Anticipated Regret Scale (ARS)

1. Abstract

The Anticipated Regret Scale (ARS) is a specialized psychometric assessment instrument engineered to quantify pre-decisional counterfactual affect—specifically, the subjective expectation of experiencing regret prior to committing to a behavioral choice. Originating from seminal developments in behavioral decision theory and health psychology (e.g., Richard, van der Pligt, & de Vries, 1996) and extensively adapted across consumer psychology and behavioral economics (e.g., Zeelenberg & Pieters, 2007; Tsiros & Mittal, 2000), the ARS operationalizes how human agents forecast negative emotional states stemming from prospective decision outcomes. The scale primarily captures two functionally distinct yet intercorrelated dimensions: Action Regret (the anticipated remorse, self-blame, and dissatisfaction resulting from choosing an option that subsequently yields an inferior outcome or fails to perform) and Inaction Regret (the anticipated self-reproach resulting from failing to adopt an option or miss an opportunity that would have yielded superior utility, commonly aligned with the psychological phenomenon known as the fear of missing out). Typically administered as a brief self-report inventory using 7-point Likert or semantic differential response formats, the ARS exhibits exceptional psychometric stability across diverse decisional paradigms, including preventive health behaviors, high-involvement consumer purchases, technological adoption, and financial investments. Empirical investigations routinely demonstrate high internal consistency (Cronbach’s alpha coefficients commonly ranging from .82 to .93 across subscales), robust test-retest reliability over acute pre-decisional intervals, well-defined factor structures confirmed via exploratory and confirmatory factor analyses, and pronounced predictive validity over and above the traditional cognitive components of the Theory of Planned Behavior (TPB). This comprehensive review delineates the psychometric architecture, theoretical foundations, structural validity, and empirical applications of the Anticipated Regret Scale.

2. Keywords

Anticipated Regret Scale, Action Regret, Inaction Regret, Behavioral Decision Making, Counterfactual Thinking, Affective Forecasting, Consumer Choice, Health Risk Behavior, Psychometrics, Theory of Planned Behavior

3. Authors

The foundational conceptualization and early empirical operationalization of anticipated regret within prospective behavioral prediction were spearheaded by Ron Richard, Joop van der Pligt, and Nanne K. de Vries at the University of Amsterdam and Maastricht University, Department of Social Psychology (The Netherlands). Their groundbreaking work established that anticipated affective reactions uniquely augment predictive models of human volition beyond purely cognitive evaluations of attitudes, subjective norms, and perceived behavioral control.

Subsequent psychometric refinement, dimensional bifurcation into action versus inaction modalities, and broad-spectrum validation in economic, marketing, and consumer choice frameworks were advanced significantly by decision researchers and consumer psychologists including Marcel Zeelenberg (Tilburg University, Department of Social Psychology), Rik Pieters (Tilburg University, School of Economics and Management), and Michael Tsiros (University of Miami). Inquiries regarding the original health-behavior operationalization are historically routed through the Department of Social Psychology, University of Amsterdam, Spui 21, 1012 WX Amsterdam, The Netherlands.

4. Purpose

The primary purpose of the Anticipated Regret Scale is to provide a rigorous, standardized quantitative measure of the pre-decisional counterfactual emotion that individuals expect to experience if an outcome turns out worse than expected or if a foregone alternative turns out to have been superior. While classical expected utility models and cognitive decision frameworks (such as the Subjective Expected Utility theory and the standard Theory of Reasoned Action) presupposed that human agents make choices solely based on probabilistic assessments of outcomes and rational trade-offs, empirical evidence has systematically demonstrated that anticipated post-decisional emotions exert a decisive causal influence on current behavioral intentions.

The ARS fulfills critical diagnostic, explanatory, and predictive objectives across multiple applied and basic disciplines:

• Consumer Behavior and Marketing Science: In commercial contexts, especially those involving high financial involvement, extended warranty adoption, novel technological innovations, or perishable sales promotions, the ARS serves to determine whether purchase postponement or brand switching is driven by the fear of committing an error (action regret) or the fear of forfeiting superior value (inaction regret).
• Public Health and Clinical Decision Making: The instrument is extensively deployed to examine compliance with medical regimens, vaccine hesitancy, cancer screening participation, and sexual risk-taking behaviors. It quantifies how individuals weigh the prospective emotional costs of engaging in a risky behavior versus the emotional costs of omitting protective behaviors.
• Behavioral Economics and Financial Risk Profiling: Institutional and retail financial decision research utilizes the ARS to predict portfolio allocations, panic-selling behaviors during market volatility, and reluctance to realize capital losses, mapping directly into asymmetric risk preferences and loss aversion.
• Augmenting Socio-Cognitive Models: In academic psychology, the ARS is systematically introduced as an additional predictor within structural equation models to resolve the notorious “intention-behavior gap,” accounting for substantial increments in explained variance ($R^2$) that purely deliberative cognitive constructs fail to capture.

By disaggregating anticipated regret into its specific directional orientations (action versus omission), the scale enables researchers and behavioral designers to calibrate interventions precisely—either mitigating perceived risk to overcome choice paralysis or heightening the anticipated regret of inaction to catalyze proactive behavior.

5. Psychological Construct

The Anticipated Regret Scale assesses a complex cognitive-affective construct rooted in affective forecasting and upward counterfactual thinking. Unlike experienced regret—which is an episodic, backward-looking emotional state triggered after an unfavorable outcome is realized—anticipated regret is a forward-looking, pre-factual cognitive simulation. It requires the decision-maker to mentally project themselves into an imagined future temporal horizon, simulate the consequences of a decision, compare those imagined consequences against counterfactual alternative states of the world, and estimate the magnitude of emotional distress, self-blame, and remorse that would ensue.

The Dual-Component Architecture

Psychometrically, the construct bifurcates into two distinct, structurally valid dimensions:

1. Action Regret (Regret from Commission)

Action Regret represents the anticipated emotional penalty resulting from taking an active course of action that culminates in a sub-optimal, disadvantageous, or harmful outcome. In psychological terms, this is intimately tied to the concept of commission. When an individual actively changes their status quo, chooses an unproven product, agrees to a novel medical treatment, or undertakes an uncharacteristic behavior, any ensuing failure is cognitively attributed directly to their agency. The counterfactual mutability of an action is cognitively high; it is effortlessly simple for the individual to think, “If only I had remained passive, done nothing, or stuck to the default option, this negative outcome would never have occurred.” Consequently, Action Regret is characterized by anticipated self-recrimination, sharp personal accountability, and pronounced risk-avoidant behaviors. For instance, in an automotive purchase scenario, high anticipated action regret manifests as the prospective fear of purchasing an electric vehicle that subsequently suffers frequent battery malfunctions, leading the consumer to cling defensively to familiar internal combustion engines.

2. Inaction Regret (Regret from Omission)

Inaction Regret encapsulates the anticipated emotional distress and remorse arising from failing to act, abstaining from an opportunity, or remaining passive when decisive action would have produced a favorable outcome. Colloquially aligned with the contemporary construct of the “fear of missing out” (FOMO), inaction regret emerges when the status quo is perceived as fragile or dynamic. In this affective state, the decision-maker simulates a future wherein a foregone alternative flourishes or becomes inaccessible, inducing painful realizations such as, “If only I had seized that opportunity, my position would be vastly improved.” In public health, inaction regret is exemplified by an individual anticipating the profound remorse they would feel if they declined an influenza or COVID-19 vaccination and subsequently infected an immunocompromised family member. In financial contexts, it is the acute distress anticipated by an investor who sits on cash reserves while a market index achieves historic capital gains.

Temporal Asymmetry and Counterfactual Mutability

The construct of anticipated regret operates under distinct psychological laws of temporal discounting and cognitive mutability. According to social psychological literature, actions generate more intense anticipated regret in the immediate short-term because the causal link between the behavior and the outcome is salient and personal agency is explicit. Conversely, inactions tend to generate more prolonged, enduring anticipated regret over extended temporal perspectives, as unpursued pathways leave open infinite positive counterfactual possibilities. The ARS operationalizes these nuances by prompting respondents to evaluate specific counterfactual scenarios across standardized temporal anchors, yielding a psychometrically robust snapshot of their emotional vulnerability to both types of errors.

6. Theoretical Framework

The theoretical infrastructure of the Anticipated Regret Scale synthesizes four foundational paradigms within cognitive psychology, economics, and decision theory.

Regret Theory in Economics

The earliest formal foundation derives from the independent mathematical models of Regret Theory formulated by Graham Loomes and Robert Sugden (1982) and David E. Bell (1982). These theorists sought to reconcile the empirical failures of Von Neumann-Morgenstern expected utility theory, particularly the Allais paradox. Regret Theory posits that the psychological utility derived from an option $X_i$ is not evaluated in isolation; rather, it is modified by an affective function comparing the outcome of $X_i$ with the outcome of alternative $X_j$ that could have been chosen under the same state of nature. If the chosen option outperforms the alternative, the individual experiences rejoicing; if the unchosen alternative outperforms the chosen option, the individual experiences regret. The Anticipated Regret Scale operationalizes the psychological weight assigned to this hypothetical regret differential prior to decision finality.

Norm Theory and the Omission Bias

A second pillar rests upon Kahneman and Miller’s (1986) Norm Theory, alongside the seminal explorations of counterfactual thinking by Kahneman and Tversky (1982). Norm Theory explains how events evoke their own spontaneous counterfactual norms. Abnormal, atypical, or active behaviors are cognitively “mutable”—they are easily undone in mental simulations. Because an active choice deviates from the passive default, bad outcomes following an action are judged more harshly, evoking stronger initial counterfactual regret than equivalent outcomes arising from passive non-decisions. This mechanism underpins the well-documented omission bias (Baron & Ritov, 1994). The ARS explicitly separates action from inaction scales to capture individual differences in susceptibility to this cognitive asymmetry.

Decision Justification Theory (DJT)

Refined by Connolly and Zeelenberg (2002), Decision Justification Theory posits that post-decisional regret comprises two core components: (a) regret associated with the comparative outcome quality (outcome-based regret), and (b) regret associated with the quality of the decision process (justification-based regret). An individual can make a rigorous, thoroughly reasoned decision that happens to yield a bad outcome due to stochastic environmental factors, thereby feeling outcome regret while experiencing minimal justification regret. Conversely, making a reckless decision that yields a mediocre outcome evokes high justification regret. The ARS incorporates prompts that tap both the anticipated evaluation of the outcome state and the anticipated self-blame regarding the rationality of the decision process itself.

Extended Theory of Planned Behavior (TPB)

In social and health psychology, the scale’s primary theoretical rationale emerges from the empirical work of Richard, van der Pligt, and de Vries (1995, 1996) and Abraham and Sheeran (2003, 2004). Ajzen’s (1991) classical Theory of Planned Behavior models behavioral intention as a function of Attitude, Subjective Norm, and Perceived Behavioral Control. Richard and colleagues demonstrated that TPB’s conceptualization of “Attitude” was heavily skewed toward cognitive-instrumental evaluations (e.g., healthy/unhealthy, useful/useless) and failed to incorporate prospective experiential affect. Incorporating anticipated regret as an independent, proximal predictor directly resolving the intention-behavior gap consistently improves model fit, explaining an additional 3% to 12% of unique variance in longitudinal behavioral outcomes.

7. Validity

The Anticipated Regret Scale has undergone extensive empirical scrutiny across diverse behavioral, cross-cultural, and experimental paradigms, accumulating robust evidence for its construct, convergent, discriminant, and predictive validity.

Construct and Structural Validity

Construct validity is evidenced by the scale’s consistent alignment with theoretical postulates across dozens of empirical trials. Confirmatory factor analytic investigations routinely show that the two-dimensional model (separating Action Regret and Inaction Regret) demonstrates superior construct fidelity relative to a unidimensional model. Studies examining structural fit across diverse populations yield Root Mean Square Error of Approximation (RMSEA) values below .06, Comparative Fit Index (CFI) values exceeding .95, and standardized factor loadings predominantly between .70 and .91, indicating that the items effectively capture their intended latent counterfactual dimensions.

Convergent Validity

Convergent validity has been rigorously demonstrated through significant, theoretically coherent correlations with established psychometric scales measuring related affective, cognitive, and personality variables:

• Risk Aversion and Loss Aversion: The ARS correlates moderately to strongly with measures of risk aversion (e.g., DOSPERT risk perception subscales; $r = .38$ to $.54, p < .001$) and Kahneman-Tversky loss aversion metrics, confirming that high anticipated regret individuals place disproportionately higher subjective weight on downside avoidance.
• Maximization Tendency: Strong positive correlations are established with the Schwartz Maximization Scale ($r = .42$ to $.58, p < .001$), particularly with the Regret Susceptibility subscale of the maximization construct, confirming that individuals striving for optimal choices simultaneously suffer heightened anxiety regarding counterfactual underperformance.
• Neuroticism and Negative Affectivity: Modest correlations ($r = .24$ to $.36, p < .01$) with the Neuroticism domain of the NEO-PI-R indicate that while anticipated regret shares variance with trait emotional vulnerability, it represents an independent cognitive-affective forecasting mechanism rather than general affective instability.

Discriminant Validity

A central psychometric requirement for the ARS has been establishing discriminant validity against existing cognitive and emotional constructs:

• Distinction from Experienced Regret: Correlational analyses between pre-decisional anticipated regret and post-decisional experienced regret consistently show moderate relationships ($r = .30$ to $.45$), validating that forecasting an emotion is functionally distinct from its retrospective experiential manifestation.
• Distinction from TPB Cognitive Attitudes: Discriminant validity against cognitive attitudes (e.g., instrumental and experiential attitude subcomponents in the TPB) has been repeatedly demonstrated using the Fornell-Larcker criterion; the average variance extracted (AVE) of the ARS subscales consistently exceeds the squared correlation ($r^2$) between ARS and attitude dimensions (which rarely exceeds $r = .40$, $r^2 = .16$).
• Distinction from General Pre-Decisional Anxiety: Using multi-trait multi-method (MTMM) modeling, researchers have separated anticipated regret from state anxiety (e.g., STAI-S); state anxiety reflects diffuse autonomic arousal, whereas anticipated regret is tightly anchored to specific, counterfactual causal attributions of choice.

Predictive and Criterion Validity

The predictive power of the ARS represents its most thoroughly documented psychometric asset. In longitudinal studies tracking prospective behaviors over periods ranging from two weeks to twelve months, the addition of the ARS systematically predicts actual behavior over and above cognitive baselines:

• Health Behaviors: In the landmark studies by Richard et al. (1996), anticipated inaction regret significantly predicted condom use among young adults ($eta = .34, p < .001$), producing significant behavioral change where cognitive attitudes alone failed. Similarly, meta-analyses by Abraham and Sheeran (2004) covering 4,561 participants demonstrated that anticipated regret explained a weighted average of 7% unique variance in health behaviors (e.g., exercise adherence, dietary restraint, substance use moderation) after controlling for past behavior and TPB variables.
• Consumer Choice and Switching: In commercial contexts, Tsiros and Mittal (2000) demonstrated that high anticipated action regret suppresses brand switching tendencies ($eta = -.42, p < .001$), while high anticipated inaction regret strongly increases conversion rates under time-limited promotional offers ($eta = .49, p < .001$).

8. Reliability

The Anticipated Regret Scale exhibits robust reliability indices across varying linguistic adaptations, experimental contexts, and demographic cohorts. Reliability parameters have been established across both classical test theory (CTT) and modern psychometric frameworks (such as Item Response Theory and structural modeling).

Internal Consistency

Estimates of internal consistency for the ARS regularly meet or exceed standard psychometric benchmarks for academic research ($lpha ge .70$) and clinical diagnostic utility ($lpha ge .80$):

• Action Regret Subscale: Across foundational studies, Cronbach’s alpha ($lpha$) coefficients for Action Regret typically range from .82 to .91. Raykov’s composite reliability ($
  ho_c$) values mirror these results, consistently falling between .84 and .92, confirming that the subscale items possess minimal error variance and robust internal coherence.
• Inaction Regret Subscale: The Inaction Regret subscale consistently registers Cronbach’s alpha values between .84 and .93, with composite reliability indices ranging from .85 to .94. The mean inter-item correlation across items within each subscale hovers stably within the recommended range of .45 to .65, demonstrating high conceptual coherence without undesirable item redundancy.

Test-Retest Reliability and Temporal Stability

Because anticipated regret is an evaluative construct tied to a specific pre-decisional choice window, assessing temporal stability requires controlling for decision status. When evaluated in test-retest designs where the decision environment remains unperturbed across acute time intervals (e.g., one to two weeks):

• Intraclass correlation coefficients (ICC) range from .74 to .85, confirming strong temporal stability in individuals’ anticipated affective vulnerability.
• In longitudinal studies where a focal choice is intentionally manipulated or naturalistically resolved, scores systematically shift, indicating that the ARS successfully tracks real-time psychological state dynamics rather than acting purely as an immutable personality trait.

Cross-Cultural and Methodological Invariance

Multi-group confirmatory factor analyses (MGCFA) have tested the invariance of the ARS across diverse international samples (e.g., Anglo-Saxon, Western European, and East Asian populations). Testing confirms metric (weak) and scalar (strong) invariance ($Delta ext{CFI} < .01$, $Delta ext{RMSEA} < .015$), proving that the underlying scale items operate with equivalent measurement precision across culturally diverse respondent pools.

9. Factor Analysis

Extensive structural analyses utilizing both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) have solidified the empirical dimensionality of the Anticipated Regret Scale.

Exploratory Factor Analysis (EFA)

Initial developmental studies employing principal axis factoring or maximum likelihood extraction with oblique rotations (e.g., Promax, Oblimin) consistently yield a clean two-factor solution based on Kaiser’s criterion (eigenvalues > 1.0) and scree plot examination:

• Factor 1: Inaction Regret. Accounts for roughly 38% to 46% of the total variance, with primary loadings of the omission items ranging between .72 and .92, and cross-loadings onto the action factor falling cleanly below .20.
• Factor 2: Action Regret. Accounts for an additional 18% to 26% of the variance, with primary loadings of the commission items ranging between .68 and .88, and negligible cross-loadings.
• The total cumulative variance explained by the two latent factors routinely exceeds 62% to 72%, satisfying robust structural standards in personality and social measurement.

Confirmatory Factor Analysis (CFA) and Model Fit

To confirm the structural integrity of the ARS, researchers routinely evaluate competing structural models using structural equation modeling software (e.g., AMOS, Mplus, lavaan in R):

• Model 1: Single-Factor Model. Collapsing all anticipated regret items into a single overarching latent construct consistently yields poor fit across literature: $\chi^2/ ext{df} > 6.50$, $ ext{CFI} < .82$,$ ext{TLI} < .78$,$ ext{RMSEA} > .12$, and$ ext{SRMR} > .09$. This empirical failure conclusively refutes the hypothesis that anticipated regret is an undifferentiated affective mass.
• Model 2: Two-Factor Orthogonal Model. Constraining the correlation between Action Regret and Inaction Regret to zero also yields suboptimal fit indices ($ ext{CFI} pprox .88$,$ ext{RMSEA} pprox .095$), demonstrating that the two dimensions are psychologically related rather than wholly independent.
• Model 3: Two-Factor Oblique Model (Target Architecture). Permitting the latent Action and Inaction factors to covary freely yields outstanding model fit indices:

The inter-factor correlation ($phi$) between latent Action Regret and latent Inaction Regret typically ranges between .25 and .48 ($p < .001$), confirming meaningful shared variance (affective counterfactual processing) combined with substantive divergent validity (directionality of agency).

10. Instrument / Measurement Tool

The structural characteristics, administration formats, and scoring mechanics of the standard Anticipated Regret Scale are detailed below:

• Measurement Paradigm: Psychometric self-report inventory designed for behavioral, experimental, consumer, and survey environments.
• Target Respondent: Adolescents and adults navigating real or hypothetically simulated pre-decisional choice scenarios.
• Administration Duration: Approximately 3 to 6 minutes for full completion.
• Dimensional Modalities: Two correlated latent subscales:
  • Action Regret Subscale (Commission)
  • Inaction Regret Subscale (Omission)
• Item Count: Typically structured as 6 to 10 standardized items across literature (most commonly 3 to 5 core items per subscale depending on context-specific adaptation).
• Response Scale Mechanics: 7-point Likert or bipolar semantic differential scales:
  • Likert Anchor: $1 = \text{Strongly Disagree}$ to $7 = \text{Strongly Agree}$.
  • Semantic Differential Anchors: Bipolar pairs such as $1 = \text{No Regret at All}$ to $7 = \text{Extreme Regret}$; $1 = \text{Not at All Worried}$ to $7 = \text{Extremely Worried}$.
• Scoring and Computational Rules:
  • Subscale Scores: Computed by calculating the arithmetic mean or summation of the respective item cluster scores.
  • Action Regret Score: $\text{ARS}_{\text{Action}} = \frac{1}{k_A} \sum_{i=1}^{k_A} A_i$, where $k_A$ is the number of action items.
  • Inaction Regret Score: $\text{ARS}_{\text{Inaction}} = \frac{1}{k_I} \sum_{j=1}^{k_I} I_j$, where $k_I$ is the number of inaction items.
  • Net Regret Differential (Decision Drive): In comparative consumer choice, researchers frequently compute the algebraic difference: $\Delta_{\text{Regret}} = \text{ARS}_{\text{Inaction}} – \text{ARS}_{\text{Action}}$. Positive values indicate a behavioral impulse to act/purchase (FOMO dominant), whereas negative values indicate risk-avoidant status quo maintenance (action-error dominant).

11. Permissions & Fee and Test Year

The structural formulation of anticipated regret stems from academic investigations initiated in the mid-1990s, with seminal publications occurring between 1995 and 1996 (e.g., Richard, van der Pligt, & de Vries, 1996) and ongoing refinements in consumer contexts appearing throughout the 2000s (e.g., Tsiros & Mittal, 2000; Zeelenberg & Pieters, 2007).

Regarding licensing, fees, and operational permissions:

• Academic and Non-Commercial Research: The Anticipated Regret Scale is widely recognized in social psychology and decision research as an open-access scientific framework. Researchers may adapt, translate, and utilize the structural prompts for non-commercial academic research without royalty fees, provided appropriate bibliographic attribution is granted to the foundational authors.
• Commercial and Proprietary Enterprise Use: Organizations incorporating the scale into proprietary consumer analytics platforms, market research engines, or commercial psychographic software must consult copyright holders of specific commercial adaptations or seek appropriate contractual licensing through institutional research offices.

12. References

• Abraham, C., & Sheeran, P. (2003). Acting on intentions: The role of anticipated regret. British Journal of Social Psychology, 42(4), 495–511. https://doi.org/10.1348/014466603322595248
• Abraham, C., & Sheeran, P. (2004). Deciding to exercise: The role of anticipated regret. British Journal of Health Psychology, 9(2), 269–278. https://doi.org/10.1348/135910704773891054
• Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
• Baron, J., & Ritov, I. (1994). Reference points and omission bias. Organizational Behavior and Human Decision Processes, 59(3), 475–498. https://doi.org/10.1006/obhd.1994.1069
• Bell, D. E. (1982). Regret in decision making under uncertainty. Operations Research, 30(5), 961–981. https://doi.org/10.1287/opre.30.5.961
• Connolly, T., & Zeelenberg, M. (2002). Regret in decision making. Current Directions in Psychological Science, 11(6), 212–216. https://doi.org/10.1111/1467-8721.00203
• Kahneman, D., & Miller, D. T. (1986). Norm theory: Comparing reality to its alternatives. Psychological Review, 93(2), 136–153. https://doi.org/10.1037/0033-295X.93.2.136
• Kahneman, D., & Tversky, A. (1982). The simulation heuristic. In D. Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment under Uncertainty: Heuristics and Biases (pp. 201–208). Cambridge University Press. https://doi.org/10.1017/CBO9780511809477.015
• Loomes, G., & Sugden, R. (1982). Regret theory: An alternative theory of rational choice under uncertainty. The Economic Journal, 92(368), 805–824. https://doi.org/10.2307/2232671
• Richard, R., van der Pligt, J., & de Vries, N. (1995). Anticipated affective reactions and health-promoting behaviors. Basic and Applied Social Psychology, 17(1–2), 217–233. https://doi.org/10.1080/01973533.1995.9646138
• Richard, R., van der Pligt, J., & de Vries, N. (1996). Anticipated regret and time perspective: Changing sexual risk-taking behavior. Journal of Behavioral Decision Making, 9(3), 185–199. https://doi.org/10.1086/209571
• Zeelenberg, M., & Pieters, R. (2007). A theory of regret regulation 1.0. Journal of Consumer Psychology, 17(1), 3–18. https://doi.org/10.1207/s15327663jcp1701_3

13. Items of the Scale