Previous research has reported shockingly (dare I say, unbelievably) high rates of food insecurity among college students. For example, here is one 2019 study reporting “60 percent of students had experienced food insecurity within the past thirty days or housing insecurity/homelessness within the past year.” Frankly, I’ve found these results a bit hard to believe given overall food insecurity rates in the U.S. are less than 10% and the data showing that young adults from wealthier households are more likely to go to college than those from poorer households. None of this is to say there isn’t any food insecurity on campus, or that we shouldn’t undertake some efforts to reduce the problem; however, we need more thoughtful discussion about what, precisely, is being measured in these studies. Particularly, when we see that 26% of college students on unlimited meal plans show up as food insecure according to traditional measures.
Thus, I was pleased to see this paper by Brenna Ellison and colleagues just released in the journal Food Policy. As they describe, there are high stakes to getting the estimates of food insecurity right:
“A growing body of research among college students has estimated remarkably high levels of food insecurity when compared to food insecurity estimates from the general population over the past decade, with recent literature reviews reporting average prevalence rates of 33–51% compared to 9.8% among U.S. adults. Given these high rates of food insecurity, policymakers at the state- and federal-levels are considering legislation to better understand and ultimately alleviate food insecurity in the college student population. Further, the National Postsecondary Student Aid Study will soon collect national surveillance data on college student food insecurity. However, emerging research suggests that existing food insecurity measurement tools may not perform as expected when used with college students. Indeed, there is no food insecurity measure that has been validated for use with college students.”
As they go on to discuss, a big issue is that food security is primarily measured in this country with a set of 18 survey questions, like “I worried whether my food would run out before I got money to buy more.” Importantly, the official government statistics on food security don’t even ask people these questions if people’s income is sufficiently high (it is assumed high income households are, by definition, food secure). Previous research has shown one can get MUCH higher food insecurity measures if you ignore these income thresholds. Ellison et al. write:
“To date, only one study has attempted to test the use of screeners in the college population (Nikolaus et al., 2019b), with results indicating that using the two-item screener produces substantially lower (13–15 percentage points lower) food insecurity rates. A recent study in the general population found similar results, with significantly higher prevalence rates of food insecurity without the use of screening protocols (Ahn et al., 2020). It should be noted that the income screener is not easily adapted to the college population due to the challenging nature of estimating income for college students, as they may have financial resources from the federal government (e.g., Pell grants), scholarships, parents/family, or other social supports in addition to any individual income. Therefore, alternative screening procedures may need to be considered to reduce respondent burden among the college population in the future.”
The authors argue for more work to derive food insecurity measures specifically designed (and validated) for college students rather than just assuming general population surveys “work” for students as well. College students are a unique population:
“For college students who are emerging adults, the ongoing transition from their parents’ households to other housing arrangements may also pose challenges for the use of “household” phrasing in questionnaires. Without clarification, students may interpret this as their parents’ households (in which they still may reside for extended periods of time) rather than their current housing situation. Students may also question whether they should include roommates, partners, or others as members of their “household” (Ames and Barnett, 2019). It is also possible that students experience housing insecurity and lack a “household” altogether (Goldrick-Rab et al., 2018).”
The whole paper has solid advice on next steps in designing appropriate food insecurity measures for college student populations.