Comparative Quality and Brand Outcomes

The link between perceived quality and shopping value, brand attitude, and purchase-related brand equity are well-documented in existing literature. The authors focus specifically on these outcomes due to their relevance

 low-involvement purchase situations. Fundamentally, achieving perceptions of comparable quality is essential for store brands to be able to compete with premium brands in the eyes of most customers (Hoch and Banerji 1993; Sayman et al. 2002; Sethuraman 1992). Beyond budget constraints, store brand purchases are arguably founded on logic (e.g., “it makes sense to purchase this product”) or congruence with retailer image (Luijten and Reijnders 2009), and most decisions within consumer packaged goods are considered routine. Thus, high comparative quality perceptions should positively (negatively) influence evaluations of private label (national) brands. Additionally, previous research suggests that the perceived general quality gap between private labels and national brands within a broader industry not only drives brand outcomes but also serves as a facilitating mechanism through which marketing factors influence these dependent variables (Steenkamp et al. 2010). Consequently, comparative quality is additionally hypothesized to mediate the relationship between dual branding information and the outcomes in this study: H2: Comparative quality perceptions significantly, positively (negatively) influence private label (national brand) (a) shopping value, (b) brand attitude, and (c) purchase-related brand equity. H3: Comparative quality mediates the relationship between dual branding and the outcomes, such that dual branding information will have a positive (negative) indirect effect on private label (national brand) (a) shopping value, (b) brand attitude, and (c) purchase-related brand equity

Method 

For hypothesis testing, 259 undergraduate students participated in an experimental study, in which they were then exposed to one of four grocery shopping scenarios (control versus dual branding manipulation), shopping for either potato chips or coffee. Two products were chosen in order to investigate generalizability of the findings and consequently were not included as manipulations for testing, although we controlled for product type in the analysis. The manipulation scenario involved a nearby customer telling the respondent that competing premium and private label brands were produced by the same manufacturer; the control scenario described a nearby customer simply reaching for a product in the respondent’s vicinity. Participants then completed multi-item seven-point Likert-type survey items measuring their comparative quality assessments, along with respective brand shopping value (e.g., “if you buy this brand, you get your money’s worth”), attitude (e.g., “this brand is appealing), and purchase-related equity (e.g., “it makes sense to buy this brand instead of others”) for each brand. All scale items were adapted from previous related literature, with the exception of comparative quality, for which items measuring individual quality were modified to capture degree of similarity between the two brands. All scales exhibited
acceptable reliabilities (α>.70, Nunnally and Bernstein 1994). A confirmatory factor analysis was performed using AMOS 23 to assess unidimensionality, convergent validity, and discriminant validity of the constructs. Results indicated an adequate fit of the model to the data (χ2=584.55, df=278, CFI=.96, TLI=.95, RMSEA=.065). All average variance extracted (AVE) calculations were greater than .50, and no shared variance between constructs exceeded the AVE per construct, indicating support for both convergent and discriminant validities (Fornell and Larcker 1981).

Hypothesis Testing Results

 H1. Analysis of variance (ANOVA) testing including a control for product type was conducted to determine the differentiating influence of dual branding information on comparative quality perceptions. Results support H1, as participants exposed to dual branding reported significantly higher comparative quality perceptions than those in the control condition (F=6.12, p=.01; μdual=4.3, μcontrol=3.8). H2. Linear regression analyses controlling for product type were subsequently conducted in order to assess the effects of comparative quality on respective premium and private label brand outcomes, and results provide support for all hypothesized relationships. Table 1 presents relevant information for each test. H3. Finally, the authors investigated the indirect effects (IEs) of dual branding information on brand-related outcomes, through comparative quality. Specifically, mediation analysis was conducted using PROCESS Model 4 with 5000 bootstrap samples and 95% confidence intervals (CIs) (Hayes 2013). Evidence of mediation is present when the confidence intervals associated with the IEs of interest do not contain zero (Hayes 2013; Zhao et al. 2010). Results of the analysis revealed significant indirect effects of dual branding exposure on all customer brand evaluations. A complete presentation of results is provided in Table 2.

The evolution of private label brand quality in part can be attributed to the observation that more than half of all private label brands are produced by dual branders (Kumar et al. 2010; Steenkamp et al. 2010). Importantly, while past research suggests that dual branding provides benefits for national brand manufacturers with respect to their relationships with retailers, these suppliers avoid expressed affiliation with retail brand production, understandably due to concerns about decreasing their own brand market share (Sethuraman and Raju 2012). The current study underscores these worries by illustrating the process through which dual branding information indirectly influences respective private label and national brand evaluations through the mechanism of compared quality perceptions. Specifically, consumers exposed to dual branding information express higher (lower) evaluations of private label (national) brands than do those not exposed to such information,