction. Older children may model language, routines, social tactics, or risky behavior, while younger siblings can also alter the older child's behavior through resistance, requests, and play. A systems account adds reciprocal feedback and the regulation of interactions across family relationships. A social-comparison account emphasizes evaluations of relative standing, achievement, affection, and fairness. These models specify different mechanisms but can operate together in a given dyad: comparison can intensify rivalry while modeled cooperation supplies a means of resolving it Whiteman et al. (2011) McHale et al. (2012).
Developmental ecological explanations emphasize contextual transitions, ages, family resources, and the distribution of attention and care. An infant sibling's birth reorganizes the firstborn's daily environment, parental availability, and caregiving routines. The transition is an opportunity to test changing roles, emotion regulation, and relationships over time rather than to presume an invariant crisis. Volling's empirical review concluded that adjustment responses vary considerably across children and contexts. This heterogeneity is theoretically informative because a good explanation should anticipate why some sibling transitions increase conflict, others promote competence, and still others produce little detectable change Volling (2005) Volling (2012).
Attachment-related and emotional-security perspectives draw attention to reliability, availability, and perceived safety in close relationships. A sibling can be a source of companionship and reassurance, but closeness does not guarantee protection from adversity. Accounts of complementary versus compensatory support make different predictions: complementary support adds benefits to already adequate parent-child relationships, whereas compensation predicts stronger sibling protection when another relationship is strained. The longitudinal postdivorce study by van Dijk and colleagues found a pattern more consistent with complementarity in its specific sample. The inference remains bounded by the observed setting, age range, and measures van Dijk et al. (2022) Feinberg et al. (2012).
Becoming a Sibling: Birth and Early Reorganization
The arrival of a new baby changes a family before the older child can sustain a reciprocal relationship with the infant. Pregnancy, preparation, altered sleeping arrangements, caregiving demands, and parental fatigue may transform routines and expectations. Observing the firstborn only after birth makes it difficult to identify whether apparent behavioral change reflects the arrival itself or an ongoing developmental trend. Prospective studies beginning during pregnancy permit more informative comparisons, although no family experiences a randomized sibling birth. The transition-to-siblinghood literature places the child's adaptation within simultaneous changes in caregivers, economic circumstances, and individual developmental capacity Volling (2005) Volling (2012).
The clinical stereotype that firstborn children inevitably regress when a sibling arrives has received insufficient support as a universal developmental account. Volling's review of thirty studies found variable patterns, including decreases in affection or responsiveness toward mothers but mixed findings for anxiety, aggression, sleep, and regression. A family may report transient distress and later adaptation, or experience substantial difficulties associated with other vulnerabilities. The key question is therefore which change occurred, relative to what baseline, for how long, and for which child. The infant's temperament and the caregiver's responsiveness can also change as reciprocal interaction emerges Volling (2012).
As the younger child develops, the relation becomes increasingly two-sided. An older sibling may gain new opportunities for teaching, protective behavior, and play leadership, while the younger sibling learns to seek attention and resist dominance. The developmental sequence from unequal competence to more reciprocal exchanges is neither mechanically tied to age spacing nor confined to competition. Family routines determine how often children interact, whether conflict is supervised, and which behaviors caregivers reward. A longitudinal account consequently needs simultaneous observations of sibling behavior and family reorganization, rather than assigning causal status to the existence of a newborn alone McHale et al. (2012) Buhrmester and Furman (1990).
Early Childhood: Play, Teaching, and Social Understanding
During early childhood, siblings encounter repeated opportunities to coordinate goals under partially shared rules. Pretend play, household tasks, toy exchanges, and disagreements require each child to predict another's intention and adjust action to an imperfectly aligned partner. Older siblings may provide a more competent model, but the interaction is rarely simple transmission. The younger child can interrupt, misunderstand, negotiate, or redirect the shared activity, thereby altering what the older child learns about patience and control. Such processes make the sibling relationship a distinctive laboratory for observing the development of turn-taking, social inference, and emotional regulation Whiteman et al. (2011) McHale et al. (2012).
A useful analysis distinguishes competence from opportunity. A child who rarely interacts with a sibling cannot display the same frequency of sibling-mediated practice as a child whose household routines create prolonged contact. Comparisons of children with and without siblings may therefore mix opportunity, family size, parental resources, and selection into parenthood. Within-family observation offers detailed behavioral descriptions, yet remains vulnerable to shared unobserved influences. Claims that sibling presence by itself causes superior theory of mind or social competence require more discriminating evidence than a contemporaneous group difference Holmes et al. (2024) McHale et al. (2012).
Early hostile exchanges also need operational precision. A frustrated argument over a toy differs from recurrent intimidation involving a power imbalance; intense conflict may coexist with repair and affiliation. Observation should code initiation, response, escalation, withdrawal, and reconciliation, including differences between the partners' power and age. Caregiver involvement can transform the sequence by supporting problem solving, rewarding demands, or unintentionally reinforcing aggression. Treating the entire interaction as one conflict score conceals these mechanisms. A multidimensional model is better suited to explaining which experiences predict later relational security or externalizing problems Furman and Buhrmester (1985) Feinberg et al. (2012).
Middle Childhood: Negotiating Power and Cooperation
In middle childhood, children generally become more capable of articulate negotiation and independent joint activity, while differences in age, knowledge, autonomy, and physical strength may continue to organize power. Companionship can become sustained through school, leisure, shared interests, and family responsibilities. These activities provide opportunities for the practice of reciprocity, negotiation, and mutual support. Yet a high volume of contact can also increase exposure to conflict. Assessment must therefore separate frequency of interaction from its quality, and affectionate contact from a child's subjective sense of fairness or coercion Buhrmester and Furman (1990) Furman and Buhrmester (1985).
Research using multi-informant and longitudinal designs illustrates why changing relational quality matters. Richmond, Stocker, and Rienks followed 133 families over three waves and examined sibling relationship quality alongside parental differential treatment and children's adjustment. Their analyses linked improvement in sibling relationships with declining depressive symptoms over time. Such results support developmental relevance but do not by themselves settle whether one process changes the other or whether a third factor drives both. The quality of inference depends on adjustment for stable individual characteristics, time-varying household conditions, and the dependence of sibling reports within families Richmond et al. (2005).
A family-systems formulation describes middle childhood as a period in which sibling rules may become increasingly explicit. Children learn which disputes can be settled privately and when adult intervention is likely; repeated patterns can stabilize either cooperation or coercion. The caregiver's role includes setting legitimate limits, protecting less powerful children, and supporting conflict resolution without turning every disagreement into a competition for favor. Randomized family programs suggest that skills can be taught in this age range, although program outcomes vary by domain. Practical implication must therefore follow measured effects rather than a generalized claim that all conflict disappears with training Feinberg et al. (2013) Feinberg et al. (2012).
Adolescence: Autonomy, Identity, and Influence
Adolescent siblings navigate new peer contexts, changing privacy expectations, school transitions, and growing autonomy from caregivers. Their relationship can serve as a source of identity comparison and confidential support while also becoming a channel for imitation of attitudes, activities, and risk behavior. Age spacing and shared schooling affect opportunity, and sibling gender composition may shape expectations within particular cultural settings. A relational approach examines which partner is influential in which domain, rather than presuming that older siblings are always models and younger siblings always followers Whiteman et al. (2011) McHale et al. (2012).
Evidence concerning influence is stronger when both siblings are followed over time and the prior levels of the relevant behaviors are measured. If two siblings report similar experiences of school difficulty or substance use, resemblance can arise from modeling, shared peers, family disadvantage, or shared predispositions. Longitudinal and within-family contrasts can constrain some explanations but require clearly described assumptions. Analysis should also consider reciprocal processes, because adolescents choose activities and identities in response to one another. The scientific significance of sibling influence rests in the interaction among exposure, opportunity, interpretation, and behavioral reinforcement McHale et al. (2012) Feinberg et al. (2012).
Adolescence can alter both closeness and conflict without implying a uniform trajectory. As private worlds expand, time together may decrease even when subjective regard remains high; conflict may diminish with less forced contact, or intensify around family obligations and unequal autonomy. Buhrmester and Furman's developmental study documented variation in children's and adolescents' perceptions. Research in Mexican American families further shows that sibling involvement can be substantial and culturally patterned, making it inappropriate to infer relationship meaning from contact frequency alone. A complete model treats developmental stage, household structure, and culturally organized obligations as interacting variables Buhrmester and Furman (1990) Updegraff et al. (2005).
Adulthood and Later-Life Sibling Bonds
Adult sibling relationships evolve as partners move, form households, become parents, experience career transitions, and assume responsibilities toward aging relatives. Their early shared history remains psychologically salient, but current intimacy depends on contact, communication, conflict, and reciprocal care. Practical assistance may become episodic rather than routine; a low-contact relationship can still carry a strong sense of obligation or belonging. Research therefore distinguishes frequency of communication, felt closeness, conflict, and concrete transfers of help. It also attends to which sibling reports each experience and how geographical distance shapes opportunity Çineli and Buyukkececi (2026) Poortman and Voorpostel (2009).
The family subsystem continues to change when parents separate later in life. In four waves of the German Family Panel, Çineli and Buyukkececi applied fixed-effects analyses to adult sibling ties and reported reductions in contact and emotional closeness after later-life parental divorce, with variation in conflict patterns by sibling gender combination. Their within-person approach controls stable individual characteristics but cannot rule out time-varying circumstances that accompany divorce. The finding illustrates how a change in one generation's couple relationship can reorganize relationships among adult children rather than remaining confined to the divorcing partners Çineli and Buyukkececi (2026).
In later life, caregiving needs, inheritance arrangements, bereavement, and family rituals can increase both cooperation and strain. Some siblings coordinate care and share knowledge of parental needs, while others disagree over responsibilities or revive longstanding interpretations of favoritism. These experiences are influenced by health, mobility, economic resources, gendered labor, and national care systems. The continuity of a kinship tie should not be mistaken for uniform emotional intimacy: relationships can be distant, ambivalent, or supportive at different times. The strongest longitudinal inferences require repeated reports from multiple siblings rather than one participant's retrospective reconstruction of a whole family Poortman and Voorpostel (2009) Çineli and Buyukkececi (2026).
Warmth, Intimacy, and Reciprocal Support
Sibling warmth is expressed through affection, shared humor, responsive attention, practical assistance, and the expectation that a sibling will remain available when circumstances change. These behaviors are distinguishable from mere co-residence or genetic relatedness. The same brother or sister can be a comforting companion in one situation and a competitor in another. Research on sibling relationship quality therefore evaluates warmth and conflict as related but separate dimensions and considers the meaning each sibling assigns to support. A family model that replaces this complexity with one global harmony score may overlook the combinations most relevant to adjustment Furman and Buhrmester (1985) Holmes et al. (2024).
Support can operate through emotional validation, concrete assistance, shared knowledge of the family's history, and opportunities for cooperative regulation. Its value depends on the recipient's needs and on whether help is welcome, reciprocated, or experienced as intrusive. After a family transition, siblings may understand one another's experience with unusual immediacy because they share elements of the household context; their interpretations, however, can diverge. Research after parental separation has found both strong warmth and hostility, sometimes in the same relationship. Such ambivalence should be modeled as a pattern requiring explanation rather than corrected into an assumption of either resilience or dysfunction Sheehan et al. (2004) van Dijk et al. (2022).
Evidence about protective effects must distinguish moderation from independent association. A supportive sibling tie may correlate with better adjustment because nurturing family conditions promote both variables. A buffering claim requires a tested interaction showing that risk is differently related to outcomes at different levels of sibling support. In a longitudinal study of children after divorce, greater sibling support was associated with more favorable self-esteem patterns and buffered some conflict-related adversity, whereas support did not substantially moderate parent-child relationship quality. This limited but informative result argues for specifying the precise outcome and context whenever protection is discussed van Dijk et al. (2022) Feinberg et al. (2012).
Conflict, Rivalry, Aggression, and Coercion
Sibling conflict encompasses disagreements over resources, attention, rules, autonomy, and perceived injustices. Rivalry is a competitive orientation toward valued standing, often involving the belief that one child's gain diminishes another's opportunity. These processes can coexist with warmth because a close relationship provides both frequent contact and emotionally meaningful stakes. Conflict becomes especially concerning when exchanges are recurrently hostile, one partner cannot safely exit, or age and power asymmetry permit intimidation. Measurement must distinguish ordinary disagreement, mutual escalation, persistent aggression, and victimization, since their psychological implications and required responses differ Furman and Buhrmester (1985) Feinberg et al. (2012).
Prospective research has linked severe sibling bullying to later adverse psychological outcomes. Bowes and colleagues studied sibling bullying in a UK cohort and examined subsequent depression, anxiety, and self-harm, illustrating the importance of distinguishing repeated aggression from ordinary conflict. Longitudinal ordering strengthens the temporal evidence but still requires attention to preexisting adversity, peer victimization, parenting, and possible common vulnerabilities. The finding warrants serious attention to coercive sibling dynamics, while specific effect estimates and causal claims must be read with the study's sampling and adjustment assumptions in view Bowes et al. (2014).
Coercive cycles can be analyzed as sequences. One child provokes or demands, the other resists, the exchange escalates, and an adult or one sibling withdraws; when withdrawal repeatedly follows escalation, the interaction may inadvertently reinforce the effective use of hostility. This explanation generates observable predictions about what happens immediately before and after an episode. It also demonstrates why simply counting arguments provides an incomplete description. Intervention trials should establish whether they alter cooperation, resolution strategies, parent responses, and bullying separately. In the Siblings Are Special initial randomized test, gains emerged in several positive domains while short-term sibling conflict measures did not significantly improve Feinberg et al. (2013) Feinberg et al. (2012).
Comparison, Birth Order, and Individual Differentiation
Children evaluate their standing relative to siblings through achievement, privileges, responsibility, physical strength, parental attention, and cultural expectations about age and gender. Perceived comparison is psychologically consequential when it changes emotions, behavior, or self-understanding, but birth order alone does not measure the content of those comparisons. An older child may be given more autonomy because of age-appropriate development, while a younger child may interpret the difference as unfair. Such interpretations are relational processes that must be observed rather than inferred from structural position Jensen et al. (2023) Furman and Buhrmester (1985).
Large samples have refined longstanding claims about ordinal position. Rohrer, Egloff, and Schmukle analyzed three national panel datasets and compared individuals within and across families. They documented small intellectual-domain patterns but did not find lasting birth-order effects on several broad personality traits. This evidence directly limits sweeping character descriptions of firstborn, middle, or youngest children. It does not answer every question about how siblings negotiate status in a particular household, because personality inventories and momentary relational roles are different objects. Scientific synthesis gains precision when it separates general trait distributions from situated family experience Rohrer et al. (2015).
Birth order also interacts statistically with family size, spacing, cohort, and age at the time of assessment. In cross-family comparisons, an older child from a two-child household is not automatically comparable with an older child in a five-child household. Within-family contrasts reduce many stable household differences but cannot eliminate age-related change or the influence of one child's behavior on another. Meta-analytic research on differential treatment shows that age and sex differences are associated with some parental treatment differences, especially domains tied to control and autonomy. A research design must therefore test specific mechanisms rather than use ordinal position as an all-purpose explanation Jensen et al. (2023) Jensen and Thomsen (2024).
Parental Differential Treatment and Perceived Fairness
Parental differential treatment refers to differences in parenting experienced by siblings within the same family. Relative differential treatment identifies who receives more or less of a particular behavior; absolute differential treatment describes how unequal the treatment is, regardless of which child receives more. Differences may concern affection, discipline, control, responsibilities, or autonomy. Some differences reflect age, ability, or need, while others may be experienced as unjustified favoritism. The child's interpretation of fairness and the domain of treatment are therefore essential to understanding the relation between parenting and sibling well-being Jensen et al. (2023) McHale et al. (2005).
Jensen and Thomsen's 2024 multilevel meta-analysis combined 2,890 effect sizes from 88 sources nested within 43 samples, representing 26,451 participants. Less-favored relative treatment was associated with more internalizing and externalizing behavior, and greater absolute differences were also associated with more difficulties. This broad evidence base supports an important association, but heterogeneity in informants, ages, samples, and parenting domains limits a simple causal conclusion. The meta-analysis reported a sample composition with substantial White and United States representation, a reason to test whether patterns replicate across different cultural and institutional settings Jensen and Thomsen (2024).
A 2026 caregiver-reported longitudinal study using random-intercept cross-lagged models separated stable between-family patterns from within-family fluctuations. It found that overall differential treatment and differential negativity were related to poorer sibling relationship quality between families, while some within-family prospective associations had directions that complicated the familiar one-way favoritism-to-rivalry story. The study's reports came from caregivers rather than two independently assessed children, and model-specific estimates should not be generalized as universal feedback laws. Taken together, meta-analytic and longitudinal evidence establishes the importance of differential treatment while keeping the pathway and direction of influence under active investigation Eckert et al. (2026) Richmond et al. (2005).
Cross-Subsystem Feedback, Boundaries, and Triangles
Sibling interactions are nested in family routines and in the relationships of caregivers with each child. A parent may intervene to stop aggression, support negotiation, or redistribute resources, and these interventions can change later sibling expectations. Conversely, frequent disputes may change a parent's stress, monitoring, and disciplinary patterns. The term feedback is useful when repeated observation shows that a change in one part of the system alters a later change elsewhere. An account that merely asserts that all family members influence one another is too broad to test; the scientific proposition must identify the direction, interval, mechanism, and observable outcome Cox and Paley (1997) Eckert et al. (2026).
A triangle occurs when a third family member is drawn into a dyadic interaction, possibly to mediate, align with one participant, or carry communication. In sibling relations, a caregiver may repeatedly side with one child, another sibling may become a confidant about parental conflict, or children may coordinate around a shared responsibility. These situations vary in their meaning and consequences. Coding must distinguish protective adult intervention from an inappropriate burden of mediation assigned to a child. Family-system hypotheses then become amenable to longitudinal assessment through repeated reports and observations of who recruits whom and how disputes resolve Feinberg et al. (2012) McHale et al. (2012).
The distinction between individual and subsystem explanations becomes especially important when children show different adjustment patterns within one family. Household-level stress may raise overall risk while differential treatment affects one sibling more strongly, and the sibling response may subsequently alter parental conduct. Statistical models that treat every child as an independent observation underestimate the dependence in these data. An appropriately designed sibling study separates stable family characteristics, variation between children, variation within each child over time, and relational responses between partners. Evidence from longitudinal differential-treatment studies illustrates why conclusions change when these levels are distinguished Richmond et al. (2005) Eckert et al. (2026).
Parental Separation, Divorce, and Reconstituted Families
Parental separation changes residence, routines, financial circumstances, caregiving patterns, and the opportunities siblings have for contact. Its impact depends on the level of parental conflict, timing of transitions, children's ages, and whether siblings continue living together. A sibling may offer emotional support during a family rupture while also expressing frustration or intensified conflict. In research comparing adolescents from separating and continuously married families, Sheehan and colleagues described relationships characterized by both pronounced warmth and hostility. This complexity undermines a universal account in which divorce either inevitably damages sibling closeness or automatically strengthens solidarity Sheehan et al. (2004).
Later research demonstrates the importance of distinguishing parental conflict from marital status. Poortman and Voorpostel examined retrospective data from the Netherlands with reports from both siblings and found that predivorce conflict helped explain sibling conflict in adulthood. A longitudinal study of recently divorced families linked sibling relationship quality with changes in externalizing problems and self-esteem, while finding less evidence for internalizing associations and for broad compensation of parent-child relationships. These designs provide complementary views: retrospective adult data can cover long intervals but have recall limitations, whereas short prospective follow-up better identifies timing yet may capture only the early adaptation period Poortman and Voorpostel (2009) van Dijk et al. (2022).
Stepfamilies, half-sibling relations, and split-residence arrangements introduce further variation in time together and in the social recognition of kinship roles. Defining a sibling solely through shared biological parentage would exclude relationships organized through sustained caregiving and shared household life. At the same time, different kinship arrangements can create different opportunities and responsibilities, so treating all pairs as equivalent also loses information. Research should report the relationship type, residential history, time since transition, and family conflicts without assigning a predetermined emotional outcome to family form. The relevant mechanism is the interaction of accessibility, family norms, caregiving, and negotiated membership McHale et al. (2012) Font and Kim (2022).
Disability, Illness, and Unequal Care Demands
When one child has a disability or chronic illness, family time, medical appointments, caregiving schedules, and expectations of siblings may change. These circumstances can encourage cooperation and mutual responsibility while also imposing emotional labor, uncertainty, or reduced parental availability. The psychological experience of a sibling cannot be inferred from the diagnosis of another child. Severity, service access, parental support, finances, communication capabilities, age, and the quality of the relationship are all relevant. Research must distinguish voluntary caring, routine family assistance, and developmentally inappropriate caregiving responsibility Levante et al. (2025) McHale et al. (2012).
Levante and colleagues' systematic integrative review examined sixty studies published between 2014 and mid-2024, encompassing more than ten thousand participants. The mapped constructs included quality of sibling relationships, sibling-focused parentification, and emotional or behavioral adjustment. The reviewed evidence was heterogeneous, and the review identified limited work specifically centered on siblings' well-being. Such findings support family-level assessment and attention to the sibling's own account, yet they do not yield a single universal adjustment profile. The presence of supportive relationships and the presence of caregiver burden must be measured separately Levante et al. (2025).
Study design is particularly consequential in this field. Families connected to disability services may differ from families with less access to support, and diagnostic labels vary substantially in their associated functional needs. Reports by caregivers can provide crucial context but may not capture private emotions of siblings who are reluctant to express distress. Longitudinal designs could examine whether changing service support or family roles predict changes in warmth, conflict, and perceived fairness. Ethical practice also requires recognizing the agency and perspectives of the disabled sibling, rather than presenting that person only as the source of another child's difficulties Levante et al. (2025) Holmes et al. (2024).
Sibling Separation, Foster Care, and Bereavement
Sibling separation may occur through parental residence decisions, adoption or foster care placements, migration, educational moves, and family death. The mechanisms differ sharply. Some children experience a loss of everyday contact despite strong emotional ties, while others experience physical safety or relief from coercion following a separation. Policy and clinical discussion should therefore distinguish separation as an event from the quality and safety of the prior relationship. The relevance of continuity depends on what was being continued: shared security, practical support, hostile interaction, or a mixture that changed across contexts Font and Kim (2022) Feinberg et al. (2012).
Using administrative data on 2,297 children entering foster care in an urban United States county, Font and Kim examined sibling placement and later placement instability. Children placed with at least one sibling showed lower hazards of placement moves, including some moves not associated with planned progress. The association is informative for placement planning but is not a randomized test: sibling groups differ in size, needs, placement availability, and prior histories. The analysis supports careful attention to sibling relationships without supporting an automatic rule that joint placement is optimal regardless of safety or children's preferences Font and Kim (2022).
Bereavement changes the sibling subsystem through the permanent absence of a shared partner and through altered family routines, narratives, and parental grief. Survivors can experience loss of an emotionally close person and disruption of their own position in the family. A psychological account needs to distinguish grief responses, available support, pre-loss relationship quality, and the meanings attached to memorialization. Because grief and contact are different from the observational outcomes of ordinary sibling-conflict studies, conclusions should not be borrowed mechanically across settings. The appropriate scientific task is to model family reorganization and individual experience while acknowledging that longitudinal evidence for specific sibling-loss pathways remains uneven McHale et al. (2012) Holmes et al. (2024).
Cultural Organization and Social Inequality
Sibling relationships are regulated by cultural understandings of duty, autonomy, kinship, respect for older relatives, gender, and the distribution of family labor. A level of caregiving that one household experiences as expected reciprocity may be experienced elsewhere as excessive responsibility. Consequently, raw frequencies of tasks or contact do not have identical psychological meaning across cultural contexts. The analysis must examine the norms through which children and adults interpret roles, and the material constraints that shape opportunity. The evidence base is especially vulnerable to overgeneralization when studies rely on relatively advantaged, two-caregiver, predominantly White samples Holmes et al. (2024) Updegraff et al. (2005).
Updegraff and colleagues studied 246 adolescent sibling pairs in two-parent Mexican American families, combining interviews and daily activity reports. They observed both intimacy and conflict, and explored connections between relationship experiences and familism values. Related work on differential treatment in Mexican American families found that adolescents' family values moderated some links between treatment differences and adjustment or fairness perceptions. These findings establish that culturally organized meanings can modify associations; they do not describe a uniform pattern for all Mexican American households or all Latino populations. Within-group variation is itself part of the research object Updegraff et al. (2005) McHale et al. (2005).
Economic inequality can affect sibling relations through housing crowding, time scarcity, parental work schedules, caregiving responsibilities, and access to supportive services. Migration and family separation can additionally create unequal educational and household roles. Research designs need reliable measures of these circumstances rather than using ethnic category as a proxy for resources or family values. The most informative comparisons test measurement equivalence and document how institutional conditions affect actual interaction opportunities. Future scholarship should include diverse kinship forms, disability contexts, foster families, and families beyond the high-income national settings dominating much of the available research Holmes et al. (2024) Levante et al. (2025).
Measurement, Sampling, and Multi-Informant Designs
A valid measure of sibling relationship quality must specify who reports, which sibling pair is described, the time frame, the domain of interaction, and the interpretation of the score. Classic questionnaire work differentiated warmth, status, conflict, and rivalry, establishing a multidimensional foundation. Contemporary studies add daily reports, caregiver observations, semi-structured interviews, and behavioral coding. Each instrument samples different experiences. An adolescent's private account may disclose hostility hidden from caregivers; a filmed task can characterize turn-taking and repair but may evoke behavior that differs from normal household life. Triangulating informants is scientifically useful when discrepancies themselves are modeled as data Furman and Buhrmester (1985) Holmes et al. (2024).
The unit of sampling can be a child, a sibling dyad, a household containing several sibling pairs, or an extended kin network. These units are nested and often nonindependent. If three siblings form three possible pairs, counting the pairs as unrelated observations creates artificial precision. Models should incorporate shared family membership, recurring observations, and directional differences between siblings. Samples recruited from schools, clinics, foster care systems, or online panels can have different selection mechanisms and external validity. A reliable report includes age and age spacing, sex composition, family structure, living arrangement, socioeconomic context, and relationship type, together with clear missing-data procedures Holmes et al. (2024) Jensen et al. (2023).
Measurement invariance is central to longitudinal and cultural comparison. A warmth item referring to play can lose relevance as siblings become adults, while caregiving or electronic communication may become more informative. A change in the total score may then represent a shift in item meaning rather than a change in underlying closeness. Researchers can use developmentally appropriate item sets, tests of factor structure and invariance, and complementary behavioral indicators. Where constructs do not remain equivalent, the honest result is a comparison of specified observed behaviors rather than an asserted identical latent quality. This protects the scientific validity of long-range claims about lifespan change Buhrmester and Furman (1990) Çineli and Buyukkececi (2026).
Longitudinal Identification, Dyadic Models, and Causality
Sibling research is particularly exposed to confounding because siblings share genes, periods of residence, parental resources, and social networks while also influencing one another. Cross-sectional correlations describe covariation within this organized system. Prospective studies establish the order of measured changes but can remain confounded by unmeasured time-varying events. Family fixed effects control stable shared characteristics under their assumptions; random-intercept cross-lagged models distinguish between-family differences from within-family changes; randomized interventions can identify effects of assigned program exposure on particular outcomes. These approaches address different estimands and cannot be treated as interchangeable proofs of one general causal mechanism Eckert et al. (2026) Feinberg et al. (2013).
Research must also confront interference: changing one sibling's behavior can alter the other's experience, parental response, or the entire household routine. This violates a simple assumption that each individual's potential outcomes are unaffected by another participant's assignment or treatment. A family-level randomized program can account for some interference by randomizing the family, but then estimates a combined household intervention rather than an isolated sibling-to-sibling pathway. Repeated measures at several intervals can clarify whether observed change precedes a later outcome, yet the temporal spacing of waves may miss fast cycles of conflict and repair. Intensive longitudinal designs offer a complementary view of these shorter processes Feinberg et al. (2013) Richmond et al. (2005).
A persuasive causal claim should identify the intervention or exposure, comparison condition, outcome, timing, target population, and alternative explanations. For instance, comparing families with larger versus smaller differences in parental treatment does not establish the result of experimentally equalizing parenting, because child needs and behavior may have influenced the original treatment. Meta-analytic associations can be highly reproducible while their directional mechanism remains unsettled. The 2024 differential-treatment meta-analysis and the 2026 longitudinal models therefore answer related but distinct questions. Scientific synthesis preserves that distinction and treats contradictory directional estimates as information about the complexity of the system Jensen and Thomsen (2024) Eckert et al. (2026).
Evidence Synthesis, Replication, and Generalizability
The evidence base is now large enough to support focused synthesis but remains unevenly distributed across populations and constructs. Holmes and colleagues mapped 277 empirical studies published from 1985 through 2022 and found extensive reliance on quantitative assessment of conflict, warmth, cohesion, power, and related domains. The map documented an emphasis on majority-White, often relatively advantaged or two-caregiver samples. This matters because a strong research tradition can still produce a narrow empirical picture of family arrangements. A claim about sibling relationships in general requires evidence from settings with different structures of co-residence, kinship, disability, work, and public support Holmes et al. (2024).
Reviews and meta-analyses serve different functions. A systematic evidence map describes what has been studied and where important gaps remain; a meta-analysis estimates a distribution of effects for a specified exposure and outcome; an original longitudinal study can illuminate time order within its sample. The Jensen and Thomsen synthesis aggregates associations between differential treatment and adjustment, while Volling's review challenges an assumed universal firstborn crisis and Levante and colleagues synthesize disability-related sibling experiences. These conclusions can coexist because they concern different populations, transitions, measures, and estimands. Any integrated explanation must retain the conditions under which each conclusion is supported Jensen and Thomsen (2024) Volling (2012) Levante et al. (2025).
Replication should establish that a finding survives independent sampling, compatible measurement, and plausible model specifications. Confidence is strengthened by preregistered hypotheses, shared analytic code, adequate attrition analysis, transparent reporting of null results, and attention to clustering. A positive finding from one intervention, especially when restricted to a particular age group and school setting, cannot be presumed to generalize to clinical populations or adult families. Correspondingly, a null average effect may conceal meaningful heterogeneity that requires prespecified subgroup analysis. The evidence standard favors clearly defined, independently repeatable findings over explanatory language detached from design Feinberg et al. (2013) Holmes et al. (2024).
Interventions, Prevention, and Family Practice
Sibling-directed interventions seek to improve constructive interaction, emotional regulation, problem solving, and caregiving responses. Their scientific evaluation should distinguish proximal skills from broader mental health, educational, and family outcomes. The Siblings Are Special program was tested in a randomized trial involving 174 families with school-aged sibling pairs. Its initial evaluation found improvements in positive sibling relationships, parenting strategies, and several child competencies, while finding no detectable short-term improvement in sibling conflict, collusion, or externalizing problems. This outcome pattern supports a selective account of intervention efficacy and the need for follow-up rather than a generic statement of program success Feinberg et al. (2013).
Intervention mechanisms must be assessed in the actual relationship. Teaching children to name emotions might improve self-report knowledge without changing coercive sequences at home. Parent coaching might alter conflict by changing adult reinforcement even if the child's expressed beliefs change little. A robust trial therefore combines observed behavior, child reports, caregiver assessment, and longer-term follow-up when feasible. Allocation concealment, attrition, blinding of observers where possible, preregistration, and transparent outcome reporting are important because participants and families often know their assignment and may change reporting behavior Feinberg et al. (2013) Feinberg et al. (2012).
Clinical and welfare contexts require additional specificity. Recurrent bullying, disability-related care burdens, and foster care placement decisions present different needs from ordinary conflict resolution in a school program. The same intervention cannot be presumed suitable for all. Protection from aggression, participation of children in decisions, availability of caregiver support, and preservation of meaningful safe ties must be evaluated together. Evidence on foster care placement stability and systematic disability reviews can inform assessment, while neither licenses an individualized treatment prescription without professional evaluation. Family practice is strongest when it relies on demonstrated outcomes and respects each sibling's distinct position within the subsystem Font and Kim (2022) Levante et al. (2025) Bowes et al. (2014).
Conclusion
Sibling relationships constitute a dynamic subsystem of family life characterized by shared history, reciprocal interaction, emotional significance, and context-sensitive asymmetries. Their quality includes warmth, cooperation, status, conflict, and rivalry, which may coexist and evolve rather than occupy opposite ends of one continuum. The historical shift from family constellation to measured relationship quality made these mechanisms scientifically visible. Family systems, social learning, social comparison, developmental ecology, and emotional-security approaches illuminate different aspects of the same organized relationship, and each must be evaluated through claims that can be operationalized and tested Furman and Buhrmester (1985) Whiteman et al. (2011) McHale et al. (2012).
The strongest current evidence establishes substantial associations among sibling quality, differential parental treatment, and child adjustment, together with important diversity in responses to family transitions. Meta-analytic results identify consistent patterns of unequal parenting and adjustment; longitudinal studies demonstrate the relevance of temporal and within-family variation; intervention experiments show that some positive skills and outcomes can be changed. Research on bullying, divorce, foster care, disability, and cultural differences expands the range of phenomena that the family subsystem must explain. These results are informative precisely because their populations, designs, and limits are described rather than compressed into a single claim about all siblings Jensen and Thomsen (2024) Eckert et al. (2026) Feinberg et al. (2013) Holmes et al. (2024).
The next scientific priorities are repeated multi-informant measurement, dyadic and household models with explicit causal estimands, more representative cultural and family sampling, and intervention follow-up that separates relationship quality from individual clinical outcomes. Digital communication and artificial systems may affect family coordination and distance-mediated contact, but their specific influence on sibling bonds remains a subject for independently designed research. The established research object remains the organization of relations among sibling partners inside changing family systems. A mature account explains the processes, measures the contingencies, and preserves the distinct experiences of every family member Çineli and Buyukkececi (2026) Levante et al. (2025) Holmes et al. (2024).
References
- Bowes, L., Wolke, D., Joinson, C., Lereya, S. T., & Lewis, G. (2014). Sibling bullying and risk of depression, anxiety, and self-harm: A prospective cohort study. Pediatrics, 134(4), e1032–e1039. https://doi.org/10.1542/peds.2014-0832
- Buhrmester, D., & Furman, W. (1990). Perceptions of sibling relationships during middle childhood and adolescence. Child Development, 61(5), 1387–1398. https://doi.org/10.1111/j.1467-8624.1990.tb02869.x
- Çineli, B., & Buyukkececi, Z. (2026). Family breakups beyond childhood: Later-life parental divorce and adult sibling ties. Journal of Family Research, 38, 159–177. https://doi.org/10.20377/jfr-1248
- Cox, M. J., & Paley, B. (1997). Families as systems. Annual Review of Psychology, 48, 243–267. https://doi.org/10.1146/annurev.psych.48.1.243
- Eckert, L. M., Zhang, J., Wade, M., & Browne, D. T. (2026). Parental differential treatment and sibling relationship quality: A longitudinal study. Journal of Social and Personal Relationships, 43(9), 2625–2647. https://doi.org/10.1177/02654075261440658
- 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. https://doi.org/10.1007/s10567-011-0104-5
- Feinberg, M. E., Solmeyer, A. R., Hostetler, M. L., Sakuma, K.-L., Jones, D., & McHale, S. M. (2013). Siblings are special: Initial test of a new approach for preventing youth behavior problems. Journal of Adolescent Health, 53(2), 166–173. https://doi.org/10.1016/j.jadohealth.2012.10.004
- Font, S. A., & Kim, H. W. (2022). Sibling separation and placement instability for children in foster care. Child Maltreatment, 27(4), 583–595. https://doi.org/10.1177/10775595211012482
- Furman, W., & Buhrmester, D. (1985). Children's perceptions of the qualities of sibling relationships. Child Development, 56(2), 448–461. https://pubmed.ncbi.nlm.nih.gov/3987418/
- Holmes, M. R., Bender, A. E., O'Donnell, K. A., Miller, E. K., & Conard, I. T. (2024). Illuminating the landscape of sibling relationship quality: An evidence and gap map. Child Development, 95(4), 1425–1440. https://doi.org/10.1111/cdev.14065
- Jensen, A. C., Jorgensen-Wells, M. A., Andrus, L. E., Pickett, J. M., Leiter, V. K., Hadlock, M. E. B., & Dayley, J. C. (2023). Sibling differences and parents' differential treatment of siblings: A multilevel meta-analysis. Developmental Psychology, 59(4), 644–654. https://doi.org/10.1037/dev0001406
- Jensen, A. C., & Thomsen, A. E. (2024). Parental differential treatment of siblings linked with internalizing and externalizing behavior: A meta-analysis. Child Development, 95(4), 1384–1405. https://doi.org/10.1111/cdev.14091
- Levante, A., Martis, C., Del Prete, C. M., Martino, P., Primiceri, P., & Lecciso, F. (2025). Siblings of persons with disabilities: A systematic integrative review of the empirical literature. Clinical Child and Family Psychology Review, 28(1), 209–253. https://doi.org/10.1007/s10567-024-00502-6
- McHale, S. M., Updegraff, K. A., Shanahan, L., Crouter, A. C., & Killoren, S. E. (2005). Siblings' differential treatment in Mexican American families. Journal of Marriage and Family, 67(5), 1259–1274. https://doi.org/10.1111/j.1741-3737.2005.00215.x
- McHale, S. M., Updegraff, K. A., & Whiteman, S. D. (2012). Sibling relationships and influences in childhood and adolescence. Journal of Marriage and Family, 74(5), 913–930. https://doi.org/10.1111/j.1741-3737.2012.01011.x
- Poortman, A.-R., & Voorpostel, M. (2009). Parental divorce and sibling relationships: A research note. Journal of Family Issues, 30(1), 74–91. https://doi.org/10.1177/0192513X08322782
- Richmond, M. K., Stocker, C. M., & Rienks, S. L. (2005). Longitudinal associations between sibling relationship quality, parental differential treatment, and children's adjustment. Journal of Family Psychology, 19(4), 550–559. https://doi.org/10.1037/0893-3200.19.4.550
- Rohrer, J. M., Egloff, B., & Schmukle, S. C. (2015). Examining the effects of birth order on personality. Proceedings of the National Academy of Sciences, 112(46), 14224–14229. https://doi.org/10.1073/pnas.1506451112
- Sheehan, G., Darlington, Y., Noller, P., & Feeney, J. (2004). Children's perceptions of their sibling relationships during parental separation and divorce. Journal of Divorce & Remarriage, 41(1–2), 69–94. https://doi.org/10.1300/J087v41n01_05
- Updegraff, K. A., McHale, S. M., Whiteman, S. D., Thayer, S. M., & Delgado, M. Y. (2005). Adolescent sibling relationships in Mexican American families: Exploring the role of familism. Journal of Family Psychology, 19(4), 512–522. https://doi.org/10.1037/0893-3200.19.4.512
- van Dijk, R., van der Valk, I. E., Buist, K. L., Branje, S., & Deković, M. (2022). Longitudinal associations between sibling relationship quality and child adjustment after divorce. Journal of Marriage and Family, 84(2), 393–414. https://doi.org/10.1111/jomf.12808
- Volling, B. L. (2005). The transition to siblinghood: A developmental ecological systems perspective and directions for future research. Journal of Family Psychology, 19(4), 542–549. https://doi.org/10.1037/0893-3200.19.4.542
- Volling, B. L. (2012). Family transitions following the birth of a sibling: An empirical review of changes in the firstborn's adjustment. Psychological Bulletin, 138(3), 497–528. https://doi.org/10.1037/a0026921
- Whiteman, S. D., McHale, S. M., & Soli, A. (2011). Theoretical perspectives on sibling relationships. Journal of Family Theory & Review, 3(2), 124–139. https://doi.org/10.1111/j.1756-2589.2011.00087.x
Social Learning, Identification, and Behavioral Transmission
Social learning within sibling pairs begins with exposure to patterned behavior and the social consequences of that behavior. A younger child may copy an older sibling's way of negotiating with parents, while an older sibling may adapt to the younger child's requests or acquire caregiving routines. Similarity can also emerge from shared household rules, temperament, neighborhood, and peer networks. The inference of transmission therefore requires more than behavioral resemblance. It becomes stronger when studies establish prior differences, document actual exposure, measure changes in both partners, and test whether proposed mediating interactions occurred Whiteman et al. (2011) McHale et al. (2012).
Identification refers to adopting some characteristics of a sibling as personally meaningful; differentiation refers to constructing distinctions in achievement, interests, habits, or social identity. Both can occur in a single family, and the same child can identify with a sibling in one domain while seeking differentiation in another. This process gives sibling comparisons a role in identity development without making competition the inevitable outcome of having a same-age or same-sex sibling. Studies need domain-specific indicators and repeated observations to decide whether a difference precedes or follows relationship strain Whiteman et al. (2011) Rohrer et al. (2015).
Behavioral risk provides a particularly difficult test of influence mechanisms. Siblings may share experiences of social disadvantage or peer networks that both influence drinking, school attendance, and other outcomes. A model that labels shared behavior as social contagion without measuring these contexts remains underidentified. Where possible, designs can compare differential timing of exposures within the same family, examine changes around school transitions, and combine sibling accounts with behavioral records. Even within-family analyses require caution when one sibling's exposure changes other family members' behavior. The subsystem is precisely the setting in which independent individual outcomes are an unrealistic assumption McHale et al. (2012) Feinberg et al. (2012).