Strategic Management for Community-Based Markets: From Consumers’ Perspectives and Experiences

: This study used a consumer survey to identify resources and services that are important to farmers’ market (FM) shoppers. The questionnaire was distributed onsite in six FMs in Tennessee, and a total of 506 FM shoppers responded. The most important resources and services in terms of a shopper’s decision to visit a FM are identiﬁed as supporting local food, quality, friendly, and diverse vendors, and food origin. Multivariate analysis of variance (MANOVA) reveals that convenience and close to home are more important to women; price is more important to younger and lower-income shoppers, while quality, convenience, interaction with farmers/producers, and food origin are more important to older shoppers. Content analysis indicated that FM shoppers were impressed with FM atmosphere and liked the quality, variety, and convenience provided by FMs, but disliked not having clear information such as product labels and websites.


Introduction
With the current globalization and industrialization of the food system, the distance between food production and consumption has grown larger and become less transparent to consumers. The large number of food safety recalls and outbreaks have led to consumer demand for more transparency in food production and information about food origin [1]. The concerns regarding the current industrial food system and food safety have raised consumer awareness and increased the demand for local and community-based food systems such as farmers' markets (FMs) [2,3].
To meet the growing demand, the number of FMs has dramatically grown over the past 15 years in the U.S. from 3706 in 2004 to 8771 in 2019 [4]. It was estimated that a total of 20,000 farmers sold their produce and a million consumers purchased fresh produce weekly from FMs [3]. The United State Department of Agriculture (USDA) [5] defined farmers' markets as "a multi-stall market at which farmer-producers sell agricultural products, particularly fresh fruits and vegetables but also meat products, dairy products, and/or grains, directly to the general public at a central or fixed location".
FMs provide benefits to both farmers/vendors and consumers. For example, small or beginning farmers and vendors can develop their business [6] and increase their revenue [7,8]. Consumers have an alternative channel to purchase fresh and nutritious produce [3,7] which can lead to a healthier lifestyle (e.g., increasing consumption of fruits and vegetables) [7,9]. Consumers can learn about product information from farmers/vendors directly, which creates social connections and trust [3,10]. Onyango, Govindasamy, and Alsup-Egbers [7] pointed out that FMs play an important role for consumers to learn firsthand knowledge of agriculture. Some barriers, including distance, transportation, higher prices, and limited hours of operation, may discourage FM shopping [11,12].
To provide better resources and services to FM shoppers and increase market share, it is important to understand FM shoppers' characteristics and views. Despite many FM consumer studies that have identified important FM resources/services that may impact FM shoppers' decisions, very little is known about how these resources/services may differ for different demographic groups and how FM shoppers' views of current FM resources/services they visit may differ depending on the frequency of visits and amount of money spent. This information will provide useful insights into FM shoppers' expectations and help market managers to allocate their resources/services more efficiently and develop better marketing strategies to better serve their customers.

Characterics of FM Shoppers
Prior research indicated that the majority of FM shoppers are more likely to be female, white, more highly educated, having higher income, and over 40 years of age [2,[13][14][15][16] and from a two-or three-person household [17,18]. Researchers found that middle-aged and older FM shoppers are more interested in supporting locally grown food. Women often do most of the cooking and grocery shopping, which might explain why a majority of FM shoppers are female [19][20][21].

Frequency of Visits, Amount of Spending, and Travel Distance
The frequency of visits and amount of spending are important to a successful FM. Studies showed that the amount of spending indicated support for locally grown food and frequency of visits. Gumirakiza and Curtis [22] used survey data collected from 819 FM consumers at four FMs in Utah. They categorized respondents into three groups based on the amount of spending. They reported that the high-spending group visited FMs 8-12 times per summer season and spent $67.97 each visit, but comprised the smallest proportion of FM shoppers. The medium-spending group was the largest group of FM shoppers. They visited FMs 4-7 times per summer season and spent $24.78 per visit. The low-spending group was also the youngest group. They attended FMs 4-7 times and spent $10.54 per visit. Govindasamy, Italia, and Adelaja [23] conducted a survey study of 21 FMs in New Jersey and reported that shoppers spent $16 on average and that the majority of shoppers visited FMs once a week, bi-weekly, or once a month. Baker, Hamshaw, and Kolodinsky [15] examined two farmers' markets in Grand Isle County, Vermont, and reported an average spending for a consumer of $13.46 at the southern site and $19.68 at the northern site. They concluded that consumers spent more if the market had a higher mix of non-food vendors. Although past research did not offer a clear conclusion on how travel distance impacts visit frequency, researchers found that FM shoppers traveled in a range of 6-17 miles or within 17-18 min [14,15,17,24].

Shoppers' Sources of Awareness of FMs
FM managers can develop better promotion plans and marketing strategies if they know how FM shoppers become aware of an FM. Studies have examined marketing strategies such as attractive visual signs and flags, newspapers, word of mouth, radio, and TV. Govindasamy et al. [23] collected data from 336 participants from New Jersey FM consumers using a survey. They found that FM shoppers learned about the FM through road signs (50%), newspapers (49%), passing by (49%), and word of mouth (42%). Onianwa, Mojica, and Wheelock [17] used a combination of face-to-face and surveys to collect data from two FMs in Alabama. They reported that respondents learned about FMs through word of mouth (69%) and drive-by (11%). Keeling Bond, Thilmany, and Bond [18] conducted a national consumer survey with 1549 respondents and reported using television ads to reach out to new customers and using e-mail newsletters, radio, and magazines to increase visit frequency for existing shoppers, also encouraging vendors to connect with their regular shoppers through setting up e-mail lists. Moreover, information booths and sampling were preferable promotion methods to provide shoppers information about food nutrition and market updates. Studies indicated that the most common and effective way to attract new customers is through attractive visual signs and flags [14,15,23].
Price has been reported as an important actor for FM consumers, and in general, consumers expect to pay lower prices at FMs compared to other stores [23]. However, prior research had different findings for price at FMs. Several studies suggested that FMs provide cheaper produce price than produce from other outlets such as supermarkets, grocery stores, or convenience stores [8,[27][28][29][30], while some studies found that food is more expensive at FMs than grocery stores [12,13,31]. Valpiani et al. [31] investigated price differences across three different outlets (34 FMs, 20 roadside stands, and 32 supermarkets) in North Carolina. They found that fruit and vegetables were comparatively expensive at FMs compared to the other two outlets; however, price factors such as being locally grown and organic should be considered. Larger FMs (more than 20 vendors) priced higher than smaller FMs, which might be due to higher infrastructure costs, having more amenities, and having a more affluent customer base. They concluded that having comparable or cheaper prices at FMs could not only attract affluent consumers but also encourage price-conscious consumers to shop at FMs. Lucan et al. [12] accessed 26 FMs in New York and 44 nearby stores. They also reported that FMs' produce was more expensive than the produce sold at nearby stores. Higher produce costs might be caused by selling more exotic/heirloom produce and a higher proportion of organic produce at FMs.
Gumirakiza, Curtis, and Bosworth [8] accessed consumers' motivations attending FMs and found that if consumers' primary motivation is purchasing produce, FM attributes such as parking and activities would be less important to these consumers. The consumer characteristics that increased the probability of attending a FM primarily for social interaction are family size, having favorite vendors, and local citizens. They also reported that consumers are seeking to purchase ready-to-eat food at FMs. Thilmany, Bond, and Keeling Bond [32] explored consumer behavior and motivations for visiting direct food markets. They found that the desire to support local businesses/producers is a better predictor of visit frequency than demographics and that frequent FM shoppers have a much higher desire to support local businesses/producers and place more value on quality, locally grown produce, and relationships with producers. Packaging and color were less important to this group of shoppers. Tsai et al. [33] conducted a FM consumer survey with closed-ended survey questions in different parts of Taiwan. They examined the relationships among institutional trust, interpersonal trust, and purchase intention. They found that consumers' purchase intention is influenced by both institutional and interpersonal trust, which suggested that providing higher-quality produce and having interaction and communication face-to-face can better earn customers' loyalty and trust.

Methodology of the Study
A questionnaire was developed based on a literature review [21][22][23][24][25]27]. Data were collected from six FMs in Knox County and Roane County in October and November 2019. Prior to conducting this study, permission to distribute the questionnaire to FM shoppers was obtained from the market managers. On the day of distributing the questionnaires, the researchers set up a booth and a sign to show FM consumers the purpose of the study, institutional affiliation of the researchers, and researchers involved in the study. When FM shoppers approached the booth, the researchers introduced themselves and explained the purpose of the study before the consent letter and questionnaire were distributed. The participants were asked to read the consent letter and instructions before answering the questionnaire.
The selected markets included a variety of market traits. Three were open two days a week (a weekday and a weekend day) and three operated one day a week (either a weekday or a weekend day). Days of operation varied from Wednesday through Saturday, and months of operation varied from April to November. Two FMs were located in downtown shopping areas, two located in church parking lots, one in a business center parking lot, and one in a park parking lot.
The questionnaire comprised five parts. Part A asked about the respondent's behavior (e.g., source of awareness of the FM, travel distance, frequency of visit, money spent, and products purchased). Part B asked the respondent to indicate the relative importance (on a seven-point Likert scale) of 13 factors on their decision to visit a FM. Part C asked the participant's impression (like or dislike) about the FM they visited that day. Part D asked the participant to indicate the level of agreement (on a seven-point Likert scale) about 15 aspects of their experience with the FM they visited that day, and Part E asked about the respondent's demographic information. The Statistical Package for the Social Sciences (SPSS) Version 26 was applied to data analysis using frequencies, exploratory factor analysis (EFA), and one-way multivariate analysis of variance (MANOVA). Excel was used for content analysis for three open-ended items.
Prior to analysis, data were examined for the accuracy of data entry, missing values, outliers, normality, linearity, and homoscedasticity. Variables were either transformed or eliminated to meet the assumptions. We applied Mahalanobi's Distance Test [34] to identify and remove 17 multivariate outliers. To evaluate the reliability and validity of the instrument, Cronbach's alpha and EFA using principal components analysis with varimax rotation were adopted for Part B (9 items) and Part D (15 items) of the instrument. A Cronbach's alpha value of 0.6 was considered acceptable [35]. Convergent validity was indicated by factor loading, average variance extracted (AVE), and composite reliability. Retention criteria for instrument items were as follows: (1) factor loading greater than 0.3 [36]; (2) communalities greater than 0.2 [37]; (3) an AVE of 0.5 or more; and (4) composite reliability (CR) greater than 0.7 [38].
MANOVA was used to detect the potential differences among FM shoppers' demographic groups with regard to 13 items on their decision to visit a FM. Similarly, we applied MANOVA to test 15 items associated with the FM the participants visited that day when controlling the frequency and money they spent on average. Analysis of variances (ANOVA) was conducted as a follow-up test to the MANOVA, with each ANOVA tested at the 0.004 level (0.05 divided by the number of ANOVAs conducted). The Bonferroni post hoc analysis was conducted if the ANOVA results revealed significant differences.

Participant Profiles
A total of 506 FM consumers participated in the survey. Table 1 presents the summary of the demographics of the participants. The majority of participants were local residents (88.4%), female (71.2%), white (85%), and married (56.9%), with a household size of 2 (52%), without children (78%), with a college degree or higher (78.4%), and with a household income of $50,000 and above (65.6%).  Table 2 summarizes the shopping experience of the participants at a given FM. The participants primarily learned about the visited FM through word of mouth (50.4%), followed by walking/driving past the location (25.9%). The majority of the participants (90%) traveled within 15 miles to the FM and visited the market at least 1-2 times per month or more. About 48% of the participants indicated that they spent between $11 and $25 on a typical trip to a FM. More than two-fifths of the participants indicated that they spent between $11 and $25 on the day of the survey, with a mean of $25.35 (minimum $0 and maximum $200). When we excluded the participants who spent more than $50, the mean was $18.12. The top three products that the participants sought to purchase on the day they visited the FM were vegetables, fresh fruit, and baked goods (77.8%, 57.1%, and 55.3%, respectively). When we asked the participants to indicate services/programs to include in the FM, more than half said that they would like to see food waste diversion, recycling collection, and free/limited usage of plastic programs implemented in the FM.

Participants' Feedback
Three open-ended questions about the participants' impression and what they liked or disliked about the FM they visited that day were included in the questionnaire. These qualitative items were organized by themes. Depending on how the participants answered the open-ended items, such as several sentences versus one word, the themes were coded based on how the participants responded. Therefore, the total number of themes might not be equal to the number of participants answering the question.

The Impression of the FM the Participants Visited
The participants were asked to describe their impression of the FM they visited. The top three most mentioned themes were atmosphere (25.9%), satisfaction (20.2%), and products (18.9%). It is worth mentioning that there was a new FM in its first year at a new location. The size of the FM was small compared to other FMs, so it is not surprising that the participants mentioned the small size and needing more vendors for this question (Table 3). Table 3. Impression of the FM (n = 454).
"love the layout", "organized", "it is setup nice", and "I'm a large guy would like wider stalls". Price 10 1.5 "some food is a little expensive", "good prices", "not inexpensive price", and "fair prices".
Products 130 18.9 "it was very diverse and filled with almost everything you could want", "lots of choices", "high quality and lots of variety", and "lots of fresh produce". Satisfactory 139 20.2 "I enjoy all of the people", "love it and have been coming regularly every month", "I wanted to stop shopping at the big box grocery to come here", "my favorite", and "love it. Strong sense of pride in product and community". Size 71 10.3 "how large it actually was", "big", "small", and "that was huge".

Reasons the Participants Liked the FM They Visited
We asked the participants to list the top three reasons they liked the FM they visited. The top three themes were quality, variety, and convenience (15.9%, 12.3%, and 12%, respectively). Many participants indicated that they liked to visit the FM because they were able to find some products that were not sold in grocery stores. Close to home and easy/free parking were the most mentioned under the convenience theme. Although social interaction was not among the top reasons, many participants indicated that having such interactions with vendors encouraged them to try new food items. Moreover, the participants also described FMs as places they can meet with friends and make friends with vendors (Table 4). "good product selection", "variety", "food options", "diversity of goods", and "various items such as arts, breakfast/lunch, VG bakery".

Reasons for Disliking the FM They Visited
The participants were asked to list the top three reasons why they disliked the FM. More than half (50.4%) of them did not answer or indicate anything they disliked. Among the participants who answered, organization (20.9%), variety (19.1%), and uncontrolled factors (18%) were the top three reasons for their disliking the FM. Under the organization theme, several participants indicated that they would like to have more clear information about nutritional facts and products such as product name, cost, and content weight, if in a bag. The participants also pointed out that plastic bag usage should be limited/reduced. They wanted to have more selections besides fresh produce (Table 5). "Website hard to understand", "too few hours", "vendors don't display if they use chemicals or not", "no nutritional facts (fat content)", "I would like demonstrations of cooking, canning etc". and "use of plastic bags/cups". Variety 66 19.1% "limited options", "need more selections", "small amount bakery options", "mostly veggies of same type", and "very little arts/crafts/pottery". Uncontrolled 62 18.0% "too many people with babies being pushed in carts in crowded area", heat", "sometimes it gets crowded with tourists", "weather-hot in summer", "outdoor", "inside option would be nice", and "street baggers homeless citizens". Price 43 12.5% "expensive", "high cost", and "I think it is important to support farmers, but prices are on the high side".

Vendors 34 9.9%
"some vendors stop coming later in the season, not enough fruit vendors", "number of vendors remaining due to lack of area support", and "meat man is not here today".
Parking traffic 29 8.4% "parking costs money due to football", "parking can be difficult, particularly on game days", "traffic", and "pay for parking". Size 27 7.8% "not large enough", "small", "would love to see it expand more", "wish it was larger and year-round", and "don't think the vendors gave it a chance, it was dwindling in size so fast". Payment 12 3.5% "credit card minimums (who carries cash), its COD but that's not anyone's fault", "need more markets that accept SNAP benefits", and "wish more vendors took credit cards".

Reliability and Validity of the Instrument
The range of Cronbach's alphas was from 0.55 to 0.84. Although one construct was below the threshold of 0.6, prior studies suggested that values between 0.45 and 0.98 were acceptable [39,40] because a low value of alpha can be improved to an acceptable level by increasing the number of questions to the construct [39,41]. Therefore, the instrument had acceptable internal consistency.
The communalities (0.36-0.8) and factor loading (0.45-0.89) of all items were above the thresholds. AVE and CR were computed for each construct. All CR values met the acceptable threshold of 0.7 (0.7-0.89). The cut-off point for AVE was 0.5 or greater. Two constructs did not meet the AVE cut-off point. However, if AVE is less than 0.5, but the CR value is greater than 0.6, the convergent validity is confirmed [38]. For Part B and Part D of the instrument, four factors were extracted for both parts and explained 67.62% and 56.84%, respectively, of the total variance.
Follow-up ANOVA results (Table 7) indicate that gender significantly differs for convenience and close to home. We found significant differences in price, quality, convenience, food origin, interaction with farmers/producers, and close to home for age groups. The Bonferroni post hoc analysis revealed that younger shoppers ranked price as more important than did older shoppers, while the older group ranked quality, convenience, interaction with farmers/producers, and food origin as of greater importance than the younger group in general. Close to home was especially important to the age groups of 35-44 and 65 and above. For income category, price and food origin were found to have significant differences. It was not surprising to learn that the lower-income group ranked price as more important than did the higher-income group, and that the higher-income group viewed food origin as more important than did the lower-income group based on post hoc analysis.

Agreement on the Services and Resources Provided by the FM
We asked the participants to indicate their level of agreement regarding the services/resources provided in the farmers' market they visited. Fifteen items were used to measure. The item "the activities organized by the market are interesting" was dropped from further analysis due to no activities organized for four out of six FMs and no activities organized on the day of survey distribution. In general, the participants strongly agreed that the FM they visited had friendly vendors and atmosphere (mean = 6.58 ± 0.69), offered high-quality food (mean = 6.44 ± 0.81), and the market location was convenient (mean = 6.25 ± 0.97). The participants disagreed that the FM was overcrowded (mean = 2.66 ± 1.51) and neither agreed nor disagreed on market needs to provide more promotional or educational activities and on whether the market should be near restaurants or cafés (mean = 3.84 ± 1.50 and mean = 3.75 ± 1.87, respectively) ( Table 8). Table 8. Participants' agreement on services and resources provided by FMs.

Variable Mean a SD
The market has friendly vendors and atmosphere 6.58 0.69 The market offers high-quality food 6.44 0.81 The location of the market is convenient 6.25 0.97 The market offers a variety of products/produce 6.04 1.03 The market is a good place to spend time with family and friend 6.00 1.17 The market has clear signs and is easy to find 5.60 1.23 The market offers affordable products 5.49 1.23 The market has ample parking spaces 5.34 1.44 The traffic flow within the market is good 5.33 1.31 The market should expand its operation hours 4.30 1.62 The market needs more diversity of vendors 4.24 1.67 The market needs to provide more promotional or educational activities 3.84 1.50 It is important for me that the market is nearby restaurants and cafes 3.75 1.87 The market is overcrowded 2.66 1.51 a Scale: 1 (strongly disagree) to 7 (strongly agree).
The results from MANOVA indicated that there was a statistically significant difference in the services and resources the FM provided based on visit frequency and average amount of money spent (Pillai's trace = 0.33, F(56, 1904.00) = 3.03, p < 0.001, multivariate η 2 = 0.082, and Pillai's trace = 0.19, F(42, 1386) = 2.19, p < 0.001, multivariate η 2 = 0.062, respectively). The follow-up ANOVA results revealed that visit frequency significantly differs if the market offers high-quality food and has friendly vendors and atmosphere, if the location of the market is convenient, if the market has ample parking, and if the market offers a variety of products. The Bonferroni post hoc analysis revealed that the first-time group had less agreement on the market offering high-quality food and having friendly vendors and atmosphere in comparison with other groups. Shoppers who visited the FM every week and first-time shoppers showed significant difference in agreement with the market offering a variety of products. Significant differences were also found in convenience of market location among those visiting the FM every week, 1-2 times per month, 1-4 times per year, and the first-time group. The market having ample parking was found to have significant differences among groups that visited every week, three times per month, and first-time shoppers.
ANOVA results indicated that the average amount of money spent significantly differed for high-quality food, friendly vendors and atmosphere, and variety of products. The post hoc analysis revealed that the group spending between $1 and $10 was significantly different for high-quality food, friendly vendors and atmosphere, and variety of products compared to all other groups (Table 9). Table 9. Visit frequency and average money spent related to agreement on services and resources.

Services and Resources
High Quality

Discussion
FMs can be used to promote a healthy diet, stimulate local economies, and support local farmers/small businesses. Studies related to FM shoppers' topics have examined consumers' perceptions, motivations, and preferences. This study examined what FM resources/services are important to Tennessee FM shoppers and the quality of FM services and resources provided by Tennessee FMs. The study found that FM shoppers' characteristics were consistent with those in previous consumer studies [2,[14][15][16]. It is not surprising to find that the majority of FM shoppers were white due to white population proportions in Knox County and Roane County (85.9% and 94.2%, respectively) [42], and women's role in cooking and grocery shopping [19][20][21].
Several studies indicated that FM shoppers tended to be over 40 years of age [2,14,16]. FM shoppers in our study formed three major groups: 37.4% were between the age of 18 and 34, 22.8% were between 35 and 54, and 39.6% were above 55. About half of the respondents reported household incomes above $75,000, while the median household income was $54,437 in Knox County and $50,003 in Roane County [42]. Moreover, while 78.7% of national FMs accepted Supplemental Nutrition Assistant Program (SNAP), a federal government program that provides food to help low-income Americans [43], only two out of the six selected FMs accepted SNAP, which might indicate that our selected FMs attract wealthier shoppers.
Although studies had identified different marketing media, word of mouth was the most mentioned by participants as to how they initially learned about the market. However, Keeling Bond, Thilmany, and Bond [44] indicated that there was no linkage between credibility of word of mouth and probability of purchase at FMs. It is important to note that more than a quarter of the participants visited the FM because of walking/driving past the location. The results suggested the importance of having attractive visual signs to attract new shoppers. According to the USDA Farmers' Market Manager study [43], fruit and vegetables, condiments and sauce, and bread and baking goods were the top three product types sold in FMs (99.6%, 94.1%, and 90.9%, respectively). There is no doubt that fresh produce was the main reason for participants to visit the FM. It is worth noting that more than half of the participants were seeking to purchase bakery products and more types of food besides fresh produce. As prior studies indicated, there were demands for refined and processed products (ready-to-eat food) at FMs [8,12,45]. Interestingly, studies in other countries have similar findings in consumer characteristics [46,47], location of FM [46], travel distance [46] and motivations [46,47]. One distinct difference between our studies' FMs and other countries' FMs is that wholesalers are prohibited at the markets, while wholesalers are allowed in other countries [46,47].
Our study indicated that FM shoppers desire to see food waste diversion in FMs. It is interesting to learn that some of vendors in our selected FMs did donate their unsold food to either food banks or non-profit organizations; however, consumers were probably unaware of this, unless when asked. In 2019, more than one-third of national FMs participated in food waste diversion programs. With those that engaged in food waste diversion programs, 66.5% of FMs engaged in food donation to food banks, and 41% participated in composting [43]. At the time of our survey, only a few vendors donated their unsold food, which was not organized by FMs. It is worth noting that the volume of food waste generated is considerably low at FMs since farmers have limited volume and selection and pick their products close to their market days [48].
Our findings indicated that most participants had a positive impression of the market. The participants were impressed with the atmosphere the FM provided, which suggests the importance of creating an inviting/welcoming atmosphere for shoppers. Studies indicated that consumers prefer product labels as their information source to learn more about their products [49] and use product labels to make their purchasing decisions [18,50]; however, our findings indicated that food products sold in our selected FMs did not display clear information on their packages. FM shoppers demand clear information about nutritional facts such as product name, cost, and content weight. Moreover, Keeling Bond, Thilmany, and Bond [44] found that frequent shoppers seek pesticide-free, nutritionally enhanced, and locally grown produce. Research indicated that about 85% of FMs used "locally grown" in market labelling, but less than half of FMs used terms such as "free-range", "grass-fed", and "gluten-free" in labelling [43]. To increase market share, FM managers can develop labelling policies to assist FM shoppers in identifying product attributes they are looking for [51]. It is interesting to learn that 46.7% of Italian producers and 37.4% of consumers favor labelling. Within those producers and consumers, only 28.6% of producers compared to 46.6% of consumers think providing descriptive tables necessary [48], and some Australian FMs had a labelling system to distinguish their producers from wholesalers and use it as a tool to assist their shoppers to make purchase decisions [46]. While our selected FMs attracted wealthier shoppers, we found that price is of greater importance to younger and lower-income shoppers. Quality, convenience, food origin, interaction with farmers/producers, and being close to home are more important to older groups. Our results show differences in visit frequency and amount of money spent regarding agreement on the resources and sources provided. These results indicate that frequent shoppers spend more and have greater agreement on product quality, variety, friendly vendors, and location of market. The difference might be that frequent FM shoppers look for high-quality fresh produce and interaction with vendors/producers, not a large variety of products or location convenience; therefore, frequent FM shoppers agree more on the market offering high-quality food, a variety of products, friendly vendors and atmosphere, convenient location, and ample parking [8,18]. Therefore, it would be easier to attract new FM shoppers if the market location has a high level of foot traffic and attractive and inviting displays [18]. To maintain frequent shoppers, it is important to promote locally grown produce and support local businesses.

Conclusions
Our results suggested several key marketing strategies for market managers to encourage shoppers to visit frequently and attract new customers. First, having clear and attractive visual signs seems to be a cost-effective way to attract new customers' attention. A majority of the participants indicated that they learned about the market via word of mouth. Market managers can develop some activities such as bring-a-friend day with distributed coupons to encourage current shoppers to bring friends to the market. Second, having a mix of nonfood and ready-to-eat items increases consumer spending such as having local food trucks. Market managers can also encourage vendors who sell ready-to-eat items to join with other vendors if they sell ingredients they need. Third, participants pointed out high prices at the selected FMs, although some participants indicated that this was understandable. However, it is important to provide competitive prices to attract both wealthier shoppers and price-sensitive shoppers to extend the FMs' customer base. Fourth, high price might also be linked to a lack of product information. Therefore, if market managers can develop clear product label policies for vendors to follow and for consumers to learn more about product such as pesticide-free, non-GMO, gluten-free, grass-fed, weight, and nutritional fact, this might explain the higher prices at the FMs. Finally, the top programs that the majority of participants would like to see being included in FMs are environment-related programs such as food waste diversion and free/limited plastic usage. Moreover, with increasing consumers' environmental consciousness, more and more consumers are willing to pay more and change their purchase habits for businesses with sustainable practices [52]. Studies also indicated that consumers with concerns about the health of the environment and well-being tend to shop at FMs [47,53]. Therefore, FM managers should consider adopting some environmental practices to attract environmentally conscious consumers.
Food waste diversion, recycling collection, and limited/free usage of plastic have been mentioned to add to FMs. Further research may focus on how consumers value those programs in FMs, and the challenges and barriers from the point of view of market managers and vendors, especially for food donation programs, since more than one-third of national FMs participated in such a program [42]. A further study is needed to determine if different types of FM organizations influence their resources and funding opportunities. At the time of our study, we asked the participants if they would like to order online, but less than a quarter of them were interested. It would be interesting to see if COVID-19 changes FM shoppers' purchasing behavior.

Limitations
The present study has limitations. The sample population focused on participants who are FM customers in Knox and Roane Counties in Tennessee only. Due to selected FM policies, the researchers were unable to randomly select the sample; instead, the researchers had to wait until FM shoppers approached; therefore, the sample population might be biased. The data were collected close to the end of the season. Some opinions might not be included for FM shoppers who shop only for certain seasonal food items such as corn.
Author Contributions: C.-J.R.C. developed the survey instrument, conducted the project, analyzed the data, and drafted the whole manuscript; T.-H.E.Y. advanced the analyses and edited the manuscript; R.J.C.F. contributed the idea of this study, co-designed the survey instrument, supported the data collection financially, edited the manuscript, and directed the project. All authors have read and agreed to the published version of the manuscript. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.