Food Quality and Preference 96 (2022) 1044052Henrik Behrens, & Klarmann, 2014). Organic labelling, in other words, can produce a “halo effect” in the same way that consumers like prod-ucts better if they think they are expensive and/or come from certain brands (Linder et al., 2010; Plassmann, O’Doherty, Shiv, & Rangel, 2008). It is not clear, however, that all results in the literature align well with this notion. For example, Schuldt and Hannahan (2013) reported the opposite effect (organic being rated liked less), possibly explained by a believed taste/healthy trade-off. Laureati, Jabes, Russo, & Pagliarini, 2013 found that the positive halo effect may only work on certain consumers, specifically those with high interest in sustainability. Thus, as concluded in a prominent review on the topic (Piqueras-Fiszman, & Spence, 2015), more research is still needed to fully understand the ef-fects of organic labelling. For one, inconsistencies in the literature suggest that the effect of organic labelling on the perceptual experience may be product- dependent. Since most of existing research comes from the food domain, a related question is whether the perceptual halo from organic labelling extends food to non-food agricultural products (e.g., textiles), whose product experience is defined by other sensory modalities than taste and flavour. Another open question pertains to how (which) in-dividual differences are relevant to explaining organic halo effects. In theory, for (organic) labelling to exert a positive effect on the perceptual experience, it is necessary that the consumers endorse the value that the label symbolizes (cf. Allen, Gupta, & Monnier, 2008). Therefore, in-dividuals with pro-environment attitudes and with stronger beliefs about the benefits of organic farming (e.g. high sustainability con-sumers) should be more likely affected by the halo effect of organic labelling, a theory that has received support in at least two studies on yoghurt (Laureati et al., 2013) and cookies (Schuldt & Schwarz, 2010), but failed to replicate in other product categories (Apaolaza et al., 2017; Lee et al., 2013). Moreover, different consumers may have different reasoning behind their preference for organic products. In the case of coffee, for instance, high sustainability consumers base their willingness to pay a premium on environmental concerns, whereas low sustain-ability consumers are driven by taste preference (S ̈orqvist et al., 2013). Moreover, there may potentially be differences in consumer attitudes whether the organic products are produced in the same country or im-ported; by purchasing local organic products, consumers are protecting the soil health of their own country, whereas purchasing imported organic products (such as coffee for Europe-based consumers) would be protecting the soil health in the country of production. Situated within this context, this short communication addresses the following two research questions (RQs): Does organic labelling lead to differences in sensory perception and willingness-to-pay for food and non-food products? (RQ1). If so, does the effect depend on whether consumers ascribe positive attributes to organic products? (RQ2). To address them, we conducted two sets of studies focusing on organic cotton (in the form of stuffed toys) and wine (Chianti DOCG). The inclusion of two separate studies was motivated by increasing the generalizability of the findings, and the two specific product categories were chosen to represent different sensory modalities (vision and touch for cotton toys, versus the chemical senses for wine) and motivation (utilitarian for toys, hedonic for wine) during the product experience. Moreover, our study is novel in that there is a dearth of research considering consumer perception of organic cotton. On the other hand, organic labelling in wine has shown a halo effect, whereby the identical wine receives a more positive sensory evaluation and higher willingness to pay in New Zealand (Forbes, Cohen, Cullen, Wratten, & Fountain, 2009), Germany (Gassler, Fronzeck, & Spiller, 2019; Wiedmann et al., 2014) and Spanish (Apaolaza et al., 2017) consumers as well as with American wine critics (Delmas et al., 2016). That said, the Danish market presents a novel context as both cotton and wine are, for all intents and purposes, imported products for Danish consumers. In addition, both grape and cotton are agricultural products with high pesticide usage (Ferrigno et al., 2017; Mailly et al., 2017), as well as major commodities, thus a better understanding of consumer perception towards these imported product categories is also consequential from a practical perspective. 2. Method 2.1.Participants and design A convenience sample of 224 Danish participants were recruited for four separate studies (see Table 1). Participants for Study 1A and 2A were recruited at a local zoo in autumn 2019, participants for Study 1B were recruited at a food festival in autumn 2019, and participants for study 2B were recruited at a workplace canteen in autumn 2020. All participants gave their informed consent in writing before the study took place. In Study 1, participants were asked to evaluate two cotton toys (1A) or wine samples (1B). Participants were instructed that one product was as organic whereas the other as conventionally produced/farmed. Unbeknownst to the participants, both products were identical. In Study 2, participants were again given two (unbeknownst to them) identical products to evaluate, but were told which was organic only until after they had made their initial preference decision. The design of both studies thus included one within-participant factor (claim: organic vs. conventional); Study 2 additionally included one between-participant factor (preference: organic preference vs. conventional preference). 2.2.Product category selection In studies 1A and 2A, a toy elephant (20 cm in height) produced from certified organic cotton was selected among the product portfolio of a partner company (Nature Planet ApS, Middelfart, Denmark) specializing in quality toys and accessories for zoos, aquariums, museums and family parks. At the time the study was conducted (fall 2019) the company only produced toys from polymer-based fabrics, with their organic cotton line being introduced to market shortly thereafter (2020). In both Study 1B and 2B, we selected a red wine from the Chianti Classico DOCG (guaranteed and controlled denomination of origin) in the region of Tuscany in Italy. Due to its warm and dry climate, Tuscany is ideal for organic viticulture (Fielden, 2009). Moreover, most Danish consumers are familiar with the region, with Italy being the leading supplier of wine to Denmark by value (Eurostat, 2020). In Study 1B, we used Castello di Querceto Chianti Classico 2015. We were unable to access the same wine for study 2B, but instead used La Misse di Candialle Chianti Classico 2015 to match the vintage and region of origin. Each wine sample contained 25 mL and was served in 21.5 cl ISO glasses. 2.3.Procedure 2.3.1.Study 1 Participants received two products (Study 1A: cotton toy, Study 1B: wine) in random order. One product was labelled as organic and the other conventional. This information was either verbally relayed by the experimenter (Study 1A) or presented on an iPad running a question-naire programmed in Qualtrics (Study 1B). Participants evaluated each product in terms of their preference, attribute pleasantness (texture in Study 1A, flavour in Study 1B), and willingness to pay (WTP). Attribute pleasantness was measured on a 1–7 scale from 1=”not very pleasant” to 7=”very pleasant”. For study 1A, WTP was anchored by a reference price of 200 DKK, corresponding to the average price of cotton toys in Table 1 Demographic information of participants. Product N % Female Mean Age (SD) Study 1A Toy 59 61% 50 (14) Study 1B Wine 75 51% 37 (14) Study 2A Toy 33 49% 42 (14) Study 2B Wine 57 35% 41 (14) Q.J. Wang et al.
Food Quality and Preference 96 (2022) 1044053that size from the producer. For study 1B, WTP was anchored by the statement “the average price of a bottle of red wine from Tuscany is about 100 kr”. Finally, participants were asked to respond to the following questions as a measure of their organic purchasing attitude (adopted from S ̈orqvist et al., 2013). “How often do you buy organic products?”, “How important is it to you to buy organic alternatives?” and “Do you feel guilty when you buy non-organic alternatives?”. Re-sponses from these three questions were averaged to create an index of organic product purchase tendency (α =0.78), and participants were classified as high- or low- organic tendency based on a median split of the index value. Data collection was carried out with a paper ques-tionnaire (Study 1A) or iPad (Study 1B). 2.3.2.Study 2 The procedure was identical to Study 1, with the exception that participants first indicated which product they preferred. Then, half the participants were randomly assigned to the “organic-preference” con-dition, where they were told that their preferred product was organic. The other half were in the “conventional-preference” condition, where they were told that their preferred product was conventional. Partici-pants then evaluated both the products in the same way as in Study 1, and organic product purchase tendency was once again calculated (α =0.84). 3. Results 3.1.Study 1 3.1.1.Product preference For cotton, 71% participants (N =59) preferred the organic labelled toy, 10% preferred the conventional label, and 19% did not have a preference. For wine, while 39% of participants (N =75) preferred the organic labelled wine, 40% preferred the conventional label, and 21% did not have a preference. There were no differences between low- and high- organic tendency participants when it came to preference for cotton toy (χ2 =3.72, p =.155) or for wine (χ2 =2.19, p =.335). 3.1.2.Attribute pleasantness For cotton products (Fig. 1A), a repeated measured ANOVA with label (organic vs. conventional) as within-participant factor and organic tendency (low vs. high) as between-participant factor revealed no sig-nificant main effect of organic tendency (F(1,57) =0.63, p =.429, ηp2 =0.01), but a significant main effect of label (F(1,57) =10.03, p =.002, ηp2 =0.15), which was driven by a significant interaction effect between label and organic tendency (F(1,57) =9.12, p =.004, ηp2 =0.14). Specifically, the high organic tendency consumers perceived the organic toy as more pleasant to the touch compared to the conventional toy (Morganic =6.14 (SE =0.22), Mconventional =4.79 (SE =0.29), p <.005). In contrast, when it comes to wine (Fig. 1B), there were no differences in flavour pleasantness of wines in terms of label (F(1,73) =0.14, p =.705, ηp2 =0.002), organic tendency (F(1,73) =1.78, p =.185, ηp2 =0.02), or interactions (F(1,73) =0.88, p =.352, ηp2 =0.01). 3.1.3.Willingness to pay For cotton (Fig. 1A), a similar ANOVA analysis revealed a significant main effect of label (F(1,57) =36.13, p <.005, ηp2 =0.39), where participants were always willing to pay more for organic cotton (Mor-ganic =183 DKK (SE =6.96), Mconventional =143 DKK (SE =6.09), p <.005); there were no main effect of organic tendency (F(1,57) =1.28, p =.263, ηp2 =0.02) or interactions (F(1,57) =3.63, p =.062, ηp2 =0.06). When it comes to wine (Fig. 1B), there were no main effects of label (F(1,73) =1.30, p =.258, ηp2 =0.02) or organic tendency (F (1,73) =0.24, p =.627, ηp2 =0.003), but there was a significant interaction effect (F(1,73) =6.29, p =.014, ηp2 =0.08) where only high organic tendency participants were willing to pay more for organic wine (Morganic =104 DKK (SE =5.23), Mconventional =94 DKK (SE =4.43), p =.015). 3.2.Study 2 3.2.1.Attribute pleasantness For cotton (Fig. 2A), a repeated measured ANOVA with label (organic vs. conventional) as within-participant factor, and preference condition (organic-preference vs. conventional-preference) and organic tendency (low vs. high) as between-participant factors revealed no sig-nificant main effects of any factors (label condition, F(1,29) =0.73, p =.399, ηp2 =0.025; preference condition, F(1,29) =1.85, p =.184, ηp2 =0.06; organic tendency, F(1,29) =4.02, p =.055, ηp2 =0.12) nor any interaction effects. In contrast, results from wine (Fig. 2B) validated the experimental manipulation whereby participants in the organic Fig. 1.Mean results from Study 1 for (A) cotton and (B) wine. Error bars indicate standard errors of the mean. Asterisks * indicate significant differences at p <.05 (Bonferroni-corrected post-hoc pairwise comparisons). Q.J. Wang et al.