Like throwing a piece of me away: How online and in-store grocery purchase channels affect consumers’ food waste

Journal of Retailing and Consumer Services 41 (2018) 20–30

Contents lists available at ScienceDirect

Journal of Retailing and Consumer Services journal homepage: www.elsevier.com/locate/jretconser

Like throwing a piece of me away: How online and in-store grocery purchase channels affect consumers’ food waste☆

T

Veronika Ilyuk Frank G. Zarb School of Business, Hofstra University, 121 Weller Hall, Hempstead, NY 11549, United States

A R T I C L E I N F O

A B S T R A C T

Keywords: Food waste Online grocery shopping Effort Psychological ownership

Consumer food waste is a significant and growing concern. As such, researchers, practitioners, and policy makers have devoted increasing attention to identifying the driving factors of this consequential consumer behavior. The present research contributes to this body of knowledge by uniquely showing how grocery purchase channels (i.e., online vs. in-store) may differentially affect food waste likelihood. Indeed, online grocery shopping is projected to rise considerably in the near future and warrants attention as a catalyst to both (normatively) positive and negative downstream outcomes. Using an experimental approach, the current research proposes and demonstrates that waste likelihood is higher (vs. lower) when consumers purchase food items online (vs. instore), and further explores the psychological mechanism underlying this effect. Three studies collectively provide evidence that online channels systematically yield lower perceptions of purchase effort, thereby reducing experiences of psychological ownership and, in turn, increasing consumers’ intentions of discarding (vs. consuming) purchased food items. Theoretical and practical implications of these findings are discussed.

1. Introduction The growth of e-commerce—and specifically the introduction of online stores by traditionally brick-and-mortar retailers (i.e., omnichannel retailing)—is one of the most transformative changes in the field of marketing to have taken place over the past several decades. While surprisingly still in its infancy compared to other consumer markets, online grocery shopping is forecasted to experience considerable growth in the near future; currently, 3% of total U.S. grocery spending occurs online and is expected to rise by 13% annually through 2023 (Steiman, 2014). Contributing to our understanding of how digital technologies will shape the retail landscape, a recent Nielsen survey of 30,000 online consumers spanning 60 countries found that nearly 25% of respondents already order grocery products online for home delivery, and 55% are willing to do so in the future (Nielsen, 2015). Without doubt, food retailers’ recent prioritization of their online platforms has fueled a new stream of consumer research. Though still relatively sparse, research in this domain has fruitfully addressed how and why consumers allocate their grocery purchases between the online and in-store grocery channels (Campo and Breugelmans, 2015), examining channel-driven purchase behavior differences such as consumers’ likelihood of buying certain categories/products (e.g., that differ in sensory properties or bulkiness; Elms et al., 2016; Degeratu et al., 2000; Chintagunta et al., 2012; Chu et al., 2010). Other work has

focused on the effects of consumer values (e.g., openness to change) on attitudes towards online grocery shopping (Hansen, 2008) and retailspecific factors like virtual grocery store layouts on consumers’ perceptions of usefulness and shopping time (Vrechopoulos et al., 2004). Past research has also delved more deeply into exploring hurdles in adoption (e.g., loss of the experiential aspect; Geuens et al., 2003) and the driving factors of online store choice—investigating, among other variables, the role of consumers’ experience with online grocery shopping (Melis et al., 2015). Notably, these research inquiries have collectively addressed the antecedents—rather than the behavioral consequences—of online food purchases. However, several important questions remain. Does postpurchase consumer behavior (i.e., during the consumption stage) differ when consumers make food purchases in-store versus online? If so, how and why? The present research takes a step in this direction by investigating a behavioral outcome that has recently received much attention from academics, practitioners, and policy makers alike: consumers’ food waste. Indeed, there is growing concern worldwide about consumers’ decisions to discard of (vs. consume) food items (Sirieix et al., 2017; Block et al., 2016; Porpino, 2016; Parfitt et al., 2010; Visschers et al., 2016; Stancu et al., 2016). As the Food and Agriculture Organization (FAO) of the U.S. (2016) notes, food waste amounts to "a major squandering of resources, including water, land, energy, labor and capital and needlessly produce greenhouse gas emissions,

☆

This research was supported by the Summer Research Grant from the Frank G. Zarb School of Business. E-mail address: [email protected].

https://doi.org/10.1016/j.jretconser.2017.11.003 Received 26 May 2017; Received in revised form 13 October 2017; Accepted 6 November 2017 Available online 21 November 2017 0969-6989/ © 2017 Elsevier Ltd. All rights reserved.

Journal of Retailing and Consumer Services 41 (2018) 20–30

V. Ilyuk

undernourished worldwide and food insecurity remains a large problem (FAO, IFAD, and WFP, 2014). Given that consumer food waste has is problematic at both the consumer and societal levels (e.g., in terms of greenhouse gas emissions; FAO, 2016; Porpino, 2016; Visschers et al., 2016; Stancu et al., 2016), emerging research has sought to identify the antecedents to consumer food waste. Prior research has shown, for example, that food waste occurs because of factors such as excessive purchasing (and stockpiling), overpreparation, avoidance of leftovers, disorganized food storage, poor knowledge of food preservation methods, and misconceptions about food safety (Porpino et al., 2015; Farr-Wharton et al., 2014; Stancu et al., 2016; Parfitt et al., 2010; Cicatiello et al., 2016). Research further suggests that the low cost of food in developed nations, “commercial pressures” (i.e., value-pricing that induce impulse or “super-sized” buying), and lack of education on expiration contribute to food waste (Godfray et al., 2010; Tsiros and Heilman, 2005; Haws and Winterich, 2013; White et al., 2016; Wansink and Wright, 2006). Moreover, demographic and cultural factors, such as consumers’ income, age, and household size/composition, have also been highlighted as possible sources (Parfitt et al., 2010; Visschers et al., 2016; Stancu et al., 2016). Of note, the limited research inquiries taking a chiefly theoretical approach to explaining food waste often rely on the Theory of Planned Behavior (Ajzen, 1991) or some modified version thereof (Visschers et al., 2016; Graham-Rowe et al., 2015; Stancu et al., 2016), largely considering food waste to be under consumers’ volitional, conscious control (see Block et al. (2016) for discussion). Graham-Rowe et al. (2015) show, for instance, that favorable explicit attitudes towards waste reduction, positive subjective norms, and perceived behavioral control (i.e., confidence that waste is avoidable) are associated with greater intentions to reduce household fruit/vegetable waste. Since consumers are often averse to waste and unused utility in many contexts and increasingly engage in environmental-friendly behaviors (Sirieix et al., 2017; Bolton and Joseph, 2012; Haws et al., 2012), it is likely that much of food waste is indeed caused by intentional acts. In other words, reasons for waste are often consciously derived and, at least in part, driven by explicit attitudes towards food waste. However, important to the present research is the idea that much of consumer behavior is also driven by factors outside of consumers’ awareness (Luomala et al., 2017). Despite the few inquiries that have found that contextual factors (i.e., visual cues like plate size, color, and disposability; Van Ittersum and Wansink, 2012; Williamson et al., 2016) unintentionally affect consumers’ food waste, the psychological underpinnings of food waste behavior—namely the catalysts that operate subconsciously and those that might not align with explicit attitudes (i.e., “food waste is bad”)—are not well understood (Block et al., 2016). As such, scholars have recently called for research on such causes (Porpino, 2016). The present research answers this call by exploring how food waste likelihood might stem from the purchase channel that consumers use, and explores the underlying psychological process: perceived effort and consequent manifestations of psychological ownership.

contributing to global warming and climate change." That is, upwards of 20% of land, 4% of energy, and 25% of water is used to produce food that ends up in the garbage—undoubtedly resulting in the futile use of valuable resources (Siegel, 2014). Moreover, when food that has been thrown away by consumers is hauled off to a landfill (constituting yet another energy expense), it ultimately breaks down to methane, an extremely potent greenhouse gas (Siegel, 2014). Indeed, consumer food waste has major detrimental economic, social, and environmental costs. The present research proposes and demonstrates that online food purchases can effectively increase consumers’ likelihood of wasting food—and explores the psychological mechanisms underlying this phenomenon. Central to the current work is the notion that the primary benefit offered by online food shopping is convenience. In contrast to physical supermarkets and grocery stores, online channels offer consumers value through time and effort savings (Campo and Breugelmans, 2015; Chu et al., 2010; Warschun, 2012; Chintagunta et al., 2012; Elms et al., 2016). In fact, competitive advantages in online food retailing require innovation in making the online buying process as easy and effortless as possible (Nielsen, 2014). Hence, one important dimension on which online and offline channels differ is the consumer effort inherent in making purchases. Building on the waste, consumer effort, and psychological ownership literature, the present research demonstrates, using an experimental approach, that reduced effort put forth in online (vs. in-store) grocery purchases may reduce psychological ownership. This, in turn, increases the likelihood that consumers throw away (vs. consume/use) food items that have been purchased through online (vs. in-store) channels. Accordingly, this work makes several important contributions. First, it uniquely implicates purchase channels (online vs. instore) as a source of food waste. Second, it demonstrates that two experiences—feelings of effort and psychological ownership—differentially emanate from these purchase channels and account for consumers’ intentions to discard of food items. Thus, in addition to filling the aforementioned gap in the online food shopping literature by identifying potential behavioral consequences of online food purchases, this research contributes to the burgeoning literature on the catalysts of consumer and household food waste—which is important, both theoretically and practically. The remainder of the paper is organized as follows. Section 2 presents a review of extant research on consumer food waste, effort, and psychological ownership. Section 3 builds on this previous work and includes a development of hypotheses, conceptual model, and an overview of the current studies. Sections 4–7 present the methodology and results of a pretest and three experiments testing the effects of purchase channel (online vs. in-store) on food waste intentions/behavior. Finally, in Section 8, the theoretical and practical implications of the findings—along with directions for future research—are discussed. 2. Theoretical background 2.1. Consumer food waste

2.2. Psychological ownership and the role of perceived effort The statistics on global food waste are staggering. Worldwide, about a third of all food produced for human consumption is wasted; in the U.S., upward of 40% of food is discarded (FAO, 2016; UNEP, 2015). Although notably food waste occurs at multiple stages of the food supply chain (i.e., by the producers, processors, and food grocers/service retailers; Cicatiello et al., 2016), significant losses occur in domestic settings (at the hands of the consumer); that is, in industrialized countries where infrastructure is developed and technical constraints in production, packaging, and shipment are limited, food is wasted primarily at later stages of the food supply chain—increasingly at the consumer (household) level (FAO, 2016; Godfray et al., 2010; Parfitt et al., 2010). Astonishingly, despite the tremendous amount of food waste occurring every year, 1 in 9 people are chronically

Psychological ownership is defined as cognitive-affective state in which people develop feelings of ownership of a variety of objects (“It is mine!”; Pierce et al., 2003). Prior research has shown that people often feel a connection between the self and their possessions, such that possessions can even become part of one's identity or the “extended self” (Belk, 1988; Dittmar, 1992). Indeed, objects that are owned are generally perceived as more valuable than objects that are not owned (Ericson and Fuster, 2014). Interestingly, psychological ownership differs from, though is related to, legal ownership; people can feel psychological ownership of objects that they never actually own, or never feel psychological ownership of possessions under their legal possession (i.e., when a target, such as a home or car, “never seems to belong to 21

Journal of Retailing and Consumer Services 41 (2018) 20–30

V. Ilyuk

3. Development of hypotheses

me”; McCracken, 1986). Thus, consumers can develop strong or weak feelings of psychological ownership towards products that they purchase (i.e., pay for and physically take hold of). Psychological ownership can emerge via three primary routes: controlling a target (e.g., having the ability to use and manage/determine the use of an object), coming to know the target intimately (e.g., having a lived relationship with, or getting to know/experience, an object), and investing the self in the target (e.g., putting forth effort and energy into the production or acquisition of an object; Pierce et al., 2003). The latter route is especially pertinent to the present research. Pierce et al. (2003) note that “As part of his political philosophy, Locke (1690) argued that we own our labor and ourselves, and therefore we are likely to feel that we own that which we create, shape, or produce. Through our labor, we invest not only our time and physical effort but also our psychic energy into the product of that labor. Sartre (1943/ 1969) even suggested that buying an object is simply another form of creating an object in that it too stems from the fruits of our labor” (p. 93). Thus, investing one's time, physical effort, and cognitive effort can effectively induce feelings of psychological ownership (Csikszentmihalyi and Rochberg-Halton, 1981). The positive link between effort and psychological ownership, and more broadly between effort and increased object valuation, has been shown in a variety of contexts. For example, laboratory rats and pigeons often exhibit preference for food that they have earned by pressing levers in a Skinner box (vs. food freely available elsewhere; Ellis, 1985). Similarly, in marketing settings, the literature on co-creation, self-design/assembly, and mass customization largely shows that consumers over-value self-made (effort-laden) products. For instance, consumers’ engagement during the product realization stage (where they invest physical effort into constructing a product) and design stage (where they invest mental effort, such as thoughts and ideas into the derivation of a product) has been found to enhance evaluations of self-made products (Atakan et al., 2014). Importantly, this effect has been attributed to associative self-anchoring (Gawronski and Bodenhausen, 2006) wherein self-productive activities lead to the establishment of ties between the self and the self-produced outcome (i.e., self-integration; Troye and Supphellen, 2012), in line with the psychological ownership account. Many related marketing inquiries have explored this general phenomenon (Buechel and Janiszewski, 2014; Franke and Schreier, 2010; Fuchs et al., 2010). Even when consumers put forth effort and labor into merely assembling products rather than customizing them (e.g., building IKEA boxes or folding origami), they too tend to over-value their creations (Norton et al., 2012). Thus, the positive effect of effort holds even despite the fact that products are never actually tailored to one's preferences, suggesting that mere labor can increase the subjective value of products (Norton et al., 2012). Though admittedly there may be other situations when invested effort does not yield higher product valuations—and may even yield lower valuations, in line with standard economic theory where effort is a cost (vs. benefit) and provides disutility during the product acquisition process (see Franke et al. (2010) for discussion)—a large body of evidence from consumer psychology and marketing suggests that effort can effectively have positive correlates with important downstream consequences (e.g., purchase intentions and willingness to pay). The present research adds to this growing body of research by empirically exploring another crucial consequence of effort perceptions and subsequent psychological ownership: consumers’ food waste.

As discussed previously, online (vs. in-store) grocery purchase channels likely yield lower (vs. higher) perceptions of invested effort, as the former are characterized by convenience: time and energy reduction. Further, given the relationship between effort and psychological ownership, it is hypothesized that lower effort perceptions will reduce psychological ownership which, in turn, will increase consumers’ food waste likelihood. Consequently, online (vs. in-store) channels should increase waste likelihood—operationalized in the present research as either intentions or decisions to throw away (rather than use/consume) food items. Formally stated, H1:. Waste likelihood will be higher (vs. lower) when grocery purchases are made through an online (vs. in-store) channel. If perceived effort indeed underlies food waste likelihood, then adjusting the level of effort inherent in these purchase channels should mitigate the effect stated in H1. More specifically, H2:. Decreasing the effort associated with in-store grocery shopping (H2a) or increasing the effort associated with online grocery shopping (H2b) will attenuate the effect such that there will be no differences in waste likelihood between purchase channels. Importantly, as described in Section 2.2, effort should have a positive effect on psychological ownership: a cognitive-affective state. The cognitive component can manifest in acts of citizenship, personal sacrifice/assumption of risk and, importantly, proactive responsibility for the target (as the assumption of accountability and obligation are common displays of this construct); the affective component, then, should come into play, as the assumption of responsibility for a target object heightens emotional responses to the possibility of losing it (i.e., leaving one's possession; Pierce et al., 2003). As such, in the present research, psychological ownership is operationalized as an emergent sense of responsibility for (cognitive component), and subsequent negative affect towards parting with (affective component), the target food items that consumers contemplate consuming or discarding. If, as hypothesized, psychological ownership proceeds effort perceptions and precedes consumers’ waste likelihood, then it should manifest in (1) higher levels of assumed responsibility for the food purchase and (2) feeling bad about disposal, when purchases are made through in-store (vs. online) channels. H3:. Perceived effort and psychological ownership (e.g., manifested by an increased sense of responsibility for, and negative feelings towards the idea of parting with, the product) will mediate, in sequence, the effect of purchase channel on food waste likelihood. See Fig. 1 for the conceptual model. A pretest and three experiments were conducted to test H1-H3, capturing both waste intentions (scenario-based studies 1–2) and actual waste behavior (laboratory study 3). In all studies testing the effects of purchase channel on food waste intentions, liking of the food product and product retention tendency (PRT; Haws et al., 2012) are included as covariates in the model, as these two variables can affect consumers’ propensities to retain (vs. waste) grocery items. PRT is positively associated with waste avoidance tendencies (frugality, creative reuse, environmental concern) and product attachment (Haws et al., 2012). Thus, PRT is measured and included in data analyses to ascertain the Fig. 1. Conceptual Model.

22

Journal of Retailing and Consumer Services 41 (2018) 20–30

V. Ilyuk

variance in waste likelihood that is explained by the focal independent variable—namely purchase channel—above and beyond that explained by this lifestyle trait.

Table 1a Perceived effort and feelings of responsibility for in-store versus online produce purchases. Consumer Activity

4. Pretest: Perceived effort and feelings of responsibility for negative outcomes (1) Hand-selecting produce (e.g., fruits and vegetables) at a supermarket (2) Ordering produce (e.g., fruits and vegetables) online from a supermarket's website for home delivery (3) Buying a pre-packaged bag of produce (e.g., fruits and vegetables) at a supermarket

Fifty-nine participants (Mage = 38.02, 20–75; 46% female) from Amazon Mechanical Turk completed this pretest in return for monetary payment. They were asked to consider various consumer activities such as purchasing food items (e.g., fruits and vegetables, snacks, take-out meals) in a physical retail location or online (see Tables 1a-1c). For each activity, they were asked to indicate how much “physical effort (physical work/energy),” “cognitive effort (mental work/energy such as thought and consideration),” and time each requires (1 = Not much at all, 7 = A great deal). These three items were combined into an “effort perceptions” measure. Subsequently, for each activity, participants were asked “how responsible you would feel if the outcome is worse than expected (e.g., the product isn’t as good as you had hoped, for example in terms of taste, texture, appearance, and general quality?”) (1 = Not at all, 7 = Very). The activities were presented in random order for each question. Finally, participants were asked to provide demographic information. Overall, paired samples t-tests revealed that in-store food purchases are seen as significantly more effortful than online food purchases (see Tables 1a-1c for results).1 In addition, participants indicated feeling significantly more responsible for a negative outcome when making the food purchase in-store than when ordering food online for home delivery—a common manifestation of psychological ownership. Given that online purchase channels yield lower perceptions of effort, and the proposed link between invested effort and consumers’ waste likelihood as described in the theoretical background section, study 1 was designed to test (1) the main hypothesis that waste intentions will be higher when grocery purchases are made through an online (vs. instore) channel (i.e., H1) and (2) that decreasing the effort associated with in-store grocery shopping should attenuate the effect of purchase channel on waste intentions, if one exists, such that there will be no differences in waste likelihood between purchase channels (i.e., H2a).

Paired [Items [Items [Items Paired [Items [Items [Items

Effort Composite

Feelings of Responsibility

α; Mean (SD)

Mean (SD)

.77; 3.88 (1.26) .73; 2.93 (1.11)

5.27 (1.42)

.84; 2.74 (1.23)

3.83 (1.54)

3.64 (1.68)

samples t-test results (Effort). (1) and (2); t(58) = 5.11, p < .001]. (1) and (3); t(58) = 7.90, p < .001]. (2) and (3); t(58) = −1.28, p = .21]. samples t-test results (Feelings of Responsibility). (1) and (2); t(58) = 6.20, p < .001]. (1) and (3); t(58) = 7.32, p < .001]. (2) and (3); t(58) = .82, p = .42].

Table 1b Perceived effort and feelings of responsibility for in-store versus online snack purchases. Consumer Activity

(1) Selecting a snack (e.g., crackers and cookies) in person at a supermarket (2) Selecting a snack (e.g., crackers and cookies) online from a supermarket's website (for home delivery)

Effort Composite

Feelings of Responsibility

α; Mean (SD)

Mean (SD)

.76; 3.40 (1.21) .77; 2.84 (1.13)

4.63 (1.46) 3.61 (1.45)

Paired samples t-test results (Effort). Items (1) and (2); t(58) = 3.57, p = .001. Paired samples t-test results (Feelings of Responsibility). Items (1) and (2); t(58) = 4.94, p < .001.

5. Study 1: Wrinkly red bell peppers—Testing the effect of purchase channel on waste intentions

Table 1c Perceived effort and feelings of responsibility for in-store versus online take-out purchases.

The purpose of study 1 was to test H1, namely that waste intentions are higher when grocery purchases are made through an online (vs. instore) channel. In addition, this study sought to explore whether, as hypothesized, effort perceptions indeed underlie the effect of purchase channel on waste intentions. If so, then decreasing the effort level associated with in-store purchases should attenuate the effect (if one exists), such that waste intentions should not significantly differ as a function of purchase channel. As shown in the pretest, effort perceptions can be effectively reduced in-store when consumers buy prepackaged (vs. individually-selected) produce. Thus, study 1 also tests H2a by including a condition in which consumers imagine shopping in a physical grocery store, but reduce their effort investment by purchasing pre-packaged food items.

Consumer Activity

Ordering food in person at a takeout restaurant (e.g., pizzeria) to bring home Ordering food online from a takeout restaurant (e.g., pizzeria) for home delivery

Effort Composite

Feelings of Responsibility

α; Mean (SD)

Mean (SD)

.80; 3.63 (1.24)

4.10 (1.54)

.81; 2.69 (1.23)

3.27 (1.46)

Paired samples t-test results (Effort). Items (1) and (2); t(58) = 6.04, p < .001. Paired samples t-test results (Feelings of Responsibility). Items (1) and (2); t(58) = 4.51, p < .001.

5.1. Design and procedure

1 The primary comparison of interest was between online and in-store food purchases, in general. However, purchasing pre-packaged produce in-store was also included as an activity for exploratory purposes (i.e., for use in a subsequent experiment), considering that pre-packaged alternatives are marketed as convenient, timesaving alternatives (Clute, 2008). Indeed, the results revealed that hand-selecting produce in-store is perceived as significantly more effortful than purchasing a pre-packaged bag of produce instore. However, there is no significant difference in effort perceptions between buying a pre-packaged bag of produce in-store and ordering the produce online.

One hundred and eighteen participants (Mage = 34.66, 20–69; 45% female) from Amazon Mechanical Turk completed this study in return for monetary payment. Participants were randomly assigned to one of three conditions: in-store (individual selection), in-store (pre-

23

Journal of Retailing and Consumer Services 41 (2018) 20–30

V. Ilyuk

packaged), or online produce purchase. Participants in the in-store (individual selection) condition read: “Imagine that several days ago, you went shopping at your local supermarket. You hand-selected and bagged produce, including 6 red bell peppers.” Participants in the instore (pre-packaged) condition read, “Imagine that several days ago, you went shopping at your local supermarket. You selected and purchased produce, including a bag of 6 pre-packaged red bell peppers.” Those in the online condition read, “Imagine that your local supermarket now offers home delivery. You can select the items that you want on the website and they are delivered straight to your door. Several days ago, you received a delivery that you had ordered online. An employee had hand-selected and bagged the produce you ordered, including 6 red bell peppers. Upon receipt, you checked your delivery and everything looked fine.” The latter statement was used to control for perceptions of product quality across conditions. Subsequently, all participants were told that, “Today—only several days after the purchase—you are planning to prepare a dish that includes red bell peppers. You notice that 2 of the peppers feel soft; the skin is starting to wrinkle. Though you know these 2 peppers are still fine to eat, they are starting to go bad. The recipe that you are making will require chopping and cooking the peppers, so their appearance and texture should not make a difference. However, the other peppers appear firm and fresh, and you have enough of them to make the dish.” Waste intentions—the main dependent variable—were assessed via a four-item scale (α = .95; 1 = Not at all, 7 = Very, with appropriately modified descriptors): “How likely are you to throw the two peppers away?”, “How likely are you to use only the firm peppers instead?”, “How likely are you to use the two peppers that have started to go bad in the dish?” (reverse-coded), and “How much do you feel like you should use the two peppers in the recipe?” (reverse-coded). Using the same 7-point scale, participants then indicated how negatively they would describe the outcome (i.e., “How “bad” did you imagine the two peppers to be?”) and their level of disappointment (“How disappointed are you in your purchase?”). Finally, participants indicated how often they buy produce at a supermarket, order produce online for home delivery, cook, buy red bell peppers, and like red bell peppers; they completed the PRT scale (α = .93) and provided their demographic information.

= 3.22 vs. Min-store[individual] = 2.14; p < .01). However, there was no significant difference in waste intentions between the in-store (pre-packaged) condition and the online condition (p = .82). PRT (F(2, 113) = 6.20, p = .01) and liking of the product (F(2, 113) = 3.95, p < .05) were significant covariates in the model; higher PRT scores and greater liking of the product yielded lower waste intentions. However, the results remained significant when these measures were not included in the analysis.

packaged

5.2.2. Other measures There were no significant differences between conditions in how “bad” participants imagined the two peppers to be (Min-store[individual] = 3.00 vs. Mpre-packaged = 3.15 vs. Monline = 3.05; F < 1) or in how disappointed participants were in the purchase (Min-store[individual] = 3.82 vs. Mpre-packaged = 3.95 vs. Monline = 3.47; F < 1). Thus, regardless of purchase channel, the wrinkly peppers yielded equal levels of dissatisfaction. This suggests that purchase channels (and invested effort) do not bias/alter perceptions of how negative the outcome is (i.e., "maybe the food isn't that bad") which, in turn, drives waste intentions, but instead affect some other psychological process that accounts for the given pattern of results (i.e., psychological ownership, addressed in the next study). Similarly, ANOVAs revealed nonsignificant effects of purchase channel on frequency of shopping for produce in-store (F(2, 115) = 2.03, p = .14) and online (F < 1), cooking (F(2, 115) = 1.65, p = .20), purchasing peppers (F < 1), and annual household income (F < 1). When included in the waste intentions analysis as covariates, these measures did not change the significance of the results. In support of H1, Study 1 demonstrates that online (vs. in-store) food purchases can systematically increase consumers’ intentions to waste food. This effect emerged despite the fact that the product was consumable and would not have a detrimental impact on the dish's final quality. Importantly, however, decreasing the effort associated with instore purchases via pre-packaged alternatives muted the effect, such that waste intentions no longer differed as a function of purchase channel. This provides support for H2a, underscoring the role of invested effort in consumers’ waste likelihood. It also demonstrates that pre-packaged produce may contribute not only to environmental waste (associated with additional packaging) but also to food waste—a point that is revisited in the general discussion section. The purpose of the next study is two-fold: (1) to provide further support for the effort-based account with a different food item and outcome-context and (2) to directly test via mediation analysis the hypothesized process through which the effect arises, by measuring both effort perceptions and psychological ownership.

5.2. Results and discussion 5.2.1. Waste intentions As hypothesized, an ANCOVA controlling for PRT and liking of the product revealed a significant overall effect on waste intentions (F(2, 113) = 4.62, p = .01; see Fig. 2). Pairwise comparisons showed that waste intentions were significantly higher among those who imagined purchasing the items online than those who imagined selecting individual items in-store (Monline = 3.13 vs. Min-store[individual] = 2.14; p < .05). Similarly, waste intentions were significantly higher among those who imagined purchasing the pre-packaged bag of produce instore than those who imagined selecting individual items in-store (Mpre-

6. Study 2: Bitter crackers—Increasing the effort of online purchases and testing the role of psychological ownership Just as reducing effort associated with in-store purchases attenuated the focal effect in the previous study, so too should increasing the effort associated with online purchases. In other words, if consumers are made to perceive their online purchase process as effortful, then their waste intentions should be similar to those in the in-store (physical grocery location) condition (H2b). Furthermore, study 2 seeks to provide empirical evidence for the underlying psychological process (H3) by explicitly gauging effort perceptions and two manifestations of psychological ownership: the assumption of responsibility for the outcome and subsequent negative affect towards parting with the purchase. As described in the theoretical background section, psychological ownership is a cognitive-affective state, and these two variables reflect the cognitive and affective components of psychological ownership, respectively (Pierce, 2003). In addition, a different food product and outcome context (snack food and unexpected negative taste) is used than in study 1 (produce and decreased freshness) to test the robustness of the focal effect.

Fig. 2. The effect of purchase channel on waste intentions (Study 1).

24

Journal of Retailing and Consumer Services 41 (2018) 20–30

V. Ilyuk

6.1. Design and procedure

[effortful] = 3.63; p = .001). Similarly, effort perceptions were significantly lower among those in the online (control) condition than among those in the in-store condition (Monline [control] = 2.56 vs. Minstore = 3.16; p < .05). However, there was no significant difference in effort perceptions between the online (effortful) and in-store conditions (p = .12).

One hundred and fifty participants (Mage = 36.85, 18–75; 48% female) from Amazon Mechanical Turk completed this study in return for monetary payment. Participants were randomly assigned to one of three conditions: in-store, online (control), or online (effortful) food purchase. Participants in the in-store condition read: “Imagine that you decide to go shopping at your local supermarket. Among other items, you hand-select a box of premium organic, multi-grain crackers that you have never tried before. Please take a moment to imagine shopping in the store and choosing a box of crackers. Imagine putting the item in your shopping cart.” Those in the online (control) condition read: “Imagine that your local supermarket now offers home delivery. You can select the items that you want on the website and they are delivered straight to your door. You decide to place an order on the store's website. Among other items, you select a box of premium organic, multi-grain crackers that you have never tried before. Please take a moment to imagine shopping online and choosing a box of crackers. Imagine putting the item in your virtual shopping cart.” Those in the online (effortful) condition read: “Imagine that your local supermarket now offers home delivery. You can select the items that you want on the website and they are delivered straight to your door. You decide to place an order on the store's website. Among other items, you select a box of premium organic, multi-grain crackers that you have never tried before. You put a lot of effort into choosing the crackers; you spend a lot of time comparing multiple brands. Please take a moment to imagine shopping online and choosing a box of crackers. Imagine putting the item in your virtual shopping cart.” Participants in both online conditions were told that “The items arrive on time; the order looks correct.” Next, all participants were asked to consider the purchase of the crackers and to indicate their effort perceptions using three scales (α = .94) anchored from (1 = Easy, 7 = Difficult; 1 = Fast, 7 = Slow; 1 = Effortless, 7 = Demanding). Afterwards, all participants read: “Now, imagine that several days later you open the box of crackers. You try one; unlike what you expected, it tastes very bitter. Eating the cracker is not very pleasant. You are disappointed with the purchase.” An image of multi-grain crackers appeared below the text. As in the previous study, waste intentions were assessed via a fouritem scale (α = .82; 1 = Not at all, 7 = Very): “How likely are you to throw the crackers away?”, “How much do you feel like you should keep the crackers?” (reverse-coded), “How much do you feel like you should eat the crackers?” (reverse-coded), and “How wasteful would throwing the crackers away be?” (reverse-coded). In the present research, two manifestations of psychological ownership were assessed: the assumption of responsibility for the outcome and negative affect towards disposal. With regard to the former (i.e., assumption of responsibility), participants were asked to indicate (1 = Not at all, 7 = Very; r = .64, p < .001): “To what extent do you feel responsible for the situation?” and “To what extent do you feel at fault for getting the bad-tasting crackers?” With regard to the latter, they were asked “How bad would you feel about throwing the crackers away?” Using the same 7-point scale, participants then indicated how negatively they would describe the outcome (i.e., “How bad did you imagine the crackers to taste?”) and their level of disappointment (“How disappointed are you in your purchase?”). Finally, participants indicated how often they buy produce at a supermarket, order produce online for home delivery, and like crackers; they completed the PRT scale (α = .92) and provided their demographic information.

6.2.2. Psychological ownership An ANOVA revealed a significant overall effect on assumption of responsibility (F(2, 147) = 3.15, p < .05). Pairwise comparisons showed that assumption of responsibility was significantly higher among those in the online (effortful) condition than among those in the online (control) condition (Monline [control] = 4.43 vs. Monline [effortful] = 5.05; p < .05). Similarly, assumption of responsibility was significantly higher among those in the in-store condition than among those in the online (control) condition (Monline [control] = 4.43 vs. Min-store = 5.08; p < .05). However, there was no significant difference in assumption of responsibility between the online (effortful) and in-store conditions (p = .93). In line with the previous results, an ANOVA revealed a marginally significant effect on negative affect that would arise due to disposal (F (2, 147) = 2.97, p = .06; see Fig. 3). More specifically, participants indicated feeling significantly worse about throwing away the product in the online (effortful) condition than in the online (control) condition (Monline [control] = 4.15 vs. Monline [effortful] = 4.96; p < .05). Likewise, they indicated feeling significantly worse about throwing away the product in the in-store condition than in the online (control) condition (Monline [control] = 4.15 vs. Min-store = 5.02; p < .05). However, there was no significant difference in negative affect towards disposal between the online (effortful) and in-store conditions (p = .88). 6.2.3. Waste intentions As hypothesized, an ANCOVA controlling for PRT and liking of the product revealed a significant overall effect on waste intentions (F(2, 145) = 5.47, p = .005; see Fig. 4). Pairwise comparisons showed that waste intentions were significantly higher among those in the online (control) condition than among those in the online (effortful experience) condition (Monline [control] = 4.33 vs. Monline [effortful] = 3.47; p = .005). Similarly, waste intentions were significantly higher among those in the online (control) condition than among those in the in-store condition (Monline [control] = 4.33 vs. Min-store = 3.47; p < .005). However, there was no significant difference in waste intentions between the online (effortful) and in-store conditions (p = .99). PRT (F(2, 145) = 18.42, p < .001) was a significant covariate in the model, while liking of crackers was not (F < 1); higher PRT scores yielded lower waste likelihood. Of note, the results remained significant when these measures were not included in the analysis.

6.2. Results and discussion 6.2.1. Effort perceptions An ANOVA revealed a significant overall effect on effort perceptions (F(2, 147) = 6.14, p < .005). Effort perceptions were significantly lower among those in the online (control) condition than among those in the online (effortful) condition (Monline [control] = 2.56 vs. Monline

Fig. 3. The effect of purchase channel on psychological ownership (Study 2).

25

Journal of Retailing and Consumer Services 41 (2018) 20–30

V. Ilyuk

6.2.5. Other measures There were no significant differences between conditions in how “bad” participants imagined the crackers to taste (F(2, 147) = 1.39, p = .25) or in how disappointed participants were in the purchase (F < 1). Similarly, ANOVAs revealed a marginally significant effect of purchase channel on frequency of shopping for produce online (F(2, 147) = 2.61, p = .08) and nonsignificant effects of purchase channel on frequency of shopping for produce in-store (F(2, 147) = 1.84, p = .16), purchasing crackers (F(2, 147) = 1.90, p = .15), and annual household income (F(2, 144) = 2.30, p = .10). When included in the waste intentions analysis as covariates, these measures did not change the significance of the results. Thus, using a different food product (snack food) and outcome context (unexpected negative taste), study 2 provides support for H2b and H3. When invested effort is increased in the online purchase channel, the focal effect of purchase channel on waste intentions is attenuated. Further, psychological ownership is shown to manifest in increased perceptions of responsibility for, and negative affect towards disposal of, the food purchase. Mediation analyses confirm that effort perceptions indeed underlie feelings of psychological ownership, which, in turn, drive consumers’ waste intentions. This study hence provides confirmatory evidence for the conceptual model presented in Fig. 1.

Fig. 4. The effect of purchase channel on waste intentions (Study 2).

6.2.4. Serial mediation Given that ANOVAs on waste intentions, effort perceptions, and both dimensions of psychological ownership revealed no differences between the in-store and online (effortful) conditions—but significant differences between these two conditions and the online (control) condition—an orthogonal contrast was used that compared the average of the in-store and the online (effortful) conditions with the online (control) condition. The dummy codes for the contrast were −1, −1, and 2 to indicate the in-store, online (effortful), and online (control) conditions, respectively. This was done to test serial mediation, namely the [purchase channel → effort perceptions → assumption of responsibility → negative affect towards disposal → waste likelihood] path, controlling for PRT and liking of the product (Hayes, 2013; PROCESS model 6). Indeed, a bootstrap test with 5000 resamples confirmed the sequential mediation process (see Table 2), as the confidence interval for the proposed path excluded zero (indirect effect = .0078; 95% CI, .0011 to .0261). Please note that mediation analyses testing each link separately (i.e., with one mediator at a time) using multicatgeorical coding and PROCESS model 4 also yielded significant results.

7. Study 3: Flavorless crackers—Capturing real waste behavior The purpose of study 3 was to replicate the effects of the previous study—namely that increased effort of online purchases effectively reduces subsequent food waste—but with actual consumption and waste behavior (versus waste intentions that were captured in the previous studies). Specifically, this lab study gave participants the opportunity to shop on a mock grocery website, select products, and receive a sample of a chosen food item. Unbeknownst to participants, the main dependent variable was their ultimate decision to keep or throw away (i.e., waste) the sample of food provided. This study also sought to use a different (more direct) measure of psychological ownership by asking the extent to participants felt the product is theirs (e.g., Fuchs et al., 2010; Van Dyne and Pierce, 2004; Peck and Shu, 2009) rather than by gauging psychological ownership through feelings of responsibility and affect. These two measurement approaches (in studies 2 and 3) were chosen to explore the underlying psychology of the focal effect through several different lenses.

Table 2 Mediation analysis results (Study 2). Variable

Purchase Channel (−1 = in store, −1 = online (effortful), 2 = online (control) PRT Liking of the Product Purchase Channel (−1 = in store, −1 = online (effortful), 2 = online (control) Effort Perceptions PRT Liking of the Product

Purchase Channel (−1 = in store, −1 = online (effortful), 2 = online (control) Effort Perceptions Assumption of Responsibility PRT Liking of the Product Purchase Channel (−1 = in store, −1 = online (effortful), 2 = online (control) Effort Perceptions Assumption of Responsibility Negative Affect Towards Disposal PRT Liking of the Product

β(se)

t

DV: Effort Perceptions −.27 (.09) −3.04** −.02 (.08) −.01 (.09) DV: Assumption −.20 (.08)

7.1. Design and procedure

−.22 −.14 of Responsibility −2.37*

Sixty-eight undergraduate students (Mage = 20.13, 18–25; 66% female) participated in this lab study in return for course credit. Participants were randomly assigned to one of two conditions: online (control) or online (effortful) food purchase. A programmer was hired to create a mock website that can be populated with product categories, product images, and product descriptions—thereby allowing flexibility in controlling which products participants are exposed to (and can select from), but at the same time creating a more real-world online shopping experience (versus, say, asking participants to simply make product choices on a screen); indeed, the mock website resembles that of national grocers. An "in-store" condition (e.g., participants approaching makeshift shelves in the lab) was not included in the experimental design, as this condition would necessarily lack the richness offered by the online platform (i.e., given the limitations in terms of setting/atmosphere, and the number of product categories and SKUs that can be stocked). This condition would not "match" the two online conditions, hence preventing a meaningful comparison between the groups. Prior to the experiment, the website was populated with 12 product categories—"Fresh Produce," "Deli," "Eggs & Dairy," "Breakfast & Cereal," "Frozen Meals," "Frozen Desserts," "Pasta, Grains, & Beans,"

.17(.07) 2.23* .29 (.07) 4.17** .01 (.08) .11 DV: Negative Affect Towards Disposal −.25 (.11) −2.24* −.02 (.10) −.19 .33 (.11) 2.96** .32 (.10) 3.20** .06 (.11) .56 DV: Waste Likelihood .08 (.07) 1.19 −.01 −.15 −.52 −.07 −.00

(.06) (.07) (.05) (.06) (06)

−.12 −2.25* −11.08** −1.15 −.03

* p < .05. ** p < .01.

26

Journal of Retailing and Consumer Services 41 (2018) 20–30

V. Ilyuk

cracker: " (1 = Worse than expected, 7 = Better than expected). They were asked to evaluate the crackers (1 = Not at all, 7 = Very much so) on flavor ("flavorful," "appetizing," "mouth-watering," and "delicious;" α = .84), texture ("crunchy"), aesthetics ("visually appealing"), and value ("worth the price"). At this point, participants were instructed to raise their hand, and indicate that they had completed the product evaluation task. The main dependent variable, however, was what participants decided to do with the remaining two crackers in the zip-lock bag thereafter. The experimenter approached each participant and asked, "Would you like to keep the crackers, or throw them away?" Each participant's decision (keep or throw away) was recorded. No participants expressed awareness that this decision (i.e., actual waste behavior) was part of the study. Participants were asked to complete the study by providing their responses to "How much effort did you put into your product selections?" (1 = Not much at all, 7 = A lot), general liking of crackers, and frequency of purchasing products, groceries, and food online. Finally, they indicated their familiarity with the cracker brands (Ryvita, Wasa, and Thinables), PRT (α = .91), and demographics.

"Gourmet Snacks," "Condiments & Sauces," "Candy & Sweets," "Beverages," and "Health & Beauty"—that were displayed as tabs across the top of the website (and on the left-hand side, vertically arranged, on the main page). Each product category was populated with between 16 and 28 unique items, displayed with 4 items per row. Each item contained a brand name, price, product image, and an "add to cart" button. This information (brand, images, etc.)—with the exception of the "crackers" (the target product, discussed subsequently)—was sourced from extant supermarkets' websites. Both well-known and less known brands were included. The "Gourmet Snacks" tab contained crackers, gourmet chocolate, nut mixes, and cookies. To limit participants' prior experience with, and preference for, the target stimulus (crackers), three brands of ostensibly gourmet (i.e., real but relatively unknown brands of crackers) were selected: Ryvita, Wasa, and Thinables. All of these brands contain an image of the crackers that resembles the crackers that participants actually received later in the study (which was held constant); the goal was to have participants select crackers based on subtle flavor variations, but believe that the crackers they later received for evaluation were indeed the ones they had selected. The packages, as such, were digitally manipulated to contain a flavor (e.g., "Hint of Salt & Pepper," "Hint of Garlic," "Sprinkle of Onion," "Mild Olive Oil Flavor," etc.); some of the other elements were also digitally manipulated (e.g., the color of several packages) to increase the perceived choice set. Accordingly, there were a total of 8 different cracker options to choose from on the website (see Appendix A). At the beginning of the experiment, all participants were asked to imagine shopping online for grocery items. They were given a shopping list such that the amount of products searched for, and selected, was held constant. All participants were then asked to browse the website and add these items to their shopping carts, as if they were actually making the purchase. However, those in the effortful purchase condition received additional instruction: "Please make sure to put significant thought and consideration into making your selections; you may be asked to explain why you added the particular items to your cart." Those in the control condition did not receive this information. Once participants finished adding the items to their carts, they were asked to open up their virtual shopping carts (as if they were proceeding to check-out), to consider the items in the cart, and to indicate their level of agreement with the following statements (used to capture psychological ownership): "Even though I haven't received the products yet, I feel like the items in the shopping cart are already mine," "I feel a sense of ownership of the products in my shopping cart," "The items in my shopping cart are a reflection of me," "I feel somewhat attached to the items in my shopping cart," and "It would be difficult for me to remove an item from my shopping cart" (1 = Strongly disagree, 7 = Strongly agree; α = .89). Thus, in this study, a different (more direct) measure of psychological ownership was used to provide further support for the underlying psychology. Next, participants were asked to raise their hand, such that the researcher could provide them with one of the items they selected for evaluation. After ostensibly noting which crackers were chosen by each participant, the experimenter went to the back of the room to get a ziplock bag containing three crackers from one of several different bins (each bin ostensibly contained different varieties/flavors). The reason why multiple bins were used (vs. all of the zip-lock bags put into one bin) was to facilitate perceptions that there were actually different crackers available for sampling. The goal was to have participants perceive the crackers that they actually received as the ones they had indeed selected on the website (i.e., not simply chosen by the researcher, or the same as those that everyone else received). However, all of the crackers (in all of the bins) were the same: bland, unseasoned, light rye. Each participant was handed the bag of their "chosen" crackers for evaluation. The survey instructions next asked participants to eat one of the crackers, and to answer "How would you describe the taste of the

7.2. Results and discussion 7.2.1. Target product evaluations An ANOVA revealed no significant difference between conditions in perceptions of the crackers' taste compared to expectations (Mcontrol = 2.97, Meffort = 2.85; F < 1). Likewise, ANOVAs revealed no significant differences between conditions in evaluations of the crackers' flavor (Mcontrol = 2.13, Meffort = 2.02; F < 1), texture (Mcontrol = 5.63, Meffort = 5.73; F < 1), aesthetic appeal (Mcontrol = 2.34, Meffort = 2.39; F < 1), or value (Mcontrol = 1.86, Meffort = 1.94; F < 1). One sample ttests revealed that responses to the taste, flavor, aesthetic appeal, and value measures were all significantly below the scale midpoint (i.e., 4; all p < .001), indicating that all participants perceived the productoutcome negatively. 7.2.2. Effort perceptions An ANOVA on the composite effort measure revealed that participants who were instructed to put significant thought and consideration into making the product selections expressed a higher level of invested effort than those in the control condition (Mcontrol = 4.03, Meffort = 5.18; F(1, 66) = 8.38, p = .005). 7.2.3. Psychological ownership As hypothesized, an ANOVA on the composite measure of psychological ownership revealed a significant effect, such that those in the effortful purchase condition felt a greater sense of psychological ownership towards the items in the virtual shopping carts (Mcontrol = 3.61, Meffort = 4.96; F(1, 66) = 17.06, p < .001). 7.2.4. Waste behavior The main dependent variable was a decision to keep or throw away the remaining crackers (0 = throw away, 1 = keep). As predicted, binary logistic regression analysis revealed a significant effect of condition (0 = control, 1 = effortful purchase) (b = 1.31, Wald χ2(1) = 4.81, p < .05). Thus, putting more thought and consideration at the product selection stage can effectively reduce subsequent food waste. 7.2.5. Mediation analysis To test whether psychological ownership drives food waste behavior, mediation analysis (Hayes, 2013; PROCESS model 4) was performed, using the bootstrapping procedure with 5000 resamples. As in the previous studies, PRT and general liking of crackers were included as covariates. As predicted, the 95% confidence interval (CI) for the path [effort condition [shopping instructions] → psychological ownership → waste decisions] excluded zero (indirect effect = 2.8474; 95% CI, .8913 to 9.4239; see Table 3). 27

Journal of Retailing and Consumer Services 41 (2018) 20–30

V. Ilyuk

8.1. Theoretical contribution

Table 3 Mediation analysis results (Study 3). Variable

Effort Condition [Shopping Instructions] (0 = Control, 1 = Effort) PRT Liking of the Product

Psychological Ownership Effort Condition [Shopping Instructions] (0 = Control, 1 = Effort) PRT Liking of the Product

β(se)

This research makes several important theoretical contributions. First, it introduces a previously unexplored antecedent to consumer food waste—grocery purchase channels (i.e., online vs. in-store)—thereby contributing to the burgeoning literature on the sources of such behavior. Second, two process (i.e., mediating) variables are shown to drive the effect of purchase channels on intentions/propensities to throw away food items: effort perceptions and psychological ownership. As such, the current research not only identifies other potential catalysts of consumers’ food waste—that can be fruitfully explored more deeply in future research—but also highlights several noteworthy behavioral outcomes of online and in-store grocery purchase channels. More specifically, as discussed in the introduction, prior research in this domain has focused on the antecedents to, rather than the behavioral consequences of, online food purchases. The present work uniquely examines post-purchase consumer behavior (during the consumption stage) and shows that it can systematically differ based on purchase channel. The findings thus also contribute to the growing literature on factors outside of consumers' awareness (e.g., Luomala et al., 2017) and the types of points of purchase (Lunardo and Saintives, 2013) that influence food-related behavior.

t/z

DV: Psychological Ownership 1.35 (.33) 4.13** .18 (.12) 1.56 −.04 (.11) −.35 DV: Waste Behavior (0 = Throw away, 1 = Keep) 2.11 (.64) 3.32** .27 (.85) .31 .34 (.31) .18 (.28)

1.09 .62

** p < .01.

Please note that effort perceptions in this study indeed reflect the instructions and hence double as a manipulation check. For this reason, the above-mentioned mediation analysis excludes the effort perceptions measure (given its "proximity" to the manipulated variable). However, serial mediation analysis (PROCESS model 6) incorporating self-reported effort perceptions revealed that effort perceptions indeed have a significant effect on psychological ownership (β = .32, t = 3.28, p < .01). Moreover, the path [effort condition (shopping instructions) → effort perceptions → psychological ownership → waste decisions] also excluded zero (indirect effect = .6355; 95% CI, .0709 to 2.7243). Accordingly, these results provide further support for the conceptual model in Fig. 1.

8.2. Practical implications This work also has several important practical/substantive implications. Online grocery purchase channels are expected to rise in the near future (Steiman, 2014; Warschun, 2012) and, accordingly, consumers are increasingly likely to use them (Nielsen, 2015). The present research suggests that these channels may have unintended social, economic, and environmental consequences because they effectively induce food waste. Indeed, food waste is a growing concern among researchers, practitioners, and policy makers alike (Sirieix et al., 2017; Block et al., 2016; Porpino, 2016; Parfitt et al., 2010; Visschers et al., 2016). Interestingly, industrialized nations (where internet penetration is high and online shopping is increasingly common) waste as much food as almost the entire net food production of sub-Saharan Africa (FAO, 2016). Though it is unlikely that such excessive waste is currently caused by online grocery purchases (and, in fact, online grocery shopping is still in its infancy worldwide and a host of other factors such as socio-economics likely play the largest role), this research presents the possibility that in the near future, geographic discrepancies in waste behavior may be in part due to internet access and the availability of online grocery purchase channels. Stated differently, this work highlights one factor that will likely contribute to particularly high waste behavior in developed (i.e., internet-ready) nations, suggesting that investment in consumer education programs and other policy initiatives may be warranted. Various methods have been proposed for waste prevention, such as encouraging consumers to use shopping lists, creatively reuse leftovers, and convert household food waste into its monetary equivalent (Schneider, 2008). Considering that consumers are unlikely to realize how online food purchases systematically affect effort perceptions, psychological ownership, and waste likelihood (i.e., the focal effect found in the current research likely operates outside of conscious awareness), initiatives to educate consumers about this phenomenon might be a worthwhile pursuit. Furthermore, study 2 uniquely provides evidence that pre-packaged produce effectively reduces effort perceptions and psychological ownership, and increases food waste intentions. This is an important finding, considering that sales of pre-packaged produce are continuously rising, and there are detrimental effects of superfluous packaging including increased (often unrecyclable) trash and "technology pollution" (Frank, 2017). The current research thus also contributes to the known negative effects of product packaging on the environment, hence warranting greater

7.2.6. Other measures ANOVAs revealed nonsignificant effects of effort condition (i.e., shopping instructions) on frequency of purchasing crackers (F < 1), liking of crackers (F < 1), frequency of purchasing products online (F < 1), frequency of making grocery purchases online (F < 1), frequency of making food purchases online (F(1, 66) = 2.15, p = .15), or familiarity with any of the three brands of crackers used in the study (all F < 1). When included in the waste behavior analysis as covariates, these measures did not change the significance of the results. Together, the results from study 3 reveal that putting forth thought and consideration into online food shopping can effectively increase consumers' attachment to the purchased items and, importantly, reduce actual food waste.

8. General discussion The present research demonstrates that waste likelihood is higher when food items are purchased via online (vs. in-store) grocery channels. Collectively, a pretest and three experiments provide empirical evidence that this effect is driven by lower effort perceptions associated with online purchases, which in turn decrease experiences of psychological ownership and, ultimately, increase intentions/decisions to discard of food items. Specifically, in brick-and-mortar grocery settings, consumers put forth more energy and time into their purchases (e.g., hand-selecting and bagging produce)—or at least perceive to do so; in online settings, much of the effort related to product acquisition is transferred to another party—namely a store's employees. Though seemingly an unequivocal benefit of online purchases, this convenience (i.e., effort reduction) is shown to negatively affect both psychological ownership (manifested in assumption of responsibility for, and negative affect towards disposal of, a purchase) and an increasingly concerning consumer behavior: food waste.

28

Journal of Retailing and Consumer Services 41 (2018) 20–30

V. Ilyuk

reduction). Might the effects documented in the present research be amplified when grocery purchases are made via mobile devices? It would be interesting to explore the focal effect for online food purchases, in general, beyond grocery items (i.e., online food delivery (OFD) services; Yeo et al., 2017). Many companies now offer consumers the ability to order from their favorite local restaurants through an easy-to-use, convenient app. The pretest in the present research revealed that consumers perceive ordering food from a take-out restaurant online as significantly less effortful, and that they would assume less responsibility for a negative outcome, compared with making the order in-store (see Tables 1a-1c). This presents the possibility—worthy of future investigation—that such services may contribute to consumers’ food waste. Future research can also investigate whether effort must be productrelated in order for the documented effects to arise. For instance, if consumers invest appreciable effort online because of website glitches (vs. invest effort online when selecting the items they want by comparing multiple brands, reading nutritional information, comparing prices, etc.), would psychological ownership still be higher and waste likelihood lower? It may be that certain types of effortful experiences elicit substantial negative affect (e.g., frustration), such that the beneficial effects of effort are attenuated. As online grocery shopping gains traction in the marketplace, scholars are encouraged to pursue the above-mentioned avenues of future research and explore other behavioral outcomes of food purchases made across different retail channels.

attention from both policy makers and food manufacturers/retailers. On the surface, it may appear that marketing managers have little incentive and/or ability to change this behavior considering that (1) increasing consumer effort in online purchases, though beneficial for waste behavior, might negatively affect some other downstream consequences/profitability, and (2) the current work shows that explicit disappointment with the purchase does not differ across channels. However, over time, consumers may become cognizant of their (excessive) food waste when making produce purchases through a retailer's website, especially given the growing attention devoted to this subject matter in the media. This, undoubtedly, may trigger dissatisfaction with the firm in the long run. Coupled with the fact that corporate social responsibility (CSR) is now more important to firms’ success than ever before (Financier Worldwide, 2015), the present research suggests that food retailers would also benefit from investing in educational initiatives, wherein consumers are made aware of waste likelihood as a function of purchase channels.

8.3. Directions for future research A fruitful avenue for future research would be to investigate other variations of online purchases. For example, omni-channel grocery retailers now offer “click and collect” services, whereby consumers place orders online for pick-up either in-store, via drive-thru, or curbside (Nielsen, 2015). It would be interesting to gauge effort perceptions when consumers utilize these alternative methods to home delivery, including subsequent experiences of psychological ownership and food waste behavior. Likewise, consumers are increasingly engaging in mobile shopping, which offers even greater convenience (i.e., effort

Appendix A Snapshot of Cracker Options on the Mock Grocery Website.

29

Journal of Retailing and Consumer Services 41 (2018) 20–30

V. Ilyuk

References

J. Consum. Psychol. 22 (2), 224–236. Haws, Kelly L., Winterich, Karen Page, 2013. When value trumps health in a supersized world. J. Mark. 77 (3), 48–64. Hayes, Andrew F., 2013. Introduction to Mediation, Moderation, and Conditional Process Analysis. The Guilford Press, New York. Lunardo, Renaud, Saintives, Camille, 2013. The effect of naturalness claims on perceptions of food product naturalness in the point of purchase. J. Retail. Consum. Serv. 20, 529–537. Luomala, Harri T., Hellén, Katarina, Jokitalo, Maijastiina, 2017. Dieting, priming, food Meanings and (Un)healthy choices: when shoppers fall for pleasure. J. Retail. Consum. Serv. http://dx.doi.org/10.1016/j.jretconser.2017.05.001. McCracken, Grant, 1986. Culture and consumption: a theoretical account of the structure and movement of the cultural meaning of consumer goods. J. Consum. Res. 13 (1), 71–84. Melis, Kristina, Campo, Katia, Breugelmans, Els, Lamey, Lien, 2015. The impact of the multi-channel retail mix on online store choice: does online experience matter? J. Retail. 91 (2), 272–288. Nielsen, 2014. Continuous Innovation: The Key to Retail Success, Nielsen, pp. 1-20. Nielsen, 2015. The Future of Grocery: E-commerce, Digital Technology and Changing Shopping Preferences Around theWorld, Nielsen, pp. 1-35. Norton, Michael I., Mochon, Daniel, Ariely, Dan, 2012. The IKEA effect: when labor leads to love. J. Consum. Psychol. 22 (3), 453–460. Parfitt, Julian, Barthel, Mark, Macnaughton, Sarah, 2010. Food waste within food supply chains: quantification and potential for change to 2050. Philos. Trans. R. Soc. B: Biol. Sci. 365 (1554), 3065–3081. Peck, Joann, Shu, Suzanne B., 2009. The Effect of Mere Touch on Perceived Ownership. J, Consum. Res. 36 (3), 434–447. Pierce, Jon L., Kostova, Tatiana, Dirks, Kurk T., 2003. The State of Psychological Ownership. Rev. General. Psychol. 7 (1), 84–107. Porpino, Gustavo, 2016. Household food waste behavior: avenues for future research. J. Assoc. Consum. Res. 1 (1), 41–51. Porpino, Gustavo, Parente, Juracy, Wansink, Brian, 2015. food waste paradox: antecedents of food disposal in low income households. Int. J. Consum. Stud. 39 (6), 619–629. Schneider, Felicitas, 2008. Wasting food: an Insistent behaviour. Proc. Waste.: Social. Context 8, 1–10. Siegel, R.P., 2014. Food waste is a Bigger problem Than you think. Triple Pundit: People, Planet, Profit. 〈http://www.triplepundit.com/2014/10/food-waste-bigger-problemthink/〉. Sirieix, Lucie, Lalá, Jan, Kocmanová, Klára, 2017. Understanding the antecedents of consumers' attitudes towards doggy bags in restaurants: concern about food waste, culture, norms, and emotions. J. Retail. Consum. Serv. 34, 153–158. Stancu, Violeta, Haugaard, Pernille, Lähteenmäki, Liisa, 2016. Determinants of consumer food waste behaviour: two routes to food waste. Appetite 96 (1), 7–17. Steiman, HannahC., 2014. Why 2014 Will Finally Be the Year of the Online Grocer, 〈https://www.bloomberg.com/news/articles/2014-01-16/why-2014-will-finally-bethe-year-of-the-online-grocer〉. Troye, Sigurd Villads, Supphellen, Magne, 2012. Consumer Participation in Coproduction: “I Made it Myself” Effects on Consumers' Sensory Perceptions and Evaluations of Outcome and Input Product. J. Mark. 76 (2), 33–46. Tsiros, Michael, Heilman, Carrie M., 2005. The effect of expiration dates and perceived risk on purchasing behavior in grocery store perishable categories. J. Mark. 69 (2), 114–129. UNEP, 2015. "Food Waste: The Facts," 〈http://www.worldfooddayusa.org/food_waste_ the_facts〉. Van Dyne, Linn, Pierce, John L., 2004. Psychological Ownership and Feelings of Possession: Three Field Studies Predicting Employee Attitudes and Organizational Citizenship Behavior. J. Organ. Behav. 25 (4), 439–459. Van Ittersum, Koert, Wansink, Brian, 2012. Plate size and color suggestibility: the Delboeuf illusion's bias on serving and eating behavior. J. Consum. Res. 39 (2), 215–228. Visschers, Vivianne H.M., Wickli, Nadine, Siegrist, Michael, 2016. Sorting out food waste behaviour: a survey on the motivators and barriers of self-reported amounts of food waste in households. J. Environ. Psychol. 45, 66–78. Vrechopoulos, Adam P., Robert, M.O.’Keefe, Doukidis, Georgios I., Siomkos, George J., 2004. Virtual Store Layout: an Experimental Comparison in the Context of Grocery Retail. J. Retail. 80 (1), 13–22. Wansink, Brian, Wright, Alan O., 2006. ‘Best If Used By?’ How Freshness Dating Influences Food Acceptance,”. J. Food Sci. 71 (4), 354–357. Warschun, Mirko, 2012. “A Fresh Look. Online Groc. 〈http://www.atkearney.com/ paper/-/asset_publisher/dVxv4Hz2h8bS/content/a-fresh-look-at-online-grocery/ 10192〉. White, Katherine, Lin, Lily, Dahl, Darren W., Ritchie, Robin J.B., 2016. "When Do Consumers Avoid Imperfections? Superficial Packaging Damage as a Contamination Cue. "J. Mark. Res. 53 (1), 110–123. Williamson, Sara, Block, Lauren G., Anand Keller, Punam, 2016. Of waste and waists: the Effect of plate material on food consumption and waste. J. Assoc. Consum. Res. 1 (1), 147–160. Yeo, Vincent Cheow Sern, Goh, See-Kwong, Rezaei, Sajad, 2017. Consumer experiences, attitude and behavioral intention Toward online food delivery (OFD) services. J. Retail. Consum. Services 35, 150–162.

Ajzen, Icek, 1991. The theory of planned behavior. Organ. Behav. Human. Decis. Process. 50, 179–211. Atakan, S. Sinem, Bagozzi, Richard P., Yoon, Carolyn, 2014. Consumer participation in the design and realization stages of production: how self-production shapes consumer evaluations and relationships to products. Int. J. Res. Mark. 31 (4), 395–408. Belk, Russell W., 1988. Possessions and the extended self. J. Consum. Res. 15, 139–168. Block, Lauren G., Keller, Punam A., Vallen, Beth, Williamson, Sara, Birau, Mia M., Grinstein, Amir, Haws, Kelly L., LaBarge, Monica C., Lamberton, Cait, Moore, Elizabeth S., Moscato, Emily M., Walker Reczek, Rebecca, Heintz Tangari, Andrea, 2016. The Squander sequence: understanding food waste at each stage of the consumer decision-making process. J. Public Policy Mark. 35 (2), 292–304. Bolton, Lisa E., Joseph, W. Alba, 2012. When less is more: consumer aversion to unused utility. J. Consum. Psychol. 22 (3), 369–383. Buechel, Eva C., Janiszewski, Chris, 2014. A lot of work or a work of art: how the structure of a customized assembly task determines the utility derived from assembly effort. J. Consum. Res. 40, 960–972. Campo, Katia, Breugelmans, Els, 2015. Buying groceries in brick and click stores: category allocation decisions and the moderating effect of online buying experience. J. Interact. Mark. 31, 63–78. Chintagunta, Pradeep K., Chu, Junhong, Cebollada, Javier, 2012. Quantifying transaction costs in online/offline grocery channel choice. Mark. Sci. 31 (1), 96–114. Chu, Junhong, Arce-Urriza, Marta, Cebollada-Calvo, Jose-Javier, Chintagunta, Pradeep K., 2010. An empirical analysis of shopping behavior across online and offline channels for grocery products: the moderating effects of household and product characteristics. J. Interact. Mark. 24 (4), 251–268. Cicatiello, Clara, Franco, Silvio, Pancino, Barbara, Blasi, Emanuele, 2016. The value of food waste: an exploratory study on retailing. J. Retail. Consum. Serv. 30, 96–104. Clute, Mitchell, 2008. Packaged Produce Sales Soar, 〈http://newhope.com/news/ packaged-produce-sales-soar〉. Csikszentmihalyi, Mihaly, Rochberg-Halton, Eugene, 1981. The Meaning of Things: Domestic Symbols and the Self. Cambridge University Press, Cambridge. Degeratu, Alexandru M., Rangaswamy, Arvind, Wu, Jianan, 2000. Consumer choice behavior in online and traditional supermarkets: the effects of brand name, price, and other search attributes. Int. J. Res. Mark. 17 (1), 55–78. Dittmar, Helga, 1992. The Social Psychology of Material Possessions: To Have is To Be. St. Martin's Press, New York. Ellis, Lee, 1985. On the rudiments of possessions and property. Social. Sci. Inf. 24 (1), 113–143. Elms, Jonathan, Kervenoael, Ronan de, Hallsworth, 2016. Internet or store? An ethnographic study of consumers' Internet and store-based grocery shopping practices. J. Retail. Consum. Serv. 32, 234–243. Ericson, Keith M. Marzilli, Fuster, Andreas, 2014. The endowment effect. Annu. Rev. Econ. 6, 555–579. FAO, 2014. IFAD, and WFP, The State of Food Insecurity in the World 2014: Strengthening the Enabling Environment for Food Security and Nutrition, 〈http:// www.fao.org/publications/sofi/2014/en/〉. FAO, 2016. Save Food: Global Inititative on Food Loss and Waste Reduction, 〈http:// www.fao.org/save-food/resources/keyfindings/en/〉. Farr-Wharton, Geremy, Foth, Marcus, Choi, JazHee-Jeong, 2014. Identifying factors that promote consumer behaviours causing expired domestic food waste. J. Consum. Behav. 13 (6), 393–402. Financier Worldwide, 2015. The Importance of Corporate Social Responsibility, 〈http:// www.financierworldwide.com/the-importance-of-corporate-socialresponsibility/#. V4KK1DkrJmA〉. Frank, Rochelle, 2017. Pre-Packaged Produce: Fresh Food Gets a "Bad Wrap" for Sustainability, 〈https://delishably.com/food-industry/Pre-Wrapped-Produce-Fruitsand-Vegetables-Are-Getting-a-Bad-Wrap〉. Franke, Nikolaus, Schreier, Martin, 2010. Why customers value self-designed products: the importance of process effort and enjoyment. J. Product. Innov. Manag. 27, 1020–1031. Franke, Nikolaus, Schreier, Martin, Kaiser, Ulrike, 2010. The “I Designed It Myself” effect in mass customization. Manag. Sci. 56 (1), 125–140. Fuchs, Christoph, Prandelli, Emanuela, Schreier, Martin, 2010. The psychological effects of empowerment strategies on consumers' product demand. J. Mark. 74 (1), 65–79. Gawronski, Bertram, Bodenhausen, Galen V., 2006. Associative and propositional processes in Eevaluation: an integrative review of implicit and explicit attitude change. Psychol. Bull. 132 (5), 692–731. Geuens, Maggie, Brengman, Malaika, Rosette, S.'Jegers, 2003. Food retailing, now and in the future. A consumer perspective. J. Retail. Consum. Serv. 10, 241–251. Godfray, H., Charles, J., Beddington, John R., Crute, Ian R., Haddad, Lawrence, Lawrence, David, Muir, James F., Pretty, Jules, Robinson, Sherman, Thomas, Sandy M., Toulmin, Camilla, 2010. Food security: the challenge of feeding 9 billion people. Science 327 (5967), 812–818. Graham-Rowe, Ella, Donna, C. Jessop, Sparks, Paul, 2015. Predicting household food waste reduction using an extended theory of planned behaviour. Resour. Conserv. Recycl. 101, 194–202. Hansen, Torben, 2008. Consumer values, the theory of planned behaviour and online grocery shopping. Int. J. Consum. Stud. 32 (2), 128–137. Haws, Kelly L., Walker Naylor, Rebecca, Coulter, Robin A., Bearden, William O., 2012. Keeping it all without being buried alive: understanding product retention tendency.

30