Introduction

A dependence on large-scale farms and the industrial food chain is the norm in the United States with few Americans relying upon small-scale, alternative forms of agriculture to stock their kitchens (Pollan 2007). Approximately 12,500 farms (0.6 percent) in the US engage in community supported agriculture (CSA) (USDA 2007a). Emerging in the mid-1980s, this model of farming consists of a cooperative agreement between farmers and members with the latter paying a seasonal fee to the former in exchange for fresh produce, and other farm products, on a weekly basis (Lang 2010). Unlike the dominant food system, CSA is predicated on local food production and consumption with an emphasis on organic and environmentally friendly practices, while sharing risks between producers and consumers (DeLind 1999; Dyck 1997; O’Hara and Stagl 2001; Tegtmeier and Duffy 2005).Footnote 1

Originally CSAs were designed to build community proximate to the farm with members collecting their goods on site, fostering community between the farmer and members. Early proponents of this model often were characterized as fringe advocates espousing communitarian farming ideals. More recently, CSAs have expanded in scope extending their reach to suburban and urban areas illustrating a metamorphosis in cooperative farming endeavors from small towns and cities to include ventures with a regional emphasis. Changes such as these demand rethinking commonly held notions of community in CSA.

Selecting for CSA farms in New York, this research poses two questions. First, what motivates members to join a CSA? Is it to meet like-minded individuals and/or share financial risks with farmers—activities constituting community or elements of community—or do members primarily desire ultra-fresh, seasonal, organic produce for reasons related to health and taste, with little connection to community? Second, are the purported benefits of community, which appear throughout much of the CSA literature, still integral to the CSA model? While some scholars (Jacques and Collins 2003; Schnell 2007) laud the role of community in CSA, scholarship in political science, sociology and psychology suggests community appears to be weak. For example, in Putnam’s (2000) much heralded study of civic engagement and community, he contends Americans are experiencing weaker ties to community, and people are “bowling alone.” Bringing together multiple disciplinary perspectives, we seek to understand members’ perceptions of community both within their CSA and beyond, reevaluating the CSA model.

Notions of community in CSA and beyond

Community in CSA

The literature highlighting the role of community in CSA appear on a continuum ranging from CSA models which involve the full support of the community (DeLind 1999; Feagan and Henderson 2008; Jacques and Collins 2003; Lass et al. 2003; Schnell 2007) to market-oriented models in which community plays a limited role (Groh and McFadden 1997; Lang 2010; O’Hara and Stagl 2002; Ostrom 1997). While the CSA model has changed over time, proponents of community (Jacques and Collins 2003; Schnell 2007) agree that CSA provides “participants with social and communal relationships with one another and the land” (Feagan and Henderson 2009, p. 205). The ideal CSA model supposedly fosters a relationship of trust between local farmers and members (Feagan and Henderson 2009) however, whether this relationship actually exists is unclear. DeLind (1999) in fact laments the dearth of community among CSA members. Given the pervasiveness of CSAs that now deliver to areas beyond a farmer’s immediate community, this study assesses which model best depicts CSA in New York.

Less optimistic, community according to some scholars (Groh and McFadden 1997; Lang 2010; O’Hara and Stagl 2002; Ostrom 1997) is anemic. One study of CSA members shows respondents do not feel that their CSA opened their eyes to the importance of community, nor are they integrated into their CSA (Lang 2010). Instead members simply desired fresh, organic, local produce (Conner 2003; Lang 2010; Oberholtzer 2004; Ostrom 2007). Earlier studies of CSA (Groh and McFadden 1997; Ostrom 1997) similarly suggest ‘‘developing community’’ was ranked weakly among CSA members. Several farmers reported not having the interest or time to engage in community building (O’Hara and Stagl 2002). This paper investigates whether and to what degree CSA members in New York view participation in their CSA as building community within their CSA.

Conceptualizing CSA more narrowly, an economic model more accurately depicts the essence of this alternative agricultural arrangement for some scholars (Lizio and Lass 2005; Oberholtzer 2004). Illustrating this, Lizio and Lass (2005) assert that even though farmers do not engage in the profit-maximizing behavior characteristic of their industrial agricultural counterparts, CSA is in fact an economically viable model. Similarly, Oberholtzer’s (2004) study shows the vast majority of farmers reported the “economic aspect” was the primary reason for initiating their projects, and only three out of 13 farmers were driven by the “social aspects” of CSA. Again, members indicated joining their CSA “for fresh, organic, and/or local produce and to support a local farmer or farm” (Oberholtzer 2004, p. 2). Our study evaluates whether or not these comments and ones similar to them, are reiterated by members who belong to CSA farms across New York, building on Oberholtzer’s (2004) work by conducting a larger study.

Perhaps best illustrating the mélange of models is Feagan and Henderson’s (2008) work, which organizes CSA on a continuum. At one of the end of the spectrum, the collaborative model incorporates elements of community through partnerships between the farmer and members and at the other end is the instrumental model consisting of an economic arrangement between the aforementioned with no elements of community. In the center, the functional model promotes greater harmony between farmers and the community than the instrumental model, however members only are partially committed to sharing risks with farmers. The authors assert in the ideal CSA, members “create relationships of trust with their CSA farmers” and it should provide alternatives to the market by sharing risk with the farm (Feagan and Henderson 2009, p. 205). The collaborative model however, is not realistic for all CSA farmers and members. Using Feagan and Henderson’s (2009) continuum, we attempt to characterize CSA in New York based on a survey of CSA members.

Local food movement

Local food and the local food movement have attracted considerable attention among scholars (DeLind 1999, 2010; Feagan and Henderson 2008; Feenstra 2002; Hamilton 2002; Lyson 2004; Ostrom 2006; Perez et al. 2003), highlighting the local nature of CSA and emphasizing community and sustainability. In a study conducted by Perez et al. (2003) of CSAs across five counties in California underlines the importance of “supporting local.” Focus group participants indicated a host of reasons for participating in their CSA including facilitating connections among local farms and farmers, other people, the land, or farming itself (Perez et al. 2003). DeLind (1999, 2010) similarly draws attention to the local nature of CSA and its place within the community, recounting her experience managing a CSA. She maintains reducing the “distance between people and their food supply” is paramount to achieving success (DeLind 1999, p. 3). The author contends locavores and the attendant local food movement focus on consumers and food rather narrowly, failing to contextualize individuals in their larger communities. With CSAs increasingly delivering produce to members not only locally but also regionally, we seek to understand how CSA members perceive community and whether CSA imbues community among its members.

Events, activities, and volunteering

Many CSAs attempt to incorporate their members into the community by holding events, planning activities and requesting that members volunteer. Despite efforts by farmers to involve members, lackluster participation is noted throughout the literature (DeLind 2003; Feagan and Henderson 2009; Lang 2010; McIlvaine-Newsad et al. 2004; Worden 2002). Feagan and Henderson (2009) adroitly describe this phenomenon. A farmer they interviewed announced a potato harvest party in the CSA newsletter and only the farmer’s sister and niece attended. Even when farmers offered sweat equity—volunteers work on the farm in exchange for a reduced subscription rate—they ran into difficulties with members finding the arrangement inconvenient (Worden 2002). As organizations become more professional, permanent staff replace volunteers. In the end, volunteers expressed higher rates of satisfaction with their CSA than members who did not volunteer (Loughridge 2002). Given the important role these activities play in fostering community, we assess the role of these elements for CSA members.

Community beyond CSA

Aside from Ostrom’s (2007) work few accounts within the CSA literature contextualize community more broadly. In contrast, political scientists (Breem 1999; Cohen 1985; Etzioni 1995) explain what constitutes community, comment on the current levels of community within society, and depict what provides individuals with a sense of community. Offering an especially useful framework for studying CSA Gusfield (1975), a sociologist, suggests community can be defined two ways, one with a focus on geography (neighborhood, town, village, etc.) and the other on social relationships regardless of location. Understanding members’ perceptions about community beyond their CSA experience is a necessary precursor for understanding community within CSA.

Several studies (Andersen et al. 2006; Putnam 1995, 1996, 2000) examine the level of community present in society at the macro-level. Participation in voluntary associations, according to Putnam (2000), has declined over the last 30 years accompanied by an aggregate loss of membership to civic organizations with individual membership not migrating to other organizations. Several researchers have challenged Putnam’s findings contending association membership remained constant, and still others reported an increase following a period of decline (Baumgartner and Walker 1988; Paxton 1999; Rotolo 1999). Given members are a critical component of CSA and many CSAs depend upon volunteers these studies provide insight into associations beyond CSA (Groh and McFadden 1997; Lang 2010; O’Hara and Stagl 2002; Ostrom 1997).

Not unlike studies of CSA membership which gauge respondents’ perceptions of community in CSA, scholarship examining community at the micro-level, conducted by psychologists, assesses individual perceptions of community. A psychological sense of community, often called a “sense of community” (SOC),Footnote 2 is defined as “a feeling that members have of belonging, a feeling that members matter to one another and to the group, and a shared faith that members’ needs will be met through their commitment to be together” (McMillan and Chavis 1986, p. 9). Four components (membership; influence; integration and fulfillment of needs; and shared emotional connection) shape one’s SOC according to McMillan and Chavis (1986). Subsequent research favors the inclusion of an additional variable related to identifying self with place (Tartaglia 2006). We explore whether elements such as membership and shared emotional connection provide CSA members with a sense of community. Assessing SOC beyond the CSA aids in understanding community within CSA.

Why study CSA members?

Studies of community in CSA tend to rely upon the farmer as the unit of analysis (Lass et al. 2003; Lizio and Lass 2005; Lyson 2004; Ostrom 2006) rather than focusing on the perceptions of CSA members (Conner 2003; Lang 2010; Perez et al. 2003). Yet surveys of CSA members provide a richer portrait of community. Investigating members’ motivations for joining a CSA and their perceptions of community within and beyond their CSA contributes to the aforementioned scholarship. Additionally, studies of CSA members are somewhat smaller in scope, typically encompassing one or a handful of farms with responses ranging from 240 to 276 members (Lang 2010; Oberholtzer 2004). These smaller studies play an important role in exploratory and descriptive research even though generalizations are not possible. Our research enlarges the scope of study, analyzing survey data culled from 565 CSA members. It also provides an opportunity to understand perceptions of CSA members across an entire state, which has not been undertaken. Finally, New York contains one of the highest concentrations of CSA farms, making it a particularly robust case study across a socio-economically diverse state. To date, there appears to be only one CSA study in New York (Conner 2003). We seek to fill this gap in the literature focusing on members who belong to CSA farms in New York.

Data and methods

Descriptive and exploratory in nature, this study investigates members’ motivations for joining their CSA, and it seeks to evaluate whether the purported benefits of community—which appear throughout much of the CSA literature—are integral components of CSA. We examine CSA membership, the unit of analysis, using an online survey distributed to members between November and December 2010.

To assess perceptions of CSA members, we use a case study design selecting for CSA farms in New York. According to Yin (2008), case studies are appropriate when embarking upon exploratory research. They are designed to focus on the unit of analysis rather than an entire system (Patton 2002). While agriculture in New York is small, relative to larger farming states, it is nevertheless vital to the state’s economy.Footnote 3 Estimates regarding the number of CSAs in the US vary considerably with the United States Department of Agriculture (USDA) ranking New York thirteenth with a total of 364 CSA farms and Local Harvest ranking New York first with 261 CSA farms.Footnote 4

To draw the population of CSA farms in New York we relied upon Local Harvest’s database. The USDA only supplies aggregate data on CSA farms, and supplements this with six links to databases designed to help visitors locate a farm. Of the six links, Local Harvest contains the most comprehensive list of CSA farms in the US. In September 2010, we generated a list of all CSA farms in New York. This was augmented by a keyword search using Google and the words “New York state CSA.” A total of 266 CSA farms were located, which appear to be well dispersed geographically throughout the state with the exception of the Adirondack Mountain region. Using a non-random sample, we engaged in purposive sampling—sending all CSA farms an invitation to participate in our study—due to constraints associated with recruitment of CSA members (described in further detail below).

Before distributing the survey, we telephoned CSA farmers/managers informing them about our study and sent an introductory e-mail asking CSA farmers/managers if they would extend an invitation to their members to participate in a voluntary online survey. In November 2010, a second e-mail was sent containing an explanation of the study and a link to the survey. A reminder to complete the survey was sent to members (via CSA farmers/managers) 2 weeks later. Several CSA farmers/managers indicated that they were unwilling or unable to forward our survey to their members. Reasons ranged from farmers/managers indicating they recently distributed their own survey to their members, to others deeming survey questions requesting salary ranges and political views too sensitive, even though respondents could opt out of answering these questions.

Consisting of one open-ended and 39 close-ended questions, the survey was divided into several sections including members’ views about their CSA, members’ views about their community beyond the CSA, and member demographics. Table 1 provides descriptive statistics for variables included in the analysis. To participate in the survey two criteria were established. First, respondents needed to be 18 years of age to participate and second, the CSA farm to which the respondent belonged needed to be located in the state of New York. Some CSA farms are located in New York, but their members are residents of neighboring states and/or the distribution occurs in a neighboring state, namely Connecticut. A total of 565 members responded to the survey.

Table 1 Summary statistics
Full size table

Data analyses consisted of Analysis of Variance (ANOVA) and descriptive statistics.Footnote 5 ANOVA is used to compare CSA members across four income groups ($0–35,000; $35,0001–$75,000; $75,001–$125,000; and $125,001 and up) to determine whether motivations for joining a CSA varied by income. Also, ANOVA is used to compare the length of time a respondent has belonged to a CSA (divided into five groups based on the year the member joined: 2010, 2009, 2008, 2007–2006, 2005 or earlier) and questions related to community including the degree to which respondents’ eyes were opened to the importance of being part of a community; the degree to which CSA members feel integrated into their CSA community; how interested CSA members are in local political and social issues; and how involved CSA members are in local political and social issues.

Limitations

This study breaks new ground because it investigates the views of CSA members from CSA farms across an entire state, and it is one of the largest studies of members to date. It contains several limitations, which are worth noting. First, because this study focuses on CSA in New York our ability to generalize in the US is limited not only by geography, but by the types of products offered and the length of seasons. Still, New York ranks thirteenth out of 50 states in terms of products marketed through CSA, making it an appropriate state for study (United States Department of Agriculture 2007b). More importantly, many of the findings from this study, including the demographics of CSA members and reasons for joining a CSA, mirror results from other CSA studies.

Second, though CSA farms are listed in the Local Harvest database, individual members are not, which limits how we collected data on members.Footnote 6 The total number of members who belong to each CSA is not available. Determining the population of CSA farms is also problematic because a comprehensive list of CSAs containing contact information is not readily available. We relied upon Local Harvest to locate CSA farms however, it is possible that some farms are not registered with Local Harvest. Similarly, farms without an online presence might be missed in a keyword search. A non-random purposive sample such as this does not allow us to generalize, though the sample size is large enough to draw some interesting conclusions.

Third, survey research contains some limitations. The length and complexity of the survey might dissuade some CSA members from responding. Close-ended questions likewise do not permit respondents to elaborate. Online surveys contain biases. Individuals without an e-mail are excluded from participating, and those who do participate in online surveys tend to be younger and highly educated. Tempering this limitation, studies of CSA members (Durrenberger 2002; Kane and Lohr 1997; Kolodinsky and Pelch 1997; Lang 2010) show respondents tend to be well educated too. Minorities with low Internet penetration rates and individuals with low levels of digital fluency might be disinclined to participate. Technical glitches such as crashes, error messages and double entry are possible. For this study, only members with e-mail and whose CSA farmers/managers forwarded the survey received an invitation to participate.Footnote 7 Since respondents are self-selecting the sample may not be representative of CSA members in New York causing unintended biases. Many of these limitations also are found with mail-in surveys (Wright 2005).

Finally, the absence of a return rate is an additional limitation. Local Harvest publishes the number of shares for each CSA farm however, this cannot be equated to the number of members, even though it might act as an approximation it is not a reliable measure. We asked farmers/mangers how many members they have with many of them offering an estimate rather than a definitive number. Still other farmers/managers did not respond to our request. As a result, a response rate cannot be calculated.

Operationalization

In this study we measure community along two dimensions. The first dimension measures notions of community traditionally associated with CSA, namely the collaborative model, which suggests farmers and members are seen as partners (Feagan and Henderson 2008). Community development theorists Wilkinson (1991) and Liepins (2000) view community as “both a physical and a social or interactive space” and increasingly communities are “dynamic social networks formed on the basis of shared interests, values, and identities” (Ostrom 2006, p. 67). This paper incorporates a range of survey questions that tap into these concepts. Respondents were asked to rate on a scale from one to five—one being no influence and five being a major influence—the degree to which the following influenced their decision to join their CSA: seasonal fruits/vegetables, freshly picked fruits/vegetables, organic fruits/vegetables, price, convenience, to eat locally produced food, to build a stronger sense of community, to share financial risks with farmers, to volunteer at the farm, to meet like-minded folks, and/or to participate in farm events. Using this same list, members were then asked to rank-order the top three items that influenced their decision to join a CSA. Respondents were asked whether or not they were required to volunteer at their CSA (1 = yes, 0 = no), and if in fact they had volunteered (1 = yes, 0 = no). Members also were asked the degree to which they agreed or disagreed with the following statements, “Joining my CSA has opened my eyes to the importance of being part of a community,” and “Since joining my CSA, I feel that I have become integrated into my CSA community” (1 = strongly agree, 2 = somewhat agree, 3 = neutral, 4 = somewhat disagree, 5 = strongly disagree, 6 = don’t know).

The second dimension measures members’ sense of community and participation in activities separate from their CSA experience by tapping into the broader literature on community. These variables further contextualize members’ perceptions about CSA. The survey asked respondents how many years they resided at their current residence. Respondents also were asked to rate on a scale from one to five—one being no sense of a community and a five being a strong sense of community—to what degree the following gave them a sense of community: work or school, their town or city of residence, associations related to residence, social organizations, and political groups. Similarly, members were asked in which groups or organizations they were involved during the last 12 months selecting from the following list, work or school related groups, neighborhood organizations, labor unions, places of worship, social organizations, political groups, and no groups or organizations. Additionally, the survey asked members whether or not they volunteered in any capacity other than their CSA in the last 12 months (1 = yes, 0 = no). Finally, the survey gauges interest in local political or social issues (1 = very interested, 2 = somewhat interested, 3 = neutral, 4 = somewhat uninterested, 5 = very uninterested, 6 = don’t know, 7 = not applicable), and involvement in local political or social issues (1 = very involved, 2 = somewhat involved, 3 = neutral, 4 = somewhat uninvolved, 5 = very uninvolved, 6 = don’t know, 7 = not applicable).

Results

Demographics of CSA members

A total of 565 CSA members responded to the survey, and of this 84 percent (477 out of 565) of respondents are women and 16 percent (88 out of 565) are men. On average, respondents are 42 years of age, ranging in age from 20 to 78. More than 80 percent of respondents (467 out of 565) identified themselves as white. Respondents who completed the survey are well educated. Almost half of CSA members (47 percent or 267 out of 565) earned a graduate degree and 11 percent of CSA members (62 out of 565) attended some graduate school, while more than a quarter of respondents (28 percent or 156 out of 565) reported earning a college degree.

A majority of the CSA members who responded to the survey appear to be well off and living in their communities for a substantial period of time. Forty-six percent of respondents (261 out of 565) indicated their annual household income is $75,000 or more, which is also the median income category. In contrast, 12 percent of CSA members (66 out of 565) reported annual household incomes less than $35,000 with the same percentage (61 out of 565) declining to provide their household income. A majority of respondents (59 percent or 334 out of 565) indicated that two individuals contributed to their annual household income, and almost one-third of respondents (31 percent or 175 out of 565) reported their household contains only one wage earner. On average, members resided at their current residence 8 years, but ranged from less than 1 year to more than 50 years.

Reasons for joining a CSA

We surveyed members’ motivations for joining a CSA asking respondents to rate the factors that motivated them to subscribe to a CSA by selecting from a list of 16 factors. On a five-point scale, approximately 80 percent of respondents rated eating freshly picked fruits and vegetables and eating locally produced food (455 out of 565 and 441 out of 565, respectively) a five—a major influence—in their decision to join a CSA. Seasonal fruits and vegetables (398 out of 565) and organic (389 out of 565) also were ranked a five by nearly 70 percent of respondents.

In contrast, few respondents indicated joining their CSA for reasons related to community. The data show 3 percent of CSA members (176 out of 565) ranked building a stronger community a five, and still fewer, 8 percent (45 out of 564), ranked meeting like-minded people similarly. Only one-fifth of the members (118 out of 564) ranked sharing financial risks with farmers a five, indicating this did not influence their decision to join a CSA.

Factors motivating respondents to join a CSA likely differed according to income. A one-way ANOVA was conducted to test factors motivating respondents to join a CSA across four income categories. The results, presented in Table 2, show the mean ratings motivating respondents to join a CSA for reasons related to sharing financial risk (F(3, 463) = 2.68, p = .046), volunteering (F(3, 463) = 6.94, p = .000), meeting like-minded people (F(3, 463) = 7.39, p = .000), and participating in farm events/activities (F(3, 463) = 5.93, p = .001) generally decline in influence from the lowest to highest income category. Tukey post hoc comparisons of the four income categories indicate that respondents with household incomes between $0 and $35,000 gave significantly higher ratings to sharing financial risk,Footnote 8 volunteering, meeting like-minded people, and participating in farm events/activities than respondents with household incomes between $75,001–$125,000 and $125,001 and up. Respondents with household incomes between $35,001 and $75,000 gave significantly higher ratings than respondents earning $125,001 or more to meeting like–minded people and participating in farm events/activities. Finally, there were also significant differences on meeting like-minded people between respondents earning $75,001–$125,000 and $125,001 and up.

Table 2 Factors motivating CSA members to join their CSA by income categories using a one-way ANOVA test
Full size table

Respondents were asked to rank-order the top three factors that influenced their decision to join a CSA. Collapsing the rankings, eating locally produced food (69 percent or 392 out of 565), followed by organic (53 percent or 303 out of 565), and finally freshly picked fruits and vegetables (48 percent or 272 out of 565) were ranked the top three factors. The least influential factors, with five percent or less ranking these items in the top-three, were price, convenience, volunteering at the farm, meeting like-minded people, and participating in farm events/activities.

Role of community

Because community is considered a central component of CSA we asked respondents about community in relation to their CSA experience. For example, we asked CSA members to what degree they agreed or disagreed with the following statement, “Joining my CSA has opened my eyes to the importance of being a part of a community.” Slightly more than one-third of respondents (38 percent or 213 out of 565) reported that they strongly or somewhat agree with the statement. A similar percentage of respondents (37 percent or 208 out of 565) are neutral. When asked whether or not they agreed or disagreed with the following statement, “Since joining my CSA, I feel that I have become integrated into my CSA community,” a slightly smaller percentage agreed. Thirty-four percent of respondents (189 out of 537) somewhat or strongly agree with the statement and a similar percentage of respondents (197 out of 537) remain neutral.

The degree to which the CSA opened respondents’ eyes to the importance of being part of a community and the extent to which respondents feel integrated into their CSA community likely varied depending on the length of time a respondent belonged to their CSA. A one-way ANOVA was conducted to test the importance of community and the extent of integration among five groups (based on the year a member joined their CSA) illustrated in Table 3. The results show the mean ratings across the five groups were not significant when compared to CSAs opening one’s eyes to the importance of community. In contrast, the mean ratings for feeling integrated into the CSA community (F(4, 532) = 4.43, p = .002) declines the longer a respondent belongs to a CSA, suggesting the longer one is a member of a CSA the more integrated one feels into their CSA. A Tukey post hoc comparison of CSA respondents shows a significant difference between respondents who joined their CSA between 2007–2006 and those who joined in 2010 with the former feeling more integrated into their community than the latter.

Table 3 Feelings of community and interest/involvement in local political or social issues by year respondent reported joined their CSA, using a one-way ANOVA test
Full size table

Respondents also were asked a variety of questions related to community distinct from their CSA experiences including the degree to which they are interested in local political or social issues, and how involved they are in the aforementioned. Half of the respondents (285 out of 565) said they are somewhat interested in local political or social issues and 28 percent of members (158 out of 565) articulated being very interested in local political or social interests. Members are not only interested in local political or social issues, but also they appear to be actively involved. Of the 443 members who are somewhat or very interested, 235 (53 percent) are somewhat or very involved in local political or social issues.

To better gauge involvement in activities typically associated with community respondents were asked to indicate whether or not they were involved in various groups or organizations during the last 12 months, illustrated in Table 4. Just over half of respondents are involved in social organizations. Participation across other activities is less robust with the exception of work or school related organizations, but still fewer than 45 percent of respondents reported involvement. Of note, 15 percent of respondents (78 out of 565) reported not being involved in any groups or organizations.

Table 4 Involvement in group/organization activities last 12 months (n = 565)
Full size table

Interest and involvement in local political and social issues likely varied depending on the length of time a respondent has belonged to a CSA. A one-way ANOVA was conducted to test interest and involvement in local political and social issues among five groups. The results, illustrated in Table 3, show the mean ratings across the five groups, when compared to interest in local political and social issues (F(4, 532) = 2.60, p = .035), decreases the longer a respondent belongs to a CSA. With respect to involvement in local political and social issues (F(4, 532) = 4.16, p = .002) the mean ratings also decreases the longer a respondent belongs to a CSA. Overall, interest and involvement in local political and social issues is highest among respondents who joined their CSA in 2005 or earlier. Respondents who joined their CSA in 2005 or earlier also are significantly more likely to be involved in local political or social issues than compared to members who joined in 2010 based on results from a Tukey post hoc test.

The survey also asked respondents what activities provide them with a sense of community or belonging, asking them to rate six items on a scale from one to five, in which one is no sense of community and five is a strong sense of community. Table 5 illustrates responses from CSA members who rated items a five, indicating a strong sense of community. In general, feelings about community or belonging are weak with few respondents maintaining they feel a strong sense of community. For example, one-third of respondents indicated the people with whom they work or attend school provides them with the strongest sense of community, while less than a quarter of the members surveyed reported feeling a strong sense of community across any of the other categories. As expected, among CSA members who reported being involved in a group or organization during the last year, they reported feeling a stronger sense of community—rating feeling a sense of community a four or five, on a scale from one to five in which a one is no sense of community and a five is a strong sense of community—for organizations and groups in which they were directly involved. More than 75 percent of respondents who reported being involved in places of worship (122 out of 157) and nearly 70 percent of CSA members (206 out of 299) involved in social organizations rated these activities a four or five. Similarly, more than half of respondents (57 percent or 94 out of 165) involved in neighborhood associations rated associations related to your place of residence as providing them with a strong sense of community, and just under half of CSA members involved in political groups (47 percent or 42 out of 89) reported a strong sense of community.

Table 5 Respondents who reported strong feelings of community (n = 565)
Full size table

Participating and volunteering is community?

Though volunteering and participating in farm events and activities are not factors that motivated respondents to join their CSA, the data suggest that a majority of respondents volunteered not only at their CSA but at other places as well. A small percentage of respondents (4 percent or 22 out of 565) reported joining their CSA to participate in farm events and activities a major influence in their decision to join their CSA. A similar percentage (4 percent or 24 out of 565) ranked joining their CSA to volunteer at the farm a five. Despite this, 60 percent of CSA respondents (326 out of 565) are required to volunteer at their CSA. Of the respondents who are required to volunteer, 94 percent (306 out of 326) stated they fulfilled this obligation. Even though two-fifths of CSA members are not required to volunteer (228 out of 564), 86 of these members (38 percent) volunteered anyway. Finally, we also asked CSA members whether or not they volunteered beyond their CSA. Nearly 70 percent of members (387 out of 565) indicated engaging in other forms of volunteering.

Discussion

Demographics of CSA members

The demographics of these CSA members are consistent with other studies of CSA members (Durrenberger 2002; Kane and Lohr 1997; Kolodinsky and Pelch 1997; Lang 2010), in which a majority of respondents are white, well-educated women. While only 30 percent of US citizens earned a bachelor’s degree (or higher degree) in 2009 (United States Census Bureau 2011a), three-quarters of CSA respondents reported earning a college degree or beyond. The household income of respondents in this sample largely mirrors household incomes in the US (United States Census Bureau 2011b), with the exception of members earning between $15,000 and $35,000 per year; however this sample is not necessarily a representative sample of members who subscribe to New York CSAs. While respondents earning between $15,000 and $35,000 per year are not well represented compared to the general population, a smaller percentage of respondents in this category is not entirely unexpected given that a subscription to a full-share of only vegetables ranges from a low of $250 to upwards of $1,000 per season. With payment due both in full and in advance of the growing season, the data on household income were not unexpected. An overwhelming percentage of respondents to the survey are female and these results are consistent with other studies of CSA membership (Lang 2010; Loughridge 2002; Oberholtzer 2004).

Reasons for joining a CSA

Few respondents reported joining their CSA for reasons related to community, and this appears to be underlined among members in higher income categories. Surprisingly, respondents in the lowest household income category rated sharing financial risk, volunteering, meeting like-minded people, and participating in farm events/activities a major influence in joining their CSA, in sharp contrast to other income categories. The disparities between the lowest and highest income categories, illustrated in the ANOVA, are significant and pronounced across variables associated with joining a CSA for reasons related to community. This finding is somewhat counterintuitive because we would expect to find respondents in higher income categories more willing to shoulder financial risk, as well as having more time to volunteer and participate in farm events/activities. That said, respondents with higher household incomes have better social networks (Lin 2000), and perhaps are less dependent on elements of CSA to garner a sense of community.

Factors influencing members’ decisions to subscribe to a CSA include tasty, healthy produce, grown locally. Highlighting this, eating locally grown food ranked ahead of eating organic fruits and vegetables, underlining the potency of the local food movement. These findings are consistent with other studies of CSA members (Cone and Kakaliouris 1995; Conner 2003; Loughridge 2002; Oberholtzer 2004; Ostrom 2007). The apparent success of “eat local” campaigns deemphasizes other elements of CSA such as sharing risk and building a stronger sense of community (DeLind 2010). Illustrating this phenomenon, DeLind (2010) questions why individuals should “Join Michael Pollan’s Army,” a national movement, to eat local when they can simply unite with people in their own neighborhoods.

Though eating locally produced food is ranked as one of the top three reasons why respondents joined their CSA, the term local is not well defined nor has it been fully conceptualized. It means different things to different people, making it difficult to assess. Perez et al. (2003), for example, find that some CSA members associate local with job opportunities, ecological benefits, and promoting connections among farmers, other people, and the land. Still others emphasize sustainability, local purchasing and local economies. Arguably more narrow, the USDA uses a geographic definition of local adopted by the 110th Congress (2007–2008), “a local or regionally produced agricultural product can be transported no more than 400 miles from the origin in which it was produced” (Martinez et al. 2010, p. 3). While local is not necessarily synonymous with community, we cannot discount that it may represent community for some respondents. Illustrating the importance of the concept a CSA member commented, “Buying local and supporting the local farmers provides my family with the best food available and has allowed me to develop deeper roots in my community.” Collecting produce from the farm’s pick-up point on a weekly basis constitutes a sense of community for some members. The number of visits a CSA member makes to the farm, including pick-ups, is considered a factor in how well integrated members are in their CSA (Loughridge 2002; Hinrichs 2000). Urban members—who may not visit the farm—may be financially, psychologically, or abstractly connected to the idea of “local farming.” According to Hinrichs (2000), CSA offers social interaction embedded in an otherwise market-oriented exchange, suggesting even urban CSA members may feel connected to farming, despite their distance to the farm.

Role of community

It appears that CSAs do not necessarily promote or facilitate community for their members. There are several plausible explanations for this. First, a sense of community is arguably more abstract than eating fresh, local, organic food. Second, Americans rely far less on their immediate community than they did previously. Results show CSA respondents resided in their communities for shorter lengths of time compared to a national study, which shows only a quarter of respondents resided at their residences between 1 and 5 years (Social Capital Community Survey 2006). Meeting like-minded people through their CSA tends to be a low priority as well. Still other factors contribute to a weakened sense of community. The ubiquity of technology fulfills the need for community for many, especially with the advent and growth of virtual communities established through social networking sites (Gruzd et al. 2011). Third, results show CSA membership does not provide many CSA members with a sense of community. For one-third of respondents who disagreed with statements about CSA opening their eyes to the importance of being part of a community and/or feeling they have become integrated into their CSA community, it may be that respondents already are well-acquainted and acclimated within their communities even before joining their CSA (DeLind 1999; Loughridge 2002). Underlining this phenomenon a respondent who belonged to two CSAs prior to 2010 noted, “Neither did much to foster a sense of community,” suggesting the member expected community to be cultivated by the farmer, rather than members assuming the role of facilitators. Supporting this idea, Ostrom (2007) asserts that the relationship between the farmer and members is central to today’s CSA, diverging from the original CSA model in which members developed relationships with each other, as well as the farmer. There is one noteworthy exception. ANOVA shows CSA members who joined their CSA between 2007 and 2006 feel significantly more integrated into their CSA community than members who joined in 2010. This is not surprising given that it probably takes times to become fully acclimated to all that a CSA offers. Still these results suggest a shift away from the more collaborative models to instrumental and functional models of CSA, prioritizing the economic components of farming above ones related to community (Feagan and Henderson 2009).

Despite feeling a rather weak sense of community derived from their CSA membership, respondents overwhelmingly expressed interest and involvement in local political or social issues. Though the survey questions differed slightly, these results mirror a national survey conducted in 2006, showing 35 percent of respondents are somewhat interested in politics and national affairs and 34 percent are very interested in politics and national affairs (Social Capital Community Survey 2006). Our results suggest members are not necessarily uninterested in community, but rather CSA membership is neither emblematic of community, nor is it a vehicle that fosters community. Illustrating this, one CSA member commented, “I see a lot of political activism in the community of people who use CSAs, but none of the CSAs I’ve been part of have participated in the activism. It’s all about the food.” The dearth of activism within CSAs is not unexpected since farmers must be concerned first and foremost with growing, harvesting, and managing the farm rather than organizing members. Studies of CSAs (e.g., Oberholtzer 2004; DeLind 1999) illustrate the challenges farmers face organizing events and building community. Expecting members to organize or mobilize on behalf of their CSA or other farm-related issues may be unrealistic. Fifteen percent of CSA members indicated not being involved with a group or organization in the year prior to the survey, compared to 21 percent of Americans (Verba et al. 1995).Footnote 9 While it appears that members are involved, these results do not indicate the frequency of involvement in activities by members. Still, ANOVA shows members who joined their CSA in 2005 or earlier are significantly more involved in local political or social issues than members who joined their CSA in 2010. It is not surprising that long term CSA members are more civically engaged than newer members. The former probably place a higher premium on not only participating in their community, but doing so in a variety of ways.

Though a substantial percentage of members are involved in groups and organizations, feelings about community—unrelated to CSA membership—are similarly weak. A 2006 national study stands in stark contrast to these results. More than half of respondents reported the people with whom they work or attend school, living in their current town or city and their place of worship is very important to them (Social Capital Community Survey 2006).Footnote 10 Weak feelings about community among CSA members are attributed to a number of factors, most notably, the frequency of participation in groups and organizations affects feelings about community and belonging (Hunter 1975; Wandersman and Giamartino 1980). Even across activities in which people might have daily contact, namely work and school, a minority of respondents reported a strong sense of community or belonging. Again, many of the reasons discussed above likely explain why feelings about community are weak.

Participating and volunteering in CSA

Members conscientiously volunteered at their CSA and other places. According to the US Census bureau, 27 percent of Americans volunteered in 2009 (Cramer et al. 2010), while more than twice as many CSA members reported volunteering for organizations unrelated to their CSA. Rates of volunteering among CSA members are higher than the general population. A strong commitment to volunteering at the CSA is evident as well. One CSA member who volunteered commented,

I have been a member of my CSA for over five years. I have been actively involved in volunteering time in helping grow the farm, and increase its membership. I designed their logo, membership brochure, as well as created large-format posters to be displayed on farm property.

Originally CSA farms were created by the community, often through a core group of volunteers working in concert with a farmer. The core group was the decision making body for the farm, responsible for budgeting, sharing costs, and communicating with members (Groh and McFadden 1997; Lamb 1994; Van En et al. 1997). Under these circumstances it is difficult to imagine a CSA not fostering a sense of community. Yet these data suggest that community is not necessarily fostered by CSAs.

Still, participation and volunteering are at times circumscribed and altogether absent. A respondent emphasized the limited nature of her CSA, noting “…our CSA is just get your share and go, volunteer when you’re supposed to, and that’s it,” underlining the limited nature of community. Another respondent acknowledged that while she had not volunteered at the farm, she helped in other ways including creating a Facebook page for her CSA. Still others desired to volunteer but were unable to do so due to work or other obligations. Some scholars contend CSA members like the idea of being involved in a CSA and supporting farmers, but busy lives and hectic schedules often prevent participation and active involvement (Cone and Kakaliouris 1995; Loughridge 2002; Oberholtzer 2004). Similarly, distance from the farm is often an obstacle to member participation, especially among urban members, a phenomenon documented by Oberholtzer (2004) and DeLind (2010). Highlighting the challenges farmers face as they deliver products further distances one farmer explained, “I know CSAs that went out of business because [the farmers] were expecting them to come out to the farm to pick up their shares, and expecting them to work on the farm—two ways to ensure your failure as a subscription grower in DC, because of everyone’s lifestyle and the traffic” (Oberholtzer 2004, p. 12).

Finally, in terms of participation and volunteering, results from this study are consistent with recent studies that show a decline of core groups (Lass et al. 2003; Loughridge 2002) and reluctance on the part of members to stay involved (DeLind 1999; Loughridge 2002; Ostrom 2007). Obviously this places a burden on farmers to maintain membership and encourage members to participate in farm activities. Coupled with this, CSA farms increasingly are initiated by farmers and not members of the community (Ostrom 2007). In short, there has been a notable shift in the nature of CSA during the last 10–15 years as CSA farms move from being truly community supported investments to alternative means of providing high-quality farm products, a transition that Loughridge (2002) investigates at length.

Conclusion

Securing fresh, local, seasonal produce are the primary reasons respondents joined their CSA, suggesting the original idealized notions of community upon which CSA was predicated is an ancillary motivation for joining a CSA. The data show that eating local surpassed organic in terms of importance. Arguably eating local can be considered an important element of community even though respondents did not necessarily link these two concepts. Though the emphasis on local is pronounced, it does not entirely discount the presence of community or social elements embedded in these alternative agriculture arrangements (Hinrichs 2000).

Contrary to claims by scholars touting the role of community in CSA, this study suggests the CSA model does not offer much in the way of community to members, even among members who are interested and involved in community. While community is desirable on many levels and much of the CSA literature accepts community as inherent and attainable, these findings underline just how idealized this concept is. Farmers and managers are more often than not farming alone without the support of community. Accordingly, the instrumental and functional models best depict the current condition between farmers and members in New York State (Feagan and Henderson 2009). In part, this is due to farmers’ need to be economically viable, as well as the difficulty they face sustaining and facilitating member participation (Loughridge 2002). While the ideal CSA is one structured around committed and involved members, CSA as an economic model is not a novel idea.

This study illustrates the importance of recognizing the broader role of community. To date, a dearth of empirical work framing community more broadly contributes to an incomplete understanding of community in CSA as evidenced by this research and Loughridge’s (2002). A weak sense of community among CSA members lends support to Putnam’s (2000) claims that Americans are experiencing a declining sense of community, likely influencing members’ perceptions of their CSA. Conceptualizing community within alternative agricultural arrangements is complex, subjective, and highly contested, but a necessary step to reconciling the ideal notion of community with practice. This is important not only for scholars but farmers too, lest they till, hoping to cultivate community without reaping any of the purported benefits.

Future studies of CSA should continue to investigate the nuances of community in CSA. To that end, we recommend the scope of study be extended to include regional and national studies of CSA members to mitigate the local bias of a single state or locality. Methodologically, enlarging sample sizes and generating random samples to yield generalizable results will enhance our understanding of member perceptions. Undertaking qualitative research, including interviews and focus groups of members and farmers, will offer a more nuanced and in-depth understanding of the role of community in CSA. For example, studying interactions between farmers/managers and members, as well as among members, at pick-up points might offer insight as to whether or not this constitutes a sense of community for some members. A qualitative approach is better suited to asking “how” and “why” questions, unlike survey research, which is particularly useful for measuring complex concepts like community. Finally, comparative research might explore differences and similarities between community in CSA and other types of alternative food arrangements.