SOKOLOVA, KRISHNA, AND D € ORING
483
for the “plastic” and the “plastic þ paper” conditions (table 12). Because our focus was on the changes in the rel- ative choice propensities of Montperal over the other brands, brand attribute was dummy coded, such that the Montperal brand was coded as 1 and the other two brands were coded as 0. Weight was dummy coded (3 oz ¼ 0, 5oz ¼ 1). Price was coded linearly ($2.00 ¼ 0; $3.00 ¼ 1; $4.00 ¼ 2). Next, we ran one-way ANOVAs to compare attribute utilities across packaging conditions. As expected, the analysis revealed that price had a sig- nificant negative utility, whereas weight had a significant positive utility (table 12). There was no significant differ- ence in the utility of price or of weight across the “plastic” and “plastic þ paper” conditions ( F price (1,400) ¼ 0.85, p ¼ .356, g p 2 ¼ 0.002; F weight (1,400) ¼ 1.57, p ¼ .211, g p 2 ¼ 0.004). Next, we analyzed the utility of brand across the two packaging conditions. In the “plastic” condition, the aver- age utility of the Montperal brand was negative and signifi- cantly different from zero, suggesting that people on average were less prone to choose it over Lays and Tyrrells ( M plastic ¼ 47.33, SD ¼ 83.06, t (200) ¼ 8.08, p < .001). In the “plastic þ paper” condition, the average util- ity of the Montperal brand was also negative and signifi- cantly different from zero ( M plastic þ paper ¼ 29.51, SD ¼ 86.45, t (200) ¼ 4.84, p < .001). Importantly, there was a significant difference in brand utilities across the two packaging conditions ( F (1, 400) ¼ 4.44, p ¼ .036, g p 2 ¼ 0.011), suggesting that the disutility of the Montperal brand was greater when this brand came in plastic packag- ing than when it came in plastic þ paper packaging. Put differently, participants were less likely to choose Montperal brand over Lays and Tyrrells when it came in plastic packaging than when it came in plastic þ paper packaging. Perceived Environmental Friendliness. Next, we ana- lyzed the PEF scores ( a ¼ 0.96) of plastic (plastic þ paper) packaging. A one-way ANOVA revealed that people per- ceived plastic packaging to be less environmentally friendly than plastic þ paper packaging ( M plastic ¼ 2.72, SD ¼ 1.35, vs. M plastic þ paper ¼ 4.44, SD ¼ 1.52; F (1, 400) ¼ 144.79, p < .001, g p 2 ¼ 0.266). Mediation Analysis. Finally, a mediation analysis with 10,000 bootstrap samples revealed that packaging type had a marginally significant indirect effect on choice via PEF [ b ¼ 10.54, SE ¼ 5.71, 90% CI: (1.50; 20.33)]. This result indicates that consumers were more likely to choose the Montperal brand when it came in plastic þ paper packag- ing than when it came in plastic packaging, in part, because they viewed the plastic þ paper packaging as more envi- ronmentally friendly than plastic packaging.
they would have otherwise. To address this concern, study 4b used three real-life brands (Lays, Tyrrells, and Montperal). We then varied the Montperal packaging between-subjects, such that half the participants saw the Montperal brand in plastic packaging and half the partici- pants saw it in plastic þ paper packaging. Critically, instead of informing the participants that the evaluated products varied in terms of price, packaging, and flavor, in this study, we informed them that the evaluated products varied in terms of price, brand, and weight. Thus, in contrast to study 4a, study 4b drew participants’ atten- tion to differences across products in terms of brands, rather than in terms of packaging, to ensure greater ecolog- ical validity of our results. In addition, to ensure greater ecological validity, the study used two versions of Montperal packaging that are in fact marketed by the brand. While study 4a presented an “all else equal” comparison, where the only difference between plastic and plastic þ paper packaging was the added layer of paper, study 4b used plastic and plastic þ paper packaging that varied on more than one dimension but represented real packaging of the focal brand. As such, this manipulation allowed us to test how changes in prod- uct packaging that may be implemented by a real company affect consumer choice and environmental friendliness per- ceptions. As in study 4a, we expected that consumers’ pro- pensity to choose the focal brand (i.e., Montperal) would be lower when it came in plastic-only packaging than when it came in plastic þ paper packaging. Method Four hundred two ProlificCo panelists ( M age ¼ 35.67, 60% female) were randomly assigned to one of two condi- tions (packaging type: plastic vs. plastic þ paper). The method was similar to that used in study 4a. Participants in the “plastic” condition saw Lays, Tyrrells, and Montperal brands, all packaged in plastic. Participants in the “plastic þ paper” condition saw Lays and Tyrrells brands packaged in plastic and the Montperal brand packaged in plastic þ paper (see table 11 for screen- shots of sample choice tasks). In addition to brand, the chips bags varied in terms of price ($2.00, $3.00, $4.00) and weight (3 oz, 5 oz). Participants made 12 choices among three bags of chips. Then, they saw the Montperal chips and rated the environ- mental friendliness of their plastic (plastic þ paper) pack- aging on the four-item PEF scale. Finally, they reported their age and gender.
Results
Conjoint Analysis. We obtained individual-specific utilities of each attribute—price, brand, and weight—using the hierarchical Bayes procedure in Sawtooth separately
Made with FlippingBook flipbook maker