Relationship Assessment Scale

1. Abstract

The Relationship Assessment Scale (RAS), originally introduced by Susan S. Hendrick in 1988, is an exceptionally concise, psychometrically robust, self-report instrument engineered to measure global romantic relationship satisfaction. Constructed to overcome the contextual and conceptual limitations of lengthier legacy inventories—most notably the 32-item Dyadic Adjustment Scale (DAS) developed by Spanier (1976)—the RAS comprises exactly seven items evaluated on a 5-point semantic-differentiated rating continuum. Unlike prior assessment batteries that were rigidly calibrated toward formally married heterosexual dyads, the RAS was intentionally conceptualized as a generic, unconstrained assessment tool applicable across an exhaustive spectrum of relational paradigms, including dating, cohabiting, married, non-cohabiting, and sexual-minority partnerships.

Psychometrically, the instrument operates as a strictly unidimensional measure that captures broad evaluations of relational well-being, perceived fulfillment of individual needs, experiential congruence with initial expectations, relational comparison against external benchmarks, subjective perception of overarching problems, and existential regret regarding partnership formation. The seven items demonstrate exemplary internal consistency across diverse empirical inquiries, yielding Cronbach’s alpha coefficients typically ranging between .85 and .91. Confirmatory factor analyses consistently substantiate an invariant single-factor infrastructure across biological sex, marital classification, and cross-cultural cohorts. The scale exhibits profound convergent validity with extensive relationship adjustment metrics, robust discriminant validity against independent affective constructs such as generalized depressive symptomatology and self-esteem, and notable predictive validity in longitudinal frameworks mapping relationship stability versus dissolution trajectories. Given its microscopic respondent burden, minimal completion latency (typically under two minutes), and profound psychometric integrity, the RAS functions as a foundational assessment benchmark in contemporary social psychology, clinical couple therapy, and epidemiological relationship science.

2. Keywords

Relationship Assessment Scale, RAS, relationship satisfaction, relationship quality, psychometrics, marital satisfaction, dyadic assessment, scale validation, Susan S. Hendrick, close relationships, couple therapy, unidimensional measurement

3. Authors

The Relationship Assessment Scale was developed by Susan S. Hendrick, Ph.D., an esteemed American clinical and social psychologist who has made foundational contributions to relationship science, romantic love taxonomies, and marital quality measurement.

• Susan S. Hendrick, Ph.D.
  • Primary Affiliation: Department of Psychological Sciences, Texas Tech University, Lubbock, Texas, United States.
  • Academic Rank: Professor Emerita of Psychology.
  • Biographical Context & Collaborative Network: Dr. Hendrick, along with her long-time collaborator and spouse Clyde Hendrick, Ph.D., formulated pioneering empirical operationalizations of love styles (Love Attitudes Scale) and relationship evaluation paradigms. Her research systematically explored relational schemas, dyadic commitment, self-disclosure dynamics, and the psychological impact of personal illness upon close partnerships.
• Significant Psychometric Re-evaluators:
  • Amy Dicke, M.A. & Susan S. Hendrick, Ph.D. (1998) — Conducted extensive psychometric validation and structural replication confirming cross-sample factor invariance.
  • Frank D. Fincham, Ph.D. & Thomas N. Bradbury, Ph.D. — Methodologists whose theoretical and empirical critiques of confounded dyadic measures supported the widespread adoption of Hendrick’s generic unidimensional paradigm over multidimensional behavioral inventories.

4. Purpose

The foundational purpose of the Relationship Assessment Scale is to provide researchers, applied clinicians, and behavioral scientists with an efficient, generic, and psychometrically pure instrument designed to measure a single subjective psychological domain: global romantic relationship satisfaction. In the decades preceding the development of the RAS, the field of marital and relationship therapy relied heavily on extensive, structurally complex instruments such as the 32-item Dyadic Adjustment Scale (DAS; Spanier, 1976) and the 280-item Marital Satisfaction Inventory (MSI; Snyder, 1979). While these legacy inventories provided granular descriptions of dyadic dynamics, they suffered from significant theoretical and practical shortcomings that severely constrained their empirical versatility.

Chief among these limitations was the profound confound between relational quality as a subjective psychological state and the specific behavioral interactions that supposedly cultivate it. For instance, the DAS includes items assessing the frequency of shared interests, household labor negotiations, joint financial management, and dyadic conflict discussions. Consequently, when investigators sought to evaluate whether effective conflict communication predicted relational satisfaction, statistical analyses suffered from extreme endogeneity, circularity, and shared-method variance, as items measuring communication behaviors were concurrently embedded within the satisfaction metric itself. Hendrick (1988) engineered the RAS explicitly to strip away behavioral interactions, conflict frequencies, and lifestyle negotiations, isolating the pure evaluative perception of how individuals subjectively appraise their relationship.

Moreover, historically dominant scales were explicitly anchored in heterosexual legal marriages, frequently incorporating terminology that assumed cohabitation, legal obligations, or shared economic infrastructure. Consequently, these measures were largely invalid or awkward when administered to dating cohorts, emerging adults, non-cohabiting lovers, or sexual-minority couples who lacked access to civil marriage frameworks. The RAS addresses this structural gap by presenting generic, emotionally evocative items that do not presuppose cohabitation status, legal standing, biological sex, or relationship longevity. Phrases such as “your partner” and “this relationship” ensure immediate, unrestricted applicability across any committed or exploratory romantic context.

From an applied clinical perspective, the purpose of the RAS encompasses rapid baseline assessment, session-by-session tracking, and intervention outcome quantification within couple and family therapy. Long psychometric batteries impose substantial cognitive overhead, fatigue, and frustration upon clinically distressed couples, precipitating high rates of missing data or perfunctory responses. The RAS eliminates respondent burden by enabling complete administration within 90 to 120 seconds, allowing clinicians to objectively track therapeutic trajectories longitudinally without consuming valuable session time. Within large-scale epidemiological, longitudinal, or cross-sectional survey research, the inclusion of a comprehensive relationship battery is frequently untenable due to item-budget constraints. The RAS solves this methodological trade-off, offering exceptional psychometric performance within an ultra-brief seven-item format.

5. Psychological Construct

The construct captured by the Relationship Assessment Scale is global relationship satisfaction, conceptualized as an overarching, subjective cognitive-affective evaluation of an ongoing romantic partnership. Grounded in social exchange principles, relationship satisfaction reflects the degree to which an individual perceives their relational experiences to be gratifying, valuable, and congruent with subjective standards, internal ideals, and comparative relationship options. The RAS approaches this construct as an integrative, unifactorial psychological state rather than a multidimensional constellation of distinct behavioral components.

Although the scale scores strictly as a unidimensional construct, its seven items systematically tap several essential evaluative facets that collectively define the subjective experience of romantic quality:

• Need Fulfillment: The first item assesses the extent to which an individual perceives their fundamental emotional, instrumental, and psychological needs as being met by their partner. Rooted in self-determination theory and interpersonal interdependence frameworks, need fulfillment reflects the partner’s responsiveness, empathy, and attunement to core psychological requirements, functioning as a primary pillar of perceived relationship adequacy.
• Global Subjective Evaluation: The second item directly queries the respondent’s overarching, holistic appraisal of relationship satisfaction. Rather than isolating specific domains (such as intimacy, finance, or shared leisure), this dimension invites the respondent to integrate all emotional, cognitive, and somatic inputs into an unconstrained, summary judgment of relational well-being.
• Comparative Social Evaluation: The third item requires the individual to benchmark their partnership against external reference points (“compared to most”). Drawing from social comparison theory, this facet recognizes that human satisfaction is fundamentally relativistic; individuals appraise their private relational experiences by assessing their relative positioning within perceived peer, familial, and cultural distributions of relationship quality.
• Existential Regret and Sunk-Cost Cognitions: The fourth item (reverse-scored) taps the frequency with which a person wishes they had never entered the partnership. This dimension identifies intrusive, second-order ruminations and covert relational dissatisfaction that may be masked by overt polite compliance or surface satisfaction. High frequencies of retrospective regret signal substantial cognitive ambivalence and profound relational erosion.
• Expectancy Congruence: The fifth item evaluates the discrepancy between an individual’s pre-relational or foundational expectations and their lived relational reality. Grounded in expectancy-disconfirmation paradigms, this dimension assesses whether the partnership has fulfilled, superseded, or fallen short of the idealized relational schemas held prior to or during the early phases of romantic attachment.
• Affective Attachment and Love: The sixth item quantifies the subjective intensity of love felt toward the partner. Although love and satisfaction are theoretically distinct constructs, phenomenological and empirical research reveals that in romantic pairings, profound positive affective investment serves as both an emotional fuel source and a defining experiential manifestation of relational fulfillment.
• Perceived Problem Density: The seventh item (reverse-scored) evaluates the perceived volume and salience of unresolved difficulties, friction points, and relational strain. Satisfaction is not merely the presence of positive inputs; it is deeply bounded by the perceived density of chronic interpersonal impediments, stressors, and unresolved conflicts.

6. Theoretical Framework

The development and interpretation of the Relationship Assessment Scale are grounded in several foundational psychological frameworks within interpersonal relationship science, primarily Interdependence Theory, Social Exchange Theory, and Social Comparison Theory.

Foremost among these theoretical foundations is Interdependence Theory, articulated by Harold Kelley and John Thibaut (1978), and further developed by Caryl Rusbult in the Investment Model of Commitment (1980). Interdependence Theory posits that relationship participants continuously evaluate the outcomes of their dyadic exchanges by computing the balance between experienced rewards (e.g., affection, emotional safety, validation, shared resources) and accrued costs (e.g., conflict, effort, compromised autonomy, behavioral irritation). Satisfaction, however, does not emerge in an absolute cognitive vacuum based solely on net outcome valence. Instead, it is evaluated against the individual’s Comparison Level ($CL$)—a subjective internal standard representing what the individual believes they deserve or have learned to anticipate based on past partnerships, parental modeling, and cultural portrayals.

When an individual perceives that their relational outcomes exceed their Comparison Level ($Outcome > CL$), high relationship satisfaction is experienced. Conversely, when relational experiences descend below this personal threshold ($Outcome < CL$), dissatisfaction occurs, irrespective of whether the partnership remains structurally intact. The RAS directly incorporates this interdependence formulation: Item 1 captures net reward value (need fulfillment), Item 7 taps the perceived severity of costs (problem volume), and Item 5 directly assesses outcome-to-$CL$ alignment by evaluating expectations against reality.

Furthermore, Interdependence Theory differentiates subjective satisfaction from structural commitment by introducing the Comparison Level for Alternatives ($CL_{alt}$). While the RAS does not directly evaluate the presence of specific external alternative partners, Item 4 (wishing one had not gotten into the relationship) captures the psychological activation of counterfactual reasoning and alternative-state evaluations. Frequent regret reflects an underlying cognition that the zero-point state (being single) or hypothetical alternative pathways possess a higher perceived outcome value than the current partnership.

The second pillar of the theoretical architecture is Social Comparison Theory (Festinger, 1954). In the absence of definitive, objective physical standards to verify relational quality, human beings inevitably engage in social comparison processes to validate their subjective experiences. Individuals observe the public behaviors, narratives, and perceived happiness of peers, friends, and family units, establishing external reference benchmarks. Item 3 (“How good is your relationship compared to most?”) explicitly activates downward or upward social comparison mechanisms. Downward social comparisons (perceiving one’s relationship as vastly superior to average partnerships) serve a crucial cognitive-enhancement and stabilization function, robustly bolstering subjective satisfaction.

Finally, the scale reflects the cognitive-affective synthesis proposed by Fincham and Bradbury (1987), who strongly advocated for the conceptual separation of marital quality evaluation from marital interaction behaviors. Hendrick (1988) embraced this cognitive-affective paradigm, operationalizing satisfaction as a unified evaluative schema. Under this framework, subjective satisfaction functions as a higher-order cognitive heuristic—a stored psychological summary index derived from countless daily interactions, emotional reactions, and cognitive appraisals, which serves to guide the individual’s global orientation toward the relationship.

7. Validity

The Relationship Assessment Scale has undergone exhaustive empirical evaluation, yielding robust evidence of construct, convergent, discriminant, and predictive validity across diverse domestic and international investigations.

Convergent Validity

Convergent validity has been repeatedly demonstrated through strong statistical associations with established, lengthier dyadic inventories. In Hendrick’s (1988) seminal validation investigation ($N = 113$ dating couples), the RAS correlated exceptionally well with the Spanier (1976) Dyadic Adjustment Scale ($r = .80$ for females, $r = .88$ for males). Subsequent investigations have documented parallel correlations with the Marital Adjustment Test (MAT; Locke & Wallace, 1959), consistently producing correlation coefficients between $r = .73$ and $r = .84$. Furthermore, the RAS exhibits moderate-to-high positive correlations with measures of secure adult attachment (e.g., the Experiences in Close Relationships scale; $r = .45$ to $.60$ with low attachment avoidance/anxiety), partner responsiveness ($r = .65$), and the Eros (passionate love) and Agape (altruistic love) subscales of the Love Attitudes Scale ($r = .50$ to $.68$).

Discriminant Validity

Discriminant validity has been demonstrated by showing that the RAS assesses relationship satisfaction as a distinct construct, separate from generalized positive affectivity, self-esteem, or psychological distress. Hendrick (1988) and Hendrick, Dicke, and Hendrick (1998) established that while the RAS correlates negatively with measures of individual psychological distress—such as the Beck Depression Inventory (BDI; $r = -.25$ to $-.38$)—these associations remain modest. This confirms that dyadic dissatisfaction is not merely an artifact of individualized depressive symptomatology or general negative affectivity. Similarly, correlations between the RAS and generalized global self-esteem (e.g., the Rosenberg Self-Esteem Scale) typically hover at weak-to-moderate levels ($r = .20$ to $.35$), illustrating that individual self-worth and relational satisfaction operate as distinct psychological systems.

Predictive and Criterion Validity

Predictive validity is particularly robust within the domain of relationship stability and dissolution trajectories. In Hendrick’s (1988) longitudinal follow-up of dating couples, the baseline RAS scores of individuals who remained together several months later were significantly higher ($M = 4.34$) compared to the baseline scores of those whose relationships had dissolved ($M = 3.32, p < .001$). Discriminant functional analysis indicated that the seven-item scale correctly classified over 80% of couples into their respective future relational categories (intact versus separated). More recent longitudinal studies have reaffirmed that low scores on the RAS reliably predict relationship termination across dating, cohabiting, and married samples, demonstrating strong utility in clinical risk assessment.

8. Reliability

The Relationship Assessment Scale demonstrates consistently high internal consistency and temporal stability across a broad spectrum of demographic groups, relationship types, and cultural contexts.

Internal Consistency

In her foundational 1988 study, Susan Hendrick reported an overall Cronbach’s alpha coefficient of $\alpha = .86$ within a collegiate dating cohort. Subsequent validation studies conducted by Hendrick, Dicke, and Hendrick (1998) involving married and community couples demonstrated even higher internal consistency, reporting alphas ranging from $\alpha = .87$ to $\alpha = .91$. Multiple independent replications across diverse samples have confirmed this reliability:

• Collegiate and emerging adult dating cohorts: $\alpha = .85 – .88$
• Cohabiting heterosexual and same-sex couples: $\alpha = .86 – .90$
• Long-term married couples: $\alpha = .88 – .92$
• Cross-cultural adaptations (including Spanish, Turkish, German, Portuguese, and Persian): $\alpha = .82 – .91$

Item-total correlations for all seven items are consistently high, rarely dipping below $r = .55$. Items 2 (general satisfaction) and 5 (met expectations) consistently exhibit the highest corrected item-total correlations, often exceeding $r = .75$, underscoring their role as central indicators of the underlying satisfaction continuum.

Test-Retest Reliability

The temporal stability of the RAS has been validated across various test-retest intervals. Hendrick (1988) observed high stability over brief test-retest windows, reporting coefficients exceeding $r = .85$ over a two-to-three-week period in stable relationship samples. In longitudinal studies spanning six to twelve weeks, test-retest reliability estimates typically range from $r = .72$ to $r = .80$. This pattern highlights an ideal balance: the instrument exhibits sufficient stability to reflect an enduring psychological orientation, while retaining the sensitivity needed to capture meaningful shifts in relationship dynamics over time or in response to clinical intervention.

9. Factor Analysis

Extensive exploratory factor analyses (EFA) and confirmatory factor analyses (CFA) have firmly established the structural unidimensionality of the Relationship Assessment Scale.

Exploratory Factor Analysis (EFA)

In the original EFA conducted by Hendrick (1988) using principal axis factoring with orthogonal and oblique rotations, a single dominant eigenvalue emerged (accounting for approximately 46% to 57% of the total variance across gender-stratified sub-analyses). The scree test clearly indicated a single-factor solution, with all seven items loading significantly onto this primary factor:

• Item 1 (Partner meets needs): Factor loading $\approx .68$ to $.73$
• Item 2 (General satisfaction): Factor loading $\approx .80$ to $.86$
• Item 3 (Compared to most): Factor loading $\approx .64$ to $.71$
• Item 4 (Wish hadn’t gotten in – reversed): Factor loading $\approx .52$ to $.61$
• Item 5 (Met expectations): Factor loading $\approx .74$ to $.81$
• Item 6 (Love for partner): Factor loading $\approx .60$ to $.68$
• Item 7 (Problems in relationship – reversed): Factor loading $\approx .50$ to $.59$

Although Items 4 and 7 (the reverse-scored items) occasionally exhibit lower loadings due to methodological variance common to negatively framed statements, they consistently exceed standard psychometric retention thresholds ($ge .40$) across samples.

Confirmatory Factor Analysis (CFA) & Model Fit

Subsequent psychometric evaluations have utilized structural equation modeling to verify this single-factor construct. Hendrick, Dicke, and Hendrick (1998) examined the unidimensional model across a large, non-clinical sample ($N = 457$), demonstrating excellent global model fit indices:

• Comparative Fit Index (CFI): $.95 – .98$
• Tucker-Lewis Index (TLI): $.94 – .97$
• Root Mean Square Error of Approximation (RMSEA): $.048 – .065$
• Standardized Root Mean Square Residual (SRMR): $.031 – .042$

Multigroup confirmatory factor analyses (MGCFA) have further confirmed structural and measurement invariance across biological sex (males vs. females) and relationship status (dating vs. cohabiting vs. married). Invariance testing demonstrates that the factor loadings ($lambda$), item intercepts ($ au$), and residual variances ($ heta$) remain stable across groups, confirming that the RAS measures the same psychological construct in identical ways across diverse populations.

10. Instrument / Measurement Tool

The operational and administrative parameters of the Relationship Assessment Scale are detailed below:

• Instrument Name: Relationship Assessment Scale (RAS)
• Developer: Susan S. Hendrick, Ph.D. (1988)
• Construct Measured: Global Romantic Relationship Satisfaction
• Format: Self-administered paper-and-pencil or digital questionnaire
• Item Count: 7 items
• Administration Time: Approximately 1 to 2 minutes
• Target Population: Adolescents and adults (ages 16+) engaged in any form of ongoing romantic partnership (dating, engaged, cohabiting, married, non-cohabiting, monogamous, non-monogamous, heterosexual, LGBTQ+)
• Response Scale: 5-point scale (ranging from 1 to 5; anchors vary by item, e.g., 1 = Poorly/Low/Never to 5 = Extremely well/High/Very often)
• Scoring and Transformation Rules:
  • Reverse Scoring: Items 4 and 7 are negatively keyed and must be reverse-scored prior to calculating final metrics ($1 \rightarrow 5$, $2 \rightarrow 4$, $3 \rightarrow 3$, $4 \rightarrow 2$, $5 \rightarrow 1$).
  • Total Score Calculation: Calculate the mean across all seven items by summing the item scores (after reversing items 4 and 7) and dividing by 7: $$\text{RAS Score} = \frac{\sum_{i=1}^{7} X_i}{7}$$ Alternatively, raw sums may be utilized (ranging from 7 to 35). The mean approach (ranging from 1.00 to 5.00) is standard practice, as it preserves direct interpretability alongside the original 5-point response continuum.
  • Score Interpretation: Higher numerical values reflect greater romantic relationship satisfaction. Mean scores $ge 4.0$ typically denote high relationship quality, while scores $le 3.5$ are commonly used in clinical screening to flag relational distress.

11. Permissions & Fee and Test Year

The Relationship Assessment Scale was formally published in 1988 in the Journal of Marriage and the Family by Dr. Susan S. Hendrick. In subsequent publications and clinical manuals, the scale was placed directly into the academic public domain for scientific, non-commercial, and therapeutic use.

• Accessibility: The scale is widely accessible without licensing fees for academic researchers, clinicians, educators, and students.
• Commercial Applications: Commercial deployments—such as integration into fee-for-service enterprise software platforms, commercial mental health apps, or corporate employee assistance systems—may require permission from the original publisher (National Council on Family Relations / Wiley-Blackwell) or the author.
• Attribution: Academic and clinical users are required to provide appropriate formal attribution citing Hendrick’s foundational 1988 publication and relevant subsequent validation literature.

12. References

The following peer-reviewed publications document the empirical development, theoretical foundations, and psychometric validation of the Relationship Assessment Scale:

• Fincham, F. D., & Bradbury, T. N. (1987). The assessment of marital quality: A reevaluation. Journal of Marriage and the Family, 49(4), 797–809. https://doi.org/10.2307/351973
• Hendrick, S. S. (1988). A generic measure of relationship satisfaction. Journal of Marriage and the Family, 50(1), 93–98. https://doi.org/10.2307/352430
• Hendrick, S. S., Dicke, A., & Hendrick, C. (1998). The Relationship Assessment Scale. Journal of Social and Personal Relationships, 15(1), 137–142. https://doi.org/10.1177/0265407598151009
• Kelley, H. H., & Thibaut, J. W. (1978). Interpersonal relations: A theory of interdependence. John Wiley & Sons.
• Locke, H. J., & Wallace, K. M. (1959). Short marital-adjustment and prediction tests: Their reliability and validity. Marriage and Family Living, 21(3), 251–255. https://doi.org/10.2307/348022
• Rusbult, C. E. (1980). Commitment and satisfaction in romantic associations: A test of the investment model. Journal of Experimental Social Psychology, 16(2), 172–186. https://doi.org/10.1016/0022-1031(80)90007-4
• Snyder, D. K. (1979). Multidimensional assessment of marital satisfaction. Journal of Marriage and the Family, 41(4), 813–823. https://doi.org/10.2307/351481
• 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
• Vaughn, M. J., & Baier, M. E. (1999). Reliability and validity of the Relationship Assessment Scale. American Journal of Family Therapy, 27(2), 137–147. https://doi.org/10.1080/019261899261971

13. Items of the Scale