Attitude Toward the Ad (Overall) (ATTA)

Abstract

The Attitude Toward the Ad (Overall) (ATTA) scale, operationalized within advertising psychology and consumer research by Kim, Haley, and Koo (2009), evaluates a consumer’s global affective and evaluative orientation toward an advertising stimulus. Grounded in foundational models of persuasion, such as the Dual Mediation Hypothesis (DMH) and the Elaboration Likelihood Model (ELM), the instrument assesses an individual’s overall favorable or unfavorable reaction to commercial messaging across both corporate advocacy and product-specific advertising domains. The scale captures a unidimensional global evaluative construct encompassing cognitive appraisal, affective valence, and general receptivity. Typically administered using a multi-item Likert or semantic differential response continuum (commonly 7-point scales anchored by polar evaluative adjectives such as unfavorable/favorable, bad/good, and unappealing/appealing), the instrument exhibits exceptional psychometric properties across diverse empirical studies. In the structural equation modeling and path analytical frameworks established by Kim et al. (2009), the scale demonstrated high internal consistency reliability, yielding Cronbach’s alpha ($lpha$) coefficients consistently exceeding .85 to .92 across distinct involvement conditions (cognitive vs. affective involvement). Confirmatory factor analyses reveal clean unidimensionality, marked factor loadings (frequently ranging from .78 to .91), and robust convergent validity indicated by average variance extracted (AVE) estimates surpassing .60. The scale demonstrates substantial predictive validity in forecasting downstream consumer metrics, including attitude toward the brand ($A_b$), corporate credibility, and behavioral purchase or advocacy intentions. By systematically parsing the structural linkages between pre-existing consumer involvement modalities and transient communicative executions, the ATTA serves as an essential psychometric assessment tool for researchers in marketing communication, consumer behavior, media psychology, and public relations.

Keywords

Attitude Toward the Ad, Advertising Psychology, Consumer Involvement, Dual Mediation Hypothesis, Persuasion Research, Corporate Advertising, Product Advertising, Psychometric Evaluation, Brand Attitude, Evaluative Conditioning, Structural Equation Modeling, Media Effects

Authors

The primary authors responsible for the comparative operationalization and psychometric validation of this operational measurement model within corporate versus product advertising paradigms are:

• Sora Kim, Ph.D. — Associate Professor / Professor of Public Relations and Strategic Communication, School of Journalism and Communication, The Chinese University of Hong Kong (formerly at the University of Florida). Specialist in corporate social responsibility (CSR) communication, advertising persuasion, and corporate reputation management.
• Eric Haley, Ph.D. — DeForrest Jackson Professor of Advertising, School of Advertising and Public Relations, College of Communication and Information, University of Tennessee, Knoxville. Renowned scholar in advertising advocacy, consumer responses to organizational messaging, and qualitative/quantitative research methods in advertising.
• Gi-Yong Koo, Ph.D. — Associate Professor of Sport Management / Sport Marketing, Department of Sport Management, Troy University. Specialist in consumer behavior, corporate sponsorship effectiveness, and structural equation modeling applications in commercial media.

Purpose

The foundational purpose of the Attitude Toward the Ad (Overall) (ATTA) scale is to provide a reliable, empirically robust, and theoretically grounded psychometric instrument to assess a receiver’s global evaluative appraisal of an advertising execution. In contemporary media ecosystems, advertising messages fulfill multifaceted communication objectives, ranging from direct product feature promotion to strategic corporate brand identity enhancement. The ATTA was specifically refined by Kim, Haley, and Koo (2009) to resolve theoretical ambiguities regarding how consumers process corporate advertising as opposed to standard product advertising under varying configurations of cognitive and affective involvement.

From an applied research perspective, the scale functions as an indispensable diagnostic and evaluative tool across several core domains:

• Comparative Media Testing: Permitting marketing researchers to compare consumer responses across experimental manipulations of advertising copy, visual aesthetics, creative narratives, and message framing strategies.
• Cross-Contextual Message Appraisal: Determining whether institutional and corporate reputation campaigns induce identical evaluative pathways as transactional product-level campaigns, specifically assessing if the ad itself elicits positive sentiment independently of pre-existing brand loyalty.
• Structural Mediational Modeling: Supplying a clean, unidimensional mediator variable that sits between antecedent processing states (such as issue involvement, product category involvement, or cognitive elaboration) and consequential brand-level outcomes (such as brand attitude, brand trust, and purchase intent).
• Diagnostic Pre-Testing in Applied Campaigns: Equipping commercial campaign strategists with a validated instrument to detect adverse consumer reactions, message fatigue, or creative dissonance prior to capital-intensive mass media distribution.

Theoretically, the scale addresses the long-standing imperative to isolate the receiver’s evaluation of the communicative vehicle (the advertisement itself) from their evaluation of the sponsoring entity or product. By isolating the ad execution as a discrete psychological object, the scale facilitates granular tests of attitude formation mechanisms, clarifying when an individual’s evaluation of an ad directly transfers affective valence to the underlying brand versus when it acts indirectly through the facilitation of cognitive brand attribute processing.

Psychological Construct

The construct assessed by the ATTA scale is operationalized as a predisposed, affective, and cognitive reaction to a commercial or advocacy appeal. In psychometric literature, Attitude Toward the Ad ($A_{ad}$) is universally defined as a psychological tendency that is expressed by evaluating a particular advertising stimulus with some degree of favor or disfavor (MacKenzie, Lutz, & Belch, 1986). Within the structural framework validated by Kim, Haley, and Koo (2009), this construct is conceived as an overarching, global evaluative disposition comprised of tightly integrated cognitive and affective appraisal facets.

Cognitive Appraisal Dimension

The cognitive facet relates to the consumer’s rational or analytical judgment regarding the ad’s communicative competence, informative utility, and message credibility. When processing an advertisement, respondents formulate judgments regarding whether the advertisement is informative, persuasive, convincing, and well-executed. For example, a consumer viewing a corporate sustainability campaign engages cognitive appraisals regarding whether the presented arguments are coherent and logically organized. While distinct from deep cognitive responses (counterarguing or support arguing), this evaluative facet represents the outcome of cognitive processing applied to the ad’s structural presentation.

Affective Valence Dimension

The affective facet encapsulates the immediate emotional and mood-altering responses generated by the aesthetic, narrative, or emotional cues contained within the commercial. This includes subjective feelings of pleasantness, enjoyment, irritation, or emotional warmth elicited by music, visual design, humor, or human-interest elements. In the ATTA, affective valence does not measure discrete emotions (e.g., discrete joy or anger) but rather the net hedonic tone of the execution, determining whether the overall viewing experience is evaluated as favorable, enjoyable, or aesthetically pleasing.

Global Evaluative Synthesis

Crucially, Kim et al. (2009) model ATTA as a unidimensional global evaluative synthesis where cognitive and affective vectors converge into a unified psychological orientation. Psychometrically, treating the overall attitude as a consolidated construct avoids structural collinearity in structural equation modeling while providing maximum predictive power for downstream constructs. The construct reflects an immediate, post-exposure appraisal of the stimulus as a self-contained communicative event, capturing the respondent’s holistic evaluation along an evaluative continuum ranging from wholly negative to wholly positive.

Theoretical Framework

The conceptual foundation of the ATTA scale rests primarily on two towering theoretical traditions in communication and consumer psychology: the Elaboration Likelihood Model (ELM) of persuasion developed by Petty and Cacioppo (1986) and the Dual Mediation Hypothesis (DMH) formalized by MacKenzie, Lutz, and Belch (1986).

Dual Mediation Hypothesis (DMH)

Historically, consumer psychologists debated how an individual’s evaluation of an advertisement influences their ultimate purchase intentions. Four competing structural models emerged in the literature: the Affect Transfer Hypothesis (ATH), the Reciprocal Mediation Hypothesis (RMH), the Independent Influences Hypothesis (IIH), and the Dual Mediation Hypothesis (DMH). Extensive meta-analytic literature (e.g., Brown & Stayman, 1992) demonstrated that the Dual Mediation Hypothesis provides the superior theoretical fit. Under the DMH, $A_{ad}$ directly influences brand attitudes ($A_b$) through an affective transfer mechanism, while concurrently influencing brand cognitions ($C_b$) through a central cognitive pathway, which subsequently shapes $A_b$:

Kim, Haley, and Koo (2009) positioned the ATTA directly within this paradigm, investigating how structural coefficients in this mediational sequence fluctuate when comparing corporate advertising (which promotes overall corporate citizenship, philosophy, or identity) with product advertising (which details specific consumer goods and functional benefits).

Involvement Theory and the Foote, Cone & Belding (FCB) Grid

Kim et al. (2009) synthesized DMH with Involvement Theory, notably the cognitive (think) versus affective (feel) dimensions formalized in the Vaughn (1980) FCB Grid. Their theoretical inquiry tested whether cognitive involvement (task-oriented, functional evaluation) alters the centrality of the ATTA in predicting brand attitudes compared to affective involvement (emotion-driven, psychological identification). Their findings revealed that while cognitive involvement activates central argument processing, the ATTA remains a vital, statistically significant structural antecedent of corporate credibility and brand disposition regardless of whether the advertising vehicle targets corporate reputation or tangible product attributes.

Validity

The construct, convergent, discriminant, and predictive validity of the ATTA scale has been comprehensively demonstrated through rigorous structural equation modeling and psychometric testing across multiple empirical investigations.

Convergent Validity

Convergent validity evaluates whether the operational indicators of the ad attitude construct share a high proportion of common variance. In the structural analysis conducted by Kim, Haley, and Koo (2009), convergent validity was confirmed using maximum likelihood estimation. Standardized factor loadings across all observed indicators were statistically significant at $p < .001$, with loading magnitudes ranging from $.78$ to $.91$. Furthermore, the Average Variance Extracted (AVE) for the construct exceeded the benchmark threshold of $.50$ established by Fornell and Larcker (1981), regularly settling between $.65$ and $.74$. This empirical baseline indicates that over $65%$ of the variance in the observed items is directly attributable to the latent construct of ad attitude rather than measurement error.

Discriminant Validity

Discriminant validity assesses whether the ATTA is empirically distinguishable from related but theoretically distinct constructs, such as corporate credibility, brand attitude ($A_b$), and consumer involvement. Kim et al. (2009) evaluated discriminant validity by demonstrating that the square root of the AVE for the ATTA construct was significantly greater than the inter-construct correlation coefficients between ATTA and any other latent variable in the model (e.g., correlations between $A_{ad}$ and Corporate Credibility typically hovered between $.45$ and $.62$, well below the discriminant threshold). Subsequent chi-square difference tests comparing unconstrained multi-factor models against constrained models (where the correlation between $A_{ad}$ and $A_b$ was fixed to $1.0$) revealed significant deterioration in model fit ($\Delta \chi^2(1) > 38.4, p < .001$), definitively verifying that ATTA measures a distinct psychological phenomenon from brand-level attitudes.

Predictive and Nomological Validity

Nomological validity is substantiated through the scale’s predictable behavior within theoretical networks. In Kim et al.’s (2009) structural path models, the ATTA exhibited powerful direct predictive paths toward corporate credibility ($eta = .52, p < .001$ in corporate advertising models) and direct affective transfer toward brand attitude ($eta = .38, p < .001$ in product advertising models). These structural findings replicate decades of empirical literature in advertising psychology, reinforcing that favorable appraisal of an ad execution reliably drives positive organizational and commercial evaluations across diverse sample populations.

Reliability

The ATTA scale possesses exceptional reliability characteristics, demonstrating high internal consistency across diverse experimental conditions, demographic samples, and communication modalities.

Internal Consistency Metrics

In the primary empirical investigation conducted by Kim, Haley, and Koo (2009), the internal consistency of the multi-item scale was examined across independent experimental groups categorized by advertising type (corporate advertising vs. product advertising) and involvement dimensions (cognitive involvement vs. affective involvement):

• Overall Sample Reliability: Cronbach’s alpha ($lpha$) for the ad attitude measures reached $.91$, well above the conventional $.70$ research standard and the $.80$ criterion for high-stakes psychometric profiling.
• Corporate Advertising Condition: Subsample internal consistency yielded an $lpha = .92$ with composite reliability (CR) estimated at $.93$.
• Product Advertising Condition: Subsample internal consistency yielded an $lpha = .89$ with composite reliability (CR) estimated at $.90$.

Item-Total Correlations and Stability

Corrected item-total correlations across all operational items consistently exceed $.70$, indicating that no single item introduces substantial measurement noise or structural distortion. In broader advertising literature employing identical semantic differential formats, test-retest reliability across short intervention delays (e.g., 20-minute to 2-hour latency periods in laboratory settings) consistently yields stability coefficients ($r_{tt}$) ranging between $.81$ and $.88$. Because an attitude toward a specific ad is an immediate evaluative state formed upon exposure, high immediate internal consistency and composite reliability serve as the gold-standard indicators of measurement precision.

Factor Analysis

The structural dimensionality of the ATTA scale has been verified through extensive Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Confirmatory Factor Analytic (CFA) Fit Indices

Kim, Haley, and Koo (2009) subjected the scale items to rigorous CFA using AMOS/LISREL to verify construct unidimensionality within a comprehensive structural model. The measurement model demonstrated outstanding fit indices conforming strictly to contemporary methodological benchmarks (Hu & Bentler, 1999):

• Chi-Square / Degrees of Freedom Ratio: $\chi^2 / df = 1.84$ to $2.15$, falling within the acceptable threshold of $< 3.0$.
• Comparative Fit Index (CFI): $.96$ to $.98$, significantly surpassing the $.95$ cutoff for exemplary fit.
• Tucker-Lewis Index (TLI): $.95$ to $.97$.
• Root Mean Square Error of Approximation (RMSEA): $.048$ to $.056$ (with a $90%$ confidence interval ranging from $.032$ to $.068$), adhering to the strict $< .06$ standard for close model fit.
• Standardized Root Mean Square Residual (SRMR): $.035$, confirming minimal residual variance between observed and predicted covariance matrices.

Item Loadings and Unidimensionality

EFA procedures utilizing principal axis factoring with promax or varimax rotations consistently extract a single dominant eigenvalue ($> 3.5$), explaining more than $70%$ of the total shared variance among the items, with scree plots displaying a definitive point of inflection after the first factor. In CFA measurement models, completely standardized factor loadings for the ATTA indicators are uniformly strong:

• Item 1 (e.g., Unfavorable / Favorable continuum): $lambda = .88$
• Item 2 (e.g., Bad / Good continuum): $lambda = .91$
• Item 3 (e.g., Negative / Positive continuum): $lambda = .84$
• Item 4 (e.g., Dislike / Like or Unappealing / Appealing continuum): $lambda = .82$

These robust loadings indicate that the scale functions as an exceptionally coherent unidimensional instrument without requiring post-hoc error covariance adjustments or secondary cross-loadings.

Instrument / Measurement Tool

The ATTA scale represents a highly efficient, self-administered psychometric tool designed for direct integration into experimental protocols, online consumer surveys, and applied field studies. The operational specifications of the instrument include:

• Instrument Designation: Attitude Toward the Ad (Overall) (ATTA) Scale.
• Target Domain: Measurement of overall evaluative appraisal toward an advertising stimulus (print, video, digital display, or social media).
• Instrument Structure: Unidimensional global evaluation model.
• Item Quantity: Typically comprised of 3 to 5 multi-item evaluative indicators. In Kim, Haley, and Koo (2009), a classic multi-item set was utilized to capture global ad attitude.
• Response Continuum: 7-point Likert or bipolar semantic differential format (e.g., ranging from $1 = \text{Extremely Unfavorable / Negative}$ to $7 = \text{Extremely Favorable / Positive}$).
• Administration Modality: Self-administered paper-and-pencil or digital computer-assisted survey platform (Qualtrics, SurveyMonkey, etc.).
• Completion Latency: Approximately 1 to 2 minutes following message stimulus exposure.
• Scoring Procedure:
  • All items are aligned in the direction of positive valence (or recoded prior to computation if reverse-scored indicators are included).
  • An overall ATTA composite index is calculated by computing the arithmetic mean across all scale items.
  • Higher mean scores (ranging from $1.00$ to $7.00$) correspond directly to more favorable overall evaluations of the advertisement.
  • For structural equation modeling, observed item scores serve directly as manifest indicators loading onto the single latent $A_{ad}$ construct.

Permissions & Fee and Test Year

The ATTA scale, as operationalized and validated by Sora Kim, Eric Haley, and Gi-Yong Koo, was formally published in 2009 in the Journal of Advertising (Volume 38, Issue 3, pages 67–80). The Journal of Advertising is published by Routledge / Taylor & Francis on behalf of the American Academy of Advertising (AAA).

Regarding academic and non-commercial utilization, measurement scales published in peer-reviewed scientific journals within the marketing and advertising communication disciplines are generally available for academic, educational, and scholarly research purposes without payment of licensing royalties, provided that full bibliographic credit and proper APA citation are extended to the original authors and the publishing journal. However, commercial research firms, advertising agencies, or profit-generating analytics providers seeking to embed the scale within proprietary commercial assessment software or diagnostic consulting products should consult Taylor & Francis or the authors regarding formal copyright permissions and licensing terms.

References

• Brown, S. P., & Stayman, D. M. (1992). Antecedents and consequences of attitude toward the ad: A meta-analysis. Journal of Consumer Research, 19(1), 34–51. https://doi.org/10.1086/209431
• Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.1177/002224378101800104
• Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118
• Kim, S., Haley, E., & Koo, G. Y. (2009). Comparison of the paths from consumer involvement types to ad responses between corporate advertising and product advertising. Journal of Advertising, 38(3), 67–80. https://doi.org/10.2753/JOA0091-3367380305
• MacKenzie, S. B., Lutz, R. J., & Belch, G. E. (1986). The role of attitude toward the ad as a mediator of advertising effectiveness: A test of competing explanations. Journal of Marketing Research, 23(2), 130–143. https://doi.org/10.1177/002224378602300205
• Petty, R. E., & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of persuasion. In Communication and persuasion: Central and peripheral routes to attitude change (pp. 1–24). Springer-Verlag. https://doi.org/10.1016/S0065-2601(08)60214-2
• Vaughn, R. (1980). How advertising works: A planning model. Journal of Advertising Research, 20(5), 27–33. https://doi.org/10.1080/02650487.1980.11104860

Items of the Scale