Abstract
Purpose
The purpose of this study is to describe the prevalence of food insecurity among college students and its relationship to on-campus food purchasing patterns and barriers to eating a healthy and sufficient diet, including the relative cost of food items on campus compared to national chain grocery stores.
Design
A campus-wide survey using stratified sampling among undergraduates and food audit.
Setting
An urban public university in New England.
Subjects
A total of 951 surveys completed by undergraduates.
Measures
Demographic characteristics, behavioral factors and food security status (measured using the USDA 6-item short form) were collected.
Method
Descriptive and multivariable analyses were conducted to describe differences between food insecure and food secure undergraduates.
Results
Overall, 35% of undergraduates experienced food insecurity in the past year (response rate = 92%). Food insecure undergraduates had different on-campus purchasing patterns than their food secure counterparts. Food insecure students were more likely to report barriers to healthy eating on all measures, including prices (AOR= 8.12, P < .0001), to experience housing insecurity (AOR= 2.64, P = .001) and to report that transportation is a barrier to buying groceries (AOR= 1.63, P = .01). After multivariable adjustment, food insecure undergraduates had higher odds of being African American (AOR= 1.57, P = .031) or other races (more than 1 race) (AOR= 3.35, P = .002) compared to white undergraduates.
Conclusions
Food insecure college students face a variety of barriers to healthy eating on campus, including high food pricing on campus and limited transportation options. Further research is needed to inform campus resource development, policies and programming focused on food insecurity prevention for college students.
The United States Department of Agriculture (USDA) defines food insecurity as reduced or disrupted quality or quantity of food intake due to limited or uncertain access to nutritious foods. 1 In 2019, approximately 10.5% of U.S. households experienced some form of food insecurity. 2
Away from the protections of home, yet often too busy with the obligations and rigors of college life to work full time and be truly independent, college students are a unique population who are not immune to the effects of food insecurity. A systematic review of studies assessing food insecurity on college campuses estimates the prevalence of food insecurity amongst college students to be about 32.9% in the United States. 3 Food insecurity amongst college students continues to be a concern as college campuses become more diverse. The percentage of Black, Latinx, and Asian undergraduate enrollment has increased steadily with a decrease of White undergraduate enrollment from 81% in 1980 to 55% in 2014 4 Race has been associated with college student food insecurity, with multiple studies finding students of color were more likely to be food insecure than their White peers.3,5 Additionally, another study amongst community colleges found that those campuses with greater proportions of students of color experienced higher rates of food insecurity than less diverse campuses. 6
The proportion of first-generation college students is also growing; first-generation college students now make up approximately 24% of the undergraduate student population. 7 A recent study conducted at an Appalachian university found that the prevalence of food insecurity amongst first-generation students was almost 16% higher than their peers. This study also found that Black first-generation students were more likely to be food insecure compared to their White peers. 8
College students are challenged with juggling competing financial priorities while seeking to attain a college degree. Many college students may not have a well-defined source of income, and may receive their income from multiple sources including employment, parental support, or financial aid and other loans, amongst many. 9 College students also commonly face debt in a variety of facets, especially student loans. One study revealed that students who owned more than $10,000 in student loans were more likely to be food insecure compared to students who owed less than $1,000. 9 Furthermore, a cross-sectional study from California revealed that the odds of food insecurity were 1.6 times higher in students who received need-based grants, scholarship, or financial-aid. 6
Credit cards are another significant source of debt for college students. A study looking at students from five colleges across three states found that the average credit card debt amongst their participants was $1,035. 9 These students had a very limited income, with the average yearly salary being $7,829. 9 Students from this study also displayed relatively low levels of financial literacy; scoring only 60% on a financial literacy assessment - these results are not that much different from high school students who have taken the same assessment. 9 During a study from the University of North Texas, interviews with students revealed that food insecure students focused first on paying for rent, school, and other bills, and used what money was leftover for food. 10 These challenges contribute to the increasing number of college students who struggle to find enough meals to eat.
The majority of other studies focusing on food insecurity on college campuses have used convenience samples, which limit the representativeness and generalizability of study findings. These studies have had low response rates, ranging from 15% to 28%.3,9,11,12 The current study adds to the literature by utilizing robust statistical methods to examine the relationship between personal characteristics and barriers to healthy eating. This study also explored the added dimension of students’ food purchasing behaviors while on campus, which helps us to better understand how limited finances can impact student selection of healthy foods. Additionally, this study explored the extent to which the cost of food on campus compared to the cost of similar items elsewhere may play a role in food insecurity among undergraduates.
Methods
Survey
Design
This study used a cross sectional design. This study analyzed data from the 2018 college health survey, which has been previously published. 13 A stratified random sample was used to collect the data for this survey for a representative sample of undergraduate students enrolled in the university (across all colleges and departments). The university course list was used as a census. Classes were randomly selected across undergraduate course levels to ensure equal representation across student classification. Data was collected during a 3-week time period during the Spring 2018 semester (February – March). Instructors were asked for permission for the student research team to attend the class session to disseminate the paper survey to students. Student researchers visited classes, explained the purpose of the survey to students and answered any questions about the survey. All surveys were anonymous to ensure students felt comfortable accurately answering the survey questions. Students provided written consent by completing the survey. When surveys were completed, students put their survey into a manila envelope collected by the student research team. Surveys were reviewed by the research team for completion, and any bubbles that were partially filled in were completely filled in prior to scanning. The scannable paper survey forms were created using Snap Survey (version 11). The research team scanned the surveys using a Canon Image Formula DR-9080C and the computer recorded the data. The scanned forms were checked for accuracy throughout the scanning process. Any errors in scanning were corrected by the research team. Survey Select was the software program used by the research team. IRB approval was obtained at the university where this research was conducted.
Sample
This study was limited to students who were enrolled in the university during Spring 2018, who were at least 18 years of age and who were in their freshman, sophomore, junior or senior years of study. Students also needed to be present in class during the days of data collection. A total of 1,336 students were invited to participate in the survey and 1,224 consented to complete the survey. A total of 951 surveys were included in this study after excluding students who were not undergraduates (n = 200) and students with missing data for food insecurity status (n = 73).
Measures
The survey instrument measured food insecurity using the US Department of Agriculture Household Food Security Survey Module: 6-item short form. 14 This measure is validated 15 and has been used in similar populations.3,16,17 Standard demographic information was collected using questions from validated sources.18,19 Aspects of dietary quality were measured by asking students about their intake of sugar sweetened beverages (soda 20 and fruit drinks 20 ) and usual fruit and vegetable consumption. 21 These questions have been validated 15 and used in similar college populations.16-19,22 Questions about barriers to food access, and locations of food purchases on campus in the past month were developed by the survey team and content experts (n=6) and checked for face validity by a group of students (n = 10) in the target group prior to use. These students included undergraduate public health students who were in a research methods class and students working (internship or research position) at the wellness center (n=24). This process included oral and written feedback including identifying questions that had wording errors, or needed additional options for response. The final version of the survey included closed option survey questions that allowed participants to select from several options (multiple choice).
Comparative Food Audit
Using a modified version of the Nutrition Environment Measures Survey in Stores (NEMS-S) food audit tool, two students were trained to collect food availability and pricing data. This measure has a high degree of inter-rater and test–retest reliability. 23 Food items for the tool were selected based upon the selection at the on-campus market. Pricing for these food items were then compared to identical food items available at two local supermarket chains, one Mega-mart and an online grocery delivery service available through one of the local supermarket chains. The stores selected for the comparative food audit were chosen based upon their close proximity to the university campus where this study took place.
Statistical Analyses
Demographic Characteristics of the Sample a .
aHolm P-Values adjusted for multiple testing. Bold values indicate P < .05.
Results
Survey Results
This survey had a response rate of 92%. 13 The undergraduates participating in this survey were similar in demographic makeup to the undergraduates enrolled in the university during the Spring 2018 semester 29 (data not shown). Students who were excluded from the analysis included graduate students, and those with missing data for key variables needed for the analysis including demographics (age, ethnicity, gender) and the six questions used to calculate food insecurity. We compare the characteristics of undergraduates overall, and by food insecurity status in Table 1. The majority of participants were female (63.1%), traditional college age (56.8%) and white (59.6%). There was a relatively even distribution in year in school, and the majority of undergraduates lived off campus (63.2%). The majority of undergraduates were enrolled in the university full time (95.0%) and 63.5% were employed full or part time.
Overall, 35% of undergraduate students in this sample were food insecure (n = 335). In bivariate analyses, no differences were observed in student food security status based upon gender or age. Food insecure students were more likely to be from nonwhite groups; these students were more likely to identify as African American (20.6 vs 14.5%). A higher proportion of food insecure students lived on than food secure students (41.2 vs 34.5%) however this difference was not statistically significant. No differences were observed in enrollment status at the university; the majority of students were enrolled full time (95%) and working part time (71%). No differences were observed in food security status based on being a member of a Greek organization, veteran status or transferring to the university in the last year. A greater proportion of food insecure students worked full time compared to food secure students (13.5 vs 8.0%) however this difference was not statistically significant. Food insecure students were more likely to be a parent to a child under 18 years of age than food secure students (6.6 vs 2.8%), or a member of a sports team (15.8 vs 11.0%), however these differences were not statistically significant. A greater proportion of students who were food insecure were also housing insecure (12.8 vs 3.9%; P < .0001). With regard to paying for school, food insecure students were more likely to report paying for school with student or parent loans than food secure students (72.3 vs 63.1%), however no statistically significant differences were observed with regard to how students paid for school.
Barriers to Eating a Healthy and Sufficient Diet are Associated with Food Insecurity a (n = 951).
aHolm P-Values adjusted for multiple testing. Bold values indicate P < .05.
Associations Between Aspects of Dietary Quality and Food Insecurity a (n = 951).
aHolm P-values adjusted for multiple testing. Bold values indicate P < .05.
Associations Between Strategies and Locations of Food Purchasing and Food Insecurity a (n = 951).
aHolm P-Values adjusted for multiple testing. Bold values indicate P < .05.
Adjusted odds of food insecurity based upon student characteristics and barriers to healthy eating (n = 951) a .
a The logistic regression model was adjusted for gender, race/ethnicity, housing insecurity, prices are a barrier to healthy eating and transportation is a barrier to buy groceries.
Comparative Food Audit Results
The food price component of this study compared the prices at this on-campus market to the prices for equivalent products (same brand and size) to items available at the local supermarket, food stores and through a grocery delivery service. Prices for all of the food items sold at the campus market were between 1.3 and 6.4 times higher compared to the off-campus locations. This data is provided in Appendix A.
Discussion
This study describes the extent of food insecurity and the reasons for it at an urban university, and is one of the first to complete a price audit of foods available on campus. More than a third (35%) of undergraduate students reported being food insecure; this finding is consistent with other studies. 19 This study identifies the disparities in diet quality and differences in food purchasing behaviors and describes the characteristics of food insecure students.
Several modifiable factors contribute to food insecurity for college students in this sample. Food insecure students were more likely to state that transportation and prices of healthy foods were a barrier to buying groceries. Food insecurity was associated with being housing insecure. All three of these factors are proxies for having a lower level of economic resources, making students more susceptible to food insecurity. 30
Cost barriers can make it difficult for students to afford healthy food items, but the reasons for cost barriers being experienced by these students is unique. Not only are healthy foods more expensive than fast food/junk foods, 31 the cost of both is more expensive for students on campus compared to off campus.
The campus food environment is an important consideration for college student food insecurity. Options to buy non-prepared food on this campus are limited primarily to one, small on-campus market. The convenience of being able to buy groceries on campus comes at a significant cost. Such intentional “captive market” or “convenience” pricing on campus not only makes it more expensive to purchase items on campus than to buy them elsewhere, but it also puts students without independent means of transportation at an increased financial disadvantage when it comes to purchasing food. Given a similar budget, and access to off campus food stores, students who are currently food insecure on this university campus might be able to acquire sufficient food so that they would no longer be categorically food insecure. Vending machines on this campus are one of the least expensive locations to make food purchases, however the healthy choices are limited 32 and are a far cry from the variety and quality of food found in markets and grocery stores.
Food insecure students brought fast food to campus more frequently than food secure students. Because of these higher prices on campus, this means food insecure students are spending more of their money on food if they make purchases on campus. It is possible that these same students would not be food insecure if they were paying less for food on campus, or had better resources to access off-campus options. Food insecure students were also less likely to bring food from home than their food secure counterparts, similar to findings from other studies. 33
Despite the urban location of this university, the particular neighborhood in which this university is located is categorically “car-dependent”, meaning that “most errands” including grocery shopping “require a car. 34 ” The urban area surrounding the university campus does not have easily walkable sidewalks, and public transportation options necessitate either taking multiple buses that do not necessarily coordinate in terms of drop off and pickup or taking a substantial walk to access a single bus to a grocery store. 34 In multivariable analysis lack of access to transportation was a significant factor that contributed to food insecurity. 27 This could be due to the fact that students with limited financial resources are less likely to be able to afford a car and the associated costs. Other research at this university has shown that while students are required to purchase a meal plan to live in the dorms, they often purchase the least expensive option, but this does not provide them with enough meals for the entire semester.17,27 Students living in on campus apartments are not required to purchase meal plans from the university (these units contain kitchens). Similar to other studies, access to a cooking preparation space, knowledge of how to prepare a meal (or cooking self-efficacy)26,35 and time to prepare a meal17,27 were all factors that were more commonly reported for food secure students, however in this study, these factors were not as important as having access to transportation or being able to afford healthy food.
This study also examined differences in dietary quality between food insecure and food secure students. Overall, students reported a very low level of fruit and vegetable consumption; 78% of undergraduates reported two or fewer servings per day. Differences in fruit and vegetable consumption were not observed by food insecurity status. Studies of other college student populations have reported that food insecure college students consume more sugar sweetened beverages23,25 and fewer servings of fruits and vegetables per day.19,36,37 At this university, other than the campus dining hall, options on campus to purchase fruits and vegetables are limited. These options are limited to single pieces of fruit, generally only apples and bananas that vary in their quality and are sold for a premium price. Students who have limited funds are unlikely to purchase these items when the same money can be spent on a greater variety of non-produce options. These findings highlight an opportunity to develop new programming with the goal of increasing dietary quality among food insecure college students.
This study developed a multivariable logistic regression model to consider the additive factors that contribute to college student food insecurity with the hopes of identifying factors that are modifiable and potential targets for interventions. This college campus is diverse, and the interventions developed for this campus may not work for others. Food insecure undergraduates had higher odds of identifying as being from a racial or ethnic minority group; specifically, food insecure students had 57% greater odds of being African American or 3.3 times the odds of being other races. This finding is consistent with research on other college campuses in the United States that racial minorities face a higher burden of food insecurity. 12 This finding highlights a need for more food justice work to address these racial disparities. Food insecure students had 2.6 times the odds of also being housing insecure compared to their food secure counterparts. Food insecure students were more likely to state that prices of healthy foods or transportation were a barrier to buy groceries. Other research has found that housing insecurity, stating that prices are a barrier to healthy eating and transportation being a barrier are all proxies for having a lower level of economic resources, making students more susceptible to food insecurity. 30 As programs are developed for food insecure students, addressing housing insecurity is an important consideration.
This study was not without limitations. This study relied on a cross-sectional design, and therefore was unable to assess the temporal relationship of the associations examined. All data collected were self-reported and it is possible that some students may feel a stigma related to food insecurity, therefore chose not to reply to these questions. To mitigate this, these questions were part of a larger survey, measuring a range of health behaviors. A total of 73 surveys were excluded due to missing data for food insecurity; no differences were observed in demographic characteristics for students with and without missing data for food insecurity status. This study did not collect data about the specific type of meal plan students have, which may be the reason why students eating in the dining hall had a higher prevalence of food insecurity. Previous research has identified that students with a block meal plan tend to run out of meals before the end of the semester. 17 Additionally, this survey asked if students had experienced food insecurity ever during the year as opposed to at a particular, identified point during the year. As food insecurity may vary by time of the year, especially but not exclusively for those with block meal plans, future studies should explore what, if any, relationship exists between rates of food insecurity and time of the academic year. Overall, this study had a high response rate (92%) and the demographic information for students participating was representative of students enrolled at the university at the time of data collection. The use of a stratified sampling method ensured a representative sample of students for this study. The findings of this study can only be generalized to other similar public universities located in urban settings with captive pricing and limited transportation to food resources in the United States.
Implications and Contribution
Food insecurity is a serious issue among university students, which has the potential to have long term impacts on students’ health and academic achievement. This large survey of food insecurity is representative of the public college campus on which it was conducted, and adds to existing knowledge about the risk factors for student food insecurity as well as students’ health related behaviors. Although the campus in which this study was conducted has a mobile food pantry, “Swipe it Forward” program and on ground food pantry, more than one third of students continue to experience food insecurity. These individual-level interventions are inadequate to meaningfully and sustainably address this issue. These programs have merit as immediate stopgap measures, 38 however they put the onus on the individual student to work around the predominant, existing system and in so doing imply that it is the student who is at fault for his or her own food insecurity. Moreover, none of these interventions address the underlying issue of food pricing on a campus with a captive audience who may be limited in their ability to access off-campus food resources. Instead, innovative, evidence-based systems-level interventions are needed to address food insecurity on college campuses similar to this one.
This study identified that the prices of groceries and transportation were barriers to accessing food. Some potential interventions include providing a free dining option on campus (John Bon Jovi Soul Kitchen, at Rutgers University - Newark),
39
providing students with transportation to local supermarkets to purchase foods off campus (Grocery Shuttle at UC San Diego)
40
and conducting food recovery on campus to make unused foods available to students (The Food Recovery Network).
41
Further research is needed to identify which interventions would be most appropriate for this specific population. Collaborations and partnerships with university food vendors are essential to ensuring that these programs are successful as they are conceptualized, developed and implemented. The Hunger Free Campus Bill
42
provides public colleges with funding who are addressing student hunger on campus; funding through this bill allows campuses to start a “Swipe Out Hunger” program, establish food pantries and create SNAP enrollment opportunities. Students in this study who were food insecure most often purchased prepared foods from the student center or snacks from vending machines on campus. While SNAP benefits cannot currently be used for hot or prepared foods, students would not get much food with a SNAP benefit if they choose to shop at the campus market. The Hunger Free Campus Bill has the potential to support students and lead to food insecurity solutions for students attending public universities in the United States. Food insecurity is a complex and serious issue that many undergraduate college students experience leading to poor health, academic outcomes. Food insecure students are more likely to drop out of college and not complete their degrees. Expensive food pricing on campus and lack of transportation to purchase groceries were factors contributing to student food insecurity. Key to health promotion practice is meeting people where they are. Innovative food insecurity solutions are needed that address the root causes for food insecurity as opposed to defaulting to generic interventions.So What?
What is already known on this topic?
What does this article add?
What are the implications for health promotion practice or research?
Footnotes
Acknowledgments
We would like to thank the students who participated in this research study. This research was supported with funding from Southern Connecticut State University. IRB approval was obtained at Southern Connecticut State University (Protocol 17-168) and the study was considered exempt; The informed consent document was waived by the university’s IRB and replaced by a very detailed cover letter to ensure that students understood the study and data remained anonymous. No author has any conflict of interest to disclose. All authors have seen and approved the final version of this manuscript. The manuscript has not been previously published nor is it being considered for publication elsewhere.
Author Contributions
V.Z. wrote the manuscript, reviewed/edited the manuscript, conducted the analyses and researched data. J.A. participated in data entry and data collection, reviewed/edited the manuscript, E.S. and P.G. reviewed/edited the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Grocery Item
Campus Market
Grocery Delivery
Supermarket A
Supermarket B
Mega Mart
Campus Market vs Off Campus Price
Tuna fish (5 oz)
$3.89
$1.50
$1.54
$0.97
$1.07
2.5–4.0 times higher
Whole wheat bread (24 oz)
$4.49
$2.99
$2.97
$2.82
$2.75
1.5–1.6 times higher
Cheerios (8.9 oz)
$6.25
$2.50
$3.73
$3.54
$3.07
1.7–2.5 times higher
Eggs (dozen)
$2.35
$1.49
$1.72
$1.72
$0.94
1.4–2.5 times higher
Canned green beans (14.5 oz)
$3.25
$1.00
$1.08
$0.92
$0.88
3.0–3.7 times higher
Canned peaches (15 oz)
$3.19
$2.09
$1.99
$2.47
$1.48
1.3–2.2 times higher
Spaghetti (16 oz)
$2.95
$1.45
$1.38
$1.40
$1.38
2.0–2.1 times higher
Marinara sauce (14 oz)
$4.65
$2.69
$2.47
$2.00
$1.77
1.7–2.6 times higher
Macaroni and cheese (14 oz)
$6.35
$2.59
$1.00
$1.53
$1.18
2.5–6.4 times higher
Peanut butter (16 oz)
$5.69
$3.39
$2.63
$2.47
$2.58
1.7–2.2 times higher
Grape jelly (18 oz)
$4.30
$2.49
$2.55
$2.29
$2.02
1.7–2.1 times higher
Apple sauce (24 oz; 6 x 4 oz)
$5.29
$1.99
$1.99
$2.11
$1.98
2.5–2.7 times higher
Total cost
$52.65
$26.17
$25.05
$24.24
$21.10
—