Attachment to Parents—Seattle Social Development Project

1. Abstract

The Attachment to Parents—Seattle Social Development Project scale is a concise, four-item psychometric instrument designed to assess the emotional bond, affective closeness, and open interpersonal communication between adolescents and their parental figures (specifically biological, adoptive, or surrogate mothers and fathers). Developed within the longitudinal framework of the Seattle Social Development Project (SSDP) and subsequently incorporated into the widely administered Communities That Care Youth Survey (CTC-YS; Arthur et al., 2002), this measurement tool operationalizes key protective mechanisms outlined by the Social Development Model (SDM). The construct measures two fundamental aspects of the parent-child relationship: perceived affective closeness and communicative disclosure. The scale utilizes a 4-point, emphatic, forced-choice response format ranging from 1 (“NO!”) to 4 (“YES!”), yielding composite sum scores between 4 and 16, where higher scores signify stronger emotional attachment and communicative safety within the primary familial unit.

Extensive psychometric investigations conducted across diverse urban, suburban, and rural youth cohorts have consistently demonstrated robust internal consistency reliability (with Cronbach’s alpha coefficients routinely ranging between α = .76 and α = .84 across statewide and multi-state representative samples). Confirmatory factor analytic investigations validate both a unidimensional parental attachment superordinate construct and a correlated two-factor dyadic structure differentiating maternal and paternal relational pathways. In terms of criterion-related, convergent, and predictive validity, empirical research shows that higher scores on this instrument are significantly associated with prosocial school engagement, emotional regulation, and academic commitment, while serving as a salient longitudinal buffer against early substance initiation, delinquency, conduct disorder, depressive symptomatology, and interpersonal youth violence. The scale is cataloged by the Centers for Disease Control and Prevention (CDC) as a primary public health surveillance measure for youth violence prevention.

2. Keywords

Attachment to parents, Seattle Social Development Project, Communities That Care, Social Development Model, adolescent attachment, parental bonding, protective factors, juvenile delinquency prevention, risk and protective factor assessment, adolescent psychometrics

3. Authors

The scale was developed and refined by a collaborative research team of developmental psychologists, criminologists, and prevention scientists affiliated with the Social Development Research Group (SDRG) within the School of Social Work at the University of Washington, Seattle, Washington, USA:

• Michael W. Arthur, Ph.D. — Social Development Research Group, School of Social Work, University of Washington; prominent investigator in public health prevention systems, epidemiological surveillance of adolescent behavioral health, and psychometric evaluation of community risk and protective factors.
• J. David Hawkins, Ph.D. — Endowed Professor Emeritus of Prevention, Social Development Research Group, School of Social Work, University of Washington; co-developer of the Social Development Model and founding principal investigator of the Seattle Social Development Project.
• John A. Pollard, Ph.D. — Social Development Research Group, School of Social Work, University of Washington; specialist in quantitative research methodology, survey design, structural equation modeling, and adolescent risk profiling.
• Richard F. Catalano, Ph.D. — Professor Emeritus, Social Development Research Group, School of Social Work, University of Washington; co-architect of the Social Development Model, international authority on youth positive development and evidence-based preventive interventions.
• Angelo J. Baglioni Jr., Ph.D. — Quantitative psychometrician and statistical consultant; specialist in multivariate behavioral modeling, measurement invariance, and structural factor validation.

Institutional Contact: Social Development Research Group (SDRG), University of Washington, 9725 3rd Ave NE, Suite 401, Seattle, WA 98115, USA. Official Web Resource: https://www.sdrg.org.

4. Purpose

The primary purpose of the Attachment to Parents—Seattle Social Development Project scale is to provide a brief, methodologically rigorous, and developmentally appropriate assessment of the affective and communicative bond connecting adolescents to their parents. From an epidemiological and developmental standpoint, the transition from childhood into adolescence is characterized by neurodevelopmental reorganization, increasing striving for autonomy, and greater exposure to peer influences. When familial bonds are fragile, alienated, or characterized by mutual hostility, youth face heightened vulnerability to a cascade of negative developmental trajectories, including early initiation of alcohol, tobacco, and illicit drug use, school dropout, association with antisocial peers, and delinquent behaviors.

In public health and community prevention, there was a historical deficit of concise assessment tools capable of capturing familial protective factors within large-scale school-based surveys. Lengthy diagnostic instruments (e.g., the Inventory of Parent and Peer Attachment [IPPA] or the Adult Attachment Interview [AAI]) are often too long for multi-construct epidemiological batteries that evaluate dozens of community, school, peer, and family domains simultaneously. The SSDP attachment scale was engineered to solve this logistical challenge, capturing the core emotional and disclosure components of parental attachment in four clear, non-burdensome questions that can be completed in under two minutes.

Beyond universal school-based epidemiological surveillance, the scale serves several practical research and clinical functions:

• Community Risk and Protective Factor Profiling: Used within the Communities That Care (CTC) framework to establish baseline community risk profiles, helping municipal prevention boards, school districts, and community coalitions identify specific deficits in family-level bonding within local student populations.
• Program Evaluation and Intervention Monitoring: Utilized as a standardized pre- and post-test outcome measure to quantify the direct impact of evidence-based family-strengthening programs (e.g., the Guiding Good Choices program, formerly Preparing for the Drug Free Years; and Strengthening Families Program).
• Longitudinal Etiological Research: Employed in structural equation modeling and latent growth curve analyses to track how fluctuating parental attachment across ages 10 to 18 mediates or moderates the transmission of genetic, temperamental, or neighborhood risk into antisocial behavior.
• Clinical and School Counseling Triage: Deployed by school social workers and adolescent behavioral specialists as a preliminary screener to detect emotional detachment or systemic communication breakdowns between struggling students and their primary caretakers.

5. Psychological Construct

The construct assessed by this scale is adolescent parental attachment, conceptualized not merely as an infant behavioral instinct, but as a cognitive-affective relational bond characterized by perceived emotional closeness, mutual trust, and willing disclosure of personal feelings. Rooted in sociological control theories and developmental attachment paradigms, this construct emphasizes two interrelated operational dimensions across maternal and paternal subsystems:

1. Affective Closeness (Perceived Emotional Bond)

Affective closeness reflects the subjective perception that a parent is a secure base, a source of genuine emotional warmth, and a vital figure in the youth’s emotional life. This dimension is tapped directly by the inquiries: “Do you feel very close to your mother?” and “Do you feel very close to your father?” Rather than assessing concrete physical co-presence, this facet assesses the internalized emotional connection. High affective closeness reflects positive emotional resonance, feelings of security, and reciprocal affection. In developmental psychopathology, an adolescent who endorses high closeness perceives parental approval and valuation, fostering self-worth and prosocial self-regulation. Conversely, an absence of affective closeness indicates emotional estrangement, emotional neglect, or relational indifference, which can sever the adolescent’s developmental tether to prosocial family norms.

2. Interpersonal Communicative Disclosure (Shared Thoughts and Feelings)

Communicative disclosure reflects the expressive, communicative component of attachment, assessed by: “Do you share your thoughts and feelings with your mother?” and “Do you share your thoughts and feelings with your father?” Adolescent psychological development involves privacy and identity differentiation; therefore, an adolescent’s voluntary choice to reveal internal psychological states, emotional vulnerabilities, and personal thoughts to a parent serves as a strong indicator of relational security. Communicative openness requires trust that the parent will respond with empathy rather than harsh punitiveness, ridicule, or boundary violations. High scores indicate a high-trust relational channel where adolescents seek guidance, share aspirations, and process emotional distress within the parent-child relationship.

3. Dyadic Differentiation: Maternal vs. Paternal Subsystems

A critical psychometric and conceptual property of the tool is the parallel formulation for maternal and paternal parental figures. Contemporary developmental psychology recognizes that maternal and paternal attachment relationships can operate via distinct attachment pathways. By assessing mother and father relationships separately, the tool allows researchers to evaluate:

• Additive Buffering Effects: Whether dual-parent attachment confers compound protection against psychological maladjustment compared to single-parent attachment.
• Compensatory Dynamics: Whether a strong, supportive bond with one parent can buffer against the negative developmental sequelae of a strained or absent relationship with the other.
• Gender-Dyadic Nuances: Gender-specific variations in disclosure patterns (e.g., mother-daughter vs. father-son dyadic communication dynamics).

6. Theoretical Framework

The development of the Attachment to Parents—Seattle Social Development Project scale is grounded in two major psychological and sociological frameworks: John Bowlby’s Attachment Theory and the Social Development Model (SDM) developed by J. David Hawkins and Richard F. Catalano (Catalano & Hawkins, 1996; Hawkins & Weis, 1985), which synthesizes Travis Hirschi’s Social Control Theory with Social Learning Theory (Bandura, 1977) and Differential Association Theory (Sutherland, 1947).

Bowlby’s Internal Working Models in Adolescence

According to John Bowlby (1969/1982), early interactions between infants and their primary caregivers are internalized into enduring cognitive representations termed internal working models of self and others. In adolescence, these working models guide interpersonal expectations, affect regulation, and autonomous exploration. Securely attached adolescents view parents as reliable sources of support. They feel comfortable using parents as a safe haven when stressed, promoting autonomous exploration and socioemotional competence. The SSDP attachment scale operationalizes these internal working models through questions on closeness and emotional disclosure.

The Social Development Model (SDM)

While classical attachment theory primarily addresses emotional security, the Social Development Model integrates this construct into a broader theory of prosocial and antisocial behavioral trajectories. The SDM posits that throughout human development, individuals form social bonds within four key socialization units: families, schools, peer groups, and neighborhoods.

The model outlines a structural developmental sequence through which parental attachment develops and inhibits delinquency:

1. Opportunities for Prosocial Involvement: Families provide adolescents with opportunities to contribute meaningful effort to familial life.
2. Interaction and Capability: Adolescents actively interact with parents, utilizing personal and cognitive skills.
3. Reinforcement and Rewards: Parents consistently reinforce positive behavior through warmth, acknowledgment, and approval.
4. Formation of the Social Bond: Consistent, positively reinforced interactions produce a strong social bond comprised of attachment (emotional closeness), commitment (investment in the family unit), and belief (internalization of parental values).
5. Inhibition of Problem Behaviors: Once an adolescent develops a strong emotional attachment to prosocial parents, they experience a psychological stake in conformity. Engaging in substance abuse or delinquency risks parental disapproval and threatens the emotional bond. Consequently, attachment functions as a strong protective mechanism against delinquent peer pressures.

7. Validity

The psychometric validity of the scale has been evaluated across multiple longitudinal cohorts from the original Seattle Social Development Project, as well as large-scale statewide validation studies of the Communities That Care Youth Survey (Arthur et al., 2002; Glaser et al., 2005).

Construct and Structural Validity

Construct validity is supported by confirmatory factor analyses showing that parental attachment items load strongly onto a distinct familial protective factor that is psychometrically separable from other family domains, such as Parental Monitoring, Family Discipline, Family Opportunities for Prosocial Involvement, and Family Rewards for Prosocial Involvement (Arthur et al., 2002). Standardized factor loadings across diverse demographic samples consistently exceed .65 for maternal items and .70 for paternal items, demonstrating strong construct representation.

Convergent Validity

The scale demonstrates expected convergent correlations with established, lengthier family assessment instruments. Statistically significant positive correlations have been observed with:

• The Inventory of Parent and Peer Attachment (IPPA; Armsden & Greenberg, 1987) Parental Trust and Parental Communication subscales (correlations typically ranging from r = .55 to r = .71, p < .001).
• The Family Assessment Measure (FAM-III; Skinner et al., 1983) Relationship Closeness indicators (r ≈ .48 to .62).
• Measures of adolescent perceived parental warmth and acceptance from the Child’s Report of Parental Behavior Inventory (CRPBI; Schaefer, 1965).

Discriminant Validity

Discriminant validity is supported by low to negligible correlations with unrelated constructs, such as neighborhood physical decay (r < .08) and general physical health status (r < .10). Furthermore, multigroup structural equation modeling demonstrates that while parental attachment shares variance with general positive affectivity, it forms an empirically distinct construct that uniquely predicts prosocial outcomes even after controlling for neuroticism, extroversion, and social desirability biases.

Predictive and Criterion-Related Validity

Longitudinal findings from the SSDP and CTC cohorts have repeatedly documented the scale’s predictive power across diverse behavioral, emotional, and social outcomes:

• Substance Abuse Deterrence: Adolescents scoring in the upper quartile of the parental attachment scale show a 35% to 50% reduced likelihood of initiating binge drinking, cigarette smoking, cannabis use, and non-medical prescription opioid use during subsequent school years (Arthur et al., 2002; Hawkins et al., 1992).
• Violence and Violent Delinquency: Cataloged in the CDC compendium of violence-related assessment tools (Dahlberg et al., 2005), longitudinal analyses indicate that high parental attachment scores in middle school significantly predict reduced participation in weapon carrying, physical fighting, and gang membership during late high school.
• Internalizing Distress: High attachment scores are significantly and inversely associated with depressive symptoms, generalized anxiety, suicidal ideation, and non-suicidal self-injury, underscoring the protective role of parental bonds for emotional regulation.

8. Reliability

The reliability of the Attachment to Parents scale has been assessed in numerous large, demographically diverse adolescent samples across the United States and internationally.

Internal Consistency Reliability

In the seminal psychometric validation study conducted by Arthur, Hawkins, Pollard, Catalano, and Baglioni (2002), which analyzed survey data from 72,130 students across Grades 6 through 12 in multiple U.S. states, the internal consistency of the overall four-item Attachment to Parents scale was robust, exhibiting a Cronbach’s alpha of α = .78. Subsequent evaluations in statewide CTC surveys across Oregon, Utah, Colorado, and Washington have reported alpha coefficients ranging from α = .76 to α = .84.

When examined as separate dyadic subscales, the maternal dyad (Items 1 and 2) yields a Spearman-Brown split-half reliability and inter-item correlation coefficient typically exceeding r = .70 (p < .001). The paternal dyad (Items 3 and 4) demonstrates an inter-item correlation coefficient typically ranging from r = .74 to r = .82 (p < .001), indicating strong internal consistency within each parental focus.

Test-Retest Stability

In short-interval test-retest reliability designs (typically evaluated over a 2-to-3 week interval under identical school-based testing conditions), the scale demonstrated acceptable temporal stability, with intraclass correlation coefficients (ICC) ranging between .72 and .79. Over longer developmental horizons (1-to-2 year longitudinal waves), stability coefficients reflect modest stability (r ≈ .42 to .55), which aligns with developmental theory regarding the normative renegotiation of adolescent-parent relationships during puberty and the transition into high school.

9. Factor Analysis

The underlying factor structure of the scale has been investigated through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA), using robust estimation methods for ordinal data (e.g., Weighted Least Squares Means and Variance adjusted, WLSMV).

Exploratory Factor Analysis (EFA)

During initial scale development, EFA conducted on adolescent responses revealed an initial eigenvalue structure with a dominant first factor (eigenvalue > 2.50) explaining over 63% of the total scale variance. When unconstrained, principal axis factoring with promax rotation revealed two distinct, highly correlated factors corresponding directly to parental gender:

• Factor 1: Maternal Attachment — Item 1 (closeness) and Item 2 (disclosure) loaded cleanly onto this factor (loadings of .81 and .84, respectively), with cross-loadings onto the paternal factor below .15.
• Factor 2: Paternal Attachment — Item 3 (closeness) and Item 4 (disclosure) loaded onto this factor (loadings of .86 and .88, respectively), with cross-loadings onto the maternal factor below .12.

Confirmatory Factor Analysis (CFA) and Model Fit

Structural equation modeling studies evaluate two competing structural models for the four items:

1. Unidimensional General Attachment Model: All four items load directly onto a single latent construct: General Parental Attachment.
2. Correlated Two-Factor Dyadic Model: Two latent factors (Maternal Attachment and Paternal Attachment) are allowed to covary freely.
3. Higher-Order Hierarchical Model: Maternal and Paternal first-order factors load onto a superordinate second-order Parental Attachment factor.

Fit statistics for the correlated two-factor dyadic model and the higher-order hierarchical model routinely demonstrate acceptable fit across large adolescent cohorts:

• Comparative Fit Index (CFI): .985 to .998
• Tucker-Lewis Index (TLI): .978 to .994
• Root Mean Square Error of Approximation (RMSEA): .032 to .048 (90% Confidence Interval: [.021, .056])
• Standardized Root Mean Square Residual (SRMR): .018 to .025

Parameter estimates and standardized item factor loadings (λ) from typical structural analyses are detailed below:

The correlation between the latent Maternal Attachment and Paternal Attachment factors typically hovers between r = .45 and r = .60, indicating that while youth often experience congruent emotional climates across parents, the maternal and paternal relational systems remain distinct.

Measurement Invariance

Rigorous multigroup CFA tests have demonstrated full metric and scalar measurement invariance across biological sex (male vs. female) and school grade bands (middle school vs. high school cohorts). Additionally, extensive testing across diverse ethnic cohorts (White, African American, Hispanic/Latino, Native American, Asian/Pacific Islander) supports configural and metric invariance, confirming that the scale functions similarly across diverse adolescent populations.

10. Instrument / Measurement Tool

• Instrument Name: Attachment to Parents—Seattle Social Development Project (also administered as the Attachment to Parents subscale within the Communities That Care Youth Survey).
• Target Population: Adolescents and youth aged 10 to 18 (typically Grades 6 through 12).
• Administration Format: Self-administered paper-and-pencil or computerized online self-report survey; suitable for individual administration or mass classroom group testing.
• Total Item Count: 4 items.
• Administration Time: Approximately 1 to 2 minutes.
• Response Scale: 4-point emphatic categorical scale:
  • YES! = 4 (Strong positive affirmation)
  • yes = 3 (Moderate positive affirmation)
  • no = 2 (Moderate negative denial)
  • NO! = 1 (Strong negative denial)
• Scoring Protocol:
  • Overall Sum Score: Sum of points for Items 1 through 4 (Theoretical range: 4 to 16).
  • Maternal Attachment Subscore: Sum of points for Item 1 and Item 2 (Theoretical range: 2 to 8).
  • Paternal Attachment Subscore: Sum of points for Item 3 and Item 4 (Theoretical range: 2 to 8).
  • Mean Scoring Approach: Alternatively, items can be averaged (sum divided by number of valid completed items) to retain the original 1 to 4 metric.
  • Directionality: Higher scores consistently designate higher levels of parental attachment, emotional closeness, and interpersonal communicative disclosure.
  • Handling Non-Applicable/Absent Parents: In standard CTC administration instructions, respondents who do not have contact with a biological mother or father are instructed to answer regarding step-parents, adoptive parents, foster parents, or primary adult guardians, or skip the relevant questions. If scored as mean composites, the available parental scores can be analyzed independently.

11. Permissions & Fee and Test Year

• Initial Publication Year: 2002 (formalized in the landmark psychometric paper by Arthur et al., 2002; foundational longitudinal items administered in SSDP protocols beginning in the mid-1980s).
• Author/Copyright Holders: Developed by Michael W. Arthur, J. David Hawkins, John A. Pollard, Richard F. Catalano, and Angelo J. Baglioni Jr., at the Social Development Research Group (SDRG), University of Washington.
• Availability and Licensing: The instrument is placed in the public domain for research, clinical, and public health evaluation purposes. It was funded in significant part through federal grants from the National Institute on Drug Abuse (NIDA), the Office of Juvenile Justice and Delinquency Prevention (OJJDP), and the Centers for Disease Control and Prevention (CDC).
• Assessment Compendium Inclusion: Published as an open-access public health assessment tool on pages 296–297 in the official CDC compendium: Measuring Violence-Related Attitudes, Behaviors, and Influences Among Youths: A Compendium of Assessment Tools (Dahlberg, Toal, Swahn, & Behrens, 2005).
• Commercial Fees: No royalty fees, registration charges, or licensing costs are required for non-commercial educational, epidemiological, or research administration. Attribution of the original authors and the Social Development Research Group is requested in academic dissemination.

12. References

• Armsden, G. C., & Greenberg, M. T. (1987). The inventory of parent and peer attachment: Individual differences and their relationship to psychological well-being in adolescence. Journal of Youth and Adolescence, 16(5), 427–454. https://doi.org/10.1007/BF02138822
• Arthur, M. W., Hawkins, J. D., Pollard, J. A., Catalano, R. F., & Baglioni, A. J. (2002). Measuring risk and protective factors for substance use, delinquency, and other adolescent problem behaviors: The Communities That Care Youth Survey. Evaluation Review, 26(6), 575–601. https://doi.org/10.1177/0193841X0202600601
• Bandura, A. (1977). Social Learning Theory. Prentice-Hall.
• Bowlby, J. (1982). Attachment and loss: Vol. 1. Attachment (2nd ed.). Basic Books. (Original work published 1969).
• Catalano, R. F., & Hawkins, J. D. (1996). The social development model: A theory of antisocial behavior. In J. D. Hawkins (Ed.), Delinquency and crime: Current theories (pp. 149–197). Cambridge University Press. https://doi.org/10.1017/CBO9780511527623.007
• Dahlberg, L. L., Toal, S. B., Swahn, M., & Behrens, C. B. (2005). Measuring violence-related attitudes, behaviors, and influences among youths: A compendium of assessment tools (2nd ed., pp. 296–297). Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. https://www.cdc.gov/violenceprevention/pdf/YV_Compendium.pdf
• Glaser, R. R., Van Horn, M. L., Arthur, M. W., & Hawkins, J. D. (2005). Measurement properties of the Communities That Care Youth Survey across demographic groups. Journal of Quantitative Criminology, 21(1), 73–102. https://doi.org/10.1007/s10940-004-1788-1
• Hawkins, J. D., Catalano, R. F., & Miller, J. Y. (1992). Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention. Psychological Bulletin, 112(1), 64–105. https://doi.org/10.1037/0033-2909.112.1.64
• Hawkins, J. D., & Weis, J. G. (1985). The social development model: An integrated approach to delinquency prevention. Journal of Primary Prevention, 6(2), 73–97. https://doi.org/10.1007/BF01325432
• Hirschi, T. (1969). Causes of Delinquency. University of California Press.
• Schaefer, E. S. (1965). Children’s reports of parental behavior: An inventory. Child Development, 36(2), 413–424. https://doi.org/10.2307/1126465
• Skinner, H. A., Steinhauer, P. D., & Santa-Barbara, J. (1983). The Family Assessment Measure. Canadian Journal of Community Mental Health, 2(2), 91–105. https://doi.org/10.5964/ejcop.v1i2.15
• Substance Abuse and Mental Health Services Administration. (2014). Communities That Care Youth Survey (CTC-YS) User’s Guide. U.S. Department of Health and Human Services. https://store.samhsa.gov/shin/content//CTC020/CTC020.pdf
• Sutherland, E. H. (1947). Principles of Criminology (4th ed.). J.B. Lippincott.

13. Items of the Scale