Abstract
This study aims to develop a comprehensive understanding of food outlet choice patterns of Alternative Food System consumers (local and organic consumers) and evaluate if these patterns differ from that of conventional consumers. We conducted a nationwide online survey collecting data from U.S. food shoppers monthly. The data utilized in this study were collected and compiled from July 2016 to November 2019, resulting in 21,135 observations. We measured choices of eight food outlet formats within four categories (high-end, traditional, broad-assortment, and limited-assortment). Further, we examined the varying effects of demographic and household characteristics on food outlet format choices. We found that relative to conventional consumers, alternative food system consumers, who are local- and/or organic-minded, tend to be diversity-seekers who patronize various formats of food outlets. Among the four food outlet categories, we identified several complementary and substitute relationships. The occurrences and strengths of these relationships vary across consumer segments identified based on their preferences for local and organic food.
1. Introduction
Alternative food systems refer to alternatives to mainstream food systems and are characterized with alternative modes of production and/or distribution, such as organic production and local food initiatives (Cleveland et al., 2014; Nousiainen et al., 2009). Consumer interest in alternative food systems (AFS) has been growing in tandem with the proliferation of food values associated with positive externalities, such as environmental friendliness and locality (see Lusk and Briggeman, 2009 for a discussion on food values). These values, together with consumer preferences for them, result in various forms of alternative food systems (Bean and Sharp, 2011; Toler et al., 2009), which coexist with the conventional food systems in a dynamic combination of competition, interaction, collaboration, and co-evolution (Mount, 2012; Sonnino and Marsden, 2006).
Although AFS are rapidly evolving and growing in popularity, they face challenges such as scale hurdles (Friedmann, 2007; Johnston and Baker, 2005; Mount, 2012). For instance, the local food movement in the U.S. has seen the number of registered farmers’ markets triple from 2,863 in 2000 to 8,804 in 2020 (USDA-AMS, 2020). However, local food sales only accounted for 3% of total food sales in the U.S. in 2017 (Johnson and Cowan, 2019). Another prominent component of AFS, the organic food system, is also facing the scale challenge. According to the Organic Trade Association, with an approximately 40% increase in sales in the last decade, organic food represented only 6% of total food sales in 2018 (McNeil, 2018). When AFS seek to scale up, scholars debated on whether the values that underpin AFS are scalable (Friedmann, 2007; Mount, 2012).
For AFS to grow, there is a need to understand AFS consumers’ food outlet choices in the conventional food system with increasing number of retailers offering organic and local foods. Additionally, for AFS to target consumer segments, there is a need to examine whether AFS consumers’ food outlet choices differ from conventional consumers significantly. However, existing literature have not examined these questions. A comprehensive understanding of how different consumers choose between food outlets will better inform the identification of marketing opportunities and challenges in a competitive environment for retailers, as well as farmers looking for effective retailing channels to sell their produce directly to consumer. To this end, this study seeks to examine food-at-home shopping outlet choice patterns of consumers with heterogeneous preferences, with a specific focus on local and organic food.
Recent studies indicate the complexity of preferences for local and organic foods through exploring the complementary or substitute relationship between these two attributes (i.e. Costanigro et al., 2011; Meas et al., 2015; Hempel and Hamm, 2016). This stream of research reveals similarities and differences in consumer attitudes and perceptions that constitute preferences for local and organic characteristics (Bean and Sharp, 2011; Lusk and Briggeman, 2009). While some studies argue that food purchase decisions and willingness to pay vary between local and organic attributes, little is known about the heterogeneity of food outlet choices among local-minded and organic-minded consumers.
To fill this gap, this study investigates the association between AFS consumer preferences and food-at-home outlet choice patterns from three perspectives. First, we evaluate the effects of consumer heterogeneity on choices of various food outlet formats. Second, we examine whether the magnitude of food outlet choice diversity (i.e. the total number of different stores a consumer visits to fulfill their food shopping task) varies across consumer segments. Third, we examine how cross-shopping behaviors within each consumer segment translate into complementary or substitute relationships among different food outlet categories. To the best of our knowledge, this is the first study that examines comprehensively how consumers’ food outlet choice patterns vary by preferences for AFS values. With a focus on local and organic foods, which represent the two most rapidly emerging alternative food systems, we elaborate on how AFS consumers’ food shopping behaviors differ from conventional consumers. We further discuss how these differences could translate into marketing strategies to potentially address the scale challenge facing AFS. We contribute to the literature by establishing a more complete understanding of the heterogeneous food outlet choice patterns among local- and/or organic-minded consumers. Furthermore, this study assists marketers and farmers in identifying marketing opportunities to grow AFS within the competitive landscape of food retailing.
Using a large nationwide consumer dataset covering the time frame from 2016 to 2019, we investigated U.S. food shoppers’ store choices. Specifically, we examined consumer choice patterns among eight food outlet formats within four categories – high-end, traditional, broad-assortment, and limited-assortment. We identified varying effects of demographic characteristics and preferences for local and/or organic foods on store choices as well as on the magnitude of store choice diversity. Further, we estimated cross-shopping behaviors and identified several complementary and substitute relationships among the four food outlet categories. The occurrences and strengths of these interrelationships vary across consumer segments identified based on their preferences for local and organic food.
2. Food outlet format and category
To formalize the bandwidth of food outlets with a comprehensive scope, we build on a broad definition (Guptill and Wilkins, 2022; Martinez, 2016) that food outlet is an omnibus term indicating any institutional retailing outlet from which consumers purchase food for at-home consumption, including, for example, grocery stores and supercenters.
In the United States, since modern retail diffusion started in the 1920s/1930s (Reardon, 2011), there have been three major evolutions in the food sector: the influx of retailers that were traditionally not involved in food retailing, such as Wal-Mart supercenters; the rapid proliferation of direct-to-consumer channels featuring locally grown and produced food; and the rise of the e-commerce sector (Hansen, 2005). As direct-to-consumer sales remain a small portion of total food sales, changes of market structure are mostly dominated by food retailers through both online and offline (physical store) channels (Gauri et al., 2021; Hansen, 2005). This indicates (1) concentration in retailing corroborated by greater market share held by fewer retailers; and (2) a shift in bargaining power from manufacturers to retailers (e.g. Guptill and Wilkins, 2002; Martinez, 2007; Volpe et al., 2018).
Internal market structural changes, as well as external pressures imposed by increased food-away-from-home consumption, have introduced increasing competition to various formats of food-at-home outlets (Volpe et al., 2018). As a result, diverse food outlet choices strive to fulfill consumers’ heterogeneous needs pertaining to two dimensions: (1) product-specific preferences, and (2) pre-determined views on attributes of food outlet format (Hsieh and Stiegert, 2012). It’s worth noting that there is no unified taxonomy of food outlets (Volpe et al., 2018). Therefore, studies exploring linkages between food outlet choices (incl. food environment) and consumer behavior usually classify food outlets differently based on distinctive research scopes or a pre-determined classification of the dataset supporting the study.
Existing literature has identified two predominant sets of food outlet classification. One is based on Nielsen Homescan data, as shown in Volpe et al. (2018), identifying several formats of food outlets, including supermarkets, drug stores, mass merchandisers, supercenters, club stores, convenience stores, and others. The other uses a more arbitrary approach focusing on two categories: traditional versus non-traditional stores (Food Marketing Institute, 2018; Leibtag et al., 2010; Martinez, 2007; Volpe et al., 2017). ‘Category’ is a broader concept than ‘format’ in the sense that it can encompass a variety of formats that share similarities in product offering or customer bases (e.g. Bodkin and Sewell, 2012; Fox et al. 2004). For example, along with the traditional versus non-traditional taxonomy, a supermarket (or grocery store) is commonly referred to as a traditional food outlet in the bulk of studies on food shopping. By contrast, all other channels are non-traditional, including emerging formats such as supercenters and convenience stores. In addition, some relevant studies consider other food outlets as additional categories, such as convenience stores in Bustillos et al. (2009), or high-end outlets in Hsieh and Stiegert (2012). In summary, what those classification and categorization methods have in common is sorting differences and similarities for each food outlet format in terms of length of the supply chain, product assortment, advertising and pricing strategies, and location (Volpe et al., 2017).
3. Literature review
3.1 Food outlet choices and consumer characteristics
Studies focusing on the linkage between consumer heterogeneities and food outlet choices primarily elaborate on two perspectives. The first one sees store attributes (incl. product attributes), as well as perceived fit between individual preferences and store attributes (Hwang and Chung, 2019), as the major antecedents to the patronage of a food outlet format (Huddleston et al., 2009; Jacobs et al., 2010; Nilsson et al., 2015; Pan and Zinkhan, 2006; Reutterer and Teller, 2009). These store attributes include, for example, proximity, product assortment/selection and quality, service and store atmosphere, and pricing strategy.
The second perspective emphasizes the effects of consumer demographic characteristics on choices of various distinctive food outlet formats (e.g. Fox et al. 2004; Pan and Zinkhan, 2006; Prasad, 2010; Taylor and Villas-Boas, 2016). For instance, Fox et al. (2004) found that household characteristics, such as family size and income, impact both store choices and expenditure. Similar results were identified in Carpenter and Moore (2006) that only income level predicts consumer patronage in high-end specialty stores while family size (also see Hsieh and Stiegert, 2012) is an important predictor of traditional supermarket’s visit. Using National Household Food Acquisition and Purchase Survey data, Taylor and Villas-Boas (2016) estimated store choices as a function of household characteristics and local food environment and found that low-income households prefer certain types of stores (e.g. superstores or supercenters) over others (e.g. convenience stores).
By combining both perspectives to examine food outlet choices, some studies imply that food outlet choice reflects the consumer’s interpretation of store attributes that appeal to their food-shopping needs (Jacobs et al., 2010; Pan and Zinkhan, 2006; Reutterer and Teller, 2009). For example, Bell and Lattin (1998) examined if price expectations affect store choices and found that large-basket shoppers prefer stores featuring the everyday low pricing strategy while small-basket shoppers are attracted by stores using high-lowing pricing strategy. Volpe et al. (2017) provide an exhaustive categorization of food outlet formats characterized by a wide range of store attributes, concluding that food outlet format connects to end consumers by offering foods and shopping experiences that cater to their aggregated preferences. To this end, addressing consumer heterogeneity is of greater significance to probing into the connection between consumer and the food retail market.
Behind consumer heterogeneity is individual’s food-related preferences, which has received limited scholarly attention in studies of food outlet choices. To illustrate, Hsieh and Stiegert (2012) argue that the variation in food quality perception (i.e. preference for organic food) accounts for differences in format choices in a way that high-quality perception drives consumers to visit high-end stores. Similarly, Prasad and Aryasri (2011) incorporate consumer perceptions (called ‘shopping motives’ in their study) regarding aspects such as product variety, local food store, and price consciousness to segment consumers in terms of food outlet choices. At the heart of this stream of research is to include consumer food preference to better understand food outlet choices.
3.2 Food outlet diversity and cross-shopping
The Food Marketing Institute (2019) found U.S. food shoppers, on average, visited 3.1 different channels of food retail outlets each month in 2019. Using various formats of food outlets is the dominant food shopping pattern (Hino, 2014; Prasad, 2014; Tran and Sirieix, 2020). However, there is scarcity of academic research on the diversity of food outlet choices and its association with consumer characteristics.
An indirect approach to examine food outlet diversity emerges as scholars show growing interests in addressing cross-shopping behavior. Initially discussed in a non-food retail context (such as apparel retailing in Cassill et al. 1994), cross-shopping studies focus on ascertaining why consumers visit different channels to better understand competition among these channels (Carpenter and Moore, 2006; Hansen, 2003; Prasad, 2014). To extend cross-shopping research in the food retail sector, two forms of competition among retail channels are identified: (1) intracategory competition, which happens when consumers cross shop at different food outlets of the same category (e.g. the competition between two supermarket brands); and (2) intercategory competition, which occurs when consumers cross shop at different categories of food outlets (e.g. the competition between a specialty store and a supercenter) (Hansen, 2003).
Intuitively, cross-shopping behavior implies potential interrelationships (i.e. complementarity or substitutability) among different types of food outlets (Bai et al., 2008). Based on cross-price estimates, Volpe et al. (2018) find ‘other formats’ (e.g. drug stores, mass merchandisers) appear to be a complementary alternative to supercenters. A complementary relationship is also observed between high-end and value-oriented food outlets in Hsieh and Stiegert (2012), indicating while formats with a broader assortment are more likely to be the ‘one-stop’ food venue, upscale outlets tend to offer ‘fill-up’ purchases. Regarding store choices within a format, Fox et al. (2004) find that spending and patronage preferences of two different grocery stores are negatively correlated, denoting a substitution relationship between the two stores within the same format. It is worth noting that exploring complementarity or substitutability among food outlets depends on how food outlet formats and categories are classified, as well as on the specific food retail context, for example, North American consumers might have different store preferences from European consumers (Nilsson et al., 2015); thus, consistent findings are lacking. And yet, what underpins this stream of research on food outlet diversity and cross-shopping behavior is whether or not, and how much, the competition within formats (or categories) is fundamentally different from that between formats (or categories) (Bodkin and Sewell, 2012; Fox et al., 2004; Reutterer and Teller, 2009).
As this literature review illustrates, a comprehensive understanding of the linkage between consumer heterogeneity and food outlet choice patterns is lacking. To fill this gap, the current study elaborates on how food outlet choice patterns vary among consumers with different preferences for organic and local food. Specifically, from the food outlet format level, we examine and compare format choices and diversity of these choices across consumer segments determined by local and organic preferences. From the food outlet category level, we examine whether different consumer segments exhibit different inter-category cross-shopping behaviors.
4. Methods
4.1 Data collection
To examine the connection between consumer heterogeneity and food shopping choice patterns, individual consumer level data are needed. However, the commonly used secondary data sources such as Nielsen homescan data do not provide information on consumer engagement in alternative food systems or food attitudes. Therefore, we are motivated to collect primary food shopper data through a nationwide online survey, focusing on food outlet choices, sociodemographic characteristics, and preference for local and organic foods. The survey instrument, a reoccurring monthly survey first implemented in July 2016, randomly draws approximately 500 primary grocery shoppers above 18 years of age from a panel demographically and geographically balanced to represent the U.S. population. The survey was distributed through the panel company Toluna. To identify inattentive respondents and control for data quality (Berinsky et al., 2014; Gummer et al., 2021; Jones et al., 2015), we incorporated two validation questions (i.e. attention checks) in the survey instrument. One example is ‘Please select Strongly Disagree for this question’ constructed in a Likert-scale format and respondents who failed this question were excluded from the survey. The cross-sectional data utilized in this study were collected and compiled from July 2016 to November 2019. By excluding missing values, the final sample size resulted in 21,135 observations.
■ Ethics statement
The data used for the article titled ‘Food Outlet Choice Patterns of Alternative Food System Consumers’ is collected via an online survey. This survey was submitted to the University of Florida’s Institutional Review Board for Human Subjects Approval and was deemed exempt under category 2, Research involving the use of educational tests (cognitive, diagnostic, aptitude, achievement), survey or interview procedures, or the observation of public behavior, so long as confidentiality is maintained. The study was assigned IRB #2016-00977. Additionally, the survey data collected was anonymous, with no personal identifiers collected.
4.2 Variable specifications
■ Segmenting local and organic consumers
The complexity of differences and overlaps between local and organic values gives rise to the situation that consumers may favor local and organic values simultaneously or exclusively (Bean and Sharp, 2011). Therefore, following the typology of decoupling local and organic proposed by Bean and Sharp (2011), we segment consumers based on their interests in local and organic food consumption. Specifically, consumer interest in local food was measured through a question on if the respondent reported shopping at direct-to-consumer channels, such as farmers’ markets and road stands, in a typical month. Similarly, consumer interest in organic food was measured through a question regarding if the respondent reported actively seeking out organic food. By cross-tabulating respondents’ answers (yes or no) to these two questions, we derived four segments: conventional consumers (i.e. who are not interested in either local or organic food), local consumers (i.e. who have interests in local food but not organic food), organic consumers (i.e. who are interested in organic food but not local food), local/organic consumers (i.e. who favor both local and organic food).
■ Classifications of food outlet format and category
This study develops the classification of food outlet formats and categories based on three references: Retail Segment Definitions by Food Marketing Institute (FMI) (2018); National Household Food Acquisition and Purchase Survey (FoodAPS) by USDA-ERS (2015); and previous studies (e.g. Carpenter and Moore, 2006; Fox et al., 2004; Volpe et al., 2018). To avoid confusion caused by interchangeably using ‘format’ (Hsieh and Stiegert, 2012), ‘category’ (Volpe et al., 2017), and ‘type’ (Bustillos et al., 2009), this study uses ‘format’ to narrate a narrower and distinctive form of food outlet (e.g. supercenters) and ‘category’ to refer to a broader group of food outlet formats with similar attributes, such as high-end category that consists of formats offering upscale food products (also see Hsieh and Stiegert, 2012).
Presented with a list of food outlet formats, survey respondents were asked to check all formats they had shopped at in the past 30 days. As summarized in Table 1, we focus on a broad range of eight food outlet formats categorized into four categories, including high-end (featured with upscale and trendy food offerings), traditional, broad-assortment (featured with a broad range of non-food product selection and rich assortments within the product category), and limited-assortment categories.
Classification and consumer patronage of food outlet formats and categories.1
Citation: International Food and Agribusiness Management Review 26, 4 (2023) ; 10.22434/ifamr2022.0095
Food outlet diversity (hereinafter referred to as FOD) is defined as the total number of formats of food outlet a respondent has patronized for food-at-home purchases during the past month. Given the eight food outlet formats considered in this study, FOD takes the value between one and eight. If a consumer’s FOD equals to one, then they exclusively shop at one format of outlet for food-at-home purchases.
4.3 Model specification and empirical analyses
To examine food outlet choices, we build on the perceived utility framework that has been broadly applied in the extant literature on brand and store format choices (e.g. Fox et al., 2004; Hsieh and Stiegert, 2012; Prasad, 2014; Reutterer and Teller, 2009). Specifically, this framework assumes that consumers choose the market outlets that maximize their latent utilities among a set of alternatives. The latent utility of consumer c for store format f consists of three components:
Citation: International Food and Agribusiness Management Review 26, 4 (2023) ; 10.22434/ifamr2022.0095
Where
Similarly, the latent utility of consumer c with preference p (i.e. conventional, local, organic, local/organic) for store category g is given by:
Citation: International Food and Agribusiness Management Review 26, 4 (2023) ; 10.22434/ifamr2022.0095
where
gj (j ≠ i, 1 ≤ i ≤ 4,1 ≤ j ≤ 4)
Assuming the unobserved terms
Citation: International Food and Agribusiness Management Review 26, 4 (2023) ; 10.22434/ifamr2022.0095
where choicecf/g is a specific choice of store format or category, and Ucf/g is the latent utility of consumer c to visit a food outlet in format f or category g. As individual consumers may purchase food at more than one format or category of stores, we first estimated likelihoods to visit a format or category individually and then combined the estimation results through a seemingly unrelated estimation with robust standard errors. This was conducted using the post-estimation command ‘suest’ in Stata, allowing for computing standard errors under the assumption of unknown error covariance across equations. This further allows for intramodel and intermodel comparisons (UCLA Statistical Methods and Data Analytics, nd). Next, post-estimation Wald tests, which focused on the consumer segment variable identified by organic and local preferences, were carried out using both intramodel and intermodel hypotheses. Within a model, the null hypothesis is that the coefficient estimates for the local, organic, and local/organic consumers are identical relative to conventional consumers. The intermodel null hypothesis is that the coefficient estimates for the local, organic, and local/organic consumers across the set of equations are identical.
Previous studies show that consumer store choice may be sensitive to income (e.g. Carpenter and Moore, 2006; Volpe et al., 2018). To be consistent and be able to compare with previous studies, we followed the approach of Volpe et al. (2018), which divides the sample by income levels to examine income’s role in food shopping patterns. By employing Poisson regressions with robust standard errors for middle-, high-, and upper-income consumers, we estimate the effects of consumer characteristics on food outlet diversity. For each regression, the assumption of equidispersion was supported by a likelihood ratio test (P-value=1.00), suggesting that overdispersion is not a concern and the Poisson model is an appropriate fit for the data. Post-estimation Wald Tests on differences of coefficients between income groups were conducted to compare the effects of local and organic preferences across income groups. In addition, to compare the effects of local and organic preferences on food outlet diversity, intramodel Wald tests were also carried out.
5. Results
5.1 Sample characteristics
Descriptive statistics of the variables discussed in this study are shown in Table 2. Respondents who actively sought out organic food constitute 38% of the sample, compared to 17% who are local food consumers using direct-to-consumer channels in a typical month. This is consistent with the fact that the organic food market is larger in scale than the local food market (Johnson and Cowan, 2019; McNeil, 2018). The four consumer segments in our sample characterized by local and organic preferences are conventional consumers (55%), local consumers (7%), organic consumers (28%), and local/organic consumers (10%).
Descriptive statistics.
Citation: International Food and Agribusiness Management Review 26, 4 (2023) ; 10.22434/ifamr2022.0095
An average consumer patronized 3.1±1.6 formats of stores for food-at-home purchases in the past 30 days. The most popular food outlet formats are traditional supermarkets or grocery stores (78%) and supercenters and warehouse clubs (73%). On the contrary, high-end stores such as fresh and specialty stores (28%) and internet grocery stores (10%) have least patronage.
5.2 Food outlet format choices
For the ease of interpretation, the estimated odds ratios for the store format choices are summarized in Table 3. The original estimated parameters are available upon request.
Logit model parameter estimates for food outlet format choices.1
Citation: International Food and Agribusiness Management Review 26, 4 (2023) ; 10.22434/ifamr2022.0095
■ Effects of demographics
Age and gender both have a positive effect on the selection of supermarkets. A one unit increase in age, and being female as opposed to male, increase the likelihood to visit supermarkets by 4% and 20%, respectively. On the contrary, for the other six formats of food outlets, being younger or being a male food shopper is associated with a higher likelihood of patronage. Being college-educated increases the likelihood of visiting fresh and specialty stores, online grocery stores, and supermarkets by 72, 60, and 14%, respectively, but decreases the likelihood of patronizing dollar stores by 19%. Employment is only significantly, and negatively, associated with visiting supercenters and club stores (odds ratio <1).
There are also significant associations between household characteristics and food outlet format choices. Compared to middle-income respondents (i.e. household income below $50,000), high- and upper-income respondents are more likely to visit high-end stores, traditional stores, and supercenters. Moreover, the odds of patronizing these four formats increases as the income level increases. The opposite trend is found for dollar stores, where the higher the income, the lower the probability of visiting. A one unit increase in household size is significantly associated with an at least 5% increase in likelihood of patronizing supercenters, mass merchandisers, convenience stores, and dollar stores, as well as an approximately 5% decrease in probabilities of visiting online stores and specialty stores. The presence of young children has mixed impacts on store format choices. Having preschool children in the household significantly decreases the likelihood to patronize fresh and specialty stores, supermarkets, and convenience and drug stores. The presence of school-age children has a negative impact only on supermarket patronage but increases the likelihood of visiting all other formats.
■ Effects of information-seeking and price consciousness
Respondents who actively search for diet and nutrition information, compared to those who don’t, are at least 14 to 75% more likely to visit store formats examined in this study except for traditional supermarkets. Taking ‘price increase doesn’t impact my food purchases’ as the baseline, consumers responding to price increases by looking for in-store deals are slightly less likely to visit fresh and specialty stores but 10 to 57% more likely to visit the traditional supermarkets, broad-assortment stores, and limited-assortment stores. Not surprisingly, consumers who switch to shop at supercenters to counteract food price increases are 163 and 61% more likely to choose supercenters and mass merchandisers as food outlets. These consumers are also more likely to shop at online channels and limited-assortment channels. For consumers who would curtail food purchases in the scenario of price increases, visiting fresh and specialty stores and supermarkets becomes less likely while visiting a dollar store becomes 68% more likely.
■ Effects of local and/or organic preferences
Compared to conventional consumers who don’t seek out organic or local foods, local consumers’ likelihoods to visit the seven food outlet formats are 53 to 207% higher. For example, local consumers are about twice as likely to use fresh and specialty stores. Organic preference has mixed effects on patronage likelihood for different formats. Specifically, relative to conventional consumers, organic consumers are slightly less likely to shop at traditional supermarkets or grocery stores but much more likely to select fresh and specialty stores, online stores, and slightly more likely to visit supercenters and warehouse clubs and convenience and drug stores. Being a consumer who favors both local and organic foods increases the likelihood to visit the seven formats of food outlets by 44 to 840%.
Post-estimation Wald tests on the equality of coefficients of local and/or organic preferences are presented in Table 4. Significant parameters denote the rejection of null-hypotheses of equal coefficients. Results indicate that estimated coefficients of local, organic, and local/organic preferences are significantly different pairwise in the models of online stores, supermarkets, convenience and drug stores, and dollar stores. Horizontally, the row of ‘organic vs local/organic’ shows that for any given food outlet formats considered, the effect of being local-minded (relative to conventional consumers) on the likelihood to visit a format is significantly different from that effect of being local organic consumers. The same result holds true for: (1) the comparisons between local and organic consumers for all formats except for fresh and specialty store; and (2) the comparisons between local and local/organic consumers for all stores except for supercenters and mass merchandisers. These results imply that the effect difference between being local-minded and being local/organic-minded on format choices is smaller than that between organic-minded and local/organic-minded. This finding is also supported by the observation of odds ratios in Table 3 that when comparing positive effects on the likelihood to patronize a format (e.g. fresh and specialty stores, supercenters), local/organic consumers have the largest odds ratio, and local consumers and organic consumers have the medium and smallest odds ratios, respectively.
Post-estimate Wald Tests on local and organic preferences.1
Citation: International Food and Agribusiness Management Review 26, 4 (2023) ; 10.22434/ifamr2022.0095
5.3 Food outlet diversity
Based on the prior studies showing that consumer store choice may be sensitive to income (e.g. Carpenter and Moore, 2006; Volpe et al., 2018), we provide estimated incidence rate ratios (IRRs) of food outlet diversity for middle-, high-, and upper-income consumers derived from Poisson regressions (Table 5). IRRs can be obtained by exponentiating the Poisson regression coefficient, allowing for comparisons of incidence rates between two or more different groups. For the middle-income consumers, being male, employed, black or African American relative to white, and having a larger household and school-age children have significant IRRs greater than 1, meaning that these consumers have larger expected counts of food outlets. In the high-income group, consumers who are male, college-educated, employed, Hispanic, and come from a larger household and have school-age children are associated with at least a 2% higher rate of food outlet diversity. For consumers with a household income level above $100,000, demographics with significant effects on food outlet diversity include female, college-educated, Asian as opposed to white, and having school-age children.
Incident rate ratios for food outlet diversity from Poisson regressions.1
Citation: International Food and Agribusiness Management Review 26, 4 (2023) ; 10.22434/ifamr2022.0095
Relative to consumers insensitive to food price increases, consumers who seek to offset increased prices by looking for in-store deals or switching to shop at value-oriented stores tend to visit at least 5% more food outlet formats, regardless of income levels. Being involved in seeking food information is associated with a 9 to 13% increase in the total number of patronized food outlets (i.e. food outlet diversity).
Compared to conventional consumers, having preferences for local food, organic food, or both are associated with significant increases in food outlet diversity (1.06 ≤ IRRs ≤ 1.49), regardless of income levels. Local consumers’ food outlet visits in a typical month are, on average, 27 to 31% more than conventional consumers. Organic consumers also tend to visit more food outlets in various formats than conventional consumers by 6 to 11%. Local/organic consumers visit 39 to 49% more food outlets than conventional consumers, showing larger incidence rate ratios than consumers who focus on only one attribute (local or organic).
5.4 Cross-shopping behavior
Our results show considerable differences of format choices and diversity between consumer segments identified by local and organic preferences. To reveal greater details on cross-category shopping behavior for each consumer segment, the patronage of a food outlet category is estimated as a function of the patronage of other food outlet categories, after controlling for covariates. Following the approach of Fox et al. (2004), we determine substitutability and/or complementarity between four food outlet categories based on the significance and magnitude of estimated coefficients. We only report estimated coefficients (i.e. odds ratios) of store category variables (Table 6) to focus our attention on disentangling relationships among food outlet categories. Full results are available upon request.
Odds ratios for food outlet category choices from logit models.1
Citation: International Food and Agribusiness Management Review 26, 4 (2023) ; 10.22434/ifamr2022.0095
A mutual substitution relationship is found between the broad-assortment and traditional categories for conventional consumers, local consumers, and organic consumers. In this case, visiting one of these two categories of food outlet decreases the likelihood of visiting the other by 86, 72 and 53%, respectively.
On the other hand, complementary relationships are also evident for each of the four consumer segments. Between the high-end and limited-assortment categories, the patronage of one increases the likelihood of visiting the other by approximately 48, 32, 28 and 107% for conventional consumers, local consumers, organic consumers, and local/organic consumers, respectively. Between the limited-assortment and broad-assortment categories, patronizing one significantly increases the likelihood to visit the other by approximately 61, 51, 84 and 205% for conventional consumers, local consumers, organic consumers, and local/organic consumers, respectively. A complementary relationship between traditional and high-end categories is only identified for local consumers, with a significant effect of an approximately 45% increase in patronage probability in both directions. The last complementary relationship identified in this study is between traditional and limited-assortment categories only for the segment of local/organic consumers, where patronage of either of the stores increases the likelihood of visiting the other by at least 37%.
Last, a one-way complementary relationship is observed between the traditional and high-end categories for organic consumers, where those who use high-end stores are 14% more likely to visit traditional stores while traditional store patronage is not significantly related to high-end store usage.
6. Discussion
U.S. consumers show growing interest in visiting various food outlets to fulfill food shopping needs. Based on an extensive review of food outlet format categorization, our nationwide data show that the most frequently used outlets for food-at-home purchases are traditional supermarkets (or grocery stores), followed by broad-assortment stores (e.g. supercenters and warehouse clubs), limited-assortment stores (e.g. convenience and dollar stores), and high-end food outlets (e.g. fresh and specialty stores). To the best of our knowledge, our study is the first to comprehensively examine the linkage between consumer heterogeneity and food outlet choice patterns characterized with three perspectives: food outlet format choice, food outlet diversity, and cross-shopping behavior.
6.1 Demographics are tied to food outlet choice patterns
Intuitively, interpreted as the total number of divergent formats of stores a consumer has patronized, a consumer’s food outlet diversity increases as the probabilities of visiting individual formats of stores collectively increase. Practically, this is verified by comparing factors related to format choices with factors related to FOD. For example, age, gender, and socioeconomic status have congruent effects on format choices and diversity of these choices; consumers with a higher socioeconomic status prefer to use multiple channels such as traditional grocery stores and specialty stores to satisfy their food acquisition needs.
Previous studies explore the association between household characteristics and store format choices (Bell and Lattin, 1998; Carpenter and Moore, 2006; Hsieh and Stiegert, 2012; Taylor and Villas-Boas, 2016). We extend the understanding of this association by targeting more distinctive store formats and differentiating households with children and comparing the effects of household characteristics on format choices and overall choice diversity. As a support to, as well as an extension of, Bell and Lattin’s (1998) finding that large-basket and small-basket shoppers have different preferences for store types, we found larger households tend to shop at value-oriented stores with a broad range of assortment (i.e. supercenters and mass merchandisers) or offering discount and convenience (i.e. convenience stores and dollar stores). However, larger households do not have a significantly higher level of food outlet choice diversity, especially for upper-income consumers. This could be explained by a conventional wisdom that with the goal of filling a large shopping basket conveniently and saving shopping time to attend to other household chores, large-household shoppers gravitate towards stores offering good dollar value and wide collection of choices. Another household feature providing additional information is the presence of school-age children, which leads to greater interest in visiting various non-traditional food outlets, and therefore, a higher level of diversity (more so for high-and upper-income families). We conjecture that these households tend to allocate their shopping tasks into different stores to satisfy heterogeneous food preferences and nutrition needs.
6.2 Local and organic preferences differentiate where to shop
Using ‘organic’ as a proxy for quality perception, an earlier study by Hsieh and Stiegert (2012) found a correlation between organic shoppers and patronage of high-end outlets. Similarly, in the extensive research on food-related lifestyle using dimensions such as ‘ways of shopping’ and ‘importance of quality’ to differentiate consumers food lifestyle (Brunsø et al., 2004), consumer preference for local and organic foods are significant factors to the identification of market segments (Chen and House, 2022). By segmenting consumers based on their local and organic preferences, our study substantially enriches the understanding of how AFS consumers decide where to shop, and how that differs from conventional consumers. Our results indicate that AFS consumers, who prefer either organic or local or both, are enthusiastic about high-end stores but less interested in traditional supermarkets. Although the difference of patronage between AFS consumers and conventional consumers regarding broad-assortment and limited-assortment food outlets is smaller than that for high-end stores, a consistent pattern emerges across formats. AFS consumers, on average, are more likely to use each of these formats than conventional consumers, and thereby cross shop more and exhibit greater FOD. Indeed, our data show in a typical month, AFS consumers, on average, shop at 3.68±1.73 food outlets with different formats while conventional consumers’ average FOD is 2.68±1.26. Meanwhile, a greater standard deviation of FOD among AFS consumers signifies greater variances. The two possible explanations for this pattern are: (1) AFS consumers are diversity-seekers who are highly involved in food shopping and planning, and prone to split their shopping tasks into different stores to make selective purchases, and (2) there exist complementary relationships among stores to avoid shoulder-to-shoulder competition and to attract consumers with diverse needs. While resonating with studies incorporating the level of involvement to account for variances in behavioral outcomes (e.g. Chen et al. 2019; Hsieh and Stiegert 2012), this finding also empirically disentangles previous conjectures about which consumers’ perceived food acquisition utility could outperform transaction cost so that they would shop at a diverse combination of stores (Hino, 2014; Klein and Schmitz, 2016).
6.3 Cross-shopping behaviors reveal interrelationships among food outlet categories
Increased food outlet diversity is in tandem with cross-shopping behavior (Cude and Morganosky, 2001; Prasad, 2014). Cude and Morganosky (2001) hypothesize that multi-purpose and multi-destination shopping motivations promote complex interrelationships between stores. Our findings decompose this complexity by empirically identifying interrelationships among four food outlet categories and comparing these relationships among consumer segments. Specifically, different from previous studies contending that consumers generally see complementarity rather than substitutability among various formats to maximize acquisition utility (Fox et al., 2004; Klein and Schmitz, 2016), we find that both complementary and substitute relationships exist among food outlet categories and they are conditional on specific categories as well as consumer segments.
Two complementary relationships, between high-end and limited-assortment stores, and between limited-assortment and broad-assortment stores, exist for all consumers, with the relationships being the strongest for local/organic consumers. On the other hand, the substitute relationship occurs between traditional and broad-assortment stores for all consumers except for local/organic consumers and is the strongest among conventional consumers. As for local/organic consumers, they see the strongest complementary ties between food outlet categories but would not substitute one for another. This finding underlines the diversity-seeking character of local/organic consumers, whose food acquisition utility derived from cross-shopping at multiple outlets considerably outweighs the inevitable transaction costs.
Local consumers appear to be the only segment that sees traditional and high-end food outlets as complements. A possible interpretation is that owing to these stores’ initiative of introducing local foods, exclusively local-minded consumers tend to use them both to fulfill their food demand.
7. Conclusions and implications
Understanding how consumers with heterogeneous preferences shop differently is critical for marketers to identify marketing opportunities and design effective marketing programs, especially in the interest of growing AFS. This study fills the gap by comprehensively investigating the linkage between AFS consumers (consumers who prefer local and/or organic foods) and food-at-home shopping patterns using a large-scale nationwide sample. Several key findings emerge. First, consumer heterogeneity differentiates food outlet choices in that AFS consumers are more likely to patronize various formats of stores than conventional consumers, and that the magnitude of the difference in store patronage varies across consumer segments identified by local and organic preferences. Second, AFS consumers tend to be diversity-seekers, using 37% more food outlets to fill their shopping baskets. Third, among four food outlet categories, several competitive interrelationships are identified, and their characteristics and strength vary across consumer segments.
These findings indicate that there is no one-size-fits-all strategy to scale up AFS. Different retailers have varying opportunities to strategically increase local and organic food supplies to cater to consumer needs. Further, these retailers represent different marketing potential to farmers who seek to identify effective retailing channels to sell their produce directly to consumers. Conventional consumers have a low food outlet diversity but predominantly use traditional and broad-assortment stores, which are strong substitutes for each other. While grocery stores and supercenters are a good market to promote local and organic food to conventional consumers, retailers need to evaluate the competition between the two outlets within the local food environment. For diversity-seekers, AFS consumers, limited-assortment stores are an important complement to high-end and broad-assortment stores. Although AFS consumers might cross-shop at these stores for different purchases, their patronage of convenience stores complementing ‘major’ food shopping outlets implies a niche opportunity to boost local and organic food visibility. Last, the competitive dynamics among food outlets not only vary across consumer segments, but also change over time. When we emerge from the COVID-19 pandemic and its aftermath, it is important to reassess how the retail food landscape and consumer food outlet choice patterns will change, especially with the rapid penetration of e-commerce in the food sector, and examine how these changes possibly impact the growth of AFS.
7.1 Limitation and future research
Our findings lay out a foundational understanding of how heterogeneous consumers shop differently, leading to several perspectives meriting future research. However, there are areas that merit further research. First, we didn’t measure shopping frequency of the stores visited and which format/category is the ‘primary’ or ‘fill-in’ outlet. We encourage future studies to build on our findings and extend the probing into the dominance of different store choices and if this dominance varies by the level of consumer involvement in alternative food systems. Second, the absence of an interrelationship between broad-assortment and limited-assortment stores implies that the complementarity between the food outlet categories is not transitive, calling for future research to further scrutinize characteristics of complementary and/or substitute relationships among food outlets. Potential research questions are, for instance, what factors determine inter-category complementarity; does the intransitivity of inter-category complementarity apply to intra-category (i.e. between formats within a category) complementarity? Third, future research that endeavors to combine food purchase scanner data with food outlet characteristics would enhance the understanding of how consumers allocate food shopping tasks.
Acknowledgements
The authors would like to thank the Florida Department of Citrus for funding. All mistakes and errors are the responsibility of the authors.
Disclaimer
The findings and conclusions in this working paper are those of the author(s) and are not purported to represent the views of the U.S. Government Accountability Office.
References
Bai, J., T.I. Wahl and J.J. McCluskey. 2008. Consumer choice of retail food store formats in Qingdao, China. Journal of International Food and Agribusiness Marketing 20(2): 89-109. https://doi.org/10.1080/08974430802186217
Bean, M. and J.S. Sharp. 2011. Profiling alternative food system supporters: the personal and social basis of local and organic food support. Renewable Agriculture and Food Systems 26(3): 243-254. https://doi.org/10.1017/S1742170511000032
Bell, D.R. and J.M. Lattin. 1998. Shopping behavior and consumer preference for store price format: why ‘large basket’ shoppers prefer EDLP. Marketing Science 17(1): 66-88. https://doi.org/10.1287/mksc.17.1.66
Berinsky, A.J., M.F. Margolis and M.W. Sances. 2014. Separating the shirkers from the workers? Making sure respondents pay attention on self-administered surveys. American Journal of Political Science 58(3): 739-753. https://doi.org/10.1111/ajps.12081
Bodkin, C.D. and E. Sewell. 2012. Assessing grocery store attraction via cross-shopping linkages. Business and Economic Research 2(2): 209-229. https://doi.org/10.5296/ber.v2i2.2572
Brunsø, K., J. Scholderer and K.G. Grunert. 2004. Closing the gap between values and behavior – a means-end theory of lifestyle. Journal of Business Research 57(6): 665-670. https://doi.org/10.1016/S01482963(02)00310-7
Bustillos, B., J.R. Sharkey, J. Anding and A. McIntosh. 2009. Availability of more healthful food alternatives in traditional, convenience, and nontraditional types of food stores in two rural Texas counties. Journal of the American Dietetic Association 109(5), 883-889. https://doi.org/10.1016/j.jada.2009.02.011
Carpenter, J.M. and M. Moore. 2006. Consumer demographics, store attributes, and retail format choice in the US grocery market. International Journal of Retail and Distribution Management 34(6): 434-452. https://doi.org/10.1108/09590550610667038
Cassill, N.L., N.C. Williamson, M. McEnally and J. Thomas. 1994. Department store cross-shoppers. Journal of Applied Business Research 10(4), 88.
Chen, L.A. and L. House. 2022. Food lifestyle patterns among contemporary food shoppers. International Journal of Consumer Studies 46(3): 944-963. https://doi.org/10.1111/ijcs.12739
Cleveland, D.A., N.M. Müller, A.C. Tranovich, D.N. Mazaroli and K. Hinson. 2014. Local food hubs for alternative food systems: a case study from Santa Barbara County, California. Journal of rural studies 35, 26-36. https://doi.org/10.1016/j.jrurstud.2014.03.008
Costanigro, M., D.T. McFadden, S. Kroll and G. Nurse. 2011. An in-store valuation of local and organic apples: the role of social desirability. Agribusiness 27(4): 465-477. https://doi.org/10.1002/agr.20281
Cude, B.J. and M.A. Morganosky. 2001. Why do consumers cross-shop between different types of food retail outlets? Journal of Food Distribution Research 32(3): 14-23.
Food Marketing Institute. (2018). Supermarket facts. Food Marketing Institute. Available at: https://www.fmi.org/our-research/supermarket-facts
Food Marketing Institute. (2019). U.S. grocery shopper trends – A look at today’s grocery shopper. Available at: https://www.fmi.org/docs/default-source/webinars/trends-a-look-at-today%27s-grocery-shopperslides-pdf.pdf
Fox, E.J., A.L. Montgomery and L.M. Lodish. 2004. Consumer shopping and spending across retail formats. Journal of Business 77(2): S25-S60. https://doi.org/10.1086/381518
Friedmann, H. 2007. Scaling up: bringing public institutions and food service corporations into the project for a local, sustainable food system in Ontario. Agriculture and Human Values 24(3): 389-398. https://doi.org/10.1007/s10460-006-9040-2
Gauri, D.K., R.P. Jindal, B. Ratchford, E. Fox, A. Bhatnagar, A. Pandey, J.R. Navallo, J. Fogarty, S. Carr and E. Howerton. 2021. Evolution of retail formats: Past, present, and future. Journal of Retailing 97(1): 42-61. https://doi.org/10.1016/j.jretai.2020.11.002
Gummer, T., J. Roßmann and H. Silber. 2021. Using instructed response items as attention checks in web surveys: Properties and implementation. Sociological Methods & Research 50(1): 238-264. https://doi.org/10.1177/0049124118769083
Guptill, A. and J.L. Wilkins. 2002. Buying into the food system: trends in food retailing in the US and implications for local foods. Agriculture and Human Values 19(1): 39-51. https://doi.org/10.1023/A:1015024827047
Hansen, T. 2003. Intertype competition: specialty food stores competing with supermarkets. Journal of Retailing and Consumer Services 10(1): 35-49. https://doi.org/10.1016/S0969-6989(01)00038-8
Hansen, T. 2005. Consumer adoption of online grocery buying: a discriminant analysis. International Journal of Retail and Distribution Management 33(2): 101-121. https://doi.org/10.1108/09590550510581449
Hempel, C. and U. Hamm. 2016. How important is local food to organic-minded consumers? Appetite 96: 309-318. https://doi.org/10.1016/j.appet.2015.09.036
Hino, H. 2014. Shopping at different food retail formats: understanding cross-shopping behavior through retail format selective use patterns. European Journal of Marketing 48(3-4), 674-698. https://doi.org/10.1108/EJM-12-2011-0764
Hsieh, M.F. and K.W. Stiegert. 2012. Store format choice in organic food consumption. American Journal of Agricultural Economics 94(2): 307-313. https://doi.org/10.1093/ajae/aar100
Huddleston, P., J. Whipple, R.N. Mattick and S.J. Lee. 2009. Customer satisfaction in food retailing: comparing specialty and conventional grocery stores. International Journal of Retail and Distribution Management 37(1): 63-80. https://doi.org/10.1108/09590550910927162
Hwang, J. and J.-E. Chung. 2019. What drives consumers to certain retailers for organic food purchase: The role of fit for consumers’ retail store preference. Journal of retailing and consumer services 47: 293-306. https://doi.org/10.1016/j.jretconser.2018.12.005
Jacobs, S., D. Van der Merwe, E. Lombard and N. Kruger. 2010. Exploring consumers’ preferences with regard to department and specialist food stores. International Journal of Consumer Studies 34(2): 169-178. https://doi.org/10.1111/j.1470-6431.2009.00844.x
Johnson, R. and T. Cowan. 2019. 2018 Farm bill primer: support for local food systems. Available at: https://fas.org/sgp/crs/misc/IF11252.pdf
Johnston, J. and L. Baker. 2005. Eating outside the box: FoodShare’s good food box and the challenge of scale. Agriculture and Human Values 22(3): 313-325. https://doi.org/10.1007/s10460-005-6048-y
Jones, M.S., L.A. House and Z.F. Gao. 2015. Respondent screening and revealed preference axioms: testing quarantining methods for enhanced data quality in web panel surveys. Public Opinion Quarterly 79(3): 687-709. https://doi.org/10.1093/poq/nfv015
Klein, A. and G. Schmitz. 2016. Main-format dominance in consumers’ FMCG cross-format shopping patterns. Journal of Retailing and Consumer Services 30: 105-115. https://doi.org/10.1016/j.jretconser.2015.11.011
Leibtag, E., C. Barker and P. Dutko. 2010. How much lower are prices at discount stores? An examination of retail food price. USDA-ERS Economic Research Report No. 105. https://doi.org/10.2139/ssrn.1711350
Lusk, J.L. and B.C. Briggeman. 2009. Food values. American Journal of Agricultural Economics 91(1): 184-196. https://doi.org/10.1111/j.1467-8276.2008.01175.x
Martinez, S.W. 2007. The U.S. food marketing system: recent developments, 1997-2006. Report ERR-42. U.S. Department of Agriculture, ERS, Washington, DC, USA.
Martinez, S.W. 2016. Hedonic analysis of US fresh produce prices at direct-to-consumer sales outlets versus competing retailers. British Food Journal 118(7): 1665-1681. https://doi.org/10.1108/BFJ-05-2015-0183
McNeil, M. 2018. U.S. organic sales break through $50 billion mark in 2018. Organic Trade Association, Washington, DC, USA. Available at: https://ota.com/news/press-releases/20699
Meas, T., W. Hu, M.T. Batte, T.A. Woods and S. Ernst. 2015. Substitutes or complements? Consumer preference for local and organic food attributes. American Journal of Agricultural Economics 97(4): 1044-1071. https://doi.org/10.1093/ajae/aau108
Mount, P. 2012. Growing local food: scale and local food systems governance. Agriculture and Human Values 29(1): 107-121. https://doi.org/10.1007/s10460-011-9331-0
Nilsson, E., T. Garling, A. Marell and A.C. Nordvall. 2015. Importance ratings of grocery store attributes. International Journal of Retail and Distribution Management 43(1): 63-91. https://doi.org/10.1108/IJRDM-12-2012-0112
Nousiainen, M., P. Pylkkänen, F. Saunders, L. Seppänen and K.M. Vesala. 2009. Are alternative food systems socially sustainable? A case study from Finland. Journal of Sustainable Agriculture 33(5): 566-594. https://doi.org/10.1080/10440040902997819
Pan, Y. and G.M. Zinkhan. 2006. Determinants of retail patronage: a meta-analytical perspective. Journal of Retailing 82(3): 229-243. https://doi.org/10.1016/j.jretai.2005.11.008
Prasad, C.J. 2010. Effect of consumer demographic attributes on store choice behaviour in food and grocery retailing-an empirical analysis. Management and Labour Studies 35(1): 35-58. https://doi.org/10.1177/0258042X1003500103
Prasad, C.J. 2014. Consumers’ cross-format shopping behavior in an emerging retail market: Multiple discriminant analysis. Journal of International Consumer Marketing 26(1): 29-57. https://doi.org/10.1080/08961530.2014.848081
Prasad, C.J. and A.R. Aryasri. 2011. Effect of shopper attributes on retail format choice behaviour for food and grocery retailing in India. International Journal of Retail and Distribution Management 39(1): 68-86. https://doi.org/10.1108/09590551111104486
Reardon, T. 2011. The global rise and impact of supermarkets: an international perspective. In: Crawford Fund 17th Annual Parliamentary Conference: The supermarket revolution in food: good, bad or ugly for the world’s farmers, consumers and retailers? Canberra, ACT, Australia, 14-16 August 2011.
Reutterer, T. and C. Teller. 2009. Store format choice and shopping trip types. International Journal of Retail and Distribution Management 37(8): 695-710. https://doi.org/10.1108/09590550910966196
Sonnino, R. and T. Marsden. 2006. Beyond the divide: rethinking relationships between alternative and conventional food networks in Europe. Journal of Economic Geography 6(2); 181-199. https://doi.org/10.1093/jeg/lbi006
Taylor, R. and S.B. Villas-Boas. 2016. Food store choices of poor households: a discrete choice analysis of the National Household Food Acquisition and Purchase Survey (FoodAPS). American Journal of Agricultural Economics 98(2): 513-532. https://doi.org/10.1093/ajae/aaw009
Toler, S., B.C. Briggeman, J.L. Lusk and D.C. Adams. 2009. Fairness, farmers markets, and local production. American Journal of Agricultural Economics 91(5): 1272-1278. https://doi.org/10.1111/j.14678276.2009.01296.x
Tran, V.H. and L. Sirieix. 2020. Shopping and cross-shopping practices in Hanoi Vietnam: An emerging urban market context. Journal of Retailing and Consumer Services 56: 102178. https://doi.org/10.1016/j.jretconser.2020.102178
U.S. Department of Agriculture – Agricultural Marketing Service (USDA-AMS). 2020. Local Food Directories: National Farmers Market Directory. Available at: https://www.ams.usda.gov/local-food-directories/farmersmarkets
U.S. Department of Agriculture – Economic Research Service (USDA-ERS). 2015. FoodAPS National Household Food Acquisition and Purchase Survey. Available at: https://www.ers.usda.gov/dataproducts/foodaps-national-household-food-acquisition-and-purchase-survey/
UCLA Statistical Methods and Data Analytics. nd. Comparing regression coefficients across groups using suest. Available at: https://stats.oarc.ucla.edu/stata/code/comparing-regression-coefficients-acrossgroups-using-suest/
Volpe, R., A. Kuhns and T. Jaenicke. 2017). Store formats and patterns in household grocery purchases. Available at: https://www.ers.usda.gov/webdocs/publications/82929/eib-167.pdf?v=42814
Volpe, R., E.C. Jaenicke and L. Chenarides. 2018. Store formats, market structure, and consumers’ food shopping decisions. Applied Economic Perspectives and Policy 40(4): 672-694. https://doi.org/10.1093/aepp/ppx033
Corresponding author
