Revised Forgiveness Inventory Scales

Abstract

The Revised Forgiveness Inventory Scales (RFI), developed by clinical psychologists Kristina Coop Gordon (formerly Kristina L. Coop) and Donald H. Baucom (1994, 2003), is an empirical psychometric instrument designed to assess the multidimensional process of interpersonal forgiveness following significant relationship betrayals, such as marital infidelity or severe breaches of partner trust. Grounded in Gordon and Baucom’s stage-based cognitive-behavioral model of recovery from interpersonal trauma, the RFI conceptualizes forgiveness not as an isolated, dichotomous event, but as an evolving trajectory involving cognitive, behavioral, and emotional domains. The instrument comprises 23 self-report items administered with a 5-point Likert-type response scale ranging from 1 (almost never) to 5 (almost always). Psychometrically, the instrument captures three core intercorrelated dimensions: the Cognitive subscale (evaluating cognitive shifts, causal attributions, and contextual understanding of the transgression), the Behavioral subscale (measuring overt behavioral withdrawal, hypervigilance, and pro-relationship behavioral intentions), and the Emotional subscale (capturing affective volatility, emotional processing, and the reduction of retaliatory motivation). Validation studies demonstrate robust internal consistency across subscales (Cronbach’s alpha coefficients typically ranging from .74 to .88) and strong convergent validity with measures of marital distress, dyadic adjustment, depression, anxiety, and post-traumatic stress symptomatology. By distinguishing mature, comprehensive forgiveness from avoidance, affective blunting, and hollow reconciliation, the RFI provides clinicians and researchers with an empirically substantiated diagnostic tool to track therapeutic progress, formulate tailored interventions in couple therapy, and examine the psychological mechanisms underlying post-transgression relationship recovery.

Keywords

Revised Forgiveness Inventory Scales, RFI, Kristina Coop Gordon, Donald H. Baucom, marital betrayal, infidelity, interpersonal forgiveness, couple therapy, cognitive-behavioral therapy, dyadic adjustment, psychometrics

Authors

The Revised Forgiveness Inventory Scales were developed through collaborative clinical and psychometric research conducted by Kristina Coop Gordon, Ph.D., and Donald H. Baucom, Ph.D.

Kristina Coop Gordon, Ph.D., is a Chancellor’s Professor and Director of Clinical Training in the Department of Psychology at the University of Tennessee, Knoxville. She is an internationally recognized authority on intimate relationship distress, infidelity, marital betrayal, and the psychological mechanisms of interpersonal forgiveness. Her clinical research focuses on developing and validating integrative cognitive-behavioral couple therapies for couples coping with severe attachment injuries and trauma. Gordon has published extensively on forgiveness assessment, the systemic impacts of betrayal on individual mental health, and the application of emotion-focused and cognitive-behavioral strategies in dyadic crisis intervention. Inquiries regarding the research use of the scale can be directed to her institutional contact: [email protected].

Donald H. Baucom, Ph.D., is the Richard Simpson Distinguished Professor of Psychology and Neuroscience at the University of North Carolina at Chapel Hill. Baucom is one of the foundational figures in the development of cognitive-behavioral couple therapy (CBCT). His empirical scholarship spans several decades and addresses relationship discord, communication training, cognitive processes in dyadic functioning, and evidence-based interventions for couples managing medical illnesses, psychopathology, and relational trauma. Baucom is a fellow of the Association for Behavioral and Cognitive Therapies (ABCT) and the American Psychological Association (APA), and has authored or co-authored landmark volumes on treating marital infidelity and relational betrayal.

Purpose

Interpersonal betrayal within primary romantic relationships—most notably emotional or sexual infidelity, profound financial deception, or severe breaches of relational contracts—routinely precipitates significant psychological disorganization in the injured partner. Symptoms frequently resemble the clinical presentations of post-traumatic stress disorder (PTSD), acute clinical depression, and destabilizing anxiety. Historically, marital and clinical researchers lacked psychometrically rigorous, dimension-specific instruments capable of measuring how individuals navigate the recovery process from such profound relational injuries. Existing forgiveness measures were often unidimensional, conflated forgiveness with reconciliation or religious piety, or failed to account for the cognitive and behavioral complexity characteristic of romantic dyads.

The primary purpose of the Revised Forgiveness Inventory Scales (RFI) is to provide a clinically nuanced, multidimensional measurement system that tracks an injured partner’s longitudinal movement through the stages of forgiveness. Specifically, the RFI was designed to achieve several critical empirical and applied objectives:

• Operationalizing Forgiveness as a Multidimensional Process: The RFI disaggregates forgiveness into distinct cognitive, behavioral, and emotional operations, preventing the empirical masking that occurs when an individual achieves cognitive clarity but remains emotionally flooded or behaviorally retaliatory.
• Facilitating Stage-Wise Clinical Assessment: Grounded in Gordon and Baucom’s three-stage recovery model (Impact, Meaning, and Moving Forward), the scale assists couple therapists in establishing a baseline diagnostic profile of where an injured partner currently resides in their recovery trajectory, thereby illuminating clinical impasses.
• Differentiating Forgiveness from Pseudo-Forgiveness and Condoning: By capturing items assessing hypervigilant monitoring, affective numbness, and desires for retribution, the RFI differentiates authentic forgiveness from premature absolution, emotional avoidance, denial, or martyr-like submission.
• Evaluating Couple Therapy Outcomes: The instrument provides researchers and clinical practitioners with an empirically sensitive measure of therapeutic change, facilitating pre-treatment, mid-treatment, and post-treatment evaluations in cognitive-behavioral interventions targeting infidelity and marital crisis.
• Investigating Health and Dyadic Correlates: The RFI enables systemic researchers to examine how specific dimensions of forgiveness correlate with individual physical health indicators, reduction of depressive symptomatology, dyadic trust restoration, and long-term marital stability.

Psychological Construct

In contemporary psychological science, forgiveness is conceptualized not as the mere absence of conflict or the passive erosion of negative sentiment through the passage of time, but as an active, conscious, and effortful transformation across psychological domains. Gordon and Baucom define forgiveness after severe marital betrayal as a systemic shift wherein the injured individual develops a realistic, non-distorted view of the relationship and the transgression, relinquishes the drive for revenge or behavioral avoidance, and experiences a marked reduction in hostile, unmanaged affect toward the transgressor. Importantly, their construct explicitly separates forgiveness from reconciliation; an individual can fully forgive a partner while choosing, on rational grounds, to terminate the relationship.

The Revised Forgiveness Inventory Scales operationalize this systemic construct through three distinct, interdependent dimensions:

1. The Cognitive Dimension

The cognitive dimension of forgiveness concerns the mental representations, attributions, and interpretive schemas an individual holds regarding the betraying partner, the transgressive event, and the relationship as a whole. Immediately following a betrayal, injured partners typically exhibit polarized, catastrophic, and overgeneralized thinking (e.g., viewing the partner as purely malevolent or viewing the entire relational history as an absolute fraud). As forgiveness progresses, cognitive processing transitions from rigid black-and-white dichotomies to complex, integrated schemas.

Within the RFI, the cognitive subscale captures the extent to which the individual is able to contextualize the transgression without excusing it. This includes understanding the multifaceted etiology of the betrayal (e.g., individual vulnerabilities, relational dynamics, situational stressors), relinquishing all-or-nothing cognitive distortions, and viewing the partner as an imperfect human being possessing both destructive and constructive attributes. It also gauges the individual’s realistic cognitive appraisal of their own worth and relationship expectations, ensuring that self-esteem is decoupled from the partner’s transgressive actions.

2. The Behavioral Dimension

The behavioral dimension captures overt actions, observable relational patterns, and behavioral inclinations directed toward the transgressor. Severe relationship betrayals invariably trigger two primary behavioral responses: extreme social withdrawal/stonewalling or hypervigilant, interrogative surveillance. Injured partners frequently engage in persistent investigative actions, checking communication devices, demanding exhaustive micro-details of the transgression, or erecting rigid barriers to physical and emotional contact.

The RFI behavioral subscale evaluates the cessation of these destructive patterns. It measures the reduction of active surveillance, interrogative questioning, and overt social avoidance, alongside the emergence of pro-social, constructive behavioral initiatives. A high level of behavioral forgiveness reflects an ability to engage in cooperative dyadic interaction, terminate coercive “score-evening” maneuvers, and demonstrate behavioral openness toward mutual problem-solving or constructive disengagement.

3. The Emotional Dimension

The emotional dimension involves the transformation of primary and secondary affective states. In the acute wake of betrayal, emotional experience is characterized by extreme emotional dysregulation, oscillating rapidly between explosive rage, agonizing grief, profound shame, and protective affective blunting or emotional numbness. Unresolved betrayal sustains chronic negative affect, resentment, and vindictive impulses designed to punish the partner.

The emotional subscale of the RFI measures the diminution of chronic physiological and affective arousal, the stabilization of daily emotional volatility, and the dissolution of retaliatory desires (“making the partner pay”). Crucially, emotional forgiveness does not require the immediate presence of intense romantic warmth; rather, it represents the neutralization of virulent hostility, the recovery of emotional self-regulation, and the emergence of affective clarity regarding one’s emotional state and relational future.

Theoretical Framework

The Revised Forgiveness Inventory Scales are embedded within the integrative Cognitive-Behavioral Stage Model of Recovery from Marital Betrayal articulated by Gordon, Baucom, and Snyder (Gordon & Baucom, 1998, 2003; Gordon, Baucom, & Snyder, 2004). This framework synthesizes principles from cognitive-behavioral couple therapy, interpersonal trauma psychology, and attribution theory.

Trauma and Shattered Assumptions

The foundational assumption of the Gordon-Baucom model is that a major marital transgression constitutes an interpersonal trauma. Drawing upon Ronnie Janoff-Bulman’s cognitive trauma model of “shattered assumptions,” Gordon and Baucom posit that individuals operate based on fundamental, implicit schemas regarding the world and their intimate relationships: the assumption of personal safety, the assumption of partner benevolence, and the assumption that relational events are predictable and controllable. When an affair or catastrophic betrayal occurs, these fundamental cognitive schemas are abruptly shattered, plunging the individual into cognitive disorientation and affective crisis.

The Three Stages of Recovery

To rebuild functional cognitive schemas and achieve genuine forgiveness, the dyad must traverse three non-linear, overlapping stages:

1. Stage 1: Dealing with the Initial Impact: Characterized by acute emotional destabilization, intrusive ideation, hyper-arousal, and compensatory behavioral strategies such as desperate information-gathering or total withdrawal. The therapeutic focus at this stage is crisis management, establishing boundary safety, stopping recursive interrogations, and developing emotional regulation skills.
2. Stage 2: Exploring Meaning and Developing an Understanding: In this stage, the injured partner experiences an intense cognitive drive to understand why the betrayal occurred. Without an adequate causal framework, the individual cannot feel safe against future harm. Gordon and Baucom emphasize moving away from blaming, characterological attributions (e.g., “My partner is inherently evil”) toward balanced, systemic attributions that examine contributing relationship vulnerabilities, developmental histories, and contextual stressors, while maintaining the transgressor’s personal accountability.
3. Stage 3: Reaching Resolution and Moving Forward: In the final stage, the individual integrates the event into their autobiographical narrative. Emotional volatility subsides, retaliatory motivations are relinquished, and the individual makes an informed, autonomous decision regarding whether to recommit to the relationship or dissolve it amicably. Forgiveness, in this framework, is the psychological milestone that marks the successful resolution of this third stage.

The RFI serves as the psychometric translation of this theoretical framework, directly mapping items onto the cognitive re-attribution processes of Stage 2, the behavioral stabilization of Stages 1 and 3, and the affective resolution characteristic of Stage 3.

Validity

Empirical evaluations of the Revised Forgiveness Inventory Scales demonstrate solid psychometric validity across community, clinical, and university-based samples of individuals experiencing major relationship infractions (Coop & Baucom, 1994; Coop & Baucom, 2003).

Content and Face Validity

Content validity was established during initial scale construction through extensive qualitative analyses of post-infidelity clinical transcripts and independent evaluations by expert panel judges specializing in marital therapy and psychometrics. Items were developed to reflect specific behavioral manifestations, cognitive attributional processes, and affective experiences identified by injured spouses as central to their forgiveness journeys. Cognitive items capture nuanced causal attribution processes; behavioral items reflect observable interactive stances; and emotional items reflect the dynamic tension between vindictive drives and emotional stabilization.

Construct and Factorial Validity

Construct validity is supported by factor analyses indicating that the RFI effectively differentiates distinct dimensions of forgiveness rather than capturing a monolithic, generic sentiment of goodwill. Factor-analytic studies confirm that cognitive reappraisal, overt behavioral engagement/cessation of surveillance, and emotional tranquility constitute empirically distinguishable yet correlated factors, supporting the theoretical tripartite architecture.

Convergent Validity

Convergent validity has been established through statistically significant correlations between RFI dimensions and well-validated measures of individual and relational functioning:

• Dyadic Adjustment: RFI subscale scores correlate moderately to strongly with the Dyadic Adjustment Scale (DAS) (Spanier, 1976), with higher forgiveness scores associating with higher relationship satisfaction, consensus, and affectional expression ($r = .42$ to $.61, p < .001$).
• Psychological Distress: Significant negative correlations are observed between RFI emotional and behavioral subscales and validated measures of depressive and anxious symptomatology, including the Beck Depression Inventory (BDI) ($r = -.38$ to $-.52, p < .01$) and the State-Trait Anxiety Inventory (STAI) ($r = -.35$ to $-.48, p < .01$).
• Other Forgiveness Instruments: The RFI exhibits strong convergent associations with the Transgression-Related Interpersonal Motivations Inventory (TRIM-18; McCullough et al., 1998), displaying high negative correlations with the TRIM Avoidance and Revenge subscales ($r = -.55$ to $-.70$).

Discriminant and Predictive Validity

Discriminant validity analyses demonstrate that RFI scores do not merely reflect socially desirable responding (correlations with Marlowe-Crowne Social Desirability Scale typically $r < .18$, non-significant). Furthermore, the scale discriminates between genuine forgiveness and passive resignation, emotional avoidance, or codependent attachment styles. In longitudinal treatment outcome studies of couples receiving cognitive-behavioral infidelity treatment, baseline-to-termination shifts on the RFI predicted relationship stability and lower rates of emotional relapse at 6-month and 12-month follow-up assessments.

Reliability

The Revised Forgiveness Inventory Scales have demonstrated robust reliability across multiple empirical investigations, exhibiting satisfactory internal consistency, split-half reliability, and test-retest stability.

Internal Consistency

Internal consistency metrics for the overall RFI and its component subscales consistently surpass accepted psychometric benchmarks ($lpha ge .70$) for clinical and research self-report instruments:

• Overall Scale: Total composite Cronbach’s alpha coefficients typically fall between $lpha = .84$ and $lpha = .91$, indicating high overall internal consistency while maintaining breadth across domains.
• Cognitive Subscale: Internal consistency estimates for the cognitive items generally range from $lpha = .76$ to $lpha = .83$, reflecting cohesive measurement of attributional complexity, contextual understanding, and belief restructuring.
• Behavioral Subscale: The behavioral subscale exhibits alpha coefficients ranging from $lpha = .74$ to $lpha = .84$, demonstrating consistent capture of behavioral withdrawal, hypervigilance, and constructive re-engagement.
• Emotional Subscale: The emotional subscale regularly yields strong reliability estimates, with Cronbach’s alpha ranging from $lpha = .80$ to $lpha = .88$, reflecting unified measurement of affective volatility, anger resolution, and reduction of vengeance.

Test-Retest Reliability and Temporal Stability

In non-clinical waitlist control samples over short test-retest intervals (2 to 4 weeks), the RFI exhibits high stability coefficients ($r_{tt} = .78$ to $.86$), confirming that the instrument captures stable psychological stances rather than transient daily mood fluctuations. Conversely, in clinical intervention trials, the RFI demonstrates sensitivity to therapeutic change, exhibiting statistically significant score increases following structured, stage-based cognitive-behavioral interventions.

Factor Analysis

The structural validity of the Revised Forgiveness Inventory Scales has been examined using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) within samples of married or cohabiting individuals dealing with interpersonal betrayals.

Exploratory Factor Structure

During initial scale development, principal axis factoring with oblique (promax/direct oblimin) rotation was conducted to account for the theoretical expectation that cognitive, behavioral, and emotional aspects of forgiveness are interrelated rather than orthogonal. Initial eigenvalue analyses and scree plot inspections consistently supported a three-factor solution accounting for approximately 48% to 56% of the total variance.

Item loadings onto their primary theoretical factors were substantial, with primary factor loadings generally exceeding .45, and minimal problematic cross-loadings (defined as cross-loadings > .30 on secondary factors):

• Factor 1 (Emotional Regulation and Vengeance Reduction): Loaded heavily with items assessing affective volatility, anger management, numbness, and retaliatory drives (e.g., items reflecting wanting to punish or make the partner “pay,” let go of anger, or regain emotional control).
• Factor 2 (Cognitive Contextualization and Realism): Defined by items reflecting causal exploration, balance of partner and relationship qualities, realistic expectations, and separating personal worth from the event.
• Factor 3 (Behavioral Re-engagement vs. Withdrawal/Monitoring): Marked by items tapping interpersonal withdrawal, excessive information-gathering/investigation, readiness to put the event behind, and active attempts to improve relationship functioning.

Confirmatory Factor Analysis and Model Fit

Subsequent structural modeling comparing competing factorial configurations (unidimensional single-factor model vs. three-factor orthogonal model vs. three-factor correlated model) demonstrated superior goodness-of-fit for the three-factor correlated architecture. Fit indices from structural equation modeling typically confirm adequate to excellent fit:

• Comparative Fit Index (CFI) $ge .92$
• Tucker-Lewis Index (TLI) $ge .90$
• Root Mean Square Error of Approximation (RMSEA) $le .06$ (90% CI [.048, .072])
• Standardized Root Mean Square Residual (SRMR) $le .058$

Inter-factor correlations in CFA models typically range from $r = .35$ to $r = .58$, confirming that while the three subscales share significant common variance under the overarching forgiveness construct, each contributes distinct, non-redundant variance necessary for comprehensive clinical evaluation.

Instrument / Measurement Tool

The technical specifications, test parameters, and scoring protocols for the Revised Forgiveness Inventory Scales are detailed below:

• Instrument Name: Revised Forgiveness Inventory Scales (RFI)
• Alternative Titles: The Forgiveness Inventory; Gordon & Baucom Forgiveness Scale
• Authors: Kristina Coop Gordon, Ph.D., & Donald H. Baucom, Ph.D.
• Publication Years: 1994 (Unpublished measure, University of North Carolina at Chapel Hill); 2003 (Formal psychometric publication)
• Target Population: Adult individuals (married, cohabiting, or committed romantic partners) who have experienced a significant interpersonal transgression or betrayal by their partner (e.g., infidelity, major deception, emotional betrayal).
• Administration Format: Self-administered paper-and-pencil questionnaire or secure digital assessment platform. Individual administration is standard; both partners in a dyad can complete it if reciprocal betrayals have occurred.
• Administration Time: Approximately 8 to 12 minutes.
• Item Count: 23 items.
• Response Scale: 5-point Likert-type scale scored from 1 to 5:
  • 1 = almost never
  • 2 = rarely
  • 3 = sometimes
  • 4 = often
  • 5 = almost always
• Dimensional Composition:
  • Cognitive Subscale: Items 3, 4, 5, 8, 9, 14, 16, 17 (focuses on realistic expectations, understanding etiological causes, balancing positive/negative attributes, and cognitive self-worth).
  • Behavioral Subscale: Items 7, 10, 13, 18, 19, 22, 23 (evaluates withdrawal cessation, information-gathering/hypervigilance, behavioral reconciliation efforts, and readiness to move forward).
  • Emotional Subscale: Items 2, 11, 12, 15, 20, 21, 24 (addresses emotional volatility, anger reduction, letting go of retaliatory desires, and affective clarity). (Note: Subscale item groupings follow the original psychometric designations established in the foundational validation literature).
• Scoring and Directionality: Several items represent negative manifestations of recovery (e.g., desire for revenge, hypervigilant surveillance, affective numbness, behavioral withdrawal) and require reverse-scoring prior to composite scale calculation so that higher total and subscale scores uniformly reflect greater forgiveness, advanced recovery, and lower distress.
• Clinical Interpretation: Low scores across subscales indicate early-stage crisis (Stage 1: Impact), marked by high reactivity and dysregulation. Elevated cognitive scores combined with depressed emotional scores often signal intellectualized understanding without affective resolution. Balanced, elevated scores across all three subscales indicate comprehensive stage 3 recovery.

Permissions & Fee and Test Year

The Revised Forgiveness Inventory Scales were initially developed in 1994 by Kristina L. Coop (Gordon) and Donald H. Baucom at the University of North Carolina at Chapel Hill, with comprehensive psychometric validation and clinical outcomes formally published in 2003 in The American Journal of Family Therapy.

Copyright and Usage Terms:

• Research and Academic Use: The RFI is generally available free of charge for non-commercial academic, clinical research, and educational purposes. Researchers wishing to utilize the instrument in empirical investigations are requested to contact Dr. Kristina Coop Gordon to obtain explicit permission, scoring keys, and normative updates.
• Clinical Use: Qualified mental health practitioners (e.g., licensed psychologists, marriage and family therapists, licensed clinical social workers) may utilize the measure for diagnostic assessment, treatment planning, and outcome monitoring in couple therapy, provided appropriate bibliographic attribution is maintained.
• Commercial Licensing: Incorporation of the RFI into commercial diagnostic software, fee-for-service digital platforms, or published assessment batteries requires formal copyright clearance and written licensing agreements from the authors and/or the copyright holders of the original publication.
• Author Contact: Inquiries regarding administration permissions, clinical usage, and current research protocols should be directed to: [email protected].

References

• Coop, K. L., & Baucom, D. H. (1994). The Forgiveness Inventory (Unpublished psychometric measure). University of North Carolina at Chapel Hill, Chapel Hill, NC.
• Coop, K. L., & Baucom, D. H. (2003). Forgiveness and marriage: Preliminary support for a measure based on a model of recovery from a marital betrayal. The American Journal of Family Therapy, 31(3), 179–199. https://doi.org/10.1080/01926180390196238
• Gordon, K. C., & Baucom, D. H. (1998). Understanding betrayals in marriage: A synthesized model of forgiveness. Family Process, 37(4), 425–449. https://doi.org/10.1111/j.1545-5300.1998.00425.x
• Gordon, K. C., Baucom, D. H., & Snyder, D. K. (2004). An integrative intervention to promote forgiveness in cases of extreme marital hurt. Clinical Psychology & Psychotherapy, 11(1), 24–34. https://doi.org/10.1002/cpp.393
• Gordon, K. C., Baucom, D. H., & Snyder, D. K. (2005). Forgiveness in couples: Divorce, infidelity, and couple therapy. In E. L. Worthington Jr. (Ed.), Handbook of Forgiveness (pp. 431–446). Routledge.
• Janoff-Bulman, R. (1992). Shattered assumptions: Towards a new psychology of trauma. Free Press.
• McCullough, M. E., Rachal, K. C., Sandage, S. J., Worthington, E. L., Jr., Brown, S. W., & Hight, T. L. (1998). Interpersonal forgiving in close relationships: II. Theoretical elaboration and measurement. Journal of Personality and Social Psychology, 75(6), 1586–1603. https://doi.org/10.1037/0022-3514.75.6.1586
• Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38(1), 15–28. https://doi.org/10.2307/350547

13. Items of the Scale (Questionnaire)