7  Brand Equity, Attachment, and the Self

The constructs of the previous two chapters describe how consumers evaluate offerings (Chapter 6) and bond with sellers (Chapter 5). This chapter turns to the construct that is, for many firms, the single most valuable intangible they own: the brand, as it lives in the consumer’s mind. We treat the brand not as the firm-side financial asset—that is the subject of Chapter 11, which develops brand valuation, signaling, and the market’s pricing of brands—but as a consumer psychological construct: the network of awareness, associations, personality, and emotional bonds that a brand name activates, and that causes a consumer to respond to the branded offering differently from how they would respond to an identical unbranded one. That differential response is the construct, and it is the thread connecting everything below.

The chapter builds from the most cognitive of these constructs to the most affective and identity-laden. We begin with customer-based brand equity (CBBE), the knowledge-based foundation; add brand personality, the brand-as-person layer of associations; then the affective bonds of brand attachment and brand love; and finally the identity constructs—self–brand connection and consumer–company identification—where the brand becomes a resource the consumer uses to construct and signal the self. A unifying concern is discriminant validity (Chapter 3): this is the most crowded construct neighborhood in marketing, and brand attitude, attachment, love, identification, and self-connection overlap enough that distinguishing them empirically is a genuine and recurring challenge rather than a formality.

7.1 Customer-Based Brand Equity

The foundational construct is customer-based brand equity, defined by Keller (1993) as the differential effect of brand knowledge on consumer response to the marketing of the brand. The definition is operational and causal: brand equity exists to the extent that consumers respond more favorably to a marketing action (a product, a price, an ad) when it is attributed to the brand than when the same action is attributed to an unnamed or fictitiously named version. The construct’s content is brand knowledge, which Keller (1993) decomposes into two dimensions:

• Brand awareness: the strength of the brand node in memory, comprising recognition (can the consumer identify the brand when cued?) and recall (does the brand come to mind in a purchase situation?).
• Brand image: the set of brand associations held in memory, evaluated on their strength (how reliably they activate), favorability (how positively valued), and uniqueness (how distinct from competitors). Associations span attributes, benefits, and attitudes.

Yoo and Donthu (2001) operationalize a multidimensional CBBE scale, recovering perceived quality, brand loyalty, and brand awareness/associations as measurable facets. Whether brand equity is best modeled as a reflective construct (a latent equity causing its indicators) or a formative index (built from distinct dimensions) is a live and unsettled dispute, exactly the reflective/formative decision of Table 3.1—and the stakes are real, since the wrong specification invalidates the reliability and validity diagnostics one then applies (Keller and Lehmann 2006). The branding chapter (Chapter 11) returns to this in the context of brand valuation.

7.1.1 Operationalizing the differential-response definition

Keller’s definition is directly simulable, and doing so clarifies what “brand equity” is as a measured quantity. Consider an identical marketing action—say, a modest price premium—presented to two matched groups of consumers, one seeing the action under a known brand and one under an unbranded control. Brand equity is the difference in response between the two, and it manifests statistically as a brand × action interaction: the slope of response on the marketing variable differs by branding condition. The chunk below simulates this and recovers brand equity as that interaction, the cleanest operational statement of Keller (1993).

set.seed(1993)
n <- 2000

branded <- rbinom(n, 1, 0.5)              # 1 = known brand, 0 = unbranded control
price_premium <- runif(n, 0, 1)           # the identical marketing action

# Unbranded buyers punish a premium; branded buyers tolerate it.
# Brand equity = the *differential* response = the interaction coefficient.
intent <- 3.0 - 1.6 * price_premium +                    # base premium aversion
          1.4 * branded * price_premium +                # equity: premium tolerated
          0.3 * branded +                                # baseline brand lift
          rnorm(n, sd = 0.8)

fit <- lm(intent ~ price_premium * branded)
co  <- summary(fit)$coefficients
cat("Differential-response (brand equity) = interaction term:\n")
#> Differential-response (brand equity) = interaction term:
print(round(co["price_premium:branded", ], 3))
#>   Estimate Std. Error    t value   Pr(>|t|) 
#>      1.390      0.126     11.008      0.000
cat(sprintf("\nPremium slope, unbranded: %.2f\nPremium slope, branded  : %.2f\n",
            co["price_premium", 1],
            co["price_premium", 1] + co["price_premium:branded", 1]))
#> 
#> Premium slope, unbranded: -1.57
#> Premium slope, branded  : -0.18

The interaction is brand equity made numeric: under the brand, the same price premium destroys far less purchase intent. A firm that measured only the level of intent (the main effect of branded) would understate the asset, because the most valuable thing the brand buys is not a constant lift but a flatter response to adverse marketing actions— price tolerance, defense against competitor moves, resilience to a misstep. That resilience is also why strong brands are double-edged (the dark-side finding of Chapter 5): the same equity that buffers small failures raises the expectation a large failure violates (J. Aaker, Fournier, and Brasel 2004).

7.2 Brand Personality

Among the associations that compose brand image is a distinctive set: the human personality traits consumers ascribe to a brand. J. L. Aaker (1997) establishes brand personality as a construct with a replicable five-factor structure—sincerity, excitement, competence, sophistication, and ruggedness—paralleling but not identical to the human Big Five. The construct matters because personality associations are self-expressive: a consumer chooses a “rugged” or “sophisticated” brand partly to express or claim those traits, which is the bridge to the self-connection constructs below. Brand personality is measured reflectively, each factor by multiple trait items, and its factor structure is a standard application of the confirmatory-factor machinery of Chapter 3. The chunk recovers the five-factor structure from simulated trait ratings, illustrating both the measurement model and the discriminant question of whether the factors are genuinely distinct.

set.seed(1997)
n <- 600
# Five latent personality factors, each with 3 trait indicators.
facs <- c("sincerity", "excitement", "competence", "sophistication", "ruggedness")
F <- matrix(rnorm(n * 5), n, 5); colnames(F) <- facs
lam <- c(0.82, 0.78, 0.72)
blocks <- lapply(1:5, function(k)
  sapply(lam, function(l) l * F[, k] + rnorm(n, sd = sqrt(1 - l^2))))
dat <- as.data.frame(do.call(cbind, blocks))
names(dat) <- paste0(rep(substr(facs, 1, 4), each = 3), 1:3)

fa <- factanal(dat, factors = 5, rotation = "varimax")
cat("Proportion of variance explained by the 5 factors:\n")
#> Proportion of variance explained by the 5 factors:
print(round(colSums(fa$loadings^2) / ncol(dat), 3))
#> Factor1 Factor2 Factor3 Factor4 Factor5 
#>   0.128   0.126   0.120   0.118   0.116
cat(sprintf("\nLikelihood-ratio test that 5 factors suffice: p = %.3f\n", fa$PVAL))
#> 
#> Likelihood-ratio test that 5 factors suffice: p = 0.895

When five factors fit, brand personality is defensible as a five-dimensional construct; the recurring critique—that the dimensions are unstable across cultures and product categories, and that some “factors” are category artifacts—is a discriminant- and generalizability-validity concern, not a refutation of the construct’s usefulness.

7.3 Brand Attachment

Cognition and personality describe what a consumer knows about a brand; brand attachment describes the emotional bond. Thomson, MacInnis, and Park (2005) define it as the strength of the affective tie connecting the consumer to the brand and develop a scale with three facets—affection, passion, and connection—the “ties that bind.” Attachment is theoretically distinct from a favorable attitude: one can evaluate a brand positively (strong attitude) without feeling bonded to it, and the distinction is not academic. Park et al. (2010) demonstrate conceptual and empirical differentiation of brand attachment from brand-attitude strength, showing that attachment predicts behaviors—actual purchase, brand-supportive effort, willingness to sacrifice—beyond what attitude strength predicts, and that attachment uniquely involves separation distress and the inclusion of the brand in the self. The managerial implication is that a brand can have many favorably-disposed customers and few attached ones, and the attached minority carries disproportionate behavioral value.

7.3.1 Attachment’s incremental prediction

The Park et al. (2010) result—attachment predicts behavior beyond attitude—is an incremental-validity claim, testable as a hierarchical regression in which attachment is entered after attitude and must add explained variance. The chunk simulates correlated attachment and attitude (they overlap, as the construct-crowding worry predicts) and shows attachment’s incremental \(R^2\) for a behavioral outcome.

set.seed(2010)
n <- 1500

# Attitude and attachment are correlated (rho ~ 0.6) but distinct.
L  <- chol(matrix(c(1, 0.6, 0.6, 1), 2))
Z  <- matrix(rnorm(n * 2), n, 2) %*% L
attitude   <- Z[, 1]
attachment <- Z[, 2]

# Behavior (e.g., willingness to sacrifice) depends MORE on attachment.
behavior <- 0.20 * attitude + 0.55 * attachment + rnorm(n, sd = 0.8)

r2_att   <- summary(lm(behavior ~ attitude))$r.squared
r2_both  <- summary(lm(behavior ~ attitude + attachment))$r.squared
cat(sprintf("R^2 with attitude only        : %.3f\n", r2_att))
#> R^2 with attitude only        : 0.261
cat(sprintf("R^2 adding brand attachment   : %.3f\n", r2_both))
#> R^2 adding brand attachment   : 0.441
cat(sprintf("Incremental R^2 of attachment : %.3f\n", r2_both - r2_att))
#> Incremental R^2 of attachment : 0.180

Attachment adds substantial explained variance over attitude, the incremental-validity signature that justifies treating it as a separate construct rather than a relabeling of a strong attitude—provided, always, that the discriminant-validity battery of Chapter 3 confirms the two are empirically separable in the data at hand.

7.4 Brand Love

Stronger and more elaborated than attachment is brand love. Batra, Ahuvia, and Bagozzi (2012) develop it not as a single emotion but as a higher-order prototype—a constellation of cognitions, emotions, and behaviors that consumers themselves classify as love. Its elements include self–brand integration, passion-driven behaviors, positive emotional connection, a long-term relationship orientation, anticipated separation distress, and overall attitude valence and strength. Carroll and Ahuvia (2006) establish the construct’s nomological network: brand love is fed by hedonic and self-expressive brand qualities and drives brand loyalty and positive word of mouth, linking it forward to the engagement and advocacy constructs and backward to the loyalty of Chapter 5. The prototype structure has a measurement consequence: because love is defined as a constellation whose elements need not be interchangeable, it sits closer to the formative pole of Table 3.1 than a unidimensional reflective scale, and forcing it into a single coefficient-\(\alpha\) logic (Equation 3.7) discards exactly the multidimensionality that distinguishes love from mere liking.

7.5 The Self: Connection and Identification

The deepest brand constructs locate the brand inside the consumer’s identity. Escalas and Bettman (2005) show that consumers form self–brand connections by using brands to construct and communicate the self—drawing brand meaning from reference groups (a brand used by an aspirational in-group acquires self-expressive value) in a way moderated by the consumer’s self-construal. The brand becomes, in Escalas and Bettman’s later phrase, part of “me”: a resource for self-definition rather than merely an object of evaluation (Edson Escalas and Bettman 2012).

Bhattacharya and Sen (2003) generalize the identity link from brands to companies with consumer–company identification, grounding it in social-identity theory: consumers identify with a company when doing so satisfies self-definitional needs—when the company’s identity is attractive, distinctive, and overlaps the consumer’s own self-concept—and that identification, not mere satisfaction, drives deep, resilient, advocacy-rich relationships. The identity constructs reframe the entire construct map of this part: where satisfaction (Chapter 4) and the relational constructs (Chapter 5) treat the consumer as an evaluator and relationship partner, the self constructs treat the consumer as someone using brands to answer the question “who am I?”—which is why these bonds are the hardest for competitors to break and the most damaging to betray.

7.6 The Construct Thicket

The progression from CBBE through personality, attachment, love, self-connection, and identification is a progression of increasing overlap, and it creates the chapter’s central measurement hazard. Brand attitude, attachment, love, self–brand connection, and identification are all positively affective, all measured by similar self-report batteries on the same survey, and all correlated—the textbook conditions for both common-method variance and discriminant-validity failure (Chapter 3). Table 7.1 distinguishes them conceptually, but conceptual distinction is necessary, not sufficient: the empirical burden is to show, via the Fornell–Larcker or HTMT criteria (Equation 3.9), that each construct’s AVE exceeds its shared variance with the others. A literature that proliferates near-synonymous brand constructs without clearing that bar is, in the language of Chapter 3, relabeling one construct as several.

Table 7.1: The brand-and-self construct family. Conceptual distinctions are clear; the empirical burden is to demonstrate discriminant validity among them (Chapter 3).

Construct	Core content	Distinguishing feature	Key source
Brand equity (CBBE)	Brand knowledge (awareness + image)	Differential response to marketing	Keller (1993)
Brand personality	Human traits ascribed to the brand	Five-factor association structure	J. L. Aaker (1997)
Brand attitude	Overall favorable evaluation	Valence/strength, no emotional bond	Park et al. (2010)
Brand attachment	Emotional bond (affection, passion, connection)	Separation distress; predicts behavior beyond attitude	Thomson, MacInnis, and Park (2005); Park et al. (2010)
Brand love	Prototype constellation	Self-integration + long-term orientation	Batra, Ahuvia, and Bagozzi (2012); Carroll and Ahuvia (2006)
Self–brand connection	Brand used to construct the self	Reference-group/self-construal driven	Escalas and Bettman (2005)
Consumer–company identification	Identity overlap with the firm	Social-identity grounding	Bhattacharya and Sen (2003)

7.7 Key Takeaways

• Customer-based brand equity is the differential effect of brand knowledge on response to marketing (Keller 1993): brand knowledge = awareness + image (associations’ strength, favorability, uniqueness), and equity manifests as a brand × action interaction (Section 7.1.1), buying price tolerance and resilience, not just a level lift. Whether it is reflective or formative is unsettled (Yoo and Donthu 2001; Keller and Lehmann 2006).
• Brand personality has a replicable five-factor structure—sincerity, excitement, competence, sophistication, ruggedness (J. L. Aaker 1997)—and is self-expressive, bridging to the identity constructs.
• Brand attachment is an emotional bond (affection, passion, connection) (Thomson, MacInnis, and Park 2005) empirically distinct from attitude strength and predicting behavior beyond it (Park et al. 2010) (Section 7.3.1).
• Brand love is a higher-order prototype constellation (Batra, Ahuvia, and Bagozzi 2012), fed by hedonic and self-expressive qualities and driving loyalty and WOM (Carroll and Ahuvia 2006); its multidimensionality pushes it toward the formative pole (Table 3.1).
• The self constructs—self–brand connection (Escalas and Bettman 2005; Edson Escalas and Bettman 2012) and consumer–company identification (Bhattacharya and Sen 2003)—locate the brand inside identity via reference groups and social-identity theory; these bonds are the most resilient and the most damaging to betray.
• The family is a construct thicket (Table 7.1): high overlap makes discriminant validity and common-method-variance control (Chapter 3) a precondition, not a formality, for any model that uses several of these constructs together.

7.8 Further Reading

The CBBE foundation is Keller (1993), with measurement in Yoo and Donthu (2001) and the research-priorities overview in Keller and Lehmann (2006); the firm-side and financial treatment is Chapter 11. Brand personality is J. L. Aaker (1997). On the affective bonds, read Thomson, MacInnis, and Park (2005) for attachment measurement and Park et al. (2010) for its differentiation from attitude, then Batra, Ahuvia, and Bagozzi (2012) and Carroll and Ahuvia (2006) for brand love. The identity constructs are developed in Escalas and Bettman (2005) and Edson Escalas and Bettman (2012) (self–brand connection) and Bhattacharya and Sen (2003) (consumer–company identification); the consumer–brand-relationship lineage runs through Fournier (1998) and the dark-side evidence of J. Aaker, Fournier, and Brasel (2004). The discriminant-validity discipline the whole family requires is the apparatus of Chapter 3.

Aaker, Jennifer Lynn. 1997. “Dimensions of Brand Personality.” SSRN Electronic Journal. https://doi.org/10.2139/ssrn.945432.

Aaker, Jennifer, Susan Fournier, and S. Adam Brasel. 2004. “When Good Brands Do Bad.” Journal of Consumer Research 31 (1): 1–16. https://doi.org/10.1086/383419.

Batra, Rajeev, Aaron Ahuvia, and Richard P. Bagozzi. 2012. “Brand Love.” Journal of Marketing 76 (2): 1–16. https://doi.org/10.1509/jm.09.0339.

Bhattacharya, C. B., and Sankar Sen. 2003. “Consumer-Company Identification: A Framework for Understanding Consumers’ Relationships with Companies.” Journal of Marketing 67 (2): 76–88. https://doi.org/10.1509/jmkg.67.2.76.18609.

Carroll, Barbara A., and Aaron C. Ahuvia. 2006. “Some Antecedents and Outcomes of Brand Love.” Marketing Letters 17 (2): 79–89. https://doi.org/10.1007/s11002-006-4219-2.

Edson Escalas, Jennifer, and James R. Bettman. 2012. “The Brand Is “Me”.” In. Routledge. https://doi.org/10.4324/9780203105337.ch37.

Escalas, Jennifer Edson, and James R. Bettman. 2005. “Self-Construal, Reference Groups, and Brand Meaning.” Journal of Consumer Research 32 (3): 378–89. https://doi.org/10.1086/497549.

Fournier, Susan. 1998. “Consumers and Their Brands: Developing Relationship Theory in Consumer Research.” Journal of Consumer Research 24 (4): 343–53. https://doi.org/10.1086/209515.

Keller, Kevin Lane. 1993. “Conceptualizing, Measuring, and Managing Customer-Based Brand Equity.” Journal of Marketing 57 (1): 1. https://doi.org/10.2307/1252054.

Keller, Kevin Lane, and Donald R. Lehmann. 2006. “Brands and Branding: Research Findings and Future Priorities.” Marketing Science 25 (6): 740–59. https://doi.org/10.1287/mksc.1050.0153.

Park, C. Whan, Deborah J. MacInnis, Joseph Priester, Andreas B. Eisingerich, and Dawn Iacobucci. 2010. “Brand Attachment and Brand Attitude Strength: Conceptual and Empirical Differentiation of Two Critical Brand Equity Drivers.” Journal of Marketing 74 (6): 1–17. https://doi.org/10.1509/jmkg.74.6.1.

Thomson, Matthew, Deborah J. MacInnis, and C. Whan Park. 2005. “The Ties That Bind: Measuring the Strength of Consumers’ Emotional Attachments to Brands.” Journal of Consumer Psychology 15 (1): 77–91. https://doi.org/10.1207/s15327663jcp1501_10.

Yoo, Boonghee, and Naveen Donthu. 2001. “Developing and Validating a Multidimensional Consumer-Based Brand Equity Scale.” Journal of Business Research 52 (1): 1–14. https://doi.org/10.1016/s0148-2963(99)00098-3.