Attitude Toward the Ad Scale (Aad)

1. Abstract

The Attitude Toward the Ad Scale (Aad) is an established psychometric instrument widely utilized in consumer psychology, marketing communications, and behavioral science. Originally conceptualized in seminal works on advertising effects (e.g., Mitchell & Olson, 1981) and systematically reviewed across empirical literature by researchers such as Muehling and McCann (1993), this brief metric assesses a consumer’s overall affective and evaluative disposition toward a specific promotional stimulus. Unlike measures of brand attitude ($A_b$) or objective message processing, Aad isolates the respondent’s immediate, non-brand-referent subjective reaction to the execution of the advertisement itself. The canonical operationalization relies on a concise 3-item semantic differential battery comprising the bipolar pairings: Unfavorable–Favorable, Bad–Good, and Dislike–Like.

Each item is administered via a 7-point semantic differential scale (ranging from 1 = strongly negative pole to 7 = strongly positive pole), with scores averaged or summed to yield a composite index of advertising evaluation. Psychometric assessments documented across hundreds of experimental and field investigations indicate exceptional internal consistency reliability, with Cronbach’s alpha coefficients routinely exceeding $\alpha = .85$ and frequently reaching $\alpha = .92$ to $.96$. Structural validity analyses consistently substantiate an essentially unidimensional affective-evaluative construct that accounts for over 75% to 85% of item variance. Aad serves as an indispensable mediating variable within prominent structural frameworks, including the Dual Mediation Hypothesis (DMH), where positive attitudes toward an advertisement directly enhance brand cognitions, attitude toward the brand, and downstream behavioral purchase intentions. This paper offers a comprehensive academic review of the scale’s historical development, theoretical mechanisms, psychometric properties, and analytical implementation.

2. Keywords

Attitude Toward the Ad, Aad, consumer psychology, advertising effectiveness, semantic differential, dual mediation hypothesis, brand attitude, affective transfer, persuasive communication, psychometrics

3. Authors

The conceptual foundation of the Attitude Toward the Ad construct as an independent mediator was pioneered by Andrew A. Mitchell and Jerry C. Olson (1981) at the Pennsylvania State University. The scale’s pervasive three-item semantic differential formulation was subsequently refined, standardized, and extensively cross-validated in seminal advertising and psychometric literature by Scott B. MacKenzie, Richard J. Lutz, and George E. Belch (1986), alongside comprehensive syntheses and theoretical systematization by Darrel D. Muehling (Washington State University) and Michelle McCann (1993), as well as meta-analytic investigations by Steven P. Brown and Douglas M. Stayman (1992).

• Darrel D. Muehling, Ph.D. — Professor Emeritus of Marketing, Carson College of Business, Washington State University, Pullman, WA, USA. Specialized in advertising effects, consumer memory, and nostalgic marketing communications.
• Michelle McCann, Ph.D. — Department of Marketing, Washington State University, Pullman, WA, USA.
• Scott B. MacKenzie, Ph.D. — Neal Gilliatt Chair and Professor of Marketing, Kelley School of Business, Indiana University, Bloomington, IN, USA.
• Richard J. Lutz, Ph.D. — J.C. Penney Professor of Marketing, Warrington College of Business, University of Florida, Gainesville, FL, USA.
• Andrew A. Mitchell, Ph.D. — Late Professor of Marketing, Rotman School of Management, University of Toronto, Toronto, ON, Canada.

4. Purpose

The primary purpose of the Attitude Toward the Ad Scale (Aad) is to provide a parsimonious, psychometrically robust measurement of an individual’s favorable or unfavorable affective evaluation of a specific advertising execution, unconfounded by prior brand familiarity, brand loyalty, or objective product attributes. Historically, early persuasive communication models assumed that advertisements influenced consumer purchase intentions solely by altering cognitive beliefs regarding product attributes. However, foundational empirical observations revealed that consumers frequently form positive behavioral inclinations toward advertised products even when advertisements convey zero factual cognitive information or when consumers process advertisements under conditions of minimal involvement.

To capture this phenomenon, the Aad instrument was engineered to assess the affective and aesthetic resonance generated by the commercial execution—including visual design, music, humor, narrative framing, pacing, and emotional tone. In applied consumer research, media testing, and commercial pre-testing, the scale functions as a diagnostic tool to ascertain whether an advertising execution succeeds in engaging the target audience or inadvertently provokes irritation, confusion, or aversion. In laboratory behavioral research, Aad operates as an indispensable dependent, mediating, or control variable. It isolates the direct affective influence exerted by promotional creative elements before those feelings cascade into brand evaluations ($A_b$) or behavioral purchase intentions ($PI$).

Beyond commercial marketing, the Aad framework is extensively deployed in public health communication, political advertising evaluation, and social marketing campaigns. For instance, investigators measuring the efficacy of anti-smoking public service announcements, vaccine adoption campaigns, or environmental sustainability appeals employ Aad to evaluate whether target audiences find the messaging appealing, persuasive, and emotionally compelling rather than alienating or reactancy-inducing. The scale’s brevity (comprising just three semantic differential items) minimizes participant fatigue and cognitive burden, rendering it well-suited for complex structural equation modeling (SEM) protocols, longitudinal repeated-exposure experiments, and high-frequency digital split-testing environments.

5. Psychological Construct

The construct of Attitude Toward the Ad ($A_{ad}$) is formally defined as a predisposition to respond in a consistently favorable or unfavorable manner to a particular advertising stimulus during a specific exposure occasion (Lutz, 1985; MacKenzie et al., 1986). It captures a state-dependent, situational affective reaction rather than an enduring cognitive personality trait. While closely related to broader constructs in social psychology, such as general attitude, $A_{ad}$ possesses unique operational boundaries that differentiate it from other communication outcomes.

Distinction from Attitude Toward the Brand ($A_b$)

A central theoretical distinction in consumer psychology is the boundary separating $A_{ad}$ from $A_b$ (Attitude Toward the Brand). An individual may evaluate an advertisement as brilliant, witty, visually stunning, and thoroughly enjoyable ($A_{ad}$ is highly positive), yet maintain reservations regarding the underlying brand’s price, corporate ethics, or product functionality ($A_b$ remains neutral or negative). Conversely, a loyal customer may hold an exceptionally high brand evaluation ($A_b$) while regarding a specific television commercial as irritating, cliché, or aesthetically uninspired ($A_{ad}$ is low). The 3-item Aad scale isolates the pure stimulus evaluation of the advertising context, removing the confounding brand heritage.

Affective versus Cognitive Facets

Although the 3-item instrument operates as a psychometrically unidimensional index, theoretical literature identifies dual underlying antecedents that inform this summary judgment: affective reactions and cognitive evaluations (Brown & Stayman, 1992; Muehling & McCann, 1993). Affective reactions represent instantaneous emotional states elicited by the ad execution—such as warmth, amusement, nostalgia, excitement, or disgust. Cognitive evaluations represent the viewer’s conscious appraisal of the commercial’s informational utility, credibility, relevance, and artistic execution. The semantic differential indicators (Unfavorable–Favorable, Bad–Good, Dislike–Like) integrate both streams into a unified evaluative continuum, capturing the holistic valence of the consumer’s reaction.

Dimensional Uniformity

Extensive psychometric investigations have affirmed that while an ad may possess multi-attribute perceptual dimensions (e.g., informative vs. uninformative, entertaining vs. boring), the global summary construct of $A_{ad}$ captured by these three items reflects an indivisible, bipolar evaluative dimension. The positive anchor represents complete valence endorsement (favorable, good, liked), whereas the negative anchor denotes global rejection (unfavorable, bad, disliked).

6. Theoretical Framework

The theoretical architecture underpinning the Attitude Toward the Ad Scale is anchored in persuasive communication theory, cognitive response modeling, and associative learning models. Among the numerous frameworks devised to explain the mechanisms through which $A_{ad}$ governs human behavior, the Dual Mediation Hypothesis (DMH) and the Elaboration Likelihood Model (ELM) represent the primary structural foundations.

The Dual Mediation Hypothesis (DMH)

Formalized by MacKenzie, Lutz, and Belch (1986) and validated across multiple replications, the Dual Mediation Hypothesis positions $A_{ad}$ as a critical nexus in the persuasion hierarchy. The DMH posits that exposure to an advertisement initiates two simultaneous, interconnected causal pathways:

• Direct Affect Transfer Route: A direct, affective link from $A_{ad}$ to $A_b$ ($A_{ad} \rightarrow A_b$). A positive feeling generated by the creative elements of the ad is transferred automatically to the brand via classical conditioning or affect transfer mechanisms.
• Indirect Cognitive Route: An indirect link mediated by brand cognitions ($A_{ad} \rightarrow C_b \rightarrow A_b$). When viewers like the advertisement, they exhibit greater perceptual receptivity, lower counterarguing, and a heightened propensity to accept the message claims, fostering favorable brand cognitions ($C_b$) that reinforce positive brand attitudes.

Competing models—such as the Affect Transfer Hypothesis (ATH), which asserts only direct transfer; the Reciprocal Mediation Hypothesis (RMH), which hypothesizes mutual influence between $A_{ad}$ and $A_b$; and the Independent Influences Hypothesis (IIH), which treats $A_{ad}$ and $A_b$ as parallel paths to purchase intention—have been extensively evaluated using structural equation modeling. Across comparative meta-analyses (e.g., Brown & Stayman, 1992), the Dual Mediation Hypothesis has consistently demonstrated superior statistical fit and explanatory power.

The Elaboration Likelihood Model (ELM)

The Elaboration Likelihood Model developed by Petty and Cacioppo provides a situational framework for how $A_{ad}$ functions across varying conditions of consumer involvement:

• Low Elaboration / Peripheral Route: When consumers possess low motivation, opportunity, or ability to scrutinize marketing claims (e.g., passive exposure to banner ads or background television spots), $A_{ad}$ acts as a potent peripheral cue. A favorable evaluation of the ad’s music, celebrity endorser, or visual aesthetics directly forms brand attitudes without intermediate cognitive processing.
• High Elaboration / Central Route: When consumers possess high involvement, $A_{ad}$ continues to exert influence, but does so primarily by biasing the processing of argument quality, enhancing attention duration, and reducing counter-attitudinal ideation.

7. Validity

The psychometric validity of the 3-item Attitude Toward the Ad Scale has been rigorously documented across numerous empirical studies, spanning laboratory experiments, field surveys, and meta-analytic syntheses.

Construct and Content Validity

Content validity was established through comprehensive semantic extraction of evaluative lexicons used in cognitive psychology and psycholinguistics (drawing directly upon Osgood, Suci, & Tannenbaum’s seminal evaluation dimension). The three pairings—Unfavorable–Favorable, Bad–Good, and Dislike–Like—capture the central evaluative core of the affective domain. In scale validation studies, expert panels and consumer focus groups confirmed that these items directly operationalize subjective appraisal of promotional creative work without confounding structural elements (e.g., ad credibility or ad novelty).

Convergent Validity

Convergent validity is corroborated by high, statistically significant correlations between the 3-item semantic differential scale and alternative affective indices. Brown and Stayman (1992) synthesized data across 43 independent studies ($N > 10,000$), demonstrating that the 3-item Aad metric converges strongly with ad-induced emotional responses (e.g., feelings of warmth, upbeatness, and joy; average $r = .58$ to $.71, p < .001$) and with single-item overall appeal ratings ($r > .85$). When modeled alongside multi-item scales measuring specific executional liking, standardized factor loadings routinely exceed $lambda = .85$, meeting established benchmarks for convergent validity.

Discriminant Validity

A critical psychometric requirement is demonstrating that $A_{ad}$ is empirically distinguishable from related constructs, particularly Attitude Toward the Brand ($A_b$), Brand Cognitions ($C_b$), and Purchase Intentions ($PI$). Confirmatory factor analyses (CFA) across numerous publications (e.g., MacKenzie et al., 1986; Muehling & McCann, 1993) show that two-factor models separating $A_{ad}$ and $A_b$ yield significantly superior fit statistics compared to single-factor models (e.g., $\Delta \chi^2 > 150.0, p < .001$). Furthermore, the Average Variance Extracted (AVE) for the Aad scale consistently exceeds $.75$, far outstripping the squared correlation ($r^2$) between $A_{ad}$ and $A_b$ (which typically ranges between $.25$ and $.49$), thereby satisfying the Fornell-Larcker criterion for discriminant validity.

Predictive and Criterion Validity

The scale possesses exemplary predictive utility. Across hundreds of marketing studies, Aad demonstrates robust direct and indirect predictive validity regarding consumer choices:

• Prediction of Brand Attitude ($A_b$): Standardized structural path coefficients typically range between $\beta = .35$ and $\beta = .65$ ($p < .001$).
• Prediction of Purchase Intention ($PI$): Indirect effect sizes via $A_b$ routinely range from $\beta = .20$ to $\beta = .45$.
• Prediction of Real Behavioral Choice: In controlled experimental contexts involving actual product selection post-exposure, higher Aad scores predict product selection with odds ratios exceeding $2.4$ ($p < .01$).

8. Reliability

The 3-item Attitude Toward the Ad Scale is characterized by outstanding internal consistency and temporal stability across diverse experimental conditions, sampling frames, and demographic cohorts.

Internal Consistency

Across published consumer research, the Cronbach’s alpha reliability coefficient of the scale consistently falls within the $.88$ to $.96$ range, comfortably exceeding the standard $.70$ psychometric benchmark:

• Mitchell & Olson (1981): Initial multi-item evaluative scales of advertising reaction yielded internal consistencies of $\alpha = .87$ to $.92$.
• MacKenzie, Lutz, & Belch (1986): Reported a Cronbach’s alpha of $\alpha = .88$ for the 3-item semantic differential scale in laboratory experimental contexts.
• Muehling & McCann (1993): Reviewed empirical literature showing median alpha coefficients of $\alpha = .91$ across dozens of published advertising investigations.
• Contemporary Digital and Social Studies: Applied testing of video, social media, and native advertisements continually reports composite reliabilities ($CR$) and McDonald’s omega ($\omega$) coefficients between $.90$ and $.95$.

Test-Retest Stability

While $A_{ad}$ measures an immediate reaction to a specific message, test-retest reliability across short retention intervals (e.g., 20 minutes to 48 hours within repeated-exposure laboratory protocols) demonstrates coefficients between $r_{tt} = .76$ and $r_{tt} = .84$. Over extended intervals (e.g., several weeks), stability coefficients naturally attenuate as memory for executional details fades, consistent with the theoretical conceptualization of $A_{ad}$ as a dynamic evaluative state.

9. Factor Analysis

The structural dimensionality of the 3-item Attitude Toward the Ad Scale has been scrutinized through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Analysis (EFA)

In exploratory analyses using principal axis factoring or maximum likelihood extraction with unrotated or orthogonal rotations, the three items consistently load onto a single dominant factor:

• The primary eigenvalue typically exceeds $2.40$, accounting for $80%$ to $90%$ of the total variance across items.
• Secondary eigenvalues routinely fail to exceed $0.35$, decisively supporting unidimensionality according to the Kaiser criterion (eigenvalues $> 1.0$) and scree plot inspections.
• Factor loadings for the three indicators are uniformly high: Unfavorable–Favorable ($lambda ge .88$), Bad–Good ($lambda ge .91$), and Dislike–Like ($lambda ge .89$).

Confirmatory Factor Analysis (CFA) and Model Fit

In structural equation modeling environments using software such as LISREL, AMOS, or Mplus, single-factor CFA specifications for the Aad construct demonstrate excellent goodness-of-fit indices when examined within broader structural measurement models:

• Comparative Fit Index (CFI): Routinely exceeds $.98$ (often saturated at $1.00$ when evaluated in isolation as a 3-indicator measurement model).
• Tucker-Lewis Index (TLI): Typically exceeds $.97$.
• Root Mean Square Error of Approximation (RMSEA): Consistently $< .05$ in fully specified structural models.
• Standardized Root Mean Square Residual (SRMR): Consistently $< .03$.
• Standardized Factor Loadings: All indicator loadings typically range from $.85$ to $.96$, with item uniqueness values ($1 – \lambda^2$) ranging between $.08$ and $.28$.

These findings demonstrate that the three semantic differential items measure a single, coherent underlying latent construct with minimal measurement error.

10. Instrument / Measurement Tool

Below is a structured overview of the technical operationalization of the scale:

• Construct Measured: Attitude Toward the Ad ($A_{ad}$) — global affective and evaluative disposition toward a specific commercial advertisement.
• Test Type: Self-report questionnaire; semantic differential scaling.
• Item Count: 3 bipolar adjective pairs.
• Administration Format: Paper-and-pencil, computer-assisted personal interviewing (CAPI), or web-based survey platforms (Qualtrics, SurveyMonkey, etc.).
• Administration Time: Less than 1 minute (approximately 15–30 seconds).
• Target Population: Consumers, general adult populations, adolescents, and research participants exposed to media advertising stimuli.
• Response Format: 7-point semantic differential scale (1 to 7 continuum anchored by opposing bipolar adjectives).
• Scoring Protocol:
  • Negative poles (Unfavorable, Bad, Dislike) are positioned or coded as 1.
  • Positive poles (Favorable, Good, Like) are positioned or coded as 7.
  • Neutral/Undecided mid-point is represented by 4.
  • Composite score calculation: Items are averaged or summed to yield an overall $A_{ad}$ index. When averaged, composite scores range from 1.00 to 7.00. Higher scores denote a more favorable attitude toward the advertisement.

11. Permissions & Fee and Test Year

The Attitude Toward the Ad Scale evolved through academic literature across several decades, with pivotal milestones in 1981 (Mitchell & Olson), 1986 (MacKenzie, Lutz, & Belch), and 1993 (Muehling & McCann’s comprehensive review in the Journal of Advertising). Because the scale was developed under university-sponsored academic grants and published in public scholarly journals for foundational scientific inquiry, the 3-item instrument resides in the public academic domain.

Fee and Licensing: There are no commercial licensing fees or royalty payments required for academic, scholarly, educational, or commercial market research applications. Researchers and practitioners may administer the scale without formal written permission, provided appropriate bibliographic attribution is given to the foundational validation literature (e.g., MacKenzie, Lutz, & Belch, 1986; Muehling & McCann, 1993).

12. 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/10.1086/209270
• 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
• Lutz, R. J. (1985). Affective and cognitive antecedents of attitude toward the ad: A conceptual framework. In L. F. Alwitt & A. A. Mitchell (Eds.), Psychological Processes and Advertising Effects: Theory, Research, and Application (pp. 45–63). Lawrence Erlbaum Associates.
• 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/10.1177/002224378602300202
• Mitchell, A. A., & Olson, J. C. (1981). Are product attribute beliefs the only mediator of advertising effects on brand attitude? Journal of Marketing Research, 18(3), 318–332. https://doi.org/10.1177/002224378101800306
• Muehling, D. D., & McCann, M. (1993). Attitude toward the ad: A review. Journal of Advertising, 22(4), 25–58. https://doi.org/10.1080/00913367.1993.10673417
• Petty, R. E., & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of persuasion. Advances in Experimental Social Psychology, 19, 123–205. https://doi.org/10.1016/S0065-2601(08)60214-2

13. Items of the Scale (Questionnaire)