Abstract
The Desirable Responding (DR) scale represents a foundational psychometric instrument utilized within psychological, organizational, and consumer research to assess individual tendencies to present oneself in an overly favorable light. In empirical investigation—such as the landmark validation studies conducted by Bearden, Hardesty, and Rose (2001) in the development of the Consumer Self-Confidence (CS-C) construct—controlling for socially desirable responding is paramount to establish discriminant validity and rule out common method variance. Desirable responding is predominantly conceptualized as a multidimensional construct comprising two distinct psychological processes: Self-Deceptive Enhancement (an honest, unconscious tendency to see oneself in a positive light) and Impression Management (deliberate, conscious tailoring of self-presentation to suit external audiences). Utilizing standardized Likert response formats ranging from 5-point to 7-point continuum scales, the instrument allows researchers to identify whether self-reported scores reflect authentic traits or artifacts of response bias. Psychometric evaluations across diverse samples consistently report robust internal consistency reliability, with Cronbach’s alpha coefficients typically ranging between .70 and .86 for subscales and total indices. Confirmatory factor analyses support both bifurcated two-factor structures and hierarchical models, exhibiting adequate goodness-of-fit indices. By decoupling substantive variance from stylistic or bias-driven variance, the Desirable Responding scale serves as a vital methodological tool for safeguarding the construct validity of psychological assessment.
Keywords
Desirable Responding, Social Desirability Bias, Impression Management, Self-Deceptive Enhancement, Response Biases, Construct Validity, Common Method Variance, Psychometrics, Consumer Research, Self-Report Validity
Authors
The formal adaptation and application of the Desirable Responding measures referenced in consumer psychology contexts were established by:
- William O. Bearden — Department of Marketing, Darla Moore School of Business, University of South Carolina, Columbia, SC, USA.
- David M. Hardesty — Department of Marketing, Gatton College of Business and Economics, University of Kentucky, Lexington, KY, USA.
- Randall L. Rose — Department of Marketing, Darla Moore School of Business, University of South Carolina, Columbia, SC, USA.
These researchers drew upon foundational psychometric frameworks established by pioneering theorists in response bias measurement, notably Delroy L. Paulhus (University of British Columbia) and Douglas P. Crowne and David Marlowe.
Purpose
The primary purpose of the Desirable Responding (DR) scale is to identify, quantify, and statistically control for social desirability bias in self-administered psychological and behavioral questionnaires. In social science and behavioral investigations, respondents frequently encounter items that invoke normative standards, moral evaluations, or perceived social competence. Consequently, participants often bias their responses, consciously or unconsciously, toward values deemed culturally acceptable or socially prestigious.
In applied research, such as the psychometric assessment of personal efficacy, self-confidence, leadership styles, or ethical decision-making, high correlations between target constructs and socially desirable responding can undermine construct validity. For instance, when Bearden et al. (2001) operationalized Consumer Self-Confidence, an indispensable phase of the scale validation protocol was proving that the newly formulated multi-dimensional construct measured consumer agency and knowledge rather than mere self-aggrandizement. The DR scale operates as a benchmark to ensure that correlations observed between focal variables are not statistical artifacts created by common method bias or systematic response distortion.
In clinical, educational, and organizational testing, the DR scale functions as an indispensable validity index. It allows practitioners to identify protocols characterized by defensive reporting, malingering in reverse, or lack of candid self-reflection. When employed as a covariate in structural equation modeling (SEM) or multiple regression analysis, the scale partials out bias variance, yielding purified parameter estimates of latent path relationships.
Psychological Construct
The Desirable Responding construct captures systematic stylistic variance in psychological inventories. Rather than representing an ephemeral behavioral anomaly, modern psychometrics conceptualizes desirable responding as an enduring individual difference variable composed of two foundational dimensions:
1. Self-Deceptive Enhancement (SDE)
Self-Deceptive Enhancement represents an unprompted, deeply held propensity to see oneself in an inflated, highly favorable light. The respondent truly believes their over-optimistic self-evaluations. SDE is closely aligned with psychological adjustment, high self-esteem, self-efficacy, and resilience. Individuals high in SDE exhibit cognitive biases such as defensive optimism, positive illusions, and the denial of negative personal attributes. For example, when responding to statements concerning whether they ever experience doubts or personal misgivings, individuals high in SDE will report an absence of negative self-perceptions, genuinely perceiving their own judgment as consistently superior.
2. Impression Management (IM)
Unlike SDE, Impression Management represents a conscious, deliberate tailoring of self-reported behaviors to satisfy social norms and gain interpersonal approval. IM is inherently situational and audience-oriented, reflecting calculated dissimulation or social conformity. Individuals with elevated IM scores intentionally overreport socially laudable behaviors (e.g., altruism, continuous civic responsibility, punctuality) and underreport minor transgressions (e.g., gossiping, mild deception, traffic infractions). IM functions as a psychological defense mechanism aimed at securing social capital and avoiding negative social evaluation.
Theoretical Framework
The conceptual underpinning of the Desirable Responding scale is anchored in the dual-component model of social desirability formulated by Delroy L. Paulhus (1984, 1991), which resolved decades of theoretical divergence between univariate formulations of social desirability. Early paradigms, such as the Marlowe-Crowne Social Desirability Scale (MCSDS; Crowne & Marlowe, 1960), treated the tendency as a single unidimensional need for social approval. However, subsequent factor-analytic explorations revealed that items clustered into distinct internal (cognitive) versus external (interpersonal) domains.
Paulhus synthesized these findings with psychoanalytic and social cognition theories, positing that self-favoring responses operate across two distinct axes:
- Cognitive Defense vs. Interpersonal Calculation: SDE reflects cognitive defense mechanisms (e.g., repressing vulnerabilities, ego defense), while IM reflects conscious impression management as conceptualized by Erving Goffman’s sociological theories of social presentation.
- Agency vs. Communion: Wiggins and Trapnell later mapped these dimensions onto the fundamental axes of interpersonal personality theory: SDE correlates with agency (dominance, ambition, self-assertion), whereas IM corresponds with communion (cooperation, agreeableness, moral conformity).
Within consumer psychology and personality research (e.g., Bearden et al., 2001), integrating this theoretical dichotomy allows researchers to determine whether observed traits correlate with healthy agentic confidence (SDE) or with an extraneous need for interpersonal conformity (IM).
Validity
The construct validity of the Desirable Responding scale has been corroborated through extensive empirical studies encompassing convergent, discriminant, and criterion-related validity assessments:
Convergent Validity
Convergent validity is substantiated by significant, positive correlations between DR subscales and alternative response bias measures. The Impression Management dimension correlates robustly with the Marlowe-Crowne scale ($r = .55$ to $.73$), indicating substantial shared variance in capturing overt social conformity. The Self-Deceptive Enhancement dimension demonstrates moderate-to-high correlations with measures of generalized self-esteem (such as the Rosenberg Self-Esteem Scale, $r = .40$ to $.55$) and generalized self-efficacy ($r = .35$ to $.50$), confirming its alignment with agentic optimism.
Discriminant Validity
Discriminant validity is demonstrated when target psychological constructs remain empirically distinct from desirable responding. In the study by Bearden et al. (2001), the authors evaluated correlations between Consumer Self-Confidence (CS-C) factors and desirable responding. While the dimensions of CS-C exhibited moderate associations with SDE ($r$ ranging between .15 and .32, consistent with the theoretical premise that confident consumers possess authentic self-positivity), they exhibited near-zero correlations with IM ($r$ ranging between -.06 and .08). This confirmed that consumer self-confidence is not a product of deliberate social dissimulation, validating the discriminant independence of the constructs.
Predictive and Criterion-Related Validity
In experimental settings with varying demand characteristics (e.g., anonymous self-administration vs. public, high-stakes evaluative conditions), IM scores increase significantly under high-demand public scrutiny, whereas SDE scores remain relatively invariant across contextual manipulations. This pattern supports the functional distinction between deliberate situational posturing and enduring cognitive self-belief.
Reliability
The Desirable Responding instrument exhibits robust reliability estimates across multiple independent investigations and demographic cohorts:
- Internal Consistency: Cronbach’s alpha coefficients for the full composite scale typically range from $\alpha = .78$ to $.88$. When analyzed by subscale, Impression Management routinely yields alpha values between $.75$ and $.86$, reflecting high item homogeneity. The Self-Deceptive Enhancement subscale displays alpha values between $.70$ and $.82$. In applied consumer samples, such as those evaluated by Bearden et al. (2001), the shortened adaptations sustained alpha reliabilities exceeding the conventional psychometric threshold of $.70$.
- Test-Retest Reliability: Longitudinal evaluations over intervals of three to five weeks demonstrate stability coefficients exceeding $r_{tt} = .75$ for Impression Management and $r_{tt} = .80$ for Self-Deceptive Enhancement, proving that the measure captures enduring response tendencies rather than transient mood states.
- Composite Reliability: In structural equation modeling frameworks, composite reliability ($CR$) metrics consistently surpass $.80$, accompanied by Average Variance Extracted ($AVE$) coefficients generally exceeding $.45$ to $.50$.
Factor Analysis
Extensive exploratory (EFA) and confirmatory factor analyses (CFA) have confirmed the multi-dimensional structure of the Desirable Responding inventory:
Exploratory Factor Analysis
Early factor analyses using principal axis factoring with oblique (e.g., Promax) rotation consistently yield a distinct two-factor solution. Items designed to capture overt denial of everyday interpersonal faults reliably load onto an Impression Management factor (factor loadings ranging from .45 to .78), while items capturing an unwavering conviction in one’s own competence and psychological stability load onto a Self-Deceptive Enhancement factor (factor loadings ranging from .40 to .72). Cross-loadings between the two factors are generally low (rarely exceeding .25), confirming dimensional orthogonality or mild correlation ($r = .15$ to $.30$).
Confirmatory Factor Analysis
Confirmatory factor analytic assessments test whether a two-factor correlated model outperforms a single-factor unidimensional model. Empirical model comparisons demonstrate superior fit for the two-factor specification:
- Comparative Fit Index (CFI): Typically exceeds $.92$ to $.95$ for the two-factor model, compared to $< .75$ for a unidimensional model.
- Tucker-Lewis Index (TLI): Consistently maintains values above $.90$.
- Root Mean Square Error of Approximation (RMSEA): Generally yields values between $.042$ and $.065$ ($90%\text{ CI } [.035, .072]$), indicating close fit to the empirical data.
- Standardized Root Mean Residual (SRMR): Remains below the $.06$ threshold.
These findings substantiate that treating desirable responding as a single undifferentiated construct obscures meaningful psychometric variance.
Instrument / Measurement Tool
The operational implementation of the Desirable Responding instrument in behavioral and consumer research utilizes the following standard parameters:
- Test Type: Standardized self-report psychometric inventory / response-bias validity scale.
- Format: Paper-and-pencil or computer-assisted self-administered questionnaire.
- Item Count: Standard full inventories comprise 40 items (20 items for SDE, 20 items for IM); shortened research adaptations (such as those employed in Bearden et al., 2001, or derived from Steenkamp, de Jong, & Baumgartner, 2010) typically feature between 10 and 20 targeted items.
- Response Scale: Typically administered using a 7-point Likert scale (ranging from 1 = Strongly Disagree to 7 = Strongly Agree), though 5-point formats are also frequently utilized.
- Scoring Protocols: Two primary scoring methodologies exist:
- Dichotomous (Extreme) Scoring: One point is awarded only for extreme responses (e.g., 6 or 7 on a 7-point scale after reversing negatively worded items). This method emphasizes overt, exaggerated self-favoring.
- Continuous Scoring: Responses across items are averaged or summed after reverse-scoring negatively keyed items. This method preserves variance and is preferred in structural equation modeling and correlational research.
- Administration Time: Approximately 5 to 10 minutes for full-length forms; 2 to 4 minutes for brief forms.
Permissions & Fee and Test Year
The foundational operationalizations of the Balanced Inventory of Desirable Responding were established by Delroy L. Paulhus (1984, 1991). The specific methodological adaptation and application to consumer self-confidence research were documented in 2001 by William O. Bearden, David M. Hardesty, and Randall L. Rose.
For non-commercial academic research and scholarly assessment, shortened desirable responding scales and public-domain adaptations are widely accessible for empirical use without licensing fees, provided the source authors are appropriately acknowledged and cited. Commercial applications or specialized diagnostic implementations of proprietary testing batteries may require formal authorization from the original copyright holders or institutional publishers.
References
- Bearden, W. O., Hardesty, D. M., & Rose, R. L. (2001). Consumer self-confidence: Refinements in conceptualization and measurement. Journal of Consumer Research, 28(1), 121–134. https://doi.org/10.1086/321948
- 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
- Paulhus, D. L. (1984). Two-component models of socially desirable responding. Journal of Personality and Social Psychology, 46(3), 598–609. https://doi.org/10.1037/0022-3514.46.3.598
- Paulhus, D. L. (1991). Measurement and control of response bias. In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of personality and social psychological attitudes (pp. 17–59). Academic Press. https://doi.org/10.1016/B978-0-12-590241-0.50006-X
- Steenkamp, J.-B. E., de Jong, M. G., & Baumgartner, H. (2010). Socially desirable response tendencies in survey research. Journal of Marketing Research, 47(2), 199–214. https://doi.org/10.1509/jmkr.47.2.199