Wong’s Emotional Intelligence Scale (WEIS)

Abstract

Wong’s Emotional Intelligence Scale (WEIS) is a specialized, forced-choice psychometric instrument developed by psychometricians and organizational behavior scholars Chi-Sum Wong, Kenneth S. Law, and Philip M. Wong (2004, 2007). Distinct from its widely recognized Likert-type sister instrument—the Wong and Law Emotional Intelligence Scale (WLEIS; Wong & Law, 2002)—the WEIS was engineered specifically to attenuate the pervasive measurement artifacts of social desirability bias, impression management, and response distortion in high-stakes organizational environments, such as personnel selection, recruitment, and executive promotion. Rooted firmly in the conceptual four-branch ability model of emotional intelligence formulated by Mayer and Salovey (1997), the WEIS operationalizes emotional intelligence across four distinct dimensions: Self-Emotion Appraisal (SEA), Others’ Emotion Appraisal (OEA), Use of Emotion to Facilitate Performance (UOE), and Regulation of Emotion in the Self (ROE).

The WEIS comprises an overall battery of 40 forced-choice items partitioned into two balanced twenty-item sections: Part I: Reactions to Various Scenarios, a situational judgment test (SJT) format presenting 20 workplace and interpersonal dilemmas with dichotomous behavioral alternatives; and Part II: Relative Strength of Abilities, a paired-comparison paradigm comprising 20 forced-choice dyads wherein participants evaluate their relative competence by contrasting an emotional intelligence capability against a matched non-emotional, cognitive, artistic, or physical ability. Psychometric investigations conducted across Hong Kong and Mainland Chinese organizational cohorts demonstrate robust criterion-related validity, predicting supervisory ratings of job performance, emotional labor efficacy, and organizational citizenship behaviors beyond the Big Five personality dimensions and general mental ability (GMA). Despite the structural ipsativity inherent to forced-choice formats, advanced Thurstonian Item Response Theory (IRT) modeling and classical test-retest assessments affirm the instrument’s structural stability, score consistency, and cross-cultural utility in Asian and global enterprise contexts.

Keywords

Wong’s Emotional Intelligence Scale, WEIS, forced-choice measurement, emotional intelligence, situational judgment test, social desirability bias, Self-Emotion Appraisal, Regulation of Emotion, personnel selection, ipsative measurement, Thurstonian IRT, organizational behavior

Authors

The WEIS was conceptualized, constructed, and empirically validated by a distinguished team of management and organizational psychology researchers based in Hong Kong:

• Chi-Sum Wong, Ph.D.: Professor Emeritus, Department of Management, Faculty of Business Administration, The Chinese University of Hong Kong (CUHK), Shatin, New Territories, Hong Kong. Dr. Wong is an internationally recognized scholar in organizational psychology, human resource management, job design, emotional labor, and psychometrics in the Asia-Pacific region. Email: [email protected].
• Kenneth S. Law, Ph.D.: Professor, Department of Management, School of Business and Management, Hong Kong University of Science and Technology (HKUST), Clear Water Bay, Kowloon, Hong Kong. Dr. Law is an expert in structural equation modeling, research methodology, leadership dynamics, and cross-cultural organizational behavior. Email: [email protected].
• Philip M. Wong, Ph.D.: Scholar and educational leadership researcher affiliated with the Hong Kong Institute of Education (now The Education University of Hong Kong), specializing in school administration, teacher emotional labor, and educational organizational dynamics.

Purpose

Self-report assessments of emotional intelligence—most notably Likert-type scales such as the TEIQue, Bar-On EQ-i, or the authors’ own WLEIS—are widely utilized in exploratory research due to their economy of administration, straightforward scoring, and strong face validity. However, in high-stakes evaluative settings—including employee recruitment, pre-employment screening, high-potential leadership identification, and performance appraisal—standard Likert items suffer catastrophic vulnerability to intentional faking, halo effects, and positive impression management. When job applicants encounter transparent Likert prompts such as “I have good control of my own emotions” or “I always set goals for myself and then try my best to achieve them,” respondents routinely endorse extreme positive response anchors regardless of their actual behavioral repertoire. Consequently, observed test scores exhibit severe negative skewness, truncated variance, and degraded criterion-related validity.

To overcome this psychometric hurdle, Wong, Law, and Wong (2004) engineered the WEIS as a robust forced-choice operationalization of Mayer and Salovey’s four-dimensional model. The principal objective was to construct a tool specifically tailored for Chinese and international organizational contexts where candid self-reporting is constrained either by competitive career motives or by cultural pressures regarding social modesty and face-saving (mianzi). The authors recognized that when forced to select between two alternative behavioral paths in a realistically depicted situational dilemma (Part I), or between an emotional capability and an equally desirable or neutral cognitive/physical skill (Part II), respondents are substantially less able to decipher the socially optimal response or successfully portray an idealized persona.

In applied organizational environments, the WEIS serves as an objective diagnostic tool for talent acquisition, leadership coaching, and organizational development. It allows human resource professionals to identify candidates possessing genuine emotional competencies required for service-oriented, high-stress, or customer-facing roles without relying on inflated self-descriptions. In research settings, the WEIS provides investigators with a methodology to test hypothesized relationships between emotional intelligence, emotional labor strategies (e.g., surface acting versus deep acting), burnout, job performance, and leadership efficacy while statistically insulating structural path coefficients from common method variance (CMV) and social desirability bias.

Psychological Construct

The WEIS operationalizes emotional intelligence not as a broad constellation of personality traits or self-esteem markers, but strictly as a set of interrelated psychological abilities encompassing the perception, assimilation, understanding, and regulation of emotion. Following the structural topology formulated by Mayer and Salovey (1997) and adapted by Wong and Law (2002), the WEIS assesses four distinct sub-constructs:

1. Self-Emotion Appraisal (SEA)

Self-Emotion Appraisal refers to an individual’s psychological capacity to accurately perceive, decode, acknowledge, and express their own internal affective states. Individuals with elevated SEA possess high emotional self-awareness, enabling them to comprehend the subtle psychological and physiological antecedents that trigger subjective feelings of anger, despondency, anxiety, or elation. Rather than experiencing diffused, undifferentiated mood states, individuals high in SEA differentiate nuanced emotional experiences and trace them back to their root cognitive appraisals. In the WEIS, SEA is assessed both through contextualized problem-solving scenarios (e.g., acknowledging one’s feelings when overwhelmed by workplace adversity) and paired comparisons requiring respondents to judge whether their capacity to comprehend their own emotional shifts is more pronounced than non-affective functional abilities, such as mastering technical machinery or physical endurance.

2. Others’ Emotion Appraisal (OEA)

Others’ Emotion Appraisal captures an individual’s empathy and interpersonal perceptual sensitivity—specifically, the facility to perceive, decode, and accurately interpret the verbal and non-verbal affective cues emitted by other people. This dimension reflects an acute sensitivity to subtle micro-expressions, posture, vocal prosody, and situational stressors affecting colleagues, clients, or subordinates. High OEA enables an employee or leader to read between the lines, recognize latent discontent or unspoken anxiety, and anticipate the behavioral reactions of others. Within the WEIS, OEA is measured through scenarios depicting distressed friends, grieving employees, or conflicted team members, evaluating whether the respondent proactively tunes into the underlying emotional reality versus remaining detached or focusing solely on task execution.

3. Use of Emotion to Facilitate Performance (UOE)

Use of Emotion denotes the capacity to consciously direct, channel, and harness one’s emotional states toward constructive, goal-directed activities, cognitive problem solving, and intrinsic personal motivation. Grounded in the concept of affective self-motivation, high-UOE individuals utilize positive affect to foster creative ideation, cultivate resilience in the face of setbacks, and sustain relentless focus on long-term objectives. Even in the presence of tedious, low-interest, or unrewarding tasks, an individual strong in UOE reframes the situation to locate internal motivation, maintaining optimism and agency. In the WEIS, UOE items compel the respondent to choose whether they actively inspire themselves to persist through professional failure and adverse assignments, or whether they succumb to frustration, procrastination, or passivity.

4. Regulation of Emotion in the Self (ROE)

Regulation of Emotion in the Self represents the executive psychological ability to modulate, manage, and recover from debilitating negative affective states, such as rage, panic, distress, and acute frustration. Operating in alignment with emotion regulation theory (Gross, 1998), ROE does not imply the emotional suppression or denial of feelings; rather, it reflects cognitive reappraisal strategies, effective distress tolerance, and the prompt restoration of homeostatic psychological functioning. High-ROE individuals preserve calm composure during high-conflict confrontations, withstand severe workplace provocations without losing temper, and rapidly rebound from traumatic professional shocks. In the WEIS, ROE is evaluated through demanding scenarios (e.g., navigating unprovoked verbal hostility from a partner or political infighting among corporate managers) and comparative ability items assessing emotional equanimity relative to cognitive or mechanical tasks.

Theoretical Framework

The architecture of Wong’s Emotional Intelligence Scale is underpinned by three convergent psychological paradigms: the Mayer-Salovey ability model of emotional intelligence, classical and modern forced-choice psychometric measurement theory, and the social-cognitive theory of emotional labor and impression management.

The Mayer-Salovey Ability Tradition

In their seminal 1997 treatise, Mayer and Salovey dismantled earlier expansive definitions of emotional intelligence that conflated emotional competence with personality variables like extraversion, persistence, and warmth. They redefined emotional intelligence as an intelligence operating over emotional information, structured hierarchically across four mental processes: (a) perception of emotion, (b) emotional facilitation of thought, (c) understanding of emotion, and (d) reflective regulation of emotion. Wong and Law (2002) and Wong et al. (2004) adopted this precise functional taxonomy, arguing that if emotional intelligence is to maintain conceptual legitimacy alongside traditional constructs like general cognitive ability (g), it must be defined strictly as an ability-centered construct rather than an amorphous personality composite.

Forced-Choice Methodology and Thurstonian Psychometrics

The standard Likert measurement approach assumes that respondents can locate their true score on an absolute continuous continuum and that their responses are linear functions of their latent trait level. However, as demonstrated by psychometrician Louis Leon Thurstone in his Law of Comparative Judgment (1927), human judgment is comparative. When respondents are presented with forced-choice pairs, their cognitive process shifts from an absolute rating to an ordinal comparison between two internal evaluations:

y_ij = 1  if  t_i > t_j;  y_ij = 0  if  t_i ≤ t_j

where t_i represents the perceived strength or utility of option i and t_j represents option j. Historically, forced-choice tests generated purely ipsative data, where an individual’s score on one dimension could only increase at the mathematical expense of another dimension, yielding constant individual sum totals, non-positive definite covariance matrices, and distorted factor structures. Wong, Law, and Wong (2004) solved this dilemma in two ways: first, by incorporating a situational judgment paradigm in Part I that evaluates choices against normative emotional effectiveness; and second, in Part II, by pairing emotional abilities against disparate non-emotional domains (e.g., singing, mechanical aptitude, athletic endurance, mental arithmetic). By contrasting an emotional ability with an unrelated general ability, the respondent’s choices do not create direct interdependencies among the four emotional intelligence dimensions themselves, thereby mitigating classic ipsativity artifacts while retaining the anti-faking benefits of forced-choice testing.

Impression Management and Cultural Context in Chinese Organizations

The development of the WEIS was deeply informed by the social psychology of Chinese workplace behavior. In Confucian-heritage cultures, social harmony, relationship management (guanxi), and preservation of collective face exert immense pressure on individuals to present themselves as emotionally composed and interpersonally accommodating. Consequently, Likert scales administered in Hong Kong, Mainland China, or Taiwan frequently suffer from profound ceiling effects and uniform acquiescence. Wong et al. (2004, 2007) constructed situational vignettes grounded in authentic contemporary Hong Kong workplace scenarios—such as dealing with public transit delays on the MTR, managing difficult cross-border relationships with foreign investors, resolving inter-departmental managerial hostility, and balancing filial responsibilities with corporate deadlines. This ecological realism ensures that the scale taps into genuine behavioral dispositions rather than abstract, socially conditioned self-conceptions.

Validity

Empirical validation of the WEIS has been established across multiple independent studies involving diverse occupational cohorts, including managerial personnel, educational administrators, frontline service workers, and university students in Hong Kong and Mainland China.

Construct and Structural Validity

Wong, Law, and Wong (2004) demonstrated that the four-factor structural model of the WEIS exhibited acceptable fit indices when modeled via structural equation techniques adapted for binary and forced-choice data. Confirmatory factor analysis supported the theoretical separation of the four dimensions (SEA, OEA, UOE, ROE), demonstrating that they represent distinct yet moderately correlated facets of an overarching second-order emotional intelligence construct. Correlation analyses confirmed that the four dimensions correlate moderately with one another (typically ranging between r = .30 and r = .55), corroborating the theoretical premise that these dimensions represent interrelated facets of a cohesive ability domain.

Convergent and Discriminant Validity

Convergent validity was substantiated by examining the associations between WEIS scores and established emotional intelligence inventories. WEIS scores correlate significantly with the Likert-based WLEIS (with correlations ranging between r = .40 and r = .62 across subscales), demonstrating shared conceptual variance while reflecting distinct method variance. Importantly, the WEIS demonstrates superior discriminant validity against social desirability measures. In comparative testing utilizing the Marlowe-Crowne Social Desirability Scale (MCSDS; Crowne & Marlowe, 1960), the WLEIS exhibited statistically significant positive correlations with social desirability (r = .28 to .38, p < .001), indicating systematic distortion. In stark contrast, the WEIS scores demonstrated trivial, statistically non-significant correlations with social desirability (r = .04 to .11, p > .05), empirically confirming that the forced-choice design successfully neutralizes positive impression management.

Furthermore, when evaluated alongside the Big Five personality traits (Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness), the WEIS shared only modest variance (multiple R² typically below .25), affirming that it does not merely replicate established personality traits. Its discriminant validity against general mental ability (GMA) was also confirmed: while displaying weak to moderate positive associations with cognitive ability (r ≈ .15 to .22), indicating a common cognitive substrate, the WEIS did not collapse into standard intelligence measures.

Criterion-Related and Predictive Validity

The practical utility of the WEIS was rigorously demonstrated by Wong, Wong, and Law (2007) in a series of field studies examining supervisory ratings of job performance, emotional labor, and employee well-being:

• Job Performance: In samples of middle-level managers and customer service personnel, WEIS dimensions—particularly UOE and ROE—demonstrated statistically significant incremental validity in predicting supervisory ratings of task performance and organizational citizenship behaviors (OCB), after controlling for cognitive intelligence and Big Five personality traits (ΔR² = .04 to .07, p < .01).
• Emotional Labor and Burnout: Consistent with Grandey’s (2000) emotional labor model, high scores on the WEIS predicted lower surface acting (faking felt emotions) and higher deep acting (genuinely modifying internal feelings to match organizational display rules), which in turn significantly buffered employees against job burnout, emotional exhaustion, and turnover intentions.
• Moderating Role in High-Stress Roles: Wong et al. (2007) observed that the predictive validity of the WEIS was moderated by the emotional demands of the job (tested using Holland’s occupational classification). In occupations requiring intensive interpersonal interaction (e.g., social workers, teachers, service personnel), the correlation between WEIS scores and job performance was substantially higher (r = .34 to .42) than in technical or solitary occupations, confirming the ecological validity of the construct.

Reliability

Because the WEIS employs a dichotomous, forced-choice response format, classical reliability estimation techniques must be interpreted with methodological rigor. When items are ipsative or semi-ipsative, traditional raw-score Cronbach’s alpha coefficients can be attenuated or mathematically perturbed due to constrained item variances and covariances.

Internal Consistency

In the primary validation cohorts reported by Wong, Law, and Wong (2004) and Wong, Wong, and Law (2007), internal consistency estimates (calculated via Kuder-Richardson Formula 20 [KR-20] for dichotomous items and standardized Cronbach’s alpha) yielded acceptable psychometric coefficients across both sections:

• Part I (Situational Judgment Scenarios): The overall scale reliability typically ranges between α = .72 and α = .81 across mixed organizational samples. Subscale reliabilities for the 5-item situational dimensions range from .65 to .74, which is robust given the contextual diversity and brevity of situational judgment subscales.
• Part II (Relative Strength of Abilities): The internal consistency coefficients for the 20 paired items hover consistently between α = .76 and α = .84. Composite reliabilities for individual dimensions (SEA, OEA, UOE, ROE) range from .68 to .78.
• Full 40-Item Composite Battery: When evaluated as an overall emotional intelligence index, composite reliability regularly exceeds .85 across studies, indicating strong measurement precision for group-level research and individual diagnostic screening.

Test-Retest Stability

To confirm that the WEIS captures stable ability-like dispositions rather than transient emotional fluctuations or state-dependent mood swings, Wong et al. (2004) conducted test-retest reliability analyses over a four-week interval with a cohort of working professionals. The stability coefficients were exceptionally high: r_tt = .82 for the full instrument, with individual subscales demonstrating temporal correlations ranging from .74 (OEA) to .85 (ROE). These findings confirm that forced-choice behavioral decisions and ability self-evaluations on the WEIS reflect enduring psychological traits rather than situational noise.

Factor Analysis

The factorial structure of the WEIS has been subject to extensive exploratory (EFA) and confirmatory factor analysis (CFA), with particular attention dedicated to managing dichotomous and comparative data structures.

Confirmatory Factor Analytic Structure

Wong, Law, and Wong (2004) tested competing structural models to determine whether the 40 forced-choice items conformed to a unidimensional model, a four-factor orthogonal model, a four-factor oblique model, or a second-order hierarchical model. Using asymptotic distribution-free (ADF) and robust weighted least squares (WLSMV) estimation specifically designed for categorical indicators, the four-factor oblique model and the second-order hierarchical model demonstrated superior fit over alternative configurations.

Model Specification	χ2 / df	CFI	TLI	RMSEA	SRMR
1. One-Factor General Model	4.82	.68	.65	.098	.105
2. Four-Factor Orthogonal Model	3.45	.79	.77	.079	.089
3. Four-Factor Oblique Model	1.86	.93	.92	.044	.051
4. Second-Order Hierarchical Model	1.92	.92	.91	.046	.053

In the second-order model, the second-order global emotional intelligence factor loaded substantially onto all four primary dimensions: SEA (γ = .71), OEA (γ = .64), UOE (γ = .78), and ROE (γ = .82). Individual standardized factor loadings for the items onto their designated latent constructs were all statistically significant (p < .001), generally ranging from .42 to .76, confirming that the forced-choice format maintains solid construct representation.

Item Response Modeling (Thurstonian IRT)

In contemporary psychometrics, forced-choice questionnaires are increasingly evaluated using Thurstonian IRT models (Brown & Maydeu-Olivares, 2011). When applied to the WEIS, Thurstonian IRT modeling reconstructs normative trait scores (θ) from binary comparative choices, eliminating the mathematical hazards of ipsativity. These analyses indicate that items in Part I exhibit high discrimination parameters across moderate difficulty levels, indicating strong diagnostic sensitivity in distinguishing average from highly emotionally intelligent employees. In Part II, the pairing of emotional traits with cognitive/physical counter-weights successfully anchors the latent trait continuum, allowing unbiased person-parameter estimation.

Instrument / Measurement Tool

• Instrument Name: Wong’s Emotional Intelligence Scale (WEIS)
• Alternative / Related Instruments: Wong and Law Emotional Intelligence Scale (WLEIS; 16-item Likert version)
• Authors: Chi-Sum Wong, Kenneth S. Law, and Philip M. Wong
• Primary Target Population: Adults, job applicants, corporate executives, managers, frontline service personnel, and university students
• Administration Format: Paper-and-pencil or interactive digital computer-based questionnaire; self-administered
• Total Item Count: 40 forced-choice items, divided into two distinct sections:
• Part I: Reactions to Various Scenarios — 20 situational judgment scenarios with dichotomous behavioral response alternatives (Option A vs. Option B).
• Part II: Relative Strength of Abilities — 20 forced-choice paired ability comparisons (Option a vs. Option b).
• Subscale Breakdown: Items across both parts correspond to the four Mayer-Salovey dimensions:
• Self-Emotion Appraisal (SEA): 10 items total (5 in Part I, 5 in Part II)
• Others’ Emotion Appraisal (OEA): 10 items total (5 in Part I, 5 in Part II)
• Use of Emotion (UOE): 10 items total (5 in Part I, 5 in Part II)
• Regulation of Emotion (ROE): 10 items total (5 in Part I, 5 in Part II)
• Completion Time: Approximately 20 to 30 minutes
• Response Format: Forced choice (dichotomous selection: A or B in Part I; a or b in Part II)
• Scoring Methodology:
• Classical Additive Scoring: Responses reflecting the emotionally intelligent alternative (predetermined through expert consensus and normative validation) receive a score of 1, whereas counter-productive, passive, or non-emotional alternative choices receive a score of 0. Dimension scores are computed by summing the endorsed keys (range 0 to 10 per dimension), yielding a total composite score ranging from 0 to 40.
• Thurstonian IRT Scoring: For high-stakes personnel selection, response matrices are modeled via Thurstonian IRT algorithms to generate normative latent trait estimates (θ) for each dimension, entirely free of ipsative constraints.
• Primary Application Areas: High-stakes personnel selection, leadership recruitment, talent management, emotional labor diagnostics, and organizational behavior research.

Permissions & Fee and Test Year

Wong’s Emotional Intelligence Scale was developed and validated in primary publications appearing between 2004 and 2007 (Wong, Law, & Wong, 2004; Wong, Wong, & Law, 2007). The authors established the WEIS as an open-access psychometric instrument for academic research and educational purposes. Researchers and non-profit educational investigators are permitted to administer the WEIS in academic empirical investigations without the payment of licensing fees, provided that appropriate scholarly attribution is cited in all resulting reports, dissertations, and publications.

For commercial usage—such as proprietary pre-employment screening systems, commercial executive coaching programs, or corporate organizational development platforms—inquiries regarding formal licensing agreements, scoring keys, and normative benchmarking datasets must be addressed directly to the primary developers, Dr. Chi-Sum Wong ([email protected]) and Dr. Kenneth S. Law ([email protected]). The full academic working manuscript describing the validation and administration protocols of the WEIS is cataloged in the institutional research repository of Shaoxing University (http://jgxy.usx.edu.cn/DAOM/084_Chi-SumWong.pdf).

References

• Brown, A., & Maydeu-Olivares, A. (2011). Item response modeling of forced-choice questionnaires. Educational and Psychological Measurement, 71(3), 460–502. https://doi.org/10.1177/0013164410387340
• Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24(4), 349–354. https://doi.org/10.1037/h0047358
• Grandey, A. A. (2000). Emotion regulation in the workplace: A new way to conceptualize emotional labor. Journal of Occupational Health Psychology, 5(1), 95–110. https://doi.org/10.1037/1076-8998.5.1.95
• Gross, J. J. (1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology, 2(3), 271–299. https://doi.org/10.1037/1089-2680.2.3.271
• Mayer, J. D., & Salovey, P. (1997). What is emotional intelligence? In P. Salovey & D. J. Sluyter (Eds.), Emotional development and emotional intelligence: Educational implications (pp. 3–34). Basic Books.
• Thurstone, L. L. (1927). A law of comparative judgment. Psychological Review, 34(4), 273–286. https://doi.org/10.1037/h0070288
• Wong, C. S., & Law, K. S. (2002). The effects of leader and follower emotional intelligence on performance and attitude: An exploratory study. The Leadership Quarterly, 13(3), 243–274. https://doi.org/10.1016/S1048-9843(02)00099-1
• Wong, C. S., Law, K. S., & Wong, P. M. (2004). Development and validation of a forced choice emotional intelligence for Chinese respondents in Hong Kong. Asia Pacific Journal of Management, 21(4), 535–559. https://doi.org/10.1023/B:APJM.0000048716.89205.e0
• Wong, C. S., Wong, P. M., & Law, K. S. (2004). The interaction effect of emotional intelligence and emotional labor on job satisfaction: A test of Holland’s classification of occupations. In C. E. J. Härtel, W. J. Zerbe, & N. M. Ashkanasy (Eds.), Emotions in organizational behavior (pp. 235–250). Lawrence Erlbaum Associates.
• Wong, C. S., Wong, P. M., & Law, K. S. (2007). Evidence on the practical utility of Wong’s emotional intelligence scale in Hong Kong and Mainland China. Asia Pacific Journal of Management, 24(1), 43–60. https://doi.org/10.1007/s10490-006-9024-1

Items of the Scale