Attitude Toward the Product (Overall Evaluation) (ATPR)

Abstract

The Attitude Toward the Product (Overall Evaluation) (ATPR) scale is a concise, psychometrically robust four-item semantic differential scale designed to assess a consumer’s global, valence-based evaluative judgment of a specific focal product. Grounded in foundational social cognitive paradigms and contemporary consumer behavior research (Han et al., 2021), the instrument operationalizes attitude as an overall psychological tendency manifested through directional evaluation along bipolar dimensional continua: unfavourable versus favourable, negative versus positive, dislike versus like, and undesirable versus desirable. The scale employs a unidimensional structure, capturing integrated cognitive appraisal and affective endorsement without imposing excessive respondent burden. Administered typically via a 7-point bipolar response metric, the ATPR exhibits exceptional internal consistency reliability across laboratory experiments, randomized controlled field trials, and online panel investigations, consistently yielding Cronbach’s alpha coefficients exceeding .90. Confirmatory factor analyses across diverse demographic and consumption contexts substantiate strict unidimensionality, marked by uniformly elevated factor loadings, low residual error variances, and strong convergent validity with proximal downstream behavioral criteria, such as purchase intentions, willingness to pay, and tangible consumption choices. This comprehensive psychometric review details the scale’s theoretical architecture, structural validity, cross-contextual reliability, administration guidelines, and standard scoring protocols, highlighting its utility as a benchmark measurement instrument in experimental psychology, behavioral economics, and marketing science.

Keywords

Attitude Toward the Product, ATPR, consumer evaluation, semantic differential, product attitude, consumer psychology, psychometrics, evaluative judgment, unidimensional measurement, purchase intention

Authors

The scale was formalized and validated in its present four-item unified configuration by:

• Dahee Han — Associate Professor of Marketing, Desautels Faculty of Management, McGill University, Montreal, Quebec, Canada. Primary research areas: consumer information processing, mindset theory, consumer health decisions, and emotional regulation.
• Adam Duhachek — Professor of Marketing and Nestle-Hustad Professor, Kelley School of Business, Indiana University, Bloomington, Indiana, USA. Primary research areas: consumer stress, coping mechanisms, affective forecasting, and persuasive communication.
• Nidhi Agrawal — Michael G. Foster Endowed Professor of Marketing, Foster School of Business, University of Washington, Seattle, Washington, USA. Primary research areas: health communication, risk perception, self-regulation, and emotions in consumer judgments.

Correspondence regarding original empirical applications and methodological protocols is typically addressed to the primary authors through their respective academic institutions or via the publishing channels of the American Marketing Association.

Purpose

The primary purpose of the Attitude Toward the Product (Overall Evaluation) (ATPR) scale is to provide researchers and behavioral scientists with a reliable, standardized, and parsimonious instrument for quantifying a consumer’s global evaluative disposition toward a specified product or consumable stimulus. In consumer psychology and decision sciences, researchers frequently introduce complex experimental manipulations—such as altered price-framing mechanisms, cognitive mindset primes, regulatory focus prompts, or health warning cues—and require a sensitive, unbiased gauge of how these interventions reshape immediate evaluative appraisals of the target entity.

Historically, measures of product attitudes have suffered from either overextension, incorporating burdensome multi-item batteries that blend general valence with idiosyncratic attributes (e.g., taste, utility, prestige, eco-friendliness), or excessive brevity, relying on single-item indicators that inherently lack quantifiable internal consistency estimates and fail to account for item-specific measurement error. The ATPR strikes a balance by providing a four-item semantic differential array that captures the common evaluative core across both affective (“dislike / like”) and cognitive-appraisal (“undesirable / desirable”, “unfavourable / favourable”, “negative / positive”) spectra.

In applied and clinical settings—such as nutritional intervention programs, public health campaigns targeting the de-escalation of ultra-processed food consumption, and consumer financial counseling—the ATPR functions as a diagnostic metric to capture shift magnitudes in product valuation following educational or behavioral nudges. In academic marketing research, it serves as a critical mediator connecting experimental independent variables (such as mindset shifts, social identity primes, or price architecture changes) to concrete behavioral endpoints, including consequential product choice, willingness to pay (WTP), brand switching, and long-term brand loyalty.

Psychological Construct

The construct captured by the ATPR scale is the overall evaluative attitude toward a focal product entity. In classical attitude theory, an attitude represents an enduring, learned psychological tendency to respond with some degree of favorableness or unfavorableness toward a psychological object (Eagly & Chaiken, 1993). In consumer research, product attitudes occupy a central position in the hierarchy of effects, serving as the cognitive-affective bridge between stimulus perception and behavioral enactment.

The scale models this construct as a global, unidimensional, valence-aligned continuum. Rather than decomposing product perception into disparate belief hierarchies, the ATPR focuses on the summative psychological verdict synthesized by the respondent. This evaluative verdict synthesizes two underlying operational domains:

• Global Cognitive Appraisal: Represented by the bipolar dimensions unfavourable / favourable and negative / positive. These items capture the respondent’s rational, comparative, and propositional assessment of whether the product aligns with generalized standards of quality, utility, and subjective value.
• Affective and Desiderative Endorsement: Represented by the bipolar dimensions dislike / like and undesirable / desirable. These items tap into visceral, emotional, and motivational appetites toward the product, reflecting immediate valence and psychological attraction or repulsion.

By integrating these four bipolar markers, the scale avoids the construct underrepresentation common in single-item gauges while excluding confounding situational dimensions (e.g., immediate purchasing constraints, social norms, or budgetary trade-offs) that belong to separate behavioral constructs. The result is an assessment of pure, focal product valence.

Theoretical Framework

The conceptual architecture of the ATPR scale draws directly from classic and modern formulations in cognitive and social psychology:

The Semantic Differential Framework

The scale relies fundamentally on the semantic differential methodology pioneered by Charles E. Osgood, George Suci, and Percy Tannenbaum (1957). In their seminal psycholinguistic research, Osgood and colleagues demonstrated that human semantic judgments of concepts across diverse linguistic cultures map predominantly onto three primary latent dimensions: Evaluation (good/bad), Potency (strong/weak), and Activity (active/passive). The Evaluation dimension consistently accounted for the largest proportion of total variance in human judgment. The ATPR deliberately isolates this foundational Evaluative dimension, utilizing antonymous adjective pairs anchored at opposite poles to quantify directional intensity along a psychological continuum.

The Expectancy-Value Model and Theory of Planned Behavior

The theoretical necessity for assessing overall product evaluation is rooted in the Expectancy-Value Model of attitudes (Fishbein & Ajzen, 1975) and the Theory of Planned Behavior (Ajzen, 1991). According to these frameworks, individuals form beliefs about an object by associating it with specific attributes. However, when making behavioral commitments, individuals spontaneously synthesize these multidimensional attribute beliefs into an integrated evaluative attitude (A_act or A_object). This integrated attitude, alongside subjective norms and perceived behavioral control, directly predicts behavioral intention. The ATPR scale operationalizes this integrated endpoint without requiring respondents to decompose and weight every individual belief, making it a reliable proxy for the latent evaluative summary.

Dual-Process and Mindset Frameworks

Recent research in consumer cognition, such as work on mindset theory and mental abstraction (Trope & Liberman, 2010; Han et al., 2021), underscores that changes in cognitive processing modes (e.g., abstract vs. concrete mindsets, deliberative vs. implemental orientations) dynamically modulate how product information is weighed. When consumers experience mindset shifts, the comparative weight of price versus hedonic pleasure changes, altering the final evaluative synthesis. The ATPR provides a clean readout of this downstream synthesis, enabling researchers to isolate how mental shifts translate into evaluative upgrades or downgrades across diverse product categories.

Validity

The ATPR scale demonstrates extensive empirical evidence supporting its construct, convergent, discriminant, and predictive validity across numerous experimental trials and survey contexts:

Construct and Convergent Validity

Construct validity has been confirmed across several empirical investigations. In the validation studies conducted by Han, Duhachek, and Agrawal (2021), the four items exhibited high inter-item correlations (typically ranging from r = .78 to r = .92, all ps < .001), indicating that each bipolar pair draws upon the same underlying psychological core. Convergent validity is evidenced by the average variance extracted (AVE), which systematically exceeds the standard benchmark of .50 (Fornell & Larcker, 1981), frequently surpassing .80 in consumer evaluations of both consumer packaged goods (CPG) and durable products.

Predictive and Nomological Validity

The scale possesses strong predictive validity within nomological networks of consumer choice. In Han et al. (2021), changes in ATPR scores significantly mediated the interactive effect of pricing strategies and consumer mindsets on actual behavioral willingness to pay (WTP) and subsequent product purchase selection. Higher ATPR scores consistently predict elevated purchase intentions (typical path coefficients β = .65 to .82, p < .001) and actual product pick rates in incentive-compatible lab lotteries.

Discriminant Validity

Discriminant validity has been established relative to conceptually distinct constructs, such as:

• General Positive/Negative Mood: Assessed via scales like the PANAS; the ATPR demonstrates low to moderate correlations (r < .25), demonstrating that the scale measures object-specific evaluation rather than transient global affect.
• Price Perceptions: Correlations between ATPR and perceived product expensiveness remain modest, confirming that consumers maintain distinct mental representations for the monetary sacrifice required versus their overarching evaluative attitude toward the product itself.
• Brand Attitude (A_brand): When assessing an unfamiliar or private-label product, ATPR diverges from corporate brand perceptions, demonstrating sensitivity to product-level attributes rather than mere corporate umbrella associations.

Reliability

The internal consistency of the ATPR scale has been extensively documented in peer-reviewed literature. Across diverse consumer sampling contexts—including undergraduate behavioral laboratory participants, Amazon Mechanical Turk panels, Prolific Academic cohorts, and nationally representative consumer panels—the scale exhibits high reliability coefficients:

• Cronbach’s Alpha (α): Consistently reported between .91 and .97. In the pivotal studies of Han, Duhachek, and Agrawal (2021), reported alpha coefficients across multiple experimental conditions examining unhealthy food choices, price-framing manipulations, and health-goal mindset shifts consistently yielded α ≥ .94, demonstrating minimal error variance.
• McDonald’s Omega (ω): Due to minor variations in item factor loadings, contemporary psychometric reporting favors McDonald’s hierarchical and total omega coefficients. Studies evaluating the ATPR report ω values spanning .93 to .96, indicating high composite reliability that does not depend on the assumption of tau-equivalence.
• Test-Retest Stability: While attitudes toward novel products can vary depending on new information or situational context, test-retest reliability across short, non-interventional intervals (e.g., 48 hours to 7 days) demonstrates temporal stability (intraclass correlation coefficients ICC > .85), confirming that the instrument captures enduring evaluative states when the target stimulus remains unchanged.

Factor Analysis

Psychometric evaluations employing both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) consistently confirm the unidimensional structure of the ATPR scale.

Exploratory Factor Analysis (EFA)

Principal axis factoring and maximum likelihood EFA procedures conducted on the four items uniformly yield a single-factor solution based on conventional extraction criteria (Kaiser-Guttman eigenvalue rule > 1.0; scree plot inflection points). The primary factor routinely accounts for 82% to 91% of the total item variance, with no secondary factor achieving an eigenvalue above 0.35. Standardized communalities (h²) for all four items routinely exceed .80.

Confirmatory Factor Analysis (CFA)

Single-factor CFA models estimated using robust maximum likelihood (MLR) demonstrate fit across varied empirical datasets:

• Standardized Factor Loadings (λ): Loadings for all four indicators are uniform and elevated:
  • Unfavourable / Favourable: λ = .88 to .95
  • Negative / Positive: λ = .90 to .96
  • Dislike / Like: λ = .86 to .93
  • Undesirable / Desirable: λ = .87 to .94
• Goodness-of-Fit Indices: In fully identified models (or models tested within broader structural equation systems where degrees of freedom permit comprehensive fit assessment), the single-factor specification yields indices well within conventional thresholds of model fit:
  • Comparative Fit Index (CFI) > .98
  • Tucker-Lewis Index (TLI) > .97
  • Root Mean Square Error of Approximation (RMSEA) < .05 (90% CI [.000, .078])
  • Standardized Root Mean Square Residual (SRMR) < .02

Measurement invariance testing across participant demographics (e.g., gender, age brackets) and experimental conditions (e.g., low vs. high price framing) confirms full metric and scalar invariance, establishing that differences in observed scores reflect true shifts in the latent evaluative construct.

Instrument / Measurement Tool

The structured attributes, administrative parameters, and scoring protocols of the ATPR are detailed below:

• Instrument Name: Attitude Toward the Product (Overall Evaluation) (ATPR)
• Test Type: Psychometric self-report rating scale (Semantic Differential)
• Target Population: Consumers, study participants, and individuals capable of reading comprehension aged 16 and older; adaptable across consumer goods, services, and digital products
• Item Count: 4 bipolar items
• Administration Format: Paper-and-pencil, computerized survey platforms (Qualtrics, Decipher, REDCap), or mobile interfaces
• Estimated Completion Time: Approximately 30 to 60 seconds
• Response Scale: Typically administered as a 7-point bipolar semantic differential scale (ranging from 1 = most negative pole to 7 = most positive pole; alternative variants occasionally employ a 9-point or -3 to +3 format)
• Scoring Procedure:
  • All items are oriented such that the left anchor represents the negative/unfavourable evaluation (coded as 1) and the right anchor represents the positive/favourable evaluation (coded as 7).
  • If the presentation order of poles is counterbalanced during survey administration to mitigate position bias, reverse-coding must be executed prior to aggregation (i.e., New Score = 8 − Recorded Score on a 7-point scale).
  • A composite Attitude Toward the Product index is derived by calculating the unweighted arithmetic mean across the four items:
    
    ATPR Score = (Item 1 + Item 2 + Item 3 + Item 4) / 4
  • Higher aggregate mean values represent a more favourable, positive, and desirable overall product evaluation.

Permissions & Fee and Test Year

The Attitude Toward the Product (Overall Evaluation) (ATPR) scale was formalized in its referenced four-item experimental configuration by Dahee Han, Adam Duhachek, and Nidhi Agrawal in their 2021 publication in the Journal of Marketing Research. The bipolar adjective pairs that constitute the instrument reflect classical semantic differential items derived from public-domain psycholinguistic traditions (Osgood et al., 1957) and consumer behavior literature.

For non-commercial academic research, empirical behavioral studies, educational applications, and university-sponsored dissertations, the scale may be utilized free of charge without formal royalty requirements, provided appropriate academic attribution and citation are accorded to Han, Duhachek, and Agrawal (2021) and the foundational measurement literature. Commercial entities, market research agencies, or developers incorporating the scale into proprietary commercial testing software should consult the publishers of the Journal of Marketing Research (American Marketing Association) or the corresponding authors to verify licensing terms.

References

• 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
• Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Harcourt Brace Jovanovich College Publishers.
• Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Addison-Wesley.
• 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
• Han, D., Duhachek, A., & Agrawal, N. (2021). Mindset shifts and the price of unhealthy products. Journal of Marketing Research, 58(4), 712–729. https://doi.org/10.1177/00222437211016010
• Osgood, C. E., Suci, G. J., & Tannenbaum, P. H. (1957). The measurement of meaning. University of Illinois Press.
• Trope, Y., & Liberman, N. (2010). Construal-level theory of psychological distance. Psychological Review, 117(2), 440–463. https://doi.org/10.1037/a0018963

Items of the Scale