Close Friend Support Subscale

Abstract

The Close Friend Support Subscale is a specialized psychometric assessment instrument developed by developmental psychologist Susan Harter in 1985 as an integral constituent of the multidimensional Social Support Scale for Children (SSSC). Designed specifically for children and early adolescents across grades three through eight (approximate chronological ages 8 to 14), the instrument operationalizes perceived emotional and instrumental support derived from an individualized, dyadic close friendship. The subscale comprises 6 self-report items formatted using Harter’s proprietary “structured alternative format.” This dual-step response architecture requires youth to first select which of two contrasting archetypes resembles them (e.g., children who have a confidant to disclose problems to versus those who do not), subsequently indicating whether that behavioral descriptor is “Sort of True for Me” or “Really True for Me.” This methodological innovation systematically attenuates social desirability bias and mitigates acquiescent responding in juvenile populations.

Psychometrically, the Close Friend Support Subscale exhibits robust internal consistency reliability across diverse community, clinical, and cross-cultural cohorts, with Cronbach’s alpha coefficients typically ranging from α = .75 to α = .88. Confirmatory factor analytic investigations consistently validate its unidimensionality and discriminant factorial independence from related social provisions, such as classmate acceptance, parental support, and teacher regard. Convergent validity is substantiated through marked positive correlations with global self-worth, perceived social competence, and positive emotional affect, alongside salient negative associations with depressive symptomatology, loneliness, and internalizing distress. The instrument holds prominent clinical, developmental, and educational utility, enabling investigators and practitioners to disentangle the psychological consequences of intimate dyadic chumships from broader classroom peer group dynamics.

Keywords

Close Friend Support Subscale, Social Support Scale for Children, Susan Harter, dyadic friendship, peer relations, structured alternative format, perceived social support, childhood psychometrics, early adolescence, self-worth, emotional disclosure, social buffering

Authors

The Close Friend Support Subscale was conceptualized, developed, and psychometrically validated by Susan Harter, Ph.D. (1939–2017), an internationally acclaimed developmental and clinical child psychologist. Dr. Harter served as Professor of Psychology and Head of the Developmental Psychology Training Program at the University of Denver in Denver, Colorado, United States. Throughout her academic career spanning more than four decades, Dr. Harter made foundational theoretical and psychometric contributions to the understanding of self-concept development, self-esteem, perceived competence, emotional regulation, and relational social support structures across the lifespan.

Inquiries regarding historic archival documentation, conceptual manuals, and related assessment batteries developed under the Harter laboratory (such as the Self-Perception Profile for Children and the Social Support Scale for Children) are preserved within academic libraries and the Department of Psychology at the University of Denver, University Park, Denver, CO 80208.

Purpose

The primary psychometric purpose of the Close Friend Support Subscale is to provide a reliable, developmentally sensitive, and structurally standardized assessment of a child’s perceived availability of intimate, supportive, and emotionally validating friendship provisions. In early childhood, peer interaction frequently centers on coordinated play, behavioral proximity, and shared activity participation. However, during the transition into middle childhood and early adolescence, peer relations undergo a profound qualitative realignment toward interpersonal intimacy, self-disclosure, loyalty, and affective mutuality. The subscale was designed to isolate this specific micro-systemic relational resource, distinguishing perceived close friend support from other concurrent support systems, including non-intimate peer network acceptance (classmate support), primary caregiver support (parent support), and adult pedagogical encouragement (teacher support).

Clinical and Developmental Applications

Within clinical psychology, developmental psychopathology, and pediatric mental health settings, the Close Friend Support Subscale serves as a crucial diagnostic and evaluative tool:

• Screening for Interpersonal Isolation and Loneliness: Dyadic alienation or the chronic inability to establish reciprocal, trusting friendships represents a potent vulnerability factor for psychological maladjustment. The subscale identifies children who lack a subjective sense of relational safety, intimacy, and reciprocal care.
• Differential Diagnosis in Internalizing Disorders: Deficits in perceived friend support frequently correlate with childhood major depressive disorder, generalized anxiety, and social withdrawal. Clinicians utilize the subscale to discern whether a child’s social difficulties stem from systemic classroom rejection or a specific absence of an intimate confidant.
• Monitoring Psychosocial and Preventive Interventions: School psychologists and clinical social workers administer the instrument pre- and post-intervention to assess the efficacy of targeted social skills training, peer-pairing initiatives, mentoring programs, and therapeutic friendship circles.

Research Applications

In academic research, the subscale is extensively utilized to test complex developmental and ecological models:

• Stress-Buffering Models: Investigating the hypothesis that intimate close friend support mitigates or buffers the adverse developmental impacts of stressful life events, such as family dissolution, bullying, academic pressure, or grade transitions.
• Self-Perception and Identity Formation: Longitudinal investigations exploring how subjective validation from a close friend directly predicts increases in global self-worth and perceived social competence over developmental time.
• Etiological Pathways of Peer Victimization: Examining how the presence of even a single highly supportive close friend can protect chronically victimized youth from developing secondary depressive symptoms or school avoidance behaviors.

Psychological Construct

The Close Friend Support Subscale measures perceived dyadic close friend support, conceptualized as a child’s cognitive appraisal of the availability, dependability, and emotional responsiveness of their primary close friend. Within developmental psychometrics, perceived support must be rigorously differentiated from structural social network variables (e.g., peer sociometric status, total number of nominated friends, or frequency of social interactions). Rather than recording the objective presence of peers, the subscale quantifies the child’s internal, subjective mental representation of whether their closest friend functions as a reliable source of validation, emotional containment, and empathetic understanding.

Dimensions of Close Friend Support

Although the subscale operates as a unidimensional latent factor within structural models, the 6 items systematically tap into four core theoretical dimensions of close friendship support:

• Intimate Disclosure and Confiding (Items 1 and 4): Assesses the respondent’s subjective freedom to communicate vulnerable internal states, private worries, and pressing personal problems to their friend without fear of betrayal, ridicule, or dismissal. Item 1 (“Some kids have a close friend who they can tell a problem to”) and Item 4 (“Some kids have a close friend who they can talk to about things that bother them”) capture this communicative safety valve.
• Empathetic Understanding and Active Listening (Items 2 and 5): Measures whether the child perceives their friend as cognitively and emotionally attuned to their inner self. Item 2 (“Some kids have a close friend who really understands them”) and Item 5 (“Some kids don’t have a close friend who really listens to what they say”) probe whether interaction goes beyond superficial talk to genuine psychological resonance.
• Affection, Benevolence, and Caring (Item 6): Focuses on perceived unconditional regard and emotional investment. Item 6 (“Some kids don’t have a close friend who cares about their feelings”) operationalizes the subjective belief that the friend genuinely values the child’s affective equilibrium and psychological well-being.
• Affiliative Companionship and Shared Time (Item 3): Evaluates the mutuality of voluntary physical presence and mutual recreational enjoyment. Item 3 (“Some kids don’t have a close friend who they like to spend time with”) ensures that intimate disclosure is anchored in positive, shared companionship.

Contrasting Close Friend Support with Classmate Support

A crucial theoretical hallmark of Harter’s conceptual framework is the empirical and psychometric boundary established between close friend support and classmate support. Classmate support assesses perceived popularity, peer group acceptance, and generalized social inclusion within the wider classroom context (“Do the other kids like me?”). In contrast, close friend support addresses psychological intimacy, loyalty, and deep dyadic bonding (“Do I have a specific person who knows, understands, and cares for me?”). Psychometric research affirms that these constructs correlate only moderately (typically r = .30 to .45), demonstrating that children can perceive themselves as widely popular yet lacking an intimate confidant, or conversely, occupy marginal positions in wider classroom popularity while possessing an exceptionally high-quality, highly protective dyadic friendship.

Theoretical Framework

The conceptual architecture of the Close Friend Support Subscale integrates several foundational paradigms from developmental psychology, psychoanalytic interpersonal theory, and social cognitive appraisal theory.

Harry Stack Sullivan’s Interpersonal Theory of Psychiatry

The most direct theoretical ancestor of the close friend support construct is the work of Harry Stack Sullivan (1953). Sullivan posited that the juvenile era gives way in preadolescence (roughly ages 8–12) to an urgent need for interpersonal intimacy that cannot be satisfied by parents or general peer groups. Sullivan termed this relational development the “chumship”—a uniquely intimate, egalitarian, same-sex dyad in which the child, for the first time, develops an intense sensitivity to what matters to another human being. Sullivan argued that chumships serve a corrective psychological function: an intimate confidant can counteract earlier interpersonal trauma, dissolve feelings of unworthiness, and facilitate accurate consensual validation of reality. Harter operationalized Sullivan’s chumship dynamics into psychometric indicators focused on mutual understanding, emotional ventilation, and reciprocal caring.

Harter’s Developmental Model of Self-Worth

The subscale is deeply embedded in Susan Harter’s broader theoretical synthesis of William James (1890) and Charles Horton Cooley (1902). Cooley’s sociological concept of the “looking-glass self” proposed that an individual’s sense of self-worth is a direct social construction derived from the perceived appraisals of significant others. Harter hypothesized that as cognitive maturation allows children to differentiate between different relational spheres, their overall global self-worth becomes selectively dependent upon the specific appraisal sources they prioritize.

In early childhood, parental approval reigns supreme. However, by middle childhood and early adolescence, close friends emerge as foundational “looking glasses.” If an adolescent perceives that a close friend validates their thoughts, protects their secrets, and genuinely cares about their emotional states, the individual internalizes these reflected appraisals as proof of their fundamental human value, elevating global self-worth and resilience.

The Stress-Buffering Hypothesis

The instrument is also grounded in the stress-buffering model of social support formulated by Sheldon Cohen and Thomas Wills (1985). According to this theory, social support operates through two distinct pathways: a main-effect model (where social integration enhances general well-being irrespective of stress) and a buffering model (where support acts as a psychological cushion between severe environmental stressors and negative psychological outcomes). In Harter’s paradigm, close friend support functions as a prime relational buffer. When an adolescent encounters family dysfunction or academic failure, having a confidant to disclose problems to (Items 1 and 4) de-escalates physiological stress reactivity and neutralizes maladaptive cognitive rumination.

Validity

Extensive psychometric investigations conducted over nearly four decades provide rigorous empirical validation for the Close Friend Support Subscale across construct, convergent, discriminant, and predictive domains.

Construct and Factorial Validity

During the initial scale development and standardization across normative samples encompassing thousands of children from elementary and middle schools (grades 3 through 8), exploratory and confirmatory factor analyses systematically isolated the Close Friend Support items onto an independent latent factor. Factor loadings for the 6 items universally exceed .55, with most values concentrating between .65 and .84. The items load cleanly on their intended factor without problematic cross-loadings onto the Parent, Teacher, or Classmate dimensions of the SSSC.

Convergent Validity

Convergent validity has been established through convergent correlations with parallel measures of friendship intimacy, relational quality, and psychological functioning:

• Friendship Quality Measures: Scores on the subscale correlate strongly (r = .60 to .74, p < .001) with the Intimate Exchange and Validation/Caring subscales of the Friendship Quality Questionnaire (FQQ; Parker & Asher, 1993) and the Network of Relationships Inventory (NRI; Furman & Buhrmester, 1985).
• Global Self-Worth: Conforming to Harter’s theoretical predictions, perceived close friend support exhibits sustained, robust positive correlations with the Global Self-Worth subscale of the Self-Perception Profile for Children (SPPC), with typical correlation coefficients ranging between r = .38 and r = .55 (p < .001).
• Positive Affect: Higher perceived friend support significantly predicts higher scores on subjective vitality, happiness scales, and adaptive emotional coping strategies in middle childhood.

Discriminant Validity

The scale effectively discriminates between close dyadic support and generalized peer group acceptance. Multiple validation studies show that the correlation between the Close Friend Support Subscale and the Classmate Support Subscale hovers at moderate levels (r = .32 to .46), confirming that youth readily distinguish between being accepted by their peer group and possessing an intimate, trustworthy friendship. Additionally, correlations with measures of social desirability (e.g., the Children’s Social Desirability Questionnaire) are negligible (typically r < .15), demonstrating that the structured alternative format successfully decouples perceived support ratings from social approval-seeking tendencies.

Predictive and Longitudinal Validity

Longitudinal studies document that baseline scores on the Close Friend Support Subscale prospectively predict changes in youth psychological adjustment:

• Mitigation of Internalizing Distress: In longitudinal cohort designs, children who report high levels of close friend support during the transition from elementary to middle school display significant declines in self-reported depressive symptoms, social anxiety, and loneliness over a 1- to 2-year tracking interval.
• Protection Against Bullying Sequelae: Research examining peer victimization demonstrates an interaction effect: victimized children with high Close Friend Support scores show significantly lower elevations in cortisol awakening responses and internalizing distress compared to victimized peers lacking friend support.

Reliability

The Close Friend Support Subscale consistently demonstrates high empirical reliability across diverse educational contexts, clinical cohorts, and linguistic adaptations.

Internal Consistency Reliability

Internal consistency metrics, assessed via Cronbach’s alpha (α) and McDonald’s omega (ω), consistently surpass standard psychometric thresholds across multiple grade levels:

• Normative Standardization Samples (Harter, 1985): In the initial standardization cohorts across grades 3 through 8, the Close Friend Support Subscale yielded alpha coefficients ranging from α = .75 to α = .84 across individual grade cohorts, with a combined cross-grade sample alpha typically averaging around .82.
• Elementary School Cohorts (Grades 3–5): Cronbach’s alpha generally ranges from α = .72 to α = .79, reflecting developing cognitive discernment of intimate friendship dimensions.
• Middle School / Early Adolescent Cohorts (Grades 6–8): Reliability estimates increase, frequently attaining α = .80 to α = .88, coinciding with the developmental crystallization of chumship structures and increased salience of peer disclosure.
• Composite Reliability (Omega): Structural equation modeling studies utilizing McDonald’s hierarchical omega have documented composite reliability values routinely exceeding ω = .83, indicating minimal measurement error within the latent construct.

Test-Retest Stability

Test-retest reliability assessments establish both temporal stability and developmental sensitivity:

• Short-Term Intervals (2 to 4 Weeks): Temporal stability coefficients range from r = .76 to r = .85, demonstrating that the subscale measures an enduring psychological perception rather than transient mood fluctuations.
• Medium-Term Intervals (3 to 6 Months): Stability coefficients average r = .65 to r = .72. Modest fluctuations over this window reflect genuine real-world shifts in childhood friendship dynamics (e.g., friendship dissolutions, classroom reassignments, or emerging new relationships).

Factor Analysis

Factor analytic evidence for the Close Friend Support Subscale within the 24-item Social Support Scale for Children (SSSC) demonstrates robust construct separation, factorial invariance, and structural validity.

Exploratory Factor Analysis (EFA)

In Dr. Harter’s foundational psychometric analyses, principal components analysis with oblique (Oblimin) and orthogonal (Varimax) rotations was conducted on the full item pool. Across four independent samples totaling over 1,500 students, four distinct eigenvalues consistently exceeded Kaiser’s criterion (λ > 1.00), corresponding exactly to the four theoretical sources of support: Parent, Classmate, Teacher, and Close Friend. The Close Friend Support items systematically loaded on a single distinct factor, with factor loadings ranging from .61 to .84, while maintaining negligible cross-loadings (rarely exceeding .20) on the remaining three dimensions.

Confirmatory Factor Analysis (CFA) and Model Fit

Subsequent contemporary structural equation modeling studies have rigorously confirmed the hypothesized four-factor first-order structure of the SSSC, confirming that Close Friend Support represents an empirically distinct latent entity. Fit indices across standard multi-group investigations consistently fulfill stringent psychometric benchmarks:

• Comparative Fit Index (CFI): Values consistently range between .94 and .98, indicating excellent comparative fit against baseline null models.
• Tucker-Lewis Index (TLI): Typical values span from .93 to .97.
• Root Mean Square Error of Approximation (RMSEA): Estimates consistently fall between .038 and .054, with 90% confidence intervals well below the .08 cutoff, denoting negligible model approximation error.
• Standardized Root Mean Square Residual (SRMR): Values reliably remain below .045.

Factor Loadings and Parameter Estimates

When evaluated within a unidimensional CFA measuring perceived close friend support, standard standardized factor loadings (λ) for the 6 individual items reflect robust structural coherence:

• Item 1 (Confiding problems): λ = .68 – .76
• Item 2 (Mutual understanding): λ = .74 – .82
• Item 3 (Spending time/companionship): λ = .58 – .67
• Item 4 (Discussing bothersome issues): λ = .75 – .84
• Item 5 (Active listening): λ = .72 – .80
• Item 6 (Caring about feelings): λ = .70 – .79

Measurement Invariance Across Demographics

Multi-group confirmatory factor analyses (MGCFA) have demonstrated full configural, metric (weak), and scalar (strong) measurement invariance across gender groups (boys vs. girls) and grade cohorts (elementary vs. middle school). While adolescent girls often report slightly higher mean latent levels of friend intimacy and disclosure than boys, the underlying factorial structure, item factor loadings, and item intercepts function equivalently across genders, confirming that observed mean differences reflect genuine psychological variances rather than psychometric measurement bias.

Instrument / Measurement Tool

The Close Friend Support Subscale is a specialized self-report psychometric instrument. Below are the structural and operational specifications of the tool:

• Construct Assessed: Perceived emotional, confiding, and companionship support derived from a dyadic close friendship.
• Respondent Target Population: Children and early adolescents aged 8 to 14 years (predominantly 3rd through 8th grades). Adapted versions have been utilized up to 10th grade.
• Total Number of Items: 6 items.
• Administration Format: Group administration in classroom settings or individual administration within clinical, pediatric, or research settings. Paper-and-pencil questionnaire or digitized self-report interface.
• Administration Duration: Approximately 3 to 5 minutes when administered as an isolated subscale; 15 to 20 minutes when administered as part of the full 24-item Social Support Scale for Children.
• Structural Alternative Response Architecture: The questionnaire does not employ traditional Likert scales. Instead, it utilizes Harter’s proprietary two-stage “structured alternative format” designed to mitigate acquiescence and social desirability:
  • Step 1: The child reads a compound sentence presenting two contrasting types of children separated by the conjunction “BUT” (e.g., “Some kids have a close friend who they can tell a problem to BUT Other kids don’t have a close friend who they can tell a problem to”). The respondent decides which group of children is most like themselves.
  • Step 2: After selecting the matching archetype side, the child designates whether that specific statement is “Sort of True for Me” or “Really True for Me” by marking a single checkbox.
• Response Scoring Framework:
  • Each item is scored on a continuous 4-point scale ranging from 1 (lowest perceived support) to 4 (highest perceived support).
  • Items with positive archetype on the left (Items 1, 2, and 4): Left side “Really True for Me” = 4; Left side “Sort of True for Me” = 3; Right side “Sort of True for Me” = 2; Right side “Really True for Me” = 1.
  • Items with negative archetype on the left (Items 3, 5, and 6): Left side “Really True for Me” = 1; Left side “Sort of True for Me” = 2; Right side “Sort of True for Me” = 3; Right side “Really True for Me” = 4.
  • Note on Source Keying: In standard psychometric scoring for Harter’s batteries, point values always reflect support level (4 = High Support, 1 = Low Support). Items where the negative stem appears first (e.g., Items 3, 5, and 6) are reverse-keyed relative to items where the positive stem appears first (Items 1, 2, and 4), ensuring that a score of 4 consistently designates maximum close friend support.
• Subscale Score Derivation:
  • Subscale Mean Score: Calculate the arithmetic mean across the 6 items (sum of item scores divided by 6). This yields a standardized scale score from 1.00 to 4.00, preserving intuitive interpretability relative to Harter’s normative profile bands.
  • Subscale Summed Score: Sum the item scores directly to produce a raw cumulative total ranging from 6 to 24.
• Normative Interpretation:
  • Low Perceived Support: Mean scores ≤ 2.00 (Summed scores ≤ 12) indicate substantial perceived relational deficit, lack of an intimate confidant, and heightened vulnerability to internalizing distress.
  • Moderate Perceived Support: Mean scores between 2.01 and 3.00 (Summed scores 13–18) indicate average, intermittent dyadic support.
  • High Perceived Support: Mean scores between 3.01 and 4.00 (Summed scores 19–24) indicate a robust, stable, and highly validating intimate friendship network.

Permissions & Fee and Test Year

The Social Support Scale for Children, containing the Close Friend Support Subscale, was formally published in 1985 by Dr. Susan Harter at the University of Denver.

Licensing and Usage Permissions: The scale is considered open-access for non-commercial academic, clinical, and educational research purposes. Formal contractual permission and licensing fees are not required for empirical investigations or educational program evaluation, provided that appropriate theoretical credit and formal bibliographic attribution are given to Susan Harter. The author historically requested that researchers utilizing the scale cite the 1985 manual and related psychometric publications accurately.

Institutional evaluators and youth development professionals can find the complete scale and administration guidelines reproduced in open educational resource handbooks, such as Assessing Outcomes in Child and Youth Programs: A Practical Handbook (University of Wisconsin-Extension, 2005).

References

• Berndt, T. J. (1982). The features and effects of friendship in early adolescence. Child Development, 53(6), 1447–1460. https://doi.org/10.2307/1130071
• Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98(2), 310–357. https://doi.org/10.1037/0033-2909.98.2.310
• Cooley, C. H. (1902). Human nature and the social order. Charles Scribner’s Sons.
• Furman, W., & Buhrmester, D. (1985). Children’s perceptions of the personal relationships in their social networks. Developmental Psychology, 21(6), 1016–1024. https://doi.org/10.1037/0012-1649.21.6.1016
• Harter, S. (1985). The Social Support Scale for Children: Manual. University of Denver.
• Harter, S. (1988). The construction and conservation of the self: Developmental and social processes. In D. K. Lapsley & F. C. Power (Eds.), Self, ego, and identity: Integrative approaches (pp. 43–70). Springer-Verlag. https://doi.org/10.1007/978-1-4615-7834-5_3
• Harter, S. (1999). The construction of the self: A developmental perspective. Guilford Press.
• Harter, S. (2012). The construction of the self: Developmental and sociocultural foundations (2nd ed.). Guilford Press.
• Parker, J. G., & Asher, S. R. (1993). Friendship and friendship quality in middle childhood: Links with peer group acceptance and feelings of loneliness and social dissatisfaction. Developmental Psychology, 29(4), 611–621. https://doi.org/10.1037/0012-1649.29.4.611
• Sullivan, H. S. (1953). The interpersonal theory of psychiatry. W. W. Norton & Company.
• University of Wisconsin-Extension. (2005). Assessing outcomes in child and youth programs: A practical handbook (pp. 135–136). Available from: http://fyi.uwex.edu/topic/youth/

13. Items of the Scale (Questionnaire)