Behavioral Affect Rating Scale (BARS)

Abstract

The Behavioral Affect Rating Scale (BARS) is a widely utilized psychometric instrument developed by Rand D. Conger (1989) in conjunction with the landmark Iowa Youth and Family Project (IYFP) at Iowa State University. Designed to evaluate the affective and behavioral climate within dyadic family relationships—most notably parent-child, interparental, and sibling subsystems—the BARS operationalizes interpersonal affect through two core latent constructs: Parental Warmth/Support (positive behavioral affect) and Parental Hostility/Rejection (negative behavioral affect). While abbreviated variations (such as 12-item versions) exist for rapid screening and multi-wave field surveys, the foundational instrument comprises a 22-item self-report or collateral-report inventory administered across mothers, fathers, and offspring. Respondents rate the frequency of specific, observable interactional behaviors on a 7-point Likert-type scale ranging from 1 (Never) to 7 (Always), with adapted 4-point and 5-point variants validated across diverse longitudinal studies. Psychometric investigations consistently indicate robust internal consistency across informants, with Cronbach’s alpha ($lpha$) coefficients typically ranging between .81 and .94 for both warmth and hostility subscales across maternal, paternal, and youth reports. Factor-analytic evaluations robustly validate a two-dimensional oblique or orthogonal factor structure, exhibiting robust longitudinal, gender, and cross-cultural measurement invariance across European-American rural cohorts and ethnic-minority populations, including Mexican-origin families. The scale has served as a foundational empirical pillar for Conger’s Family Stress Model, elucidating the transactional pathways through which contextual stressors, such as economic hardship, undermine parenting practices and exacerbate adolescent internalizing and externalizing psychopathology. This article provides an exhaustive psychometric, theoretical, and practical review of the BARS, detailing its structural validity, longitudinal reliability, factor architecture, administrative guidelines, and full inventory items.

Keywords

Behavioral Affect Rating Scale, BARS, Rand D. Conger, Parental Warmth, Parental Hostility, Family Stress Model, Parent-Child Interaction, Dyadic Affect, Psychometrics, Adolescent Development, Observational-Self-Report Convergence, Measurement Invariance

Authors

The Behavioral Affect Rating Scale was originated and standardized under the direction of Rand D. Conger, Ph.D., Distinguished Professor Emeritus of Psychology and Human Development at the University of California, Davis, and former Director of the Center for Family Research in Rural Mental Health at Iowa State University. Dr. Conger’s pioneering contributions to developmental psychopathology, family sociology, and behavioral economics include the formalization of the Family Stress Model (FSM), which empirically links macro-level socioeconomic hardship to micro-level family emotional dynamics and adolescent outcomes.

The development, refinement, and psychometric validation of the BARS occurred in collaboration with key colleagues and methodologists associated with the Iowa Youth and Family Project (IYFP), the Family Transitions Project (FTP), and subsequent cross-cultural projects, including:

• Glen H. Elder Jr., Ph.D. – Research Professor of Sociology and Psychology at the University of North Carolina at Chapel Hill, renowned for his life-course perspective on human development.
• Ronald L. Simons, Ph.D. – Distinguished Research Professor of Sociology at the University of Georgia, whose research focuses on criminological mechanisms and family-of-origin parenting styles.
• Frederick O. Lorenz, Ph.D. – Emeritus Professor of Statistics and Psychology at Iowa State University, instrumental in developing multi-informant structural equation modeling (SEM) strategies validating the scale.
• Xiaojia Ge, Ph.D. – Late Professor of Human Development at the University of Minnesota and UC Davis, who extensively examined pubertal timing and parenting affect.
• Zoe E. Taylor, Ph.D. – Associate Professor of Human Development and Family Studies at Purdue University, who with Conger and Keith F. Widaman evaluated the cultural validity, factor structure, and measurement invariance of the BARS among Latino/Mexican-origin family systems.

Purpose

The primary diagnostic and empirical purpose of the Behavioral Affect Rating Scale (BARS) is to quantify the overt behavioral manifestations of positive and negative affect exchanged within primary dyadic interpersonal relationships, specifically parent-child and interparental configurations. Prior to the development of the BARS in the late 1980s, family assessment literature was heavily polarized between two methodological extremes: broad, global self-report questionnaires measuring general parental attitudes (e.g., “I love my child,” which suffered from pervasive social desirability and poor situational specificity) and micro-coded observational systems (which yielded high ecological validity but imposed substantial administrative burdens, prohibitive coding costs, and low feasibility for large epidemiological cohorts).

Conger conceived the BARS to bridge this methodological divide. Grounded in behavioral interaction paradigms, the BARS prompts respondents to report on concrete, observable behavioral acts occurring over a designated reference timeframe (typically the past 12 months, or the preceding 30 days in high-frequency assessment designs). Rather than asking respondents to evaluate ambiguous psychological constructs, items assess precise operational behaviors such as “Shout or yell at you because he/she was mad at you” or “Help you do something that was important to you.”

In clinical and developmental research, the BARS fulfills several critical functions:

• Testing Transactional Family Models: Serving as the operational core of the Family Stress Model, the scale maps the functional deterioration of positive behavioral warmth and the escalation of hostile, coercive behaviors under the influence of parental depression, marital distress, and financial strain.
• Multi-Informant Dyadic Assessment: Because the BARS items were engineered to possess parallel forms across informants, the scale allows researchers to collect reciprocal perspectives (e.g., mother-about-child, child-about-mother, father-about-child, child-about-father, and partner-about-partner). This facilitates the application of structural equation modeling (SEM) to extract latent traits free from single-reporter variance and shared method bias.
• Clinical Screening and Intervention Evaluation: In family therapy and parent-training programs (such as the Strengthening Families Program), the BARS provides sensitive, pre- and post-intervention outcome metrics. It reliably detects reductions in coercive, verbally harsh, and physically intimidating behaviors alongside increases in supportive, communicative, and validating interactions.
• Longitudinal Psychopathology Trajectory Mapping: The scale enables developmental psychopathology researchers to trace the prospective influence of early, chronic exposures to parental hostility versus warmth on adult outcomes, including substance use, antisocial behavior, relational insecurity, metabolic syndrome, and intergenerational transmission of harsh parenting.

Psychological Construct

The BARS captures interpersonal interaction through a dual-construct architectural framework. Rather than conceiving family climate along a single, bipolar continuum (ranging from severe hostility to high warmth), contemporary psychometric and developmental theory treats positive affect and negative affect as distinct, quasi-orthogonal behavioral axes. The BARS explicitly measures these two primary latent dimensions:

1. Parental Warmth / Support / Involvement (Positive Affect)

The Parental Warmth dimension assesses the frequency and intensity with which a parent exhibits emotional nurturance, validation, active empathy, instrumental assistance, and relational availability. Drawing upon attachment theory and Diana Baumrind’s classical typology of parenting styles (specifically the “responsiveness” dimension of authoritative parenting), this construct reflects actions that build a child’s psychological safety, felt security, and emotional competence. Sub-behavioral markers operationalized within this construct include:

• Affective Nurturance and Love: Direct physical and verbal demonstrations of love, closeness, and tenderness (e.g., “Act lovingly and affectionate toward you,” “Tell you that he/she loves you”).
• Instrumental and Cognitive Support: Active assistance with tasks, problems, or developmental milestones deemed important by the child (e.g., “Help you do something that was important to you”).
• Validation and Respect for Autonomy: Soliciting the youth’s perspective, active listening, and affirming competence (e.g., “Listen carefully to what you had to say,” “Ask you for your opinion or advice,” “Tell you you did a good job on something”).
• Shared Positive Hedonic States: Engaging in spontaneous joint mirth and humor (e.g., “Have a good laugh with you about something that was funny”).

2. Parental Hostility / Rejection / Conflict (Negative Affect)

The Parental Hostility dimension assesses the manifestation of aversive, coercive, emotionally punitive, and physically aggressive behavioral exchanges. Informed by Gerald Patterson’s coercion model and social learning formulations of family violence, this construct operationalizes the negative valence of relational interaction, encompassing behaviors that elicit fear, shame, anger, and hyperarousal in the offspring. Crucially, the BARS differentiates normative limit-setting from affective hostility. Sub-dimensions operationalized across items include:

• Overt Verbal Aggression: Expressive outbursts involving yelling, swearing, and insults (e.g., “Shout or yell at you because he/she was mad at you,” “Insult you or call you names when he/she was mad at you”).
• Severe Dysregulated Hostility: Explosive expressions of anger reflecting emotional dysregulation (e.g., “Get so mad at you that he/she seemed to lose control”).
• Coercive Domination and Psychological Control: Invalidation of the child’s agency, arbitrary blaming, and rigid authoritarian imposition (e.g., “Boss you around and make you do things his/her way,” “Blame you for things that were not your fault,” “Tell you she/he is right and you are wrong about things”).
• Physical Intimidation and Coercion: Mild to moderate physical victimization and threats of harm (e.g., “Hit, push, grab, or shove you,” “Throw something at you when he/she was mad”).
• Emotional Rejection and Stonewalling: Active withdrawal of attention and passive-aggressive dismissiveness (e.g., “Ignore you when you tried to talk to him/her”).

Theoretical Framework

The architectural configuration and operational rationale of the BARS are grounded in an integration of five foundational theoretical paradigms in developmental psychology and family sociology:

1. The Family Stress Model (FSM)

Formulated primarily by Rand D. Conger and colleagues during the agricultural financial crisis of the 1980s, the Family Stress Model posits that the psychological impact of objective economic hardship (e.g., income loss, heavy debt-to-asset ratios) on child adjustment is not direct, but is instead entirely mediated by a cascade of intra- and interpersonal disruptions. Macroeconomic strain triggers emotional distress and depressive symptoms in parents, which destabilizes the marital or co-parenting relationship and systematically degrades parenting behavior. In this framework, compromised parenting manifests precisely as decreased positive affect (warmth, support, and responsiveness) and escalated negative affect (hostility, irritability, erratic discipline, and coercion). The BARS serves as the explicit measurement engine designed to detect these disruptions in the parenting subsystem.

2. Coercion Theory and Behavioral Interactionism

Rooted in the seminal work of Gerald R. Patterson at the Oregon Social Learning Center, coercion theory conceptualizes family conflict as a transactional feedback loop reinforced by negative reinforcement dynamics. A child exhibits aversive behavior, the parent responds with escalating hostility (yelling, threatening), and the cycle culminates when one party capitulates, thereby conditioning coercive interpersonal habits. The BARS operationalizes these exact functional behavioral units, capturing the frequency of hostile exchanges rather than abstract emotional states.

3. Attachment Theory and the Secure Base Phenomenon

According to John Bowlby and Mary Ainsworth, optimal socioemotional adaptation requires an environment characterized by sensitive parental responsiveness, creating a secure base from which the child explores the world. Parental warmth, as operationalized within the BARS, directly tracks parental sensitivity and availability, whereas the hostility scale quantifies the environmental unpredictability, threat, and rejection that generate insecure (avoidant, anxious, or disorganized) internal working models of relationships.

4. Social Learning Theory and Observational Modeling

Albert Bandura’s social cognitive theory emphasizes that children acquire interpersonal scripts, conflict-resolution mechanisms, and emotion regulation competencies by observing primary caregivers. Exposure to elevated parental hostility modeled on the BARS provides direct observational templates for aggressive, hostile behavior in the child’s subsequent peer and romantic interactions. Conversely, sustained parental warmth models prosocial reciprocity and emotional self-regulation.

5. Belsky’s Process Model of Parenting

Jay Belsky’s ecological model of parenting determinants posits that parenting practices are shaped by three primary forces: parental psychological resources, child contextual characteristics, and contextual sources of stress and support. The BARS functions as an ecologically sensitive index sensitive to fluctuations across these three domains, serving as a dynamic outcome variable when assessing marital quality, parental psychopathology, or community support systems.

Validity

The psychometric validity of the BARS has been examined extensively over three decades across prospective longitudinal designs, epidemiological surveys, and cross-cultural validation initiatives:

Construct and Structural Validity

Construct validity has been established through exhaustive exploratory and confirmatory factor analyses. CFA across diverse datasets has repeatedly proven that a two-factor model (Warmth vs. Hostility) fits the observed data significantly better than a unidimensional affective model ($Deltachi^2, p < .001$). The correlation between the two latent factors is typically moderate and negative (ranging between$r = -.30$ and $r = -.52$), affirming that warmth and hostility represent distinct functional dimensions rather than opposite ends of a single continuum. Parents can display low warmth without being actively hostile, or oscillate between high hostility and intermittent warmth in dysregulated environments.

Convergent and Multi-Method Validity

A gold standard in psychometrics is demonstrating convergence between subjective questionnaires and objective observational coding. In the Iowa Youth and Family Project, researchers cross-validated BARS self- and child-reports against the Iowa Family Interaction Rating Scales (IFIRS), an intensive, macro-level coding system of videotaped family interactions scored by highly trained, independent observational raters. Latent constructs of parental warmth and hostility derived from BARS questionnaires correlated strongly ($r = .45$ to $.68, p < .001$) with observed behavioral ratings of parental supportive behavior and parental hostility, demonstrating exceptional convergent validity across completely distinct measurement modalities.

Predictive and Criterion-Related Validity

The predictive utility of the BARS has been substantiated in hundreds of developmental studies. Longitudinal structural equation models demonstrate that:

• Internalizing Symptoms: High scores on BARS Parental Hostility and low scores on Parental Warmth during early adolescence prospectively predict clinical elevations in youth depressive symptoms, generalized anxiety, and suicidal ideation across 2- to 5-year intervals, even after controlling for baseline distress.
• Externalizing and Delinquency Trajectories: In line with coercion theory, BARS hostility scores reliably forecast adolescent conduct problems, contact with law enforcement, substance experimentation, and association with deviant peer groups.
• Adult Relational Competence: Long-term follow-ups from the Family Transitions Project (FTP) revealed that BARS scores during adolescence predicted the quality of the youth’s romantic relationships at ages 25 to 30, including romantic conflict resolution strategies and observational hostility toward romantic partners.

Discriminant and Cross-Cultural Validity

Research conducted by Zoe E. Taylor, Conger, and Widaman (2012) examined the BARS within low-income Mexican-origin families, testing for construct equivalence across cultural and language barriers. Using multi-group confirmatory factor analysis (MGCFA), the researchers demonstrated full metric and partial scalar invariance across English- and Spanish-speaking cohorts. Furthermore, the scale demonstrated discriminant validity from measures of general negative affectivity, neuroticism, socioeconomic status, and broad parental stress indices.

Reliability

The BARS demonstrates high internal consistency, test-retest stability, and cross-informant reliability across varied sampling frames, socioeconomic strata, and age groups.

Internal Consistency (Cronbach’s Alpha)

Extensive reporting from the Iowa Youth and Family Project, the Family Transitions Project, and the Latino Family and Youth Project yields consistently high Cronbach’s alpha ($lpha$) coefficients:

• Mother report of father hostility: $lpha = .88$
• Mother report of father warmth: $lpha = .94$
• Father report of mother hostility: $lpha = .81$
• Father report of mother warmth: $lpha = .91$
• Child report of mother warmth: $lpha = .82$ to $.90$
• Child report of father warmth: $lpha = .87$ to $.92$
• Child report of maternal hostility: $lpha = .84$ to $.89$
• Child report of paternal hostility: $lpha = .85$ to $.89$

These values demonstrate excellent internal consistency, exceeding standard psychometric criteria for both research and clinical evaluation ($lpha ge .80$). Composite reliability coefficients ($\omega$) derived from structural equation modeling consistently match or slightly exceed these alpha parameters.

Test-Retest Stability and Cross-Informant Agreement

In multi-wave longitudinal assessments, the BARS displays substantial rank-order stability across one-year intervals (typically $r = .55$ to $.70$ for warmth, and $r = .50$ to $.65$ for hostility), capturing both enduring relationship traits and dynamic behavioral shifts in response to acute familial stressors. Cross-informant concordance (e.g., correlation between adolescent report of maternal warmth and the mother’s own self-reported warmth) ranges between $r = .35$ and $r = .55$, an expected and psychometrically healthy convergence level in multi-informant family systems literature that affirms informant-specific perspective without reflecting measurement error.

Factor Analysis

Empirical verification of the internal architecture of the 22-item BARS has been executed using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Structure

Initial principal axis factoring with promax (oblique) and varimax (orthogonal) rotations consistently extracts two primary components with eigenvalues substantially exceeding the Kaiser-Guttman criterion of 1.0 (typically $\lambda_1 > 6.5$ representing Hostility, and $\lambda_2 > 4.2$ representing Warmth), explaining over 52% to 64% of the total variance depending on informant type. Factor loadings for designated subscale items systematically range between $.50$ and $.86$, with minimal cross-loadings ($< .20$).

Confirmatory Factor Analysis and Model Fit

In advanced structural evaluations (e.g., Conger et al., 1994; Taylor et al., 2012), a two-factor CFA specification consistently yields superior model fit across various demographic groups. Typical structural fit indices satisfy rigorous psychometric criteria:

• $\chi^2 / ext{df}$ ratio: $< 2.5$
• Comparative Fit Index (CFI): $ge .93 – .97$
• Tucker-Lewis Index (TLI): $ge .92 – .96$
• Root Mean Square Error of Approximation (RMSEA): $le .045 – .058$ ($90%\text{ CI } [.038, .064]$)
• Standardized Root Mean Square Residual (SRMR): $le .042 – .051$

Factor Loadings and Measurement Invariance

Standardized item factor loadings for the primary dimensions are consistently robust:

• Parental Warmth Loadings: Items such as “Act loving and affectionate” ($lambda pprox .84$), “Let you know she/he really cares about you” ($lambda pprox .82$), and “Tell you she/he loves you” ($lambda pprox .78$) demonstrate the highest saturation on the latent positive affect factor.
• Parental Hostility Loadings: Items such as “Shout or yell at you because she/he was mad” ($lambda pprox .79$), “Criticize you or your ideas” ($lambda pprox .71$), “Get angry at you” ($lambda pprox .68$), and overt aggression items like “Hit, push, grab, or shove you” ($lambda pprox .60 – .74$) anchor the latent negative affect factor.

Multi-group CFA evaluations testing for invariance across youth gender, parental gender, and racial/ethnic groups have confirmed configural invariance (identical factor structures across groups), metric invariance (equal factor loadings, permitting comparison of regression slopes and structural paths), and scalar invariance (equal item intercepts, allowing valid comparisons of latent means across groups).

Instrument / Measurement Tool

• Test Type: Multi-informant behavioral rating scale / Dyadic affect inventory
• Target Population: Adolescents (ages 10–19), emerging adults, and adult parents/partners reporting on dyadic family interactions
• Informant Versions: Child-report on mother; Child-report on father; Mother self-report / Mother report on father; Father self-report / Father report on mother
• Administration Time: Approximately 5 to 8 minutes for the 22-item inventory; 3 minutes for the 12-item inventory
• Item Formats: 22 items (standard complete form); 12 items (validated brief screening form)
• Authentic Response Scale: 7-point Likert scale: 1 = Never, 2 = Almost never, 3 = A few times, 4 = About half the time, 5 = More than half the time, 6 = Almost always, 7 = Always (also commonly administered with a 4-point or 5-point frequency scale: 1 = Never to 4 = Always / 1 = Never to 5 = Always)
• Primary Dimensions & Subscales:
  • Parental Warmth / Support / Involvement: Evaluates positive affect, instrumental assistance, verbal and physical affection, and active listening.
  • Parental Hostility / Rejection / Conflict: Evaluates verbal aggression, criticism, physical intimidation, emotional dismissiveness, and coercive control.
• Scoring Procedures:
  • Items are grouped according to subscale and averaged or summed.
  • No reverse scoring is applied to merge the two subscales into a single composite; Warmth and Hostility are maintained as independent continuous dimensions.
  • In structural equation modeling (SEM), items are frequently grouped into three or four empirical parcels per latent construct to minimize measurement error and preserve degrees of freedom.

Permissions & Fee and Test Year

The Behavioral Affect Rating Scale was developed in 1989 by Dr. Rand D. Conger at Iowa State University for the Iowa Youth and Family Project (IYFP). The instrument was established under research grants supported by the National Institute of Mental Health (NIMH), the National Institute on Drug Abuse (NIDA), and the National Institute of Child Health and Human Development (NICHD).

As an instrument developed through federally funded academic research, the BARS resides in the public domain for non-commercial academic, research, and clinical assessment purposes. It is published openly across numerous academic texts and public health compendia, including the Pan American Health Organization (PAHO) and the World Health Organization (WHO) clinical diagnostic compilations (e.g., Latino Families and Youth: A Compendium of Assessment Tools). No royalty fees or proprietary user licenses are required for scholarly or pedagogical use. Clinicians and researchers wishing to employ the scale in institutional programs are advised to cite the primary source literature (Conger, 1989; Conger et al., 1994; Taylor et al., 2012) accurately in resulting publications.

References

• Conger, R. D. (1989). Behavioral Affect Rating Scale (BARS). Developed from diverse sources for the Iowa Youth & Family Project. Ames, IA: Iowa State University.
• Conger, R. D., Conger, K. J., Elder, G. H., Jr., Lorenz, F. O., Simons, R. L., & Whitbeck, L. B. (1992). A family process model of economic hardship and adjustment of early adolescent boys. Child Development, 63(3), 526–541. https://doi.org/10.2307/1131344
• Conger, R. D., Ge, X., Elder, G. H., Jr., Lorenz, F. O., & Simons, R. L. (1994). Economic stress, coercive family process, and early adolescent depression. The Journal of Early Adolescence, 14(2), 184–200. https://doi.org/10.1177/027243169401400204
• Conger, R. D., & Conger, K. J. (2002). Resilience in Midwestern families: Selected findings from the first decade of a prospective, longitudinal study. Journal of Marriage and Family, 64(2), 361–373. https://doi.org/10.1111/j.1741-3737.2002.00361.x
• Elder, G. H., Jr., Nguyen, T. V., & Caspi, A. (1985). Linking family hardship to children’s lives. Child Development, 56(6), 1561–1575. https://doi.org/10.2307/1130474
• Lorenz, F. O., Conger, R. D., Simons, R. L., Whitbeck, L. B., & Elder, G. H., Jr. (1991). Economic pressure and marital quality: An illustration of the method variance problem in the causal modeling of family processes. Journal of Marriage and the Family, 53(2), 375–388. https://doi.org/10.2307/352906
• Melby, J. N., & Conger, R. D. (2001). The Iowa Family Interaction Rating Scales: Instrument summary. In P. K. Kerig & K. M. Lindahl (Eds.), Family Observational Coding Systems: Resources for Systemic Research (pp. 33–58). Lawrence Erlbaum Associates.
• Orpinas, P. (2009). Latino Families and Youth: A Compendium of Assessment Tools. Pan American Health Organization (PAHO / WHO). https://iris.paho.org/handle/10665.2/28415
• Patterson, G. R. (1982). Coercive Family Process. Eugene, OR: Castalia Publishing Company.
• Simons, R. L., Whitbeck, L. B., Conger, R. D., & Wu, C. I. (1991). Intergenerational transmission of harsh parenting. Developmental Psychology, 27(1), 159–171. https://doi.org/10.1037/0012-1649.27.1.159
• Taylor, Z. E., Larsen-Rife, D., Conger, R. D., & Widaman, K. F. (2012). Familism, interparental conflict, and parenting in Mexican-origin families: A cultural–contextual framework. Journal of Marriage and Family, 74(2), 312–327. https://doi.org/10.1111/j.1741-3737.2012.00961.x

Items of the Scale