482
JOURNAL OF CONSUMER RESEARCH
TABLE10
STUDY 4A: ATTRIBUTE UTILITIES a
Plastic condition
Plastic þ paper condition
M (SD)
t -Stat.
p -Value
M (SD)
t -Stat.
p -Value
(Test value ¼ 0)
(Test value ¼ 0)
Zero-centered utility scores b
Price
56.11 (26.09) 37.54 (67.07) 8.42 (140.48)
30.42 7.92 0.85
< .001 < .001
55.75 (31.99) 15.38 (46.57) 21.39 (167.38)
24.64 4.67 1.81
< .001 < .001
Packaging c
Flavor .072 a As a robustness check, we also estimated attribute utilities across packaging conditions using an aggregate logit approach for this study and the remaining choice-based conjoint studies 4b and 5. The effect of packaging type on packaging utility was replicated. The results of the analysis are reported in web appendix F. Since logit estimation does not produce individual-level utility estimates, we could not test mediation through PEF scores with the logit approach. b $3.00 price, paper packaging, and dark chocolate flavor served as the baseline attribute levels for utility estimates. c On average, plastic is 37.54 utils worse than paper, whereas plastic þ paper is 15.38 utils worse than paper. As such, plastic is 22.16 utils worse than plastic þ paper. .398
were not more likely to choose the dark chocolate flavor over the dark chocolate with strawberries flavor. There was no significant difference in the utility of flavor across the two packaging conditions ( F (1, 398) ¼ 0.71, p ¼ .402, g p 2 ¼ 0.002). Next, we analyzed the utility of packaging across the “plastic” and “plastic þ paper” conditions. In the “plastic” condition, the average utility of plastic packaging was neg- ative and significantly different from zero, suggesting that, on average, people were more likely to choose chocolate in paper packaging compared to chocolate in plastic packag- ing ( M plastic ¼ 37.54, SD ¼ 67.07, t (199) ¼ 7.92, p < .001). Similarly, in the “plastic þ paper” condition, the average utility of plastic þ paper packaging was negative and significantly different from zero, suggesting that peo- ple on average were more likely to choose chocolate in paper than chocolate in plastic þ paper packaging ( M plastic þ paper ¼ 15.38, SD ¼ 46.57, t (199) ¼ 4.67, p < .001). Importantly, plastic packaging had a greater disutil- ity compared to plastic þ paper packaging ( F (1, 398) ¼ 14.72, p < .001, g p 2 ¼ 0.036), meaning that participants’ propensity to choose chocolate in plastic packaging was lower than their propensity to choose chocolate in plastic þ paper packaging. Perceived Environmental Friendliness. Next, we ana- lyzed participants’ PEF scores ( a ¼ 0.96) of plastic (plastic þ paper) packaging. A one-way ANOVA revealed that people perceived plastic packaging to be less environmen- tally friendly than plastic þ paper packaging ( M plastic ¼ 2.37, SD ¼ 1.40, vs. M plastic þ paper ¼ 2.82, SD ¼ 1.57, F (1, 398) ¼ 9.29, p ¼ .002, g p 2 ¼ 0.023). Mediation Analysis. For mediation analysis, we used our estimated individual-specific utilities as the dependent variable and participants’ PEF scores as the mediator. Mediation with 10,000 bootstrap samples revealed that packaging type had a significant indirect effect on packag- ing utilities via PEF [ b ¼ 3.81, SE ¼ 1.47, 95% CI: (1.42; 7.36); see web appendix G for a full summary of mediation
results for studies 4a–5]. Thus, consumers were more likely to choose chocolate in plastic þ paper (vs. plastic) packag- ing, because they viewed plastic þ paper packaging as more environmentally friendly than plastic-only packaging. Discussion This study replicates the effect of packaging type on the environmental friendliness perceptions and demonstrates the implications of this effect for consumers’ choice. In this study, we used a choice-based conjoint setup to com- pare the impact of product packaging on product choice. We observed that addition of a layer of paper to plastic packaging made people more likely to select a chocolate over a chocolate packaged in paper. In addition, this study addresses the possible limitation of study 3 wherein consumers could have made additional inferences about product quality across the two packaging conditions. In this study, we informed consumers upfront that the evaluated products only differed in terms of their price, flavor, and packaging to minimize possible infer- ences regarding product quality. As our effect was repli- cated, we would suggest that our results are not likely driven by quality inferences. STUDY 4B: IMPLICATIONS FOR CHOICE OF EXISTING BRANDS Study 4a showed that consumers are less prone to choose a chocolate bar packaged in plastic than to choose a choco- late bar packaged in plastic þ paper. It also demonstrated that this effect is mediated by PEF. One possible limitation of study 4a is that the two chocolate bar images only varied in terms of their packaging. While this manipulation allowed us to control for possible confounding effects of brand name and minimized possible additional inferences regarding product quality, it may have prompted our partic- ipants to pay more attention to product packaging than
Made with FlippingBook flipbook maker