Final Report for GS13-124
Project Information
A study evaluating how NC’s farmers’ market managers’ motivations influence SNAP/EBT availability and business opportunities for farms found no association between managers’ motivation scores and SNAP/EBT availability. The study did find a significant, positive association between business motivation and the number of local farmers participating at markets. Findings suggest logistical constraints may preclude offering SNAP/EBT at markets despite managers’ motivations.
Introduction
Farmers’ markets are important components of local food systems and they convey benefits to farmers and consumers. Farmers’ market customers benefit from the opportunity to interact directly with local farmers, and perceive that food sold at farmers’ markets tends to be fresher, less expensive, and higher quality than food purchased in supermarkets.1
Since 1970, the number of farmers’ markets in the US has grown from 340 to 7000, and the number more than tripled from 2000 to 2012.2 While they are gaining popularity, not all farmer’s markets are successful business operations. For example, while Oregon experienced the creation of 62 new farmers’ markets from 2000 to 2007, the state had a net gain of only 30 markets due to concurrent markets failures.Young markets are especially at risk for failure as they work to attract an appropriate balance of farms and customers.3
Farmers’ markets are endorsed by the Centers for Disease Control and Prevention (CDC) for their potential to increase access to healthful food for groups at risk for poor diet quality, like low-income households.4 Intervention studies have documented improved dietary intake in low-income individuals participating in the United States Department of Agriculture’s (USDA) Special Supplemental Nutrition Program for Women, Infants and Children (FMNP) and Supplemental Nutrition Assistance Program (SNAP) at farmers’ markets.5 Farms benefit from participating in these programs, as they are able to attract a customer base that would otherwise be unable to purchase their products.6 In 2011, farms nationwide redeemed $11.7 million in SNAP benefits, representing .02% of all SNAP redemptions in the US.7
Despite the potential for simultaneously increasing access to healthful food and expanding business opportunities for small- and moderate-size farms, only about 25% of farmers’ markets offer SNAP Electronic Benefit Transfer (EBT) (the mechanism by which SNAP participants can use their benefits) at their markets.8 Combined with other barriers to farmers’ market access for low-income consumers, such as transportation issues, cultural barriers, and lack of awareness of the benefits of farmers’ markets, the limited availability of opportunities to use SNAP/EBT at markets could prevent a large contingent of the population from shopping at farmers’ markets.6
While the methods for accomplishing the two goals of 1) improving business opportunities for small- and moderate-size farms and 2) expanding access to healthful foods among low-income households are well documented, the impact of managers’ perceptions of their roles in addressing these goals has not been evaluated.9 To develop and implement comprehensive strategies for ensuring farmers’ markets’ success in facilitating business opportunities for small- and moderate-size farms and expanding access to healthful food for low-income households, it is important managers’ motivations to achieve these goals, and the barriers and facilitators to achieving them.
The purpose of this project was to evaluate how managers' motivations influence 1) business opportunities for small and moderate size farms and 2) access to healthful food for low income households. The specific aims of this study follow: Aim 1: Examine associations between managers' motivations related to being facilitators of business opportunities for small and moderate size farms and farmers' market vendor recruitment, sales, and customer counts. Aim 2: Examine associations between managers' motivations related to being promoters of healthful foods and farmers' market Supplemental Nutrition Assistance Program Electronic Benefit Transfer (SNAP/EBT) availability and participation.
We hypothesized that compared to their counterparts: 1.) Managers who are motivated by providing business opportunities for small and moderate size farms foster greater business opportunities for these farms evidenced by farmers' market vendor recruitment, sales, and customer counts; and 2.) Managers who identify as promoters of healthful foods in their communities are more likely to operate programs that facilitate access to their markets for low income households evidenced by SNAP/EBT availability and participation.
Objective 1: The first objective of this study was to develop survey items assessing North Carolina's managers' motivations related to facilitating business opportunities for small and moderate size farms and access to healthful food for low income households. While there is extensive literature describing different farmers' market organizational structures and managerial roles, and farmers' market manager surveys are widely available, there are no surveys or studies exploring managers' motivations.
Objective 2: The second objective of this study was to administer the surveys to all farmers' market managers identified through the North Carolina CTGP.
Objectives 3 & 4: The final objectives of this study were to use the results from the NC manager survey to examine 1) associations between managers' motivations related to being facilitators of business opportunities for small and moderate size farms and indicators of these goals reported by the managers (i.e., market vendor recruitment, sales, and customer counts) (Objective 3) and 2) associations between motivations related to being promoters of healthful foods and SNAP/EBT availability and participation(Objective 4).
Research
The purpose of this study was to explore how farmers’ market managers’ motivations impact business opportunities for small and moderate-size farmers in their community and access to healthful foods for low-income consumers. This study was conducted from September 2013 to August 2014. All components of this study were approved by the East Tennessee State University Institutional Review Board.
Survey development:
The first step in this study was to develop a survey assessing farmers’ market managers’ motivation related to facilitating business opportunities for small and moderate-size farmers and expanding access to healthful foods for low-income consumers (Objective 1). To gather formative data, focus groups with 5 farmers’ market managers from southwest Virginia and northeast Tennessee were held immediately before and after the Appalachian Farmers Market Association meeting in Bristol, Tennessee in January 2013. To meet our recruitment goal, 3 additional farmers’ market managers participated in focus groups in Asheville, North Carolina in February 2013. Focus groups lasted for 30-60 minutes. To gain information on the farmer’s perspective, telephone interviews with 8 farmers participating in farmers’ markets in the same regions were conducted in February and March 2013. Interviews lasted 11-90 minutes. Participants received $25 for helping with the study.
The focus groups and interviews were recorded and transcribed verbatim and imported into a qualitative analysis software (QSR N’Vivo, version 10). A codebook was developed by the investigator and reviewed by a graduate student who assisted with thematic analysis. Relevant codes were identified in the text and discussed by the investigator and graduate assistant until agreement was met.
Items for a quantitative survey on farmers’ market managers’ motivations were developed using the themes resulting from the qualitative analysis. An exhaustive list of possible survey items (218 items) was developed and reviewed by a panel of experts, including 2 members of the investigator’s dissertation committee, and 3 colleagues with research backgrounds in food environment and farmers’ market research. A draft survey was developed using motivation items selected by the investigator and panel of reviewers. These items were combined with USDA Farmers’ Market Manager Survey questions assessing the outcome variables of interest (i.e., customer count, SNAP/EBT participation, vendor participation, etc). This draft was sent to 5 farmers’ market managers who participated in the focus groups for piloting. Their feedback was used to refine the survey, and a final version was programmed into SurveyMonkey (Palo Alto, CA) for distribution to the NC farmers’ market managers.
Survey of NC farmers' market managers:
As a consultant on the North Carolina CTG farmers’ market development and enhancement evaluation grant, the investigator had access to the North Carolina CTG Fruit and Vegetable Outlet Inventory (FVOI), which includes a comprehensive listing of all 692 farmers’ markets, produce stands, and Community Supported Agriculture (CSA) sites in the state, their location, and their manager’s contact information. In May 2015, all farmers’ markets managers in the state (n=309) were contacted by e-mail (or telephone if e-mail addresses were not provided) and invited to participate in the online survey. They were sent 2 reminder e-mails and given 2 weeks to complete the survey. To improve the response rate, a second wave of surveys was conducted in July and August 2014. This involved hard copy mailings of the survey in addition to the SurveyMonkey link to all managers who had not completed the survey. All participants received $10 for their assistance with this portion of the study.
Data analysis:
Survey results were analyzed by the investigator in summer 2014. Analyses were reviewed by the biostatistician on the dissertation committee. A food access motivation score was created by averaging the participants’ responses to the following items:
- It’s important that low income people feel welcome at my market (1 – strongly disagree to 4- strongly agree)
- It’s important that everyone can shop at my market (1- strongly disagree to 4-strongly agree)
- Low income people feel welcome at my market (1-strongly disagree to 4-strongly agree)
- Ranking variable My role making food affordable (1-least important to 6-most important)
- Ranking variable My role making food accessible (1-least important to 6-most important)
A business motivation score was created by averaging the participants’ responses to the following items:
- It’s important to support local agriculture (1-strongly disagree to 4-strongly agree)
- It’s important to support local business (1-strongly disagree to 4-strongly agree)
- Ranking variable My role supporting local agriculture (1-least important to 6-most important)
- Ranking variable My role supporting local artists (1-least important to 6-most important)
- Ranking variable My role supporting the local economy (1-least important to 6-most important)
Limitations addressed in analysis:
To address limited variance in participants’ responses to the Likert scale items, binary food access motivation and business motivation scores were created by assigning participants a “high” or “low” motivation score. Specifically, participants were assigned a “high” food access motivation score if they ranked “My role making food affordable” or “My role making food accessible” as at least 5 or 6 (most important or 2nd most important), and a “high” business motivation score if they ranked “My role supporting local agriculture”, “My role supporting local artists”, or “My role supporting the local economy” as at least 5 or 6 (Table 4).
Several outcome variables of interest were not examined in this study due to limited reliability and insufficient responses. Specifically, customer count, SNAP/EBT participation count, and SNAP/EBT value were found to be unreliable and therefore not used as outcome variables in the inferential analyses.
Statistical analysis:
Descriptive statistics were used to summarize participant characteristics. Potential covariates were selected based on characteristics of farmers’ market managers and markets that were hypothesized to influence market outcomes based on the literature. Manager characteristics included: whether or not the manager was paid (“Are you paid to manage the market? Yes or No”), the manager’s age (“What is your age in years?”), and the mangers’ years of experience (“Including 2014, how many years have you managed this market?”). Market characteristics included: the number of years in operation (“Including 2014, how many years has the market been in operation?”), and the number of volunteers (“Including you, how many volunteers work at this market?”).
Binary logistic regression was used to examine the association between the likelihood that participants have SNAP/EBT (yes/no) at their farmers’ markets (dependent variable) and their community food access motivation score (independent variable) in crude (model 1) models. Backwards selection of covariates was used to find the most parsimonious models. Models were adjusted for manager characteristics (age, pay status, and years managing the market, Model 2) and further adjusted for market characteristics (volunteers and years in operation, Model 3). Separate, crude (Model 1) multiple linear regression models were used to examine associations between continuous business outcomes (total vendor count, average weekly vendor count, and local vendor count, all dependent variables) and the business motivation score (independent variable). These models were also adjusted for manager characteristics (age, pay status, and years managing the market, Model 2) and further adjusted for market characteristics (volunteers and years in operation, Model 3). Final, adjusted models retained only the covariates that were significantly associated with the dependent variable (p<.05).
Eight managers participated in the farmers’ market managers’ focus groups; of these, 5 were from rural farmers’ markets in Southwest Virginia and East Tennessee, and 3 were from urban farmers’ markets in Asheville, North Carolina. Eight farmers participated in phone interviews; 4 participants were from Western North Carolina, and 4 were from Southwest Virginia. Eight themes emerged from the focus group, and 5 themes emerged from the interviews. A summary of themes and sub-themes from the focus groups and interviews can be found in Tables 1 and 2.
In the first wave of data collection, only 66 managers completed the survey; only 60 of these responded beyond the first 2 questions. The second wave of data collection resulted in a final sample of 80 participants. Upon removing duplicate markets and erroneous entries, the final study sample was 70 market managers. The average participant age was 48 years (range: 22-88 years) (The majority of managers were paid to operate their markets (56%) and had an average of 5 years (range: 1-20 years) of experience managing the market. The average market had operated for 11 years (range: 1-41 years), and had an average of 8 volunteers (range: 0-300 volunteers). Markets had an average of 31 vendors in 2013 (range: 3-150 vendors), 19 (range: 1-65) vendors per week, and 17 local vendors per season (range: 0-125 vendors). An average of 353 (range: 0-3000 customers) customers visited the markets each week. Thirteen participants (19%) reported having the SNAP/EBT program at their markets (Table 3).
The results of the regression analyses can be found in Table 5. The association between food access motivation score and SNAP/EBT availability was not significant in the crude or adjusted models, suggesting there was no association between community food access motivation score and SNAP/EBT availability (Objective 3). The association between business motivation score and total vendor count and average vendor count was not significant in the crude or adjusted models. The association between business motivation score and local vendor count, however, was significant in the unadjusted model and the model adjusted for manager and market characteristics (include ORs, etc) (Objective 4). In the final, adjusted model (adjusted for manager pay status) a "high" business score among managers was associated with an increase of 13 local vendors per market (B: 12.55, SE: 4.05, p<.05).
Market years of operation and manager pay status were the only covariates retained in final models. Specifically, the covariate market years of operation was significantly associated (p<.05) with SNAP/EBT availability and the average number of vendors per week. Manager pay status was significantly associated (p<.05) wtih total vendor count, average vendor count, and local vendor count.
Educational & Outreach Activities
Participation Summary:
Findings from this study were presented at the North Carolina Department of Public Health’s NCCTGP Evaluation Collaborative Meeting on September 23, 2014.
A manuscript of this study is currently being drafted by the investigator and co-authors from the NCCTGP and ETSU dissertation committee. The manuscript will be submitted to the Journal of Nutrition Education and Behavior in October 2014.
A written summary of findings will be shared with the director of the Appalachian Farmers Market Association group and the Appalachian Sustainable Agriculture Project in fall 2014.
Project Outcomes
This study provides new information on how farmers’ market managers may influence access to healthful foods for low-income households and business opportunities for small and moderate size farmers. Currently, there is only one published study of farmers’ market managers’ influence on SNAP/EBT participation, which describes the positive economic impact of accepting SNAP/EBT at farmers’ markets, and the need for increased training of market managers around food security among market managers in Michigan. The findings presented here provide starting point for deeper investigation of behavioral factors that determine how managers effect market outcomes, and could be used to inform interventions aiming to maximize managements’ impact local agriculture economies and making healthful foods more affordable in communities.
This study recruited a small yet diverse sample of North Carolina’s farmers’ market managers. Only 27% of known farmers’ market managers in NC participated. The relationships between business motivation and local vendor participation was significant. This finding indicates that managers who are interested in and motivated to influence the economic benefits of their markets can impact and improve certain economic outcomes at the market. It may also be that mangers with higher business motivation scores are also more likely to know and accurately report market vitality data like local vendor participation.
It is important to consider how to improve managers’ business motivation scores in order to identify strategies for improving farmers’ market vitality indicators as well as approaches for addressing the goal of connecting markets to the SNAP/EBT program and its direct financial benefit to farmers. Data from the qualitative portion of this study could provide insight on how to address managers’ business motivation. Farmers participating in the study’s interviews gave practical examples of how they felt managers impacted their sales. For example, one North Carolina-based farmer described the importance of interactive promotions like cooking demonstrations and sampling in attracting customers to buy products. A farmer in Southwest Virginia discussed the impact of the market layout on foot traffic. Providing managers with case studies of real-life examples of how their efforts impact sales and practical tips on how to recruit customers could encourage them to bolster their efforts.
Managers’ motivations to improve access to healthful foods in their communities was not associated with SNAP/EBT placement at their markets. This suggests that the relationship between being motivated by community food access issues and providing SNAP/EBT is not as straightforward as was hypothesized. There are a number of necessary steps between recognizing and being motivated to mitigate food access barriers in the community and actually implementing SNAP/EBT. Market finances, vendor pay, manpower, and the community context are just several of many factors that contribute to the introduction of SNAP/EBT to their markets. An example of this is found in the focus groups with market managers. One urban manger, in particular, was very passionate about local food access and business; unfortunately, she could do very little about improving access to low-income households because her market did not use SNAP/EBT for financial reasons.
Managers who are not highly motivated by food access, but are motivated by business, could be motivated to offer federal nutrition benefit programs, like SNAP/EBT, if they are more knowledge about the economic benefits of program participation. Most farmers participating in the interviews reported good sales outcomes from participating in the SNAP/EBT program, with many stating that participation expanded their customer base. By highlighting the economic potential of participating in SNAP/EBT, managers who are more motivated by business than food access and public health issues could be encouraged to offer SNAP/EBT at their markets.
A key limitation of this study was the small sample size. Only 27% of potential farmers’ market managers participated in the study. This may be due to incomplete or outdated contact information in the FVOI and the university’s requirement of participants to provide their social security number to receive payment. Another limitation of the study was the poor reliability of certain indicators of SNAP/EBT participation and business vitality. Specifically, vendor and SNAP/EBT sales, and SNAP/EBT customer counts would have been important outcomes to examine for this study, but they tend to be unreliable as most market managers do not document them.
Economic Analysis
N/A
Areas needing additional study
Future work should attempt to introduce a standard method for collecting these important indicators of market reach and impact, as these metrics could be useful for longitudinal evaluations of farmers’ market based-strategies to improve community nutrition and local economic outcomes.
Future studies should further explore the barriers to offering SNAP/EBT at markets among a larger sample of market managers. A better understanding of the number of managers who desire to offer SNAP/EBT at their markets and the reasons they do not currently offer SNAP/EBT would highlight key areas for intervention for initiatives aiming to increase SNAP/EBT availability. Improving awareness of the SNAP/EBT program and existing supports for managers could increase the number of managers with high community food access motivation who offer SNAP/EBT.
More work should be done to understand the interplay between market manager characteristics and the different environmental and organizational factors that influence market outcomes. As this study suggests, public health interventions should consider the market’s broader mission of creating business opportunities and leverage it to meet their goals. Addressing managers’ motivations will be critical to improving the food environment through farmers’ markets.