Am J Community Psychol (2018) 61:488–499

DOI 10.1002/ajcp.12245

ORIGINAL ARTICLE

Predictors of Farmers’ Market Shopping among People Receiving

Supplemental Nutrition Assistance Program Benefits
Darcy A. Freedman,1 Eunlye Lee,1 Punam Ohri-Vachaspati,2 Erika Trapl,1 Elaine Borawski,1 Kimberly Bess,3
and Susan Flocke1

Highlights
• Study finds factors influencing farmers’ market use among low income.
• Awareness of farmers’ markets and incentive programming were related to increased market use.
• Farmers’ market use associated with multiple dimensions of healthy food access among low income.
• Findings reveal modifiable leverage points to enhance reach of farmers’ markets.

© Society for Community Research and Action 2018

Abstract Promoting use of farmers’ markets (FMs) is a Keywords Community interventions Farmers’ markets,


promising community-level strategy to increase access to nutrition Poverty Health promotion Supplemental
  

nutritious foods such as fruits and vegetables. Yet, FM Nutrition Assistance Program
shopping among people with Supplemental Nutrition
Assistance Program (SNAP) benefits remains low. This
research examined predictors of FM shopping among
SNAP recipients living within 1 mile of a FM. A cross- Introduction
sectional survey of SNAP participants (N = 270) was
conducted in 2015 in Cleveland and East Cleveland, OH, Farmers’ markets (FMs) are a promising community-level
USA. Multinomial regression and zero-truncated Poisson strategy to increase access to nutritious foods such as fruits
regression analyses were conducted to examine factors and vegetables especially in neighborhoods with limited
associated with FM shopping. Results indicate 48% access to other healthy food retailers (Khan et al., 2009;
reported shopping at a FM at least once in the past year, U.S. Department of Agriculture [USDA], 2013; Centers
26% had shopped at a FM before, but not in the last year, for Disease Control and Prevention [CDC], 2011). One
and 26% had never shopped at a FM. The multivariable targeted low-income population for these strategies is
analyses found awareness of FMs and a healthy food households receiving Supplemental Nutrition Assistance
incentive program, and four dimensions of healthy food Program (SNAP) benefits. SNAP is the largest food assis-
access are significantly associated with FM shopping tance program in the U.S. providing financial resources for
among SNAP recipients. The food access dimensions food purchases to more than 40 million low-income Amer-
included service delivery, spatial-temporal, personal, and icans each year (U. S. Department of Agriculture, Food
social access. Findings highlight modifiable leverage and Nutrition Service, 2017b). Among this population,
points for improving the reach of FMs among low-income FM shopping is less common (U. S. Department of Agri-
populations. culture, Food and Nutrition Service, 2016; USDA, 2017;
U. S. Department of Agriculture, Food and Nutrition Ser-
vice, 2017a). The goal of this study was to identify factors
associated with FM shopping among households with chil-
✉ Darcy A. Freedman dren that are receiving SNAP benefits.
[email protected]
1
Food, Health, and Community-level Change
Case Western Reserve University, Cleveland, OH, USA
2
Arizona State University, Phoenix, AZ, USA Most Americans consume fewer fruits and vegetables per
3
Vanderbilt University, Nashville, TN, USA day than recommended by dietary guidelines (Lee-Kwan,
Am J Community Psychol (2018) 61:488–499 489

Moore, Blanck, Harris, & Galuska, 2017). Subpopulations and preferences, food preparation knowledge and skills,
categorized by social positions related to class and race and personal health status that influence food-related
experience disproportionately worse dietary patterns and behaviors.
diet-related chronic disease trends (Dong & Lin, 2009; Importantly, the multicomponent framework of nutri-
Lee-Kwan et al., 2017; Wang et al., 2014). The persis- tious food access illuminates that changes to spatial con-
tence of class- and race-based inequities in diet and texts through the introduction of a new food retailer is
diet-related chronic disease motivated research on commu- only one part of a community-level intervention approach.
nity-level factors that may be at the root of these dispari- Developing a new food retailer in a food desert context
ties, including growing attention to spatial differences in may be the necessary first step to community-level
healthy food access (Beaulac, Kristjansson, & Cummins, change. However, additional steps are needed to address
2009; Dutko, Ver Ploeg, & Farrigan, 2012; Walker, economic access to promote affordability, service delivery
Keane, & Burke, 2010). The term “food desert” is now access to improve availability, social access to support
commonly used in public and professional discourse acceptability, and personal access to realize alignment and
reflecting mainstream awareness of healthy food access accommodation to individual need. The aim in this
gaps more common in low-income and racial and ethnic research was to assess different domains of nutritious food
minority communities. access to understand which factors are related to utiliza-
There is increasing interest in community-level inter- tion of FMs located within urban communities that have
ventions to improve diet-related health conditions. This is limited access to full-service supermarkets and grocery
in contrast to individual-level interventions that focus on stores offering a wide variety of healthy foods.
changing personal attributes such as knowledge, attitudes,
and beliefs considered to shape lifestyle choices. Commu- Reach of Farmers’ Market Interventions
nity-level health interventions include changes to environ-
ments to make the healthy choice the easy choice Farmers’ markets are a recommended community-level
(Frieden, 2010). This approach draws attention to limita- approach to increase healthy food access with the
tions of individual-level interventions by highlighting broader goal of improving diet (CDC, 2011; Khan et al.,
environmental constraints on lifestyle choices. Commu- 2009; USDA, 2013). The number of FMs in the U.S.
nity-level interventions such as FMs have the potential for has grown exponentially over the past two decades
broader population health impact because exposure is gen- (U. S. Department of Agriculture, Agricultural Marketing
erally available to the public versus a select group. Service, 2016). While FM growth is important, FM reach
among populations with high rates of diet-related health
Nutritious Food Access Framework conditions is critical to achieve population health goals.
In 2015, <1% of all SNAP dollars were spent at FMs;
A multicomponent framework of nutritious food access these purchases were made by SNAP recipients repre-
draws attention to five domains to guide community- senting <0.02% of the households receiving SNAP dur-
level strategies designed to improve healthy food access ing this timeframe (U. S. Department of Agriculture,
(Freedman, Blake, & Liese, 2013). The five domains of Food and Nutrition Service, 2016; U. S. Department of
nutritious food access include (a) spatial-temporal, (b) Agriculture, Food and Nutrition Service, 2017a). Results
economic, (c) service delivery, (d) social, and (e) per- of a systematic review revealed barriers to FM use
sonal. Spatial-temporal access takes into account the among low-income populations related to the five
stores available in the local food environment, travel domains of the multicomponent nutritious food access
time and transportation resources needed for food shop- framework (Freedman et al., 2016). We are unaware of
ping, and time costs for food preparation. Economic any study that examined collectively all five domains of
access represents financial constraints and resources nutritious food access to better understand FM shopping
influencing nutritious food access including food prices, behaviors.
household finances, perceived value of foods in stores,
and the availability of financial incentive programs to Study Goals
reduce food costs. Service delivery access is defined as
retailer factors including the quality and variety of foods To realize population health goals through FM implemen-
sold, staff and customer service, and the organization tation, it is important to identify factors associated with
and cleanliness of the retail space. Social access takes FM shopping, particularly among those who might benefit
into account familial, racial, and ethnic foodways and the most. Thus, this study examined the association of
traditions, as well as social connections shaping food multiple measures of access associated with FM shopping
habits. Personal access includes food-related identities among SNAP recipients who live within 1 mile of a FM.
490 Am J Community Psychol (2018) 61:488–499

Methods Trained community researchers approached 1,182 indi-

viduals and 910 (77.0%) completed the initial eligibility
Study Context screening. Five of the study personnel were community res-
idents who lived and/or worked in the targeted geographic
This research used data collected from FreshLink, a 5-year area and one was a research assistant who was bilingual in
study seeking to increase the reach, adoption, and impact English and Spanish. Of the 910 screened, 360 (39.6%)
of FMs among people receiving SNAP benefits using a met the inclusion criteria. Most ineligible individuals lived
peer-to-peer outreach intervention approach. Building on outside the targeted geographic area or did not receive
methods used to develop other social network-based dis- SNAP. A total of 355 individuals (98.6% of eligible) pro-
semination interventions (Kegeles, Hays, & Coates, 1996; vided written consent to join the study and 322 (89.4%)
Li, Weeks, Borgatti, Clair, & Dickson-Gomez, 2012), we completed the survey. The analyses here are focused on
conducted two formative studies including a cross-sec- 270 participants with data for relevant variables.
tional survey and in-depth interviews to inform the peer-
to-peer outreach intervention. Procedure
This manuscript is focused on findings from the cross-
sectional survey conducted between June and August 2015 Recruitment took place at community locations with high
in Cleveland and East Cleveland, Ohio. These adjacent access to SNAP or SNAP eligible populations. Recruit-
metropolitan areas were selected for several reasons. First, ment of study participants and data collection was simul-
more than a third of residents received SNAP benefits taneous (June to August 2015). Over the 12-week period,
(U. S. Census Bureau, 2010). Second, these cities had a sig- the team conducted 37 recruitment events at 17 commu-
nificant number of census tracts that are low income with nity-based sites to raise awareness about the study and
low access to full-service supermarkets (i.e., food deserts) directly recruit. Thirty-seven percent of the people who
(U. S. Department of Agriculture, Economic Research Ser- joined the study were recruited at county offices were
vice, 2016). Third, since 2010, FMs have been implemented SNAP benefits and other social services were provided.
as a community-level strategy to address food access chal- Of the remaining participants, 23.3% were recruited at
lenges in these cities (Walsh, Taggart, Freedman, Trapl, & emergency food assistance sites, 15.2% at neighborhood
Borawski, 2015). At the time of data collection, these cities centers, and 3.1% at farmers’ markets. Additionally,
had higher access to FMs compared to state and national 21.4% of the participants joined by calling the study
trends (CDC, 2013). Fourth, since 2010, a FM healthy food phone line in response to fliers and word-of-mouth. All
incentive program has been implemented to promote eco- interested individuals were screened by study personnel.
nomic access among SNAP recipients. The program pro- Participants completed an approximately 40-minute
vides a dollar for dollar match on SNAP benefits used at a close-ended survey administered orally in-person or on
FM for a maximum of $10 in matching benefits per day that the phone by trained researchers. Data were directly
can only be used to purchase fruits and vegetables. entered into survey software (Qualtrics, 2015). Participants
received a $25 supermarket gift card for their participa-
Participants tion. All study procedures were approved by the Institu-
tional Research Board at Case Western Reserve
To select the geographic boundaries for sampling, we first University.
identified all FMs in Cleveland and East Cleveland that
were open between 2014 and 2015 and were located in or Measures
adjacent to at least one census tract with a SNAP partici-
pation rate of ≥30%. Second, a 1-mile radius around these Outcome Variables: Farmers’ Market Shopping
FMs was created resulting in a 32.5 square mile target
area including 17 FMs that accepted SNAP benefits for Participants were asked if they ever shopped at a FM. If
payment. Eleven of the 17 FMs offered healthy food yes, then they were asked to indicate the last time they
incentives (Freedman et al., 2017). Within these geo- shopped at a FM, and responses were recorded as less
graphic boundaries, individuals were eligible for the study than 1 year ago, 1–2 years ago, 3–4 years ago, and 4+
if they: (a) resided in targeted geographic area ≥1 year; years ago. Those who reported “less than 1 year ago”
(b) were current SNAP recipients; (c) had child(ren) were asked to provide the name of the FM(s) and how
≤18 years in household; (d) were responsible for house- often they went to each FM during the past 12 months.
hold food shopping; (e) spoke English or Spanish; and (f) All responses were coded as a frequency in the past
were ≥18 years old. One adult per household could 12 months (e.g., once every month = 12 times). These
participate. variables were used to create two outcome measures. A
Am J Community Psychol (2018) 61:488–499 491

categorical variable of FM shopping included four groups: bus stop or other public transportation” and “It is easy to
never, not in the last year, 1–2 times in past year, three or get there.” Responses were recorded on a 5-point Likert
more times in past year. Among those who went to a FM scale ranging from poor (1) to very good (5). An average
in the past year (n = 129), a continuous outcome variable score was calculated with higher scores indicating higher
was calculated. perceived spatial access to FMs.
Time costs of food purchased and consumed is a four-
Predictor Variables item scale (Cronbach’s alpha = .74) adapted from prior
research (Steptoe, Pollard, & Wardle, 1995). The four-
Predictor variables included measures of awareness of item scale asked participants how much they agreed or
FMs and the healthy food incentive program developed disagreed with statements such as: “I never have enough
for this study and six measures of access to healthy foods time to shop for fruits and vegetables” and “My food
that are described in full detail in a prior study by Flocke shopping is always rushed.” Responses were on a 5-point
et al. (2017). Nutritious food access measures aligned Likert scale ranging from strongly disagree (1) to strongly
with the framework developed by Freedman et al. (2013) agree (5). For each scale, an average score was calculated
including indicators to assess service delivery, economic, with higher scores reflecting greater levels of time burden.
spatial-temporal, personal, and social factors. Personal access measure. Personal access was
Awareness of farmers’ markets and healthy food measured using a four-item fruit and vegetable preparation
incentive program. Two single items were developed to self-efficacy scale (Cronbach’s alpha = .86) adapted from
assess awareness of FMs near home and the healthy food an existing measure (Condrasky, Williams, Catalano, &
incentive program. Participants were asked, “Is there a Griffin, 2011). Participants were asked to rate their
farmers’ market located near where you stay or live?” and confidence in preparing foods such as “fresh green
“Have you heard about the [name of healthy food vegetables (e.g., broccoli, spinach)” and “root vegetables
incentive] Program?” and were shown a flyer for the (e.g., potatoes, beets, sweet potatoes)” on a 5-point Likert-
incentive program. Response options were Yes (1), No type scale from not at all confident (1) to extremely
(0), or Don’t Know (0). confident (5). Scores on each item were averaged to create
Service delivery measure. A single item assessed a total scale score with higher scores indicating higher
perceptions of the quality of fruits and vegetable at FMs. levels of self-efficacy.
The questions stated, “How does the quality of the fresh Social access measure. A seven-item measure of
fruit and vegetables at farmers’ markets compare to the social connectedness to FMs (Cronbach’s alpha = .86)
quality of fresh fruits and vegetables you buy from other assessed feelings of welcome, familiarity, and social
stores? In general would you say that it is. . .?” Response relationships with people at FMs; the measure was
option ranged from a lot worse quality (1) to a lot better informed by other studies (Leone et al., 2012; Liese et al.,
quality (5). A higher score indicates better perceived 2013; Steptoe et al., 1995; Thompson, Haziris, & Alekos,
quality of fruits and vegetables available at FMs 1994). Five questions asked participants to rate the
compared to other stores. features of the FM they go to most often or about FMs
Economic measure. A single item assessed more generally using items such as “You feel welcome
perceptions of prices of fruits and vegetables at FMs: when you shop there” or “Other customers at the farmers’
“How do the prices of the fresh fruits and vegetables at market will be friendly.” Response options ranged from
farmers’ markets compare to the prices of fresh fruits and poor (1) to excellent (5). Participants rated agreement on
vegetables you buy from other stores? In general would a 5-point Likert scale ranging from strongly disagree (1)
you say that it is. . .?” Response option ranged from a lot to strongly agree (5) on two items: “Members of my
higher prices than other stores (1) to a lot lower prices family think that it is a good idea to buy food at farmers’
than other stores (5). This variable was reverse-coded so markets” and “Most of my friends and acquaintances
a higher score reflected higher perceived prices of fruits think that shopping for food at a farmers’ market is a
and vegetables at FMs compared to other stores. good idea.” Item scores were averaged to create a total
Spatial-temporal measures. Two measures of spatial- score; higher scores were indicative of greater social
temporal access were included. Ease of access to FMs is a connectedness to FMs.
four-item scale (Cronbach’s alpha = .85) informed by a
prior study (Liese et al., 2013). Participants rated the Analytic Approaches
features of the FM they shopped at most often. If they
had never been to a FM or did not have a most frequent Frequency distributions were examined for demographics,
FM, then they were asked to respond based on what they predictors, and the outcome variables. Given our modest
think about FMs. Sample items included: “It is near the sample size and goal to develop a parsimonious model,
492 Am J Community Psychol (2018) 61:488–499

one-way ANOVA and chi-square tests were conducted to at a FM 3+ times during the past 12 months had signifi-
assess if demographic variables were significantly associ- cantly higher perceptions of the quality of fruits and veg-
ated with the outcome variables and were potential con- etables at a FM compared to the other categories
founding variables (p < .10). (p = .03). Ease of access to FMs scores were highest
Two multivariable statistical analyses were conducted among those shopping at a FM 1 or 2 times and 3+ times
to examine factors associated with FM use among SNAP in the past year, and the difference across groups was sig-
recipients. A multinomial logistic regression model inves- nificant (p < .001). Participants shopping at a FM 1 or 2
tigated factors associated with the four groups of FM times and 3+ times in the past year had significantly
shoppers: never, not in last year, 1–2 times in past year, higher levels of confidence about their fruit and vegetable
and (reference category) three or more times in past year. preparation skills (p < .001). There was also a significant
A zero-truncated Poisson regression examined which fac- difference in social connectedness to FMs by FM shop-
tors influenced more frequent use of FMs and was limited ping groups. Specifically, the mean score of social con-
to the subsample who reported shopping at a FM at least nectedness to FMs was the highest among those in 3+
once in the past 12 months (n = 129). All preliminary times, whereas participants in the “not in the last year”
analyses and the multinomial regression were conducted group had the lowest level of social connectedness to FM
in SPSS (v. 24) [IBM Corp, 2016] and STATA (v. 13) (p < .001).
[StataCorp, 2013] was used for the zero-truncated regres- Table 3 reports the results of the multinomial logistic
sion analysis. All results were considered statistically sig- regression model that compared the four groups of FM
nificant at p < .05. shoppers. Perception of prices of fruits and vegetables at
FMs, ease of access to FMs, and time costs of food pur-
chased and consumed were not significant predictors for
Results any comparisons in the model. Additionally, there were
no significant differences between the two groups of cur-
Most participants (N = 270) were female (88%) and rent FM shoppers (1–2 times v. 3+) for any of the vari-
African American (84%) (see Table 1), representing a ables in the model. The odds of being aware of FM near
higher proportion of females and African Americans home were 75% lower among the never group than for
compared to the total population of SNAP recipients the 3+ times group (RR = 0.25, p < .01, 95% CI (0.11,
from the same time period and within the same census 0.56)). Similarly, the odds of being aware of FM near the
tracts (data source: 2015 aggregate data by census tract home were 54% lower for the not in the last year group
from Cuyahoga County Jobs and Family Services). On compared to the 3+ times group (RR = 0.46, p < .05,
average, households included 2.4 children and 1.5 95% CI (0.21, 0.98)). The odds of being aware of the
adults. Roughly 25% did not have a high school degree, healthy food incentive program were 59% lower for the
70% reported their annual household income was less never group compared to the 3+ times group (RR = 0.41,
than $10,000, and 70% were unemployed. Almost 67% p < .05, 95% CI (0.17, 0.96)). Perceptions of quality of
had received SNAP benefits for more than 5 years. Less fruits and vegetables at FMs were 35% lower for the
than half (43%) had access to their own form of trans- never group than for the 3+ times group (RR = 0.65,
portation for food shopping. No significant relationships p < .05, 95% CI (0.44, 0.98)). Similarly, the odds of fruit
between the demographic characteristics and the out- and vegetable preparation self-efficacy were 51% lower
come variables were identified, therefore all demo- among the never group compared to the 3+ times group
graphic variables were excluded from further analyses (RR = 0.49, p < .05, 95% CI (0.27, 0.89)). The odds of
(see Table 1). social connectedness to FMs was 55% lower in the not in
Table 2 presents descriptive statistics of predictors by last year group compared to the 3+ times group
the four groups of FM shoppers. Among the total sample, (RR = 0.45, p < .05, 95%).
26% had never shopped at a FM, 26% had shopped at a Table 4 presents the results of the Zero-truncated Pois-
FM before but not in the last 12 months, 20% went 1–2 son regression model to investigate factors associated with
times in the past 12 months, and 28% went 3+ times. more frequent FM use among SNAP recipients who
More than half (57%) were not aware of a FM near their shopped at a FM at least once during the past 12 months.
home. There were significant differences in awareness of Awareness of the healthy food incentive program had a
FM near the home by the four groups of FM shoppers. significant and positive relationship with frequency of FM
Awareness was lower among participants who never use in the past year, yielding an 18% increase in FM
shopped at a FM, did not shop at a FM in the past shopping frequency (IRR = 1.18, p < .05, 95% CI (1.01,
12 months, and shopped at a FM 1–2 times compared to 1.39)). However, awareness of FMs near the home was
those who went 3+ times (p < .001). Those who shopped not a significant predictor of frequency of FM use. A
Am J Community Psychol (2018) 61:488–499 493

Table 1 Demographic characteristics of participants (N = 270) by farmers’ market shopping categories

Total Never Not in Last One or Two Three or
(N = 270) (n = 70) year (n = 71) Times (n = 54) More Times (n = 75)
Chi-square/
n (%) or Mean (SD) F value p value

Gender
Female 237 (88.1) 61 (22.7) 66 (24.5) 46 (17.1) 64 (23.8) 2.6 .5
Race
Black 228 (84.4) 58 (21.5) 63 (23.3) 45 (16.7) 62 (23.0) 1.4 .7
White, Hispanic/ 42 (15.6) 12 (4.4) 8 (3.0) 9 (3.3) 13 (4.8)
Latino, or Other
Age
18–27 58 (21.6) 23 (8.6) 14 (5.2) 11 (4.1) 10 (3.7) 14.9 .1
28–37 92 (34.2) 15 (5.6) 25 (9.3) 21 (7.8) 31 (11.5)
38–47 56 (20.8) 15 (5.6) 18 (6.7) 11 (4.1) 12 (4.5)
48 or older 63 (23.4) 16 (5.9) 14 (5.2) 11 (4.1) 22 (8.2)
Education attainment
Less than high school 67 (24.8) 19 (7.0) 21 (7.8) 11 (4.1) 16 (5.9) 2.1 .5
High school degree or More 203 (75.2) 51 (18.9) 50 (18.5) 43 (15.9) 59 (21.9)
Annual household income
Less than $ 10,000 185 (69.8) 54 (20.4) 50 (18.9) 38 (14.3) 43 (16.2) 6.7 .1
$ 10,000 or more 80 (30.2) 15 (5.7) 19 (7.2) 16 (6.0) 30 (11.3)
Length of time on SNAP
Less than 1 year 11 (4.2) 3 (1.1) 1 (0.4) 2 (0.8) 5 (1.9) 14.9 .1
1–2 years 30 (11.3) 10 (3.8) 7 (2.6) 8 (3.0) 5 (1.9)
3–4 years 47 (17.7) 16 (6.0) 9 (3.4) 14 (5.3) 8 (3.0)
5 or more years 177 (66.8) 39 (14.7) 52 (19.6) 30 (11.3) 56 (21.1)
WIC recipient
Yes 99 (36.7) 33 (12.2) 22 (8.1) 17 (6.3) 27 (10.0) 4.9 .2
Employment status
Employed for wages 80 (29.6) 24 (8.9) 16 (5.9) 20 (7.4) 20 (7.4) 4.2 .2
Transportation to primary food shopping
Has own car 115 (42.6) 25 (9.3) 33 (12.2) 21 (7.8) 36 (13.3) 2.9 .4
Number of adultsa 1.5 (0.8) 1.7 (0.8) 1.4 (0.7) 1.6 (0.8) 1.4 (0.7) 1.5 .2
Number of childrena 2.4 (1.6) 1.9 (1.2) 2.5 (1.9) 2.5 (1.4) 2.5 (1.9) 2.4 .1
No significant differences between each demographic variable and farmers’ market use were found. The numbers of sample varied for some
variables due to missing values.
SNAP = Supplemental Nutrition Assistance Program.
WIC = Special Supplemental Nutrition Program for Women, Infants, and Children.
a
Means and standard deviations are reported.

one-unit increase in perception of quality of fruits and Every one-unit increase in social connectedness to FMs is
vegetables at FMs was associated with 23% increase in associated with 15% reduction in the frequency of FM
FM shopping frequency (IRR = 1.23, p < .001, 95% CI use (IRR = 0.85, p < .05, 95% CI (0.75, 0.97)).
(1.11, 1.36)). Ease of access to FMs significantly pre-
dicted frequency of FM shopping; every one-unit increase
in ease of access to a FM is associated with a 14% Discussion
increase in frequency of FM shopping (IRR = 1.14,
p < .01, 95% CI (1.03, 1.26)). Time costs of food pur- This study is the first, to our knowledge, to quantita-
chased and consumed were inversely related to frequency tively assess multiple domains of nutritious food access
of FM shopping indicating that for every one-unit increase influencing FM shopping patterns among people receiv-
in time costs, there is 24% reduction in FM shopping in ing SNAP benefits using robust measures. Informed by a
the past 12 months (IRR = 0.76, p < .001, 95% CI (0.67, multicomponent framework of nutritious food access
0.86)). For every one-unit increase in fruit and vegetable (Freedman et al., 2013) and by prior research on FM use
preparation self-efficacy, the frequency of FM shopping (Brown, 2002; Byker, Misyak, Shanks, & Serrano, 2013;
increased by 32% (IRR = 1.32, p < .001, 95% CI (1.12, Freedman et al., 2016; McCormack, Laska, Larson, &
1.55)). Lastly, social connectedness to FMs was signifi- Story, 2010), this study extends research, policy, and
cantly, but inversely related to FM shopping frequency. practice by identifying modifiable leverage points for
494 Am J Community Psychol (2018) 61:488–499

Table 2 Descriptive statistics of predictors for farmers’ market shopping categories among participants receiving SNAP benefits
Three or
Total Never Not in last One or two more
(N = 270) (n = 70) year (n = 71) times (n = 54) times (n = 75)
Chi-square/
n (%) or Mean (SD) F value p value

Measures of awareness
Awareness of FM near home 117 (43.3%) 18 (6.7%) 22 (8.1%) 33 (12.2%) 44 (16.3%) 27.4 <.001
Awareness of healthy 73 (27.0%) 11 (4.1%) 19 (7.0%) 17 (6.3%) 27 (10.0%) 8.1 .05
food incentive program
Service delivery access
Perceived quality of 4.2 (1.0) 3.9 (0.9) 4.1 (1.0) 4.2 (0.9) 4.4 (0.9) 3.0 .03
fruits and vegetables at FM
Economic access
Perceived prices of fruits 2.5 (1.2) 2.4 (1.1) 2.6 (1.2) 2.5 (1.3) 2.3 (1.1) 0.6 .62
and vegetables at FM
Spatial-temporal access
Ease of access to FM 3.5 (1.1) 3.4 (1.1) 3.1 (1.2) 3.8 (1.1) 3.8 (0.9) 6.8 <.001
Time costs of food 2.0 (0.8) 2.2 (0.9) 1.9 (0.8) 2.0 (0.7) 1.9 (0.7) 2.6 .06
purchased and consumed
Personal access
Fruit and vegetable 4.4 (0.7) 4.1 (0.8) 4.3 (0.7) 4.5 (0.6) 4.5 (0.5) 5.4 <.001
preparation self-efficacy
Social access
Social connectedness to FM 3.8 (0.8) 3.7 (0.9) 3.5 (0.9) 4.0 (0.7) 4.1 (0.7) 9.8 <.001
FM, Farmers’ Market; SNAP, Supplemental Nutrition Assistance Program.

Table 3 Multinomial logistic regression predicting farmers’ market shopping categories among participants receiving SNAP benefits
Not in last year One or two times
Never (n = 70) (n = 71) (n = 54)
RR 95% CI RR 95% CI RR 95% CI

Measures of awareness
Awareness of FM near home 0.25** 0.11, 0.56 0.46* 0.21, 0.98 1.15 0.53, 2.50
Awareness of healthy food incentive program 0.41* 0.17, 0.96 0.82 0.38, 1.75 0.82 0.38, 1.76
Service delivery access
Perceived quality of fruits and vegetables at FM 0.65* 0.44, 0.98 0.88 0.59, 1.32 0.83 0.54, 1.27
Economic access
Perceived prices of fruits and vegetables at FM 0.95 0.69, 1.32 1.06 0.79, 1.44 1.14 0.84, 1.55
Spatial-temporal access
Ease of access to FM 1.38 0.88, 2.15 0.92 0.61, 1.38 1.10 0.70, 1.71
Time costs of food purchased and consumed 1.36 0.85, 2.18 0.82 0.51, 1.33 1.16 0.71, 1.89
Personal access
Fruit and vegetable preparation self-efficacy 0.49* 0.27, 0.89 0.86 0.47, 1.56 1.05 0.55, 1.98
Social access
Social connectedness to FM 0.61 0.33, 1.13 0.45* 0.25, 0.79 0.83 0.45, 1.54
Risk ratios (RR) are presented. Reference category = three or more times.
FM, Farmers’ Market; SNAP, Supplemental Nutrition Assistance Program.
2 LL = 667.43 (df = 24, p < .001), Chi-square (D 2LL) = 77.20, df = 24, p < .001.
Deviance statistic = 667.43 (df = 783, p = .99); Naglekerke Pseudo R2 = 0.27.
*p < .05. **p < .01.

improving the reach of FMs among low-income popula- Importantly, the only domain of nutritious food access
tions. The sampling approach, which included partici- that was not associated with FM shopping patterns in both
pants with similar spatial access to FMs, allowed for a regression models was perceptions of the prices of fruits
more nuanced examination of the influence of other and vegetables available at FMs. On average, the SNAP
domains of nutritious food access including economic, recipients participating in this study perceived prices at
service delivery, social, and personal factors on FM FMs to be about the same as or slightly higher than prices
shopping. at other food stores. This finding is important because
Am J Community Psychol (2018) 61:488–499 495

Table 4 Zero-truncated Poisson regression model predicting farm- study participants who never shopped at a FM or had not
ers’ market shopping frequency among SNAP participants who vis- shopped at one in the past year were significantly less
ited farmers’ markets at least once during the past year (N = 129)
likely to know they had a FM near their home even
IRR 95% CI though all participants lived within 1 mile of a FM. This
Measures of awareness finding corroborates other research that found coordinated
Awareness of FM near home 1.00 0.84, 1.18 marketing and outreach efforts to be lacking in many FM
Awareness of healthy food 1.18* 1.01, 1.39 implementation models (Colasanti, Conner, & Smalley,
incentive program 2010; Flamm, 2011; Leone et al., 2012).
Service delivery access
Perceived quality of fruits 1.23*** 1.11, 1.36 Additional factors that differentiated the never group
and vegetables at FM from those shopping at FMs more often were fruit and
Economic access vegetable preparation self-efficacy and perceptions of the
Perceived prices of fruits 0.96 0.90, 1.02 quality of fruits and vegetables available at FMs. Those in
and vegetables at FM
Spatial-temporal access the never group had significantly lower levels of self-effi-
Ease of access to FM 1.14** 1.03, 1.26 cacy indicating less confidence in preparing foods such as
Time costs of food purchased 0.76*** 0.67, 0.86 fresh green vegetables or root vegetables. Additionally,
and consumed those in the never group had less favorable views of the
Personal access
Fruit and vegetable 1.32*** 1.12, 1.55 quality of fruits and vegetables at FMs compared to these
preparation self-efficacy same foods available at other food stores. While there is
Social access consistent evidence that low-income consumers perceive
Social connectedness to FM 0.85* 0.75, 0.97 the quality of fruits and vegetables at FMs to be high
Incident rate ratios (IRR) for zero-truncated Poisson models are pre- (Freedman et al., 2016), factors such as the balance of
sented. Likelihood ratio Chi-square = 83.23, df = 8, p < .001. vendors at a FM (e.g., more arts and craft vendors than
FM, Farmers’ Market; SNAP, Supplemental Nutrition Assistance
Program.
produce vendors) may influence perceptions of quality
Pseudo R2 = 0.08. (Colasanti et al., 2010).
*p < .05. **p < .01. ***p < .001. Examining FM shopping patterns as a continuous mea-
sure with Poisson regression reinforced several of the
results of the multinomial logistic regression. Awareness
food prices are a major driver of food-related habits of the healthy food incentive program, perceptions of
(Drewnowski, 2010; Glanz, Basil, Maibach, Goldberg, & quality of fruits and vegetables at FMs, and fruit and veg-
Snyder, 1998). etable preparation self-efficacy were all positively associ-
Healthy food incentive programs are one approach to ated with increased frequency of FM shopping. Two
address economic barriers, such as food prices, to using additional factors emerged as significant predictors of
FMs. These programs represent a behavioral economic increased FM use. First, as perceptions of ease of access
approach to motivate FM use by providing a small finan- to a FM increased (e.g., FM is easy to get to), FM shop-
cial incentive to reduce the costs of purchasing fruits and ping frequency also increased. Second, as perceptions
vegetables. There is strong evidence that healthy food about the time costs of foods purchased and consumed
incentive programs result in improvements in diet (Afshin increased (e.g., do not have time to cook), FM shopping
et al., 2017; Pearson-Stuttard et al., 2017). In the present frequency decreased. These factors related to spatial-tem-
study, participants had low levels of awareness of the poral domains of nutritious food access highlight com-
healthy food incentive program that had been available in plexity within community-level interventions. While a FM
the study community for at least 4 years prior to data col- was located near all participants, access to transportation
lection (Cleveland-Cuyahoga County Food Policy Coali- may influence ease of access to this community resource.
tion, 2014). This program provides a dollar for dollar In our study, only 43% of the sample had access to a per-
match for SNAP benefits that are transacted at a FM (Cle- sonal vehicle for food shopping. Furthermore, integration
veland-Cuyahoga County Food Policy Coalition, 2014). of FMs into food procurement routines was less likely for
Only 10% of the study participants who shopped at a FM those with constrained access to time needed for food
three or more times in the past year reported awareness of shopping and preparation. These findings illuminate the
the program; awareness was significantly less among need for additional research into factors that improve con-
those who never shopped at a FM. These findings reveal venience of FMs for low-income populations.
the need for dissemination strategies aimed at raising The influence of social access to FMs on shopping pat-
awareness about healthy food incentive programs as well terns resulted in findings that were contrary to our predic-
as the need for general outreach targeting SNAP partici- tions. The role of FMs as spaces for supporting social
pants about FM availability in the neighborhood. Notably, interaction is a salient theme in FM literature, although
496 Am J Community Psychol (2018) 61:488–499

this social benefit seems to be less pronounced in FM settings, contexts with limited or no access to FMs, or
studies including low-income populations (Freedman communities with low rates of poverty.
et al., 2016). Study participants who went to a FM, but
not in the last year, reported significantly lower percep- Implications
tions of social connectedness to FMs compared to those
who shopped at FMs three or more times in the past year. Lewin’s concept that behavior is a function of personal
For this group, there is a chance that visiting the FM and environmental interactions is at the core of commu-
resulted in a negative social experience contributing to nity psychology and has relevance for interpreting the
their decisions to discontinue FM shopping. Among the findings of this study (Lewin, 1935; Lewin, 1997). Addi-
current FM shoppers, increased social connectedness to tionally, social ecological theory, another dominant frame-
FMs was predictive of decreased FM shopping frequency. work guiding the field of community psychology, posits
The unexpected social connectedness findings may be multiple levels of influence on behaviors (Bronfenbrenner,
explained in several ways. Prior research suggests people 1977). These two frameworks highlight the dynamic rela-
receiving SNAP may feel stigmatized shopping at FMs tionship between community-level interventions such as
because of the unique payment method required to trans- FMs and individual behavior. Building on these frame-
act SNAP benefits (Haynes-Maslow, Auvergne, Mark, works, findings can be used to inform strategies designed
Ammerman, & Weiner, 2015). At many FMs, SNAP ben- to increase access to and use of FMs among SNAP recipi-
efits are often transacted at a central point-of-purchase ents by focusing on individual, market, and community-
where these benefits are transferred into tokens that are level factors.
transacted directly with a vendor. The use of a token dis- At the individual level, results illuminate the opportu-
tinguishes a SNAP customer from other customers using nity for FM interventions to operate in tandem with edu-
cash or check with a vendor. Other research suggests FMs cational and skill-based training aimed at promoting fruit
can be exclusionary spaces because of limited diversity and vegetable preparation self-efficacy. Prior intervention
among customers and vendors (Alkon & McCullen, 2011; research that used tailored email-based information
Larchet, 2014). However, there is evidence that FMs pro- exchange to deliver messages to promote fruit and veg-
vide a space to bridge social divides and build social capi- etable self-efficacy resulted in improvements in consump-
tal (Alia, Freedman, Brandt, & Browne, 2014). Finally, tion over time (Luszczynska, Tryburcy, & Schwarzer,
there is a chance that those who frequent FMs more often 2007). Future research may explore connecting these types
are motivated by utilitarian factors more than social con- of interventions within the context of community-level
nectedness. Prior research on motivations for food shop- FM interventions.
ping reveals differences between utilitarian versus At the market level, findings reveal opportunities for
hedonic shoppers (Guido, 2006). Decisions to shop at a FMs to strategically implement communication and out-
particular food store are based not only on expectations reach efforts as a core element of the program model.
about what is available at the store but also on desires Social marketing theory suggests active dissemination will
related to inner motivations, interests, and goals (Guido, require development of salient messages and inclusion of
2006). These differences may explain why social interac- relevant messengers to reach different target populations
tions at FMs are inversely related to FM shopping fre- (Grier & Bryant, 2005). These dissemination efforts may
quency. Further research is needed to unpack the take into account different categories of FM shoppers such
complexity of this phenomenon. as frequent versus less frequent (Freedman et al., 2017).
This study has strengths and limitations. Strengths Findings also highlight the importance of quality and
include the sample population of SNAP recipients that all price in terms of FM operations. Efforts to promote qual-
had access to a FM, use of robust measures, assessment ity through diversification of vendors as well as through
of multiple domains of nutritious food access, orally actions to help potential customers experience high-quality
administered data collection to promote engagement products through taste testing may enhance reach. Integra-
across different levels of literacy, and use of community tion of healthy food incentive programs may reduce price
researchers who may be considered more trustworthy than concerns related to FMs, though our study offers clear
university-based research staff. Limitations include the evidence that incentive programs must be marketed to
risk of recall bias since participants were asked to provide potential customers.
feedback about FM shopping in the past, social desirabil- At the community level, findings highlight FM use is
ity, lack of information about the specific FMs where par- facilitated or constrained by some factors that are beyond
ticipants were shopping, and cross-sectional nature of the the control of an individual that may be addressed through
research. Results of this research may not be generalizable macro-level change fostered by local food policy coali-
to populations not receiving SNAP benefits, non-urban tions (Calancie et al., 2017). Our study offers evidence to
Am J Community Psychol (2018) 61:488–499 497

support efforts to promote ease of access to FMs through impact of food pricing on improving dietary consumption: A
strategic location, such as aligning FMs with public trans- systematic review and meta-analysis. PLoS ONE, 12,
e0172277.
portation or within settings where low-income consumers Alia, K. A., Freedman, D. A., Brandt, H. M., & Browne, T. (2014).
reside and/or work. Findings illuminate the dynamic rela- Identifying emergent social networks at a federally qualified
tionship between time costs and FM shopping. Factors health center-based farmers’ market. American Journal of Com-
that make food shopping or preparation a rushed experi- munity Psychology, 53, 335–345.
Alkon, A. H., & McCullen, C. G. (2011). Whiteness and farmers
ence such as irregular or extensive work hours or child markets: Performances, perpetuations . . . contestations? Anti-
care duties should be considered in FM implementation. pode, 43, 937–959.
Finally, results highlight that social climate may influence Beaulac, J., Kristjansson, E., & Cummins, S. (2009). A systematic
FM use especially among SNAP recipients shopping at review of food deserts, 1966-2007. Preventing Chronic Dis-
ease, 6, 1–10.
FMs less frequently. Bronfenbrenner, U. (1977). Toward an experimental ecology of
human development. American Psychologist, 32, 513–531.
Brown, A. (2002). Farmers’ market research 1940-2000: An inven-
tory and review. American Journal of Alternative Agriculture,
Conclusion 17, 167–176.
Byker, C. J., Misyak, S., Shanks, J., & Serrano, E. L. (2013). Do
Farmers’ markets have the potential to increase access to farmers’ markets improve diet of participants using federal
nutrition assistance programs? A literature review. Journal of
fruits and vegetables in communities where healthy food Extension, 51, 1–11.
access is limited. Findings from this research provide evi- Calancie, L., Allen, N. E., Ng, S. W., Weiner, B. J., Ward, D. S.,
dence to support a multicomponent approach to improve Ware, W. B., & Ammerman, A. S. (2017). Evaluating food pol-
FM reach among low-income populations. Farmers’ mar- icy councils using structural equation modeling. American Jour-
nal of Community Psychology, 61, 251–264.
ket interventions that include strategies focused on indi-
Cleveland-Cuyahoga County Food Policy Coalition (2014). Double
viduals, markets, and the community should be tested to value produce perks 2014 report. Available from: Cleveland-
further the impact of these community-level interventions. Cuyahoga County Food Policy Coaltion website: http://cccf
oodpolicy.org/sites/default/files/resources/2014_dvpp_report_final.
pdf [last accessed March 18, 2018].
Acknowledgments We thank the data collection team, study par- Colasanti, K. J. A., Conner, D. S., & Smalley, S. B. (2010). Under-
ticipants, and community partners who supported this research, standing barriers to farmers’ market patronage in Michigan:
which is a product of a Health Promotion and Disease Prevention Perspectives from marginalized populations. Journal of Hunger
Research Center supported by Cooperative Agreement Number and Environmental Nutrition, 5, 316–338.
1U48DP005030 from the Centers for Disease Control and Preven- Condrasky, M. D., Williams, J. E., Catalano, P. M., & Griffin, S. F.
tion. Findings and conclusions in this publication are those of the (2011). Development of psychosocial scales for evaluating the
authors and do not necessarily represent the official position of the impact of a culinary nutrition education program on cooking
Centers for Disease Control and Prevention. The authors do not have and healthful eating. Journal of Nutrition Education and Behav-
any conflicts of interest, real or perceived, related to this research. ior, 43, 511–516.
Dong, D., & Lin, B. (2009). Fruit and vegetable consumption by
low-income Americans. Washington, DC: U. S. Department of
Agriculture.
Conflict of Interest Drewnowski, A. (2010). The cost of US foods as related to their
nutritive value. The American Journal of Clinical Nutrition, 92,
This research was funded through a grant from the Cen- 1181–1188.
ters for Disease Control and Prevention, Cooperative Dutko, P., Ver Ploeg, M., & Farrigan, T. (2012). Characteristics
and influential factors of food deserts. Available from: the
Agreement Number 1U48DP005030. Findings and con- U. S. Department of Agriculture, Economic Research Service
clusions in this publication are those of the authors and website: https://www.ers.usda.gov/publications/pub-details/?pubid=
do not necessarily represent the official position of the 45017 [last accessed March 28, 2018].
Centers for Disease Control and Prevention. The authors Flamm, L. J. (2011). Barriers to EBT use at farmers’ markets: Les-
sons in empowerment evaluation from rural Ohio. Journal of
do not have any conflicts of interest, real or perceived, Hunger and Environmental Nutrition, 6, 54–63.
related to this research. The study complied with all ethi- Flocke, S. A., Ohri-Vachispati, P., Shon, E.-J., Borawski, E., Mat-
cal standards of human subjects research. It was reviewed lack, K., & Freedman, D. A. (2017). Developing multidimen-
and approved by the Institutional Review Board at Case sional measures of healthy food access among low-income
adults in Cleveland, Ohio, USA. Public Health Nutrition, 20,
Western Reserve University (IRB-2015-1102). 2859–2868.
Freedman, D. A., Blake, C. E., & Liese, A. D. (2013). Developing a
multicomponent model of nutritious food access and related
implications for community and policy practice. Journal of
References Community Practice, 21, 379–409.
Freedman, D. A., Flocke, S., Shon, E. J., Matlack, K., Trapl, E.,
Afshin, A., Pe~nalvo, J. L., Del Gobbo, L., Silva, J., Michaelson, M., Ohri-Vachaspati, P., Osborne, A., & Borawski, E. (2017).
O’Flaherty, M., . . . & Mozaffarian, D. (2017). The prospective Farmers’ market patterns among Supplemental Nutrition
498 Am J Community Psychol (2018) 61:488–499

Assistance Program recipients with high access to farmers’ mar- McCormack, L. A., Laska, M. N., Larson, N. I., & Story, M.
kets. Journal of Nutrition Education and Behavior, 49, 397– (2010). Review of the nutritional implications of farmers’ mar-
404. kets and community gardens: A call for evaluation and research
Freedman, D. A., Vaudrin, N., Schneider, C., Trapl, E., Ohri- efforts. Journal of the American Dietetic Association, 110,
Vachaspati, P., Taggart, M., . . . & Flocke, S. (2016). System- 399–408.
atic review of factors influencing farmers’ market use overall Pearson-Stuttard, J., Bandosz, P., Rehm, C. D., Penalvo, J., Whitsel,
and among low-income populations. Journal of the Academy of L., Gaziano, T., Conrad, Z., Wilde, P., Micha, E., Lloyd-Wil-
Nutrition and Dietetics, 116, 1136–1155. liams, F., Capewell, S., Mozaffarian, D., & O’Flaherty, M.
Frieden, T. R. (2010). A framework for public health action: The (2017). Reducing US cardiovascular disease burden and dispari-
health impact pyramid. American Journal of Public Health, ties through national and targeted dietary policies: A modelling
100, 590–595. study. PLOS Medicine, 14, e1002311.
Glanz, K., Basil, M., Maibach, E., Goldberg, J., & Snyder, D. Qualtrics (2015). Qualtrics software. Provo, UT: Author.
(1998). Why Americans eat what they do: Taste, nutrition, cost, StataCorp. (2013). Stata statistical software: Release 13. College
convenience, and weight control concerns as influences on food Station, TX: Author.
consumption. Journal of the American Dietetic Association, 98, Steptoe, A., Pollard, T. M., & Wardle, J. (1995). Development of a
1118–1126. measure of the motives underlying the selection of food: The
Grier, S., & Bryant, C. A. (2005). Social marketing in public health. food choice questionnaire. Appetite, 25, 267–284.
Annual Review of Public Health, 26, 319–339. Thompson, K. E., Haziris, N., & Alekos, P. J. (1994). Attitudes and
Guido, G. (2006). Shopping motives, big five factors, and the hedo- food choice behavior. British Food Journal, 96, 9–13.
nic/utilitarian shopping value: An integration and factorial U. S. Census Bureau (2010). Decennial census. Washington, DC:
study. Innovative Marketing, 2, 57–67. Author.
Haynes-Maslow, L., Auvergne, L., Mark, B., Ammerman, A., & U. S. Department of Agriculture (USDA) (2013). SNAP-Ed strate-
Weiner, B. J. (2015). Low-income individuals’ perceptions gies and interventions: An obesity prevention toolkit for states.
about fruit and vegetable access programs: A qualitative study. Washington, DC: Author.
Journal of Nutrition Education and Behavior, 47, 317–324. U. S. Department of Agriculture (USDA) (2017). Comparison of
IBM Corp. (2016). IBM SPSS statistics for windows, version 24.0. SNAP authorized farmers’ and markets FY2012 and FY2017.
Armonk, NY: Author. Available from: https://fns-prod.azureedge.net/sites/default/files/
Kegeles, S. M., Hays, R. B., & Coates, T. J. (1996). The Mpower- snap/SNAP-Farmers-Markets-Redemptions.pdf [last accessed
ment Project: A community-level HIV prevention intervention March 18, 2018].
for young gay men. American Journal of Public Health, 86, U. S. Department of Agriculture, Agricultural Marketing Service
1129–1136. (2016). National count of farmers’ market directory listings:
Khan, L. K., Sobush, K., Keener, D., Goodman, K., Lowry, A., graph: 1994-2017. Available from: https://www.ams.usda.gov/
Kakietek, J., & Zaro, S. (2009). Recommended community sites/default/files/media/NationalCountofFMDirectory17.JPG
strategies and measurements to prevent obesity in the United [last accessed March 18, 2018].
States. Morbidity and Mortality Weekly Report, 58(RR07), U. S. Department of Agriculture, Economic Research Service
1–26. (2016). Food environment atlas. Available from: https://www.
Larchet, N. (2014). Learning from the Corner Store. Food, Culture ers.usda.gov/data-products/food-environment-atlas [last accessed
and Society, 17, 395–416. March 18, 2018].
Lee-Kwan, S. H., Moore, L. V., Blanck, H. M., Harris, D. M., & U. S. Department of Agriculture, Food and Nutrition Service (2016).
Galuska, D. (2017). Disparities in state-specific adult in fruit Fiscal year 2015 at a glance. Available from: Food and Nutrition
and vegetable consumption – United States, 2015. Morbidity Service websiteFood and Nutrition Service website: https://fns-
and Mortality Weekly Report, 66, 1241–1247. prod.azureedge.net/sites/default/files/snap/2015-SNAP-Retailer-
Leone, L. A., Beth, D., Ickes, S. B., Macguire, K., Nelson, E., Management-Year-End-Summary.pdf [last accessed March 18,
Smith, R. A., Tate, D. F., & Ammerman, A. S. (2012). Atti- 2018].
tudes toward fruit and vegetable consumption and farmers’ mar- U. S. Department of Agriculture, Food and Nutrition Service (2017a).
ket usage among low-income north Carolinians. Journal of Fiscal year 2016 at a glance. Food and Nutrition Service web-
Hunger and Environmental Nutrition, 7, 64–76. site. Available from: https://fns-prod.azureedge.net/sites/default/
Lewin, K. (1935). A dynamic theory of personality. New York: files/snap/2016-SNAP-Retailer-Management-Year-End-Summary.
McGraw-Hill. pdf [last accessed March 20, 2018].
Lewin, K. (1997). Resolving social conflicts: Field theory in social U. S. Department of Agriculture, Food and Nutrition Service
science (pp. 289–300). Washington, DC: American Psychologi- (2017b). Supplemental Nutrition Assistance Program: As of
cal Association. May 5, 2017. Available from: https://www.fns.usda.gov/sites/
Li, J., Weeks, M. R., Borgatti, S. P., Clair, S., & Dickson-Gomez, J. default/files/pd/34SNAPmonthly.pdf [last accessed March 28,
(2012). A social network approach to demonstrate the diffusion 2018].
and change process of intervention from peer health advocates U. S. Department of Health and Human Services, Centers for
to the drug using community. Substance Use and Misuse, 47, Disease Control and Prevention (2011). Strategies to prevent
474–490. obesity and other chroinic diseases: The CDC guide to
Liese, A. D., Bell, B. A., Barnes, T. L., Colabianchi, N., Hibbert, J. strategies to increase the consumption of fruits and vegeta-
D., Blake, C. E., & Freedman, D. A. (2013). Environmental bles. Available from: http://www.cdc.gov/obesity/downloads/
influences on fruit and vegetable intake: Results from a path FandV_2011_WEB_TAG508.pdf [last accessed March 18,
analytic model. Public Health Nutrition, 17, 2595–2604. 2018].
Luszczynska, A., Tryburcy, M., & Schwarzer, R. (2007). Improving U. S. Department of Health and Human Services, Centers for Dis-
fruit and vegetable consumption: A self-efficacy intervention ease Control and Prevention (2013). State indicator report on
compared with a combined self-efficacy and planning interven- fruits and vegetables, 2013. Atlanta, GA: Centers for Disease
tion. Health Education Research, 22, 630–638. Control and Prevention.
Am J Community Psychol (2018) 61:488–499 499

Walker, R. E., Keane, C. R., & Burke, J. G. (2010). Disparities and Wang, D. D., Leung, C. W., Li, Y., Ding, E. L., Chiuve, S. E., Hu,
access to healhty food in the United States: A review of food F. B., & Willett, W. C. (2014). Trends in dietary quality among
deserts literature. Health and Place, 16, 876–884. adults in the United States, 1999 through 2010. Journal of
Walsh, C. C., Taggart, M., Freedman, D. A., Trapl, E. S., & Bor- American Medical Association Internal Medicine, 174, 1587–
awski, E. A. (2015). The Cleveland-Cuyahoga county food pol- 1595.
icy coalition: “we have evolved”. Preventing Chronic Disease,
12, E86.
Copyright of American Journal of Community Psychology is the property of Wiley-Blackwell
and its content may not be copied or emailed to multiple sites or posted to a listserv without
the copyright holder's express written permission. However, users may print, download, or
email articles for individual use.