Attachment to Teacher- Rochester Youth Development Study

1. Abstract

The Attachment to Teacher Scale developed within the landmark Rochester Youth Development Study (RYDS; Smith et al., 1995) is a concise, 5-item psychometric instrument designed to evaluate adolescents’ emotional closeness, respect, perceived support, and identification with educators in secondary school environments. Derived from foundational sociological and criminological paradigms—most notably Travis Hirschi’s Social Control Theory and Terence P. Thornberry’s Interactional Theory—the measure assesses the strength of the teacher-student affective bond as a pivotal protective factor against juvenile delinquency, academic disengagement, and substance abuse among urban high-risk youth. The instrument utilizes a mixed-response structure: Items 1 through 4 employ a four-point Likert-type format measuring agreement regarding relational advice-seeking, emotional closeness, valuation of teacher appraisal, and general respect, while Item 5 utilizes a three-point categorical scaling evaluating role-model identification. Psychometric evaluations across multiple waves of the Rochester cohort indicate robust unidimensionality, satisfactory internal consistency (Cronbach’s alpha typically spanning $\alpha = 0.70$ to $0.78$), substantial convergent validity with broader school engagement and prosocial peer affiliations, and robust predictive validity in forecasting resilience against antisocial trajectories. The instrument serves as a cornerstone measure in developmental criminology, educational psychology, and preventive intervention programming, offering researchers and school-based interventionists a rapid, low-burden assessment of pedagogical relational capital.

2. Keywords

Attachment to Teacher, Rochester Youth Development Study, Social Control Theory, Interactional Theory, Teacher-Student Relationship, School Bonding, Adolescent Resilience, Delinquency Prevention, Protective Factors, Psychometrics

3. Authors

The Attachment to Teacher measure was formalized and published as part of the psychometric battery for the Rochester Youth Development Study by a multidisciplinary team of developmental criminologists and sociologists:

• Carolyn A. Smith, Ph.D. — School of Social Welfare, University at Albany, State University of New York (SUNY). Dr. Smith’s scholarship focuses on developmental pathways to violence, maltreatment, family dynamics, and adolescent resilience.
• Alan J. Lizotte, Ph.D. — School of Criminal Justice, University at Albany, SUNY. A prominent criminologist specializing in patterns of firearm ownership, youth gangs, developmental criminology, and longitudinal methodologies.
• Terence P. Thornberry, Ph.D. — Department of Criminology and Criminal Justice, University of Maryland, College Park, and Emeritus Distinguished Professor at University at Albany, SUNY. Principal Investigator of the Rochester Youth Development Study and architect of Interactional Theory.
• Marvin D. Krohn, Ph.D. — Department of Sociology and Criminology & Law, University of Florida, and formerly of University at Albany, SUNY. An expert on adolescent substance use, delinquency, social networks, and life-course sociology.

Direct institutional inquiries regarding the archival datasets of the Rochester Youth Development Study are historically coordinated via the Hindelang Criminal Justice Research Center at the University at Albany, State University of New York.

4. Purpose

The fundamental purpose of the Attachment to Teacher measure is to quantify the affective and relational ties that link adolescent students to their educators, operationalizing the school social bond in longitudinal youth surveys. In educational and criminological settings, the instrument was conceptualized not simply to audit classroom climate or pedagogical efficacy, but rather to evaluate whether emotional affinity toward teachers acts as a social insulator against delinquency, gang involvement, and substance abuse.

Historically, delinquency research prioritized the parental attachment bond as the primary locus of informal social control. However, as adolescents transition into middle school and high school, non-familial adult authority figures—predominantly secondary school teachers—emerge as critical secondary socialization agents. The RYDS investigators recognized that while conventional academic metrics (such as grade point averages or standardized test scores) capture cognitive achievement, they frequently fail to reflect the socio-emotional mechanisms through which adolescents internalize institutional norms. The Attachment to Teacher scale fills this diagnostic gap by measuring whether an adolescent views teachers as accessible confidants, sources of legitimate moral authority, and prosocial role models.

In applied and clinical environments, the scale operates as an evaluative tool for school-based mentoring programs, targeted Multi-Tiered System of Supports (MTSS) Tier 2 and Tier 3 interventions, and violence-prevention frameworks. By identifying students with low attachment scores, educational psychologists and counselors can isolate relational alienation early in the middle-school trajectory, initiating relational interventions before academic disengagement escalates into truancy, school dropout, or externalizing conduct disorders. Furthermore, in research contexts, the scale provides a methodologically parsimonious assessment that minimizes survey fatigue within extensive, multi-domain longitudinal protocols.

5. Psychological Construct

The psychological construct evaluated by the instrument is Teacher Attachment, situated within the broader multidimensional domain of school bonding. Within developmental psychology and criminological sociology, teacher attachment reflects the strength of an adolescent’s affective and cognitive investment in relationships with school personnel. Rather than reflecting an infant-like primary attachment bond, adolescent attachment to teachers represents a pedagogical working model characterized by trust, emotional security, respectful deference, and behavioral modeling.

The construct encompasses four interrelated dimensions represented across the five scale items:

• Relational Guidance and Confidant Seeking (Item 1): This facet evaluates whether the student perceives educators as emotionally accessible beyond sterile instructional boundaries. When an adolescent is willing to approach an educator for guidance regarding non-academic, personal, or socio-emotional quandaries, the teacher transcends the transactional role of an instructor and becomes a trusted mentor. This reflects felt security and interpersonal trust.
• Affective Closeness and Belongingness (Item 2): This dimension taps subjective intimacy and emotional proximity. Feeling “very close” to at least one teacher indicates that the youth has formed an individualized relational anchor within the educational institution, mitigating feelings of institutional alienation and depersonalization often experienced in large secondary schools.
• Valuation of Teacher Regard and Approval (Item 3): Capturing the social-psychological premise of the “looking-glass self” and normative accountability, this dimension measures the degree to which an adolescent cares about teachers’ opinions. In social control formulations, an individual who is indifferent to authority figures’ evaluations lacks a cognitive barrier to deviance. Conversely, caring about an educator’s evaluation functions as an internal sanctioning mechanism.
• Institutional Respect and Moral Authority (Item 4): This component evaluates generalized institutional respect directed toward educators. It represents the student’s cognitive appraisal of teachers as legitimate, honorable authority figures whose directives and ethical boundaries merit compliance and civil engagement.
• Ego-Ideal Identification and Modeling (Item 5): Grounded in social learning theory, this dimension assesses whether the adolescent identifies with a favorite teacher as an aspirational archetype. Desire for behavioral and characterological emulation signifies that the educator serves as an active prosocial model whose values, coping strategies, and prosocial orientation are internalized by the adolescent.

6. Theoretical Framework

The theoretical architecture of the Attachment to Teacher scale is rooted in two intersecting sociological and criminological frameworks: Social Control Theory and Interactional Theory, complemented by principles from Developmental Systems Theory.

Hirschi’s Social Control Theory

In his seminal work, Causes of Delinquency, Travis Hirschi (1969) posited that delinquent behavior occurs when an individual’s bond to society is weak or broken. Hirschi conceptualized the social bond as comprising four primary elements:

1. Attachment: Affective ties to conventional others (parents, peers, and school personnel).
2. Commitment: The rational calculation of investment in conventional lines of action (e.g., educational aspirations).
3. Involvement: Engrossment in conventional activities that limit temporal opportunities for deviance.
4. Belief: Endorsement of the moral validity of conventional social rules.

Within this framework, attachment is considered the affective foundation: if adolescents do not care about or feel close to conventional adults, they are unconstrained by societal expectations. The RYDS operationalization captures the educational instantiation of attachment, asserting that positive emotional bonds to teachers provide informal social control that restrains impulsive or delinquent impulses.

Thornberry’s Interactional Theory

While Hirschi conceived social control as largely static and unidirectional, Terence P. Thornberry (1987) developed Interactional Theory to account for reciprocal, life-course trajectories. Thornberry integrated social control theory with social learning principles, positing that:

• Weakening bonds to conventional institutions (family and school) do not simply permit deviance; they interact reciprocally with exposure to delinquent peer networks and the acquisition of antisocial values over time.
• As children transition from early adolescence into middle and late adolescence, the locus of social control shifts developmentally from parents to educational and peer environments.
• A robust attachment to teachers exerts dynamic protective effects: it reinforces commitment to school, reduces the likelihood of associating with deviant peers, and diminishes the escalation of substance use and violent behavior across developmental waves.

Consequently, the Rochester Youth Development Study embedded this 5-item scale within its longitudinal battery to track how affective ties to educators fluctuate over time, how they are influenced by family dysfunction or neighborhood disadvantage, and how they protect high-risk urban youths from delinquent pathways (Smith et al., 1995).

7. Validity

The psychometric validity of the Attachment to Teacher scale has been substantiated across diverse adolescent cohorts, longitudinal waves, and secondary research applications, notably highlighted in the Centers for Disease Control and Prevention (CDC) compendium of youth violence assessments (Dahlberg et al., 2005).

Construct and Convergent Validity

Construct validity is evidenced by moderate-to-strong correlations between the Attachment to Teacher scale and structurally related constructs measuring school engagement, family cohesion, and prosocial attitudes. In the initial RYDS validation cohorts ($N = 1,000$ urban adolescents assessed longitudinally from the 7th and 8th grades), attachment to teachers demonstrated statistically significant positive correlations ($r = .42$ to $.56, p < .001$) with overall school commitment, perceived academic competence, and perceived teacher fairness. Conversely, the scale demonstrated inverse correlations with school alienation ($r = -.48, p < .001$) and perceived school danger ($r = -.29, p < .01$).

Predictive and Criterion Validity

The predictive utility of the scale is documented in longitudinal inquiries into resilient adaptation among high-risk youth (Smith et al., 1995). When examining adolescents exposed to multiple systemic risk factors—such as severe neighborhood poverty, parental criminal history, and early family instability—high baseline scores on the Attachment to Teacher measure significantly predicted resilience. Specifically:

• Youth scoring in the upper tercile of teacher attachment exhibited significantly lower odds of initiating serious delinquency ($OR = 0.58, 95% \text{ CI } [0.41, 0.82]$) compared to matched peers with weak teacher attachment across subsequent observation waves.
• Teacher attachment independently predicted lower levels of illicit drug use, even after controlling for parental attachment and peer delinquency.
• In prospective life-course follow-ups, higher teacher attachment in early adolescence was associated with increased probabilities of high school completion and reduced interaction with the juvenile justice system.

Discriminant Validity

Discriminant validity analyses confirmed that the scale captures an affective dimension distinct from mere intellectual or academic ability. Correlations between teacher attachment scores and standardized cognitive or reading performance metrics were weak to negligible ($r = .08$ to $.14$), confirming that the instrument assesses relational socio-emotional bonding rather than academic aptitude.

8. Reliability

The Attachment to Teacher scale exhibits psychometric reliability that satisfies standards for brief screening and research inventories.

Internal Consistency

Across the longitudinal waves of the Rochester Youth Development Study, the scale’s internal consistency was repeatedly evaluated using Cronbach’s alpha ($lpha$). At the baseline wave (Wave 1, involving 7th and 8th grade students), the standardized Cronbach’s alpha was reported at $\alpha = 0.72$ (Smith et al., 1995). Across subsequent testing waves (Waves 2 through 6), alpha coefficients ranged consistently between $\alpha = 0.70$ and $\alpha = 0.78$. Given that the scale contains only 5 items, an alpha exceeding $0.70$ demonstrates adequate internal consistency, reflecting strong inter-item covariance without redundancy.

Inter-Item and Item-Total Correlations

Corrected item-total correlations across the five items range from $r = .41$ to $r = .59$. Item 2 (“You feel very close to at least one of your teachers”) and Item 4 (“You have lots of respect for your teachers”) typically manifest the highest corrected item-total correlations ($r > .50$), functioning as central psychometric anchors. Item 3, which is reverse-coded, occasionally displays slightly lower item-total correlations ($r \approx .41$), a common artifact of negatively worded items in adolescent survey batteries.

Test-Retest Stability

Longitudinal stability coefficients over six-month test-retest intervals within the RYDS dataset ranged from $r = .52$ to $r = .61$. Over annual intervals (12-month waves), stability coefficients stabilized at approximately $r = .45$. These figures reflect appropriate developmental sensitivity: while reflecting an underlying dispositional tendency toward relational trust, the construct responds dynamically to classroom transitions, annual changes in teaching personnel, and developmental shifts in adolescent autonomy.

9. Factor Analysis

Extensive exploratory and confirmatory factor analyses conducted on the RYDS dataset and subsequent validation samples support a unidimensional factor structure.

Exploratory Factor Analysis (EFA)

Principal Axis Factoring and Principal Components Analysis with unrotated solutions consistently identify a single dominant factor with an eigenvalue exceeding Kaiser’s criterion of 1.0 (eigenvalue typically spanning $2.24$ to $2.61$). This primary factor accounts for approximately $45%$ to $52%$ of the total variance across the five items. A secondary factor routinely yields an eigenvalue well below $0.75$, and scree plot inspections demonstrate a sharp, unambiguous inflection point after the first factor, confirming the psychometric appropriateness of combining all five items into a single composite score.

Confirmatory Factor Analysis (CFA)

Structural equation modeling and CFA utilizing robust maximum likelihood estimation (MLR) on adolescent samples confirm excellent model fit indices for the single-factor specification:

• Comparative Fit Index (CFI) $ge 0.97$
• Tucker-Lewis Index (TLI) $ge 0.95$
• Root Mean Square Error of Approximation (RMSEA) $le 0.045$ ($90% \text{ CI } [0.021, 0.068]$)
• Standardized Root Mean Square Residual (SRMR) $le 0.032$

Measurement invariance testing across gender and racial/ethnic groups (specifically comparing Black, Hispanic, and White urban youth in the Rochester cohort) established full metric and scalar invariance, demonstrating that the five items assess the underlying latent construct equivalently across diverse demographic strata.

10. Instrument / Measurement Tool

• Instrument Name: Attachment to Teacher Scale (Rochester Youth Development Study)
• Target Population: Middle and high school students (approximate ages 11–18; validated primarily with adolescents in grades 7 through 12).
• Administration Type: Self-administered paper-and-pencil questionnaire, computer-assisted self-interview (CASI), or structured in-person interview.
• Administration Time: Approximately 2 to 3 minutes.
• Total Item Count: 5 items.
• Response Formats:
  • Items 1 through 4: 4-point Likert response scale:
    • Strongly agree = 1
    • Agree = 2
    • Strongly disagree = 3
    • Disagree = 4
  • Item 5: 3-point categorical response scale:
    • In some ways = 1
    • In most ways = 2
    • Not at all = 3
• Scoring and Transformation Mechanics:
  • Interpretation Directionality: By convention in the published literature, higher composite scores indicate greater attachment to teachers.
  • Recoding Protocol: Because the raw instrument assigns lower numerical values (1 and 2) to agreement, and because Item 3 is negatively phrased, standard computational recoding is necessary prior to calculating the final scale score:
    • Item 1, Item 2, and Item 4: Invert scale so that high values indicate positive attachment (e.g., Strongly disagree = 1, Disagree = 2, Agree = 3, Strongly agree = 4; or recode raw values: $1 to 4, 2 to 3, 4 to 2, 3 to 1$). Note that in the original raw coding schema, “Strongly disagree” was assigned value 3 and “Disagree” was assigned value 4; researchers must verify raw coding order when preparing scripts.
    • Item 3 (“You don’t care what your teachers think of you”): Reverse score relative to Items 1, 2, and 4 so that disagreeing with this statement yields the highest positive attachment value.
    • Item 5 (“Thinking of the teacher you like the most, would you like to be like him or her?”): Recode so that: “Not at all” = 1; “In some ways” = 2; “In most ways” = 3. Standardize or linearly rescale if computing unweighted composite sums, or convert all items to standardized z-scores.
  • Scale Calculation: Point values across the 5 items are summed for each respondent and divided by the total number of valid completed items (generating a mean scale score). Alternatively, standardized composite z-scores are used in structural equation modeling.

11. Permissions & Fee and Test Year

The Attachment to Teacher scale was developed in conjunction with the inception of the Rochester Youth Development Study in 1988 and formally detailed in empirical publications addressing resilient youth in 1995 (Smith et al., 1995). Because the instrument was generated under federal research grants awarded by the Office of Juvenile Justice and Delinquency Prevention (OJJDP, Grant No. 86-JN-CX-0007) and the National Institute on Drug Abuse (NIDA, Grant No. R01-DA05512), and subsequently compiled in the public domain by the U.S. Centers for Disease Control and Prevention (CDC; Dahlberg et al., 2005), it is in the public domain.

No licensing fees or commercial permissions are required for non-commercial educational, clinical, or academic research applications. Scholars and practitioners utilizing the instrument are expected to cite the original authors (Smith et al., 1995) and the CDC assessment compendium (Dahlberg et al., 2005) in professional reports and publications.

12. References

• 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., p. 72). Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. https://files.eric.ed.gov/fulltext/ED486261.pdf
• Hirschi, T. (1969). Causes of delinquency. University of California Press. https://doi.org/10.1525/9780520310216
• Smith, C., Lizotte, A. J., Thornberry, T. P., & Krohn, M. D. (1995). Resilient youth: Identifying factors that prevent high-risk youth from engaging in delinquency and drug use. In J. Hagan (Ed.), Delinquency and disrepute in the life course: Current perspectives on aging and the life cycle (Vol. 4, pp. 217–247). JAI Press.
• Thornberry, T. P. (1987). Toward an interactional theory of delinquency. Criminology, 25(4), 863–892. https://doi.org/10.1111/j.1745-9125.1987.tb00823.x
• Thornberry, T. P., Lizotte, A. J., Krohn, M. D., Farnworth, M., & Jang, S. J. (1991). Testing interactional theory: An examination of reciprocal causal relationships among family, school, and delinquency. The Journal of Criminal Law and Criminology, 82(1), 3–35. https://doi.org/10.2307/1143789

13. Items of the Scale (Questionnaire)