Parent Affect Test (PAT)

Abstract

The Parent Affect Test (PAT) is an influential psychometric instrument developed by Marsha M. Linehan, Elise Paul, and Kelly J. Egan in 1983 to assess parental emotional responses and behavioral action tendencies toward everyday child behaviors. Designed to address the clinical and empirical limitations of retrospective parental reports and uncontextualized affective inventories, the PAT utilizes an analogue vignette format comprising 40 standardized child behavior descriptions (20 positive or prosocial behaviors and 20 negative, noncompliant, or disruptive behaviors). For each situation, parents evaluate their immediate internal emotional states and overt behavioral action tendencies across six 7-point semantic differential bipolar rating scales: feel angry versus feel pleased, feel bad versus feel good, feel tense versus feel relaxed, want to hit/spank versus want to hug/kiss, want to yell versus want to praise, and want to send child to room versus want to be with child. Psychometric analyses demonstrate that the PAT reliably yields distinct indices of positive and negative parental affect across affective-emotional and behavioral-disciplinary dimensions. The scale exhibits high internal consistency, with Cronbach’s alpha coefficients typically ranging from .89 to .96 for overall positive and negative affect composite scores, robust test-retest reliability across multiple assessment intervals, and strong convergent, construct, and discriminant validity. The PAT discriminates effectively between non-referred parents and those referred for physical abuse, harsh disciplinary practices, or significant parent-child relational conflict. It serves as a vital clinical assessment instrument, progress monitor for behavioral parent training programs, and research tool in developmental and clinical psychology.

Keywords

Parent Affect Test, parental affect, child behavior, parenting practices, parent-child interaction, affective responses, behavioral action tendencies, parent training, child abuse assessment, psychometrics

Authors

The Parent Affect Test was conceptualized, developed, and empirically validated by a team of clinical psychologists specializing in behavioral assessment, parent training, and emotional regulation:

• Marsha M. Linehan, Ph.D., ABPP: Professor Emeritus of Psychology and Director of the Behavioral Research and Therapy Clinics (BRTC) at the University of Washington, Seattle, Washington. World-renowned as the developer of Dialectical Behavior Therapy (DBT), Dr. Linehan’s early scholarship focused extensively on behavioral assessment, social competence, and the emotional mechanisms underlying familial conflict and severe behavioral disorders.
• Elise Paul, Ph.D.: Clinical and research psychologist who collaborated on parent-child interaction paradigms, behavioral observation systems, and clinical family evaluation instruments at the University of Washington.
• Kelly J. Egan, Ph.D.: Clinical psychologist and researcher known for work in behavioral medicine, pain management, and parent-child transactional dynamics, affiliated during the development of the PAT with the University of Washington School of Medicine.

Purpose

The central purpose of the Parent Affect Test is to quantify and delineate a caregiver’s conditioned emotional and behavioral action tendencies elicited by specific, prototypical child behaviors. In both developmental psychopathology and clinical child psychology, parental affect serves as a critical mediating variable between child conduct and parental disciplinary behavior. While traditional parenting assessments historically emphasized self-reported parenting styles (e.g., authoritarian, authoritative, permissive) or global perceptions of child deviance, they frequently failed to capture the immediate, contextualized affective reactions that directly precede parental reinforcement, punishment, hostility, or emotional withdrawal.

Clinical applications of the PAT encompass:

• Child Maltreatment and Risk Assessment: Identifying parents at high risk for physical abuse or coercive disciplinary escalation by measuring excessive negative affect and elevated disciplinary action tendencies (e.g., strong impulses to yell, hit, or spank) in response to minor or normative child misbehaviors.
• Pre- and Post-Treatment Evaluation in Parent Training: Assessing shifts in parental emotional reactivity following evidence-based parent management interventions, such as Parent Management Training (PMT), Parent-Child Interaction Therapy (PCIT), or the Triple P – Positive Parenting Program.
• Clinical Diagnosis of Parent-Child Relational Problems: Disentangling whether a parent experiences blunted positive affect in response to prosocial child behaviors, hyper-reactivity to normative child indiscretions, or a pervasive negative attributional-affective bias.

From an empirical research perspective, the PAT provides investigators with a standardized laboratory analogue instrument capable of simulating parental affective reactivity without introducing the logistical complexities, demand characteristics, or safety concerns associated with in vivo stress induction. It allows researchers to investigate reciprocal parent-child coercive cycles, parental depression-related emotional blunting, and the psychophysiological correlates of parental anger.

Psychological Construct

The PAT measures parental affective reactivity, operationalized across two major behavioral domains (positive child behaviors and negative child behaviors) and two response modalities (subjective emotional experience and overt behavioral action tendencies):

Negative Affect and Disciplinary Urges (Negative Child Behaviors)

This dimension evaluates the caregiver’s subjective emotional distress and punitive behavioral impulses when confronted with child noncompliance, defiance, messiness, carelessness, or emotional volatility. Items depicting negative situations (e.g., “My child talks back to me”, “I ask my child to do something and he/she gets really angry”, “My child carelessly spills something”) elicit parental ratings across:

• Subjective Emotional Dysphoria/Arousal: Measured by feelings of anger, badness, and tension. Elevated scores on these dimensions reflect heightened physiological arousal, low distress tolerance, and cognitive appraisals that view the child’s misbehavior as intentional, hostile, or threatening.
• Coercive Behavioral Impulses: Measured by the parental urge to yell, hit/spank, and send the child away. These action tendencies capture the mobilization of harsh, authoritarian control mechanisms designed to immediately suppress child behavior through intimidation or isolation.

Positive Affect and Prosocial Approach Urges (Positive Child Behaviors)

This dimension measures the caregiver’s capacity for positive emotional resonance, reward responsiveness, and affectionate behavior when observing child competence, affection, and compliance. Items capturing positive situations (e.g., “My child comes to me and gives me a kiss”, “My child makes a good decision”, “My child helps out with the household chores without being asked”) evaluate:

• Subjective Positive Valence: Indicated by feelings of pleasure, goodness, and relaxation. Parents who score low in this domain exhibit anhedonia or emotional detachment, failing to derive intrinsic satisfaction from their child’s prosocial development.
• Nurturant Behavioral Impulses: Indicated by the desire to praise, hug/kiss, and spend time with the child. This reflects parental approach motivation, social reinforcement readiness, and emotional availability, which are vital components of secure caregiver-child attachment.

Theoretical Framework

The Parent Affect Test is grounded in social learning theory, behavioral family systems, and transactional models of developmental psychopathology. In particular, it operationalizes key tenets from Gerald Patterson’s Coercive Family Process Model and cognitive-behavioral formulations of child abuse and family conflict.

Patterson’s coercion theory posits that severe parent-child conflict evolves through repetitive, reciprocally reinforced behavioral interactions. When a child engages in aversive behaviors (e.g., whining, noncompliance, yelling), parents often experience acute negative emotional arousal. If the parent lacks effective emotion regulation skills or nonviolent disciplinary strategies, this negative affect translates into coercive behavioral action tendencies, such as yelling, threatening, or corporal punishment. If the child temporarily ceases the aversive behavior, the parent’s aggressive reaction is negatively reinforced through relief from distress. Conversely, if the parent capitulates, the child’s noncompliance is positively reinforced. The PAT specifically captures the early affective and impulse-generation phases of this coercive cascade.

Furthermore, the PAT integrates classical affective and appraisal theories of emotion, such as those articulated by Richard Lazarus and Nico Frijda. According to these frameworks, emotions are not merely internal subjective feelings; they represent action tendencies that prepare the organism for adaptive behavioral responses (e.g., fight, flight, approach). The inclusion of both affective feeling states (angry-pleased, bad-good, tense-relaxed) and overt behavioral inclinations (hit-hug, yell-praise, isolate-approach) reflects this theoretical dual-structure of affective states.

Validity

The psychometric validity of the Parent Affect Test has been empirically established through multiple validation studies across both clinical and community populations:

• Construct Validity: Initial validation by Linehan, Paul, and Egan (1983) demonstrated that the PAT reliably differentiates emotional reactions to positive versus negative stimuli. Parents systematically report high positive affect and approach impulses to positive child scenarios and varying degrees of negative affect and disciplinary impulses to negative scenarios. Factor analytic findings further confirm that emotional ratings and action tendencies load consistently onto congruent overarching affective dimensions.
• Criterion and Discriminant Validity: The PAT exhibits remarkable sensitivity in distinguishing between non-abusive control parents and parents confirmed or identified as at-risk for child physical abuse. Linehan et al. (1983) and subsequent investigators found that physically abusive and highly stressed parents reported significantly higher levels of anger, tension, and urges to hit, spank, or yell in response to normative negative child behaviors compared to matched community controls. Furthermore, abusive parents frequently exhibited attenuated positive affect and reduced urges to praise or embrace their children in response to positive behaviors.
• Convergent Validity: PAT scores correlate significantly with related standardized parenting and family functioning instruments. Strong positive correlations are observed between PAT negative affect scores and the Child Abuse Potential (CAP) Inventory, the Parenting Stress Index (PSI), and maternal reports of child externalizing behaviors on the Child Behavior Checklist (CBCL). Significant inverse correlations exist between PAT positive affect scores and measures of parental depression, parental burnout, and marital distress.
• Treatment Sensitivity: The PAT has demonstrated robust sensitivity to therapeutic change. Following behavioral parent training programs, mothers and fathers exhibit statistically significant decreases in negative affect and coercive disciplinary urges, alongside significant increases in positive affect and praise tendencies when presented with identical vignette stimuli.

Reliability

The reliability of the Parent Affect Test has been confirmed across several indices of psychometric precision:

• Internal Consistency: The PAT demonstrates exceptional internal consistency across its primary summary dimensions. In the original validation cohort of Linehan, Paul, and Egan (1983), Cronbach’s alpha values exceeded .90 for all overarching scales. Specifically, alpha coefficients for the Positive Affect Composite (ratings across positive situations) typically range from .92 to .96, while alpha coefficients for the Negative Affect Composite (ratings across negative situations) range from .89 to .95. Individual subscale alphas (e.g., the anger-pleased scale across all 20 negative vignettes) consistently exceed .85, indicating strong item homogeneity.
• Test-Retest Reliability: Temporal stability assessments conducted over intervals ranging from 2 weeks to 2 months have yielded test-retest correlation coefficients ($r$) between .74 and .88 for the negative behavior scales and .71 to .85 for the positive behavior scales in stable, untreated community cohorts. This demonstrates that while the PAT captures situational reactions, it measures stable trait-like parental affective dispositions.
• Inter-Item Correlations: Item-total correlations within the designated subscales consistently exceed .40, with minimal redundant overlap, indicating that each distinct vignette provides unique clinical variance to the composite construct.

Factor Analysis

Exploratory factor analyses (EFA) and subsequent confirmatory factor analyses (CFA) conducted on the Parent Affect Test confirm a robust two-tier structural architecture:

First-Order Factor Structure

When factor analyses are conducted separately across the response dimensions within situation types, the six semantic differential scales reliably coalesce into two correlated but distinct first-order factors:

1. Emotional/Internal Affective State: Defined by strong, salient factor loadings (typically ranging from .65 to .88) on Scale 1 (feel angry – feel pleased), Scale 2 (feel bad – feel good), and Scale 3 (feel tense – feel relaxed).
2. Behavioral/Action Tendency: Defined by robust factor loadings (ranging from .60 to .85) on Scale 4 (want to hit/spank – want to hug/kiss), Scale 5 (want to yell – want to praise), and Scale 6 (want to send child to room – want to be with child).

Higher-Order Factor Structure

At the macro level, higher-order factor analysis reveals that variance is primarily governed by the valence of the stimulus vignettes:

• Factor I: Negative Parental Affect / Disciplinary Arousal: Accounts for the largest proportion of total variance (often between 35% and 48%). All responses across the 20 negative child behavior vignettes load heavily onto this dimension. CFA models evaluating this construct demonstrate acceptable goodness-of-fit indices (e.g., Comparative Fit Index [CFI] > .91, Root Mean Square Error of Approximation [RMSEA] < .06).
• Factor II: Positive Parental Affect / Prosocial Nurturance: Accounts for an additional 18% to 26% of variance, capturing parental emotional resonance and approach tendencies across the 20 positive child behavior vignettes.

Importantly, empirical studies have consistently noted that the correlation between Factor I and Factor II is modest to low ($r = -.15$ to $-.32$), indicating that parental positive affect and parental negative affect are largely orthogonal dimensions rather than mutually exclusive endpoints of a single continuum. A parent may possess moderate positive warmth alongside dangerously elevated punitive reactivity, or conversely, exhibit an absence of both hostility and warmth.

Instrument / Measurement Tool

• Instrument Name: Parent Affect Test (PAT)
• Acronym: PAT
• Authors: Marsha M. Linehan, Elise Paul, and Kelly J. Egan
• Publication Year: 1983
• Assessment Type: Standardized analogue self-report questionnaire / vignette rating inventory
• Format: Written questionnaire; can be administered via paper-and-pencil or computerized clinical assessment software
• Total Number of Vignette Items: 40 child behavior situations
• Distribution of Situations:
  • Negative Situations (20 items): Items 1, 2, 3, 4, 7, 8, 10, 13, 14, 17, 20, 22, 25, 26, 29, 32, 36, 37, 39, and 40.
  • Positive Situations (20 items): Items 5, 6, 9, 11, 12, 15, 16, 18, 19, 21, 23, 24, 27, 28, 30, 31, 33, 34, 35, and 38.
• Response Scale: For each of the 40 situations, the parent completes six 7-point bipolar semantic differential scales:
  1. feel angry (1) — feel pleased (7)
  2. feel bad (1) — feel good (7)
  3. feel tense (1) — feel relaxed (7)
  4. want to hit/spank (1) — want to hug/kiss (7)
  5. want to yell (1) — want to praise (7)
  6. want to send child to room (1) — want to be with child (7)
• Total Number of Responses: 240 individual data points (40 situations × 6 rating scales).
• Scoring Procedures:
  • Each bipolar rating is scored from 1 (extreme negative affect/punitive impulse) to 7 (extreme positive affect/nurturant impulse), or conversely reverse-scored depending on clinical preference.
  • Negative Situation Subscales: Scores are computed across the 20 negative situations. Clinicians commonly compute an Internal Negative Affect score (average or sum of scales 1, 2, and 3) and a Negative Action Tendency score (scales 4, 5, and 6), or a global Negative Affect Composite. Lower raw scores (toward 1) reflect high anger, tension, dysphoria, and punitive impulses; in reverse-scored protocols, higher scores indicate greater parental hostility and distress.
  • Positive Situation Subscales: Scores are computed across the 20 positive situations to derive an Internal Positive Affect score (scales 1, 2, and 3) and a Positive Action Tendency score (scales 4, 5, and 6), or a global Positive Affect Composite. Higher scores (toward 7) indicate warmth, pleasure, relaxation, and affectionate behavioral impulses.
• Completion Time: Approximately 20 to 30 minutes.

Permissions & Fee and Test Year

The Parent Affect Test was first published in 1983 in the Journal of Clinical Child Psychology by Dr. Marsha M. Linehan and colleagues. It is reproduced in standard clinical measurement compendiums, most notably Fischer and Corcoran’s seminal sourcebook, Measures for Clinical Practice and Research: A Sourcebook (Oxford University Press).

The instrument is generally categorized as an open-access academic measurement tool for non-commercial clinical, educational, and research purposes. Researchers and practicing clinicians wishing to use the PAT in formal investigative protocols or diagnostic batteries may typically do so without licensing fees, provided that appropriate scholarly attribution is cited according to APA guidelines. Commercial republishing, inclusion in proprietary commercial diagnostic software, or mass distribution requires formal permission from the copyright holders or original journal publisher (Taylor & Francis / American Psychological Association Division 53).

References

• Fischer, J., & Corcoran, K. J. (2007). Measures for clinical practice and research: A sourcebook (4th ed., Vol. 1, pp. 371–377). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195304381.001.0001
• Linehan, M. M., Paul, E., & Egan, K. J. (1983). The Parent Affect Test: Development, validity, and reliability. Journal of Clinical Child Psychology, 12(2), 161–166. https://doi.org/10.1080/15374418309533127
• Milner, J. S. (1994). Assessing physical child abuse risk: The Child Abuse Potential Inventory. Clinical Psychology Review, 14(6), 547–583. https://doi.org/10.1016/0272-7358(94)90017-5
• Patterson, G. R. (1982). Coercive family process. Castalia Publishing Company.
• Whisman, M. A., & McClelland, G. H. (2005). Designing, testing, and interpreting interactions and moderator effects in family research. Journal of Family Psychology, 19(1), 111–120. https://doi.org/10.1037/0893-3200.19.1.111

13. Items of the Scale (Questionnaire)