The Association among Emotions and Food Choices in First-Year College Students Using mobile-Ecological Momentary Assessments

The transition from adolescence to young adulthood, or “emerging adulthood,” is said to be an especially vulnerable period for excess weight gain [1, 2]. During this critical time period, many emerging adults in college make poor food choices [2‐6], which may affect long-term health outcomes [7], particularly among first-year college students.

The transition into college is also an emotional experience for emerging adults [8]. Stress is a commonly cited negative emotion among college students [8‐11] and is frequently associated with emotional eating [8‐12]. Emerging adults have identified negative emotions as contributors to their own overeating [3]. When feeling stressed, they may turn to food for psychological comfort [10, 12]. There is a significant body of literature to support that stress is linked to binge eating [13‐15]. however, the causal relationship is still unknown [15], even in this more developed area of the field. At the same time, research suggests that for some, stress may not change eating habits [16], or may even cause decreases in eating [17]. Eating in response to stress is characterized by eating more food and/or consuming more calories [10], though there is only limited evidence regarding which types of foods are consumed from various food groups. Some research indicates that the foods people eat when experiencing stress are likely to be energy-dense, specifically high in sugar and fat [10, 18, 19]. For instance, a study examining female college students found that over half of participants reported an increase in appetite from stress, and those participants chose to consume significantly more types of sweet foods and mixed dishes which included high-fat items like casseroles, burgers, pizza, and fast food [10]. In these studies, stress was the only emotional aspect assessed, and healthy food choices were not examined.

Negative emotions that have been studied in relation to eating include sadness, anger/frustration, anxiety/fear, and boredom [20‐22]. Some have argued that boredom should not be considered a negative emotion, but rather its own distinct emotional state, especially in relation to eating [23, 24]. Indeed, in open-ended questionnaire items, emerging adults in college have reported eating more in response to boredom than in response to negative emotions like anger and anxiety [24].

Certain gaps and inconsistencies remain in the study of emotion-related eating. For example, positive emotions are rarely mentioned in the eating behavior literature. Although, the very small group of studies that have examined positive emotions indicate that positive and negative emotions do differ in their associations with specific eating behaviors [8, 21, 22, 25]. In general, negative emotions are linked to adverse eating behaviors (e.g., overeating), while happiness is not associated with changes in eating behaviors [8]. Further, the motivation to eat is higher for negative emotions like anger and tension than for positive emotions like joy and relaxation [22]. Of the studies examining associations between emotions and eating, no studies have included a sample of first-year college students who happen to be at risk for weight gain and changes in their eating behaviors, and only a few studies have addressed associations between emotions and types of food consumed [10, 12, 25‐27]. While not first-year students, a study of low-income adolescents found that consuming both sweet and salty snacks as well as sweetened beverages were associated with feeling lonely or bored [28]. The majority of these studies have used retrospective surveys for data capture [10, 12, 25, 26], though, Oliver et al. conducted a study experimentally manipulated stress (resulting in negative mood) and found higher intakes of sweet high-fat foods and a more energy-dense meal [27]. A meta-analysis examining effects of positive and negative moods on food choices and eating behaviors in laboratory settings, suggested that negative mood was associated with greater food intake, while positive mood was linked to higher caloric intake [29]. Retrospective self-reports and laboratory studies fail to capture the temporal associations between rapidly unfolding phenomena like the experience of emotion and food choices in ecologically valid settings. As such, these methods pose limits on understanding under what circumstances first-year college students make particular food choices.

Ecological momentary assessment (EMA) is a data collection technique in which a participant’s thoughts, feelings, and behaviors are repeatedly assessed in real time or near-real time in the contexts which they are occurring [30, 31]. Thus, EMA mitigates concerns regarding retrospection-related biases and lack of ecological validity. Historically, EMA studies of health behaviors have used palmtop computers [13, 32‐34] and, more recently, mobile phones [31, 35, 36] allowing for self-reported assessment of health behaviors as they occur naturally in real life situations [35, 37, 38]. EMA via mobile phones (mEMA) appears to be a feasible and acceptable strategy for measuring physical activity behaviors [33, 35]. As yet, links between emotions and food choices have been assessed only in a limited fashion via mEMA methods. Given that first-year college students are vulnerable to weight gain and unhealthy lifestyle habits, our study aimed to use mEMA methods to determine associations among negative, positive, and apathetic emotions and food choices in first-year college students. Our specific hypotheses for this study were:

The results will provide an understanding of these relationships in the context of how emotions might be involved the healthy and unhealthy food choices that are occurring so that healthy food choices can be better promoted in this vulnerable population of emerging adults.

This study examined how emotions were associated with food choices in first-year college students. Using mEMA methodology, this study yielded tests of within- and between-person associations between emotions and food choices, which is unique in the literature. Negative and positive emotions had unique relationships with food choices; apathetic emotions were not associated with any food choices. Characterizing the relationships among different types of emotions and food choices is helpful in understanding the motivations behind healthy and unhealthy food choices. These findings can be used to guide effective interventions for promoting healthy food choices among first-year college students while setting the stage for future longitudinal research to examine how food choices and emotions impact health outcomes over time among emerging adults. Additionally, these findings set the stage for future research to examine how under-and over-eating of certain food choices is related to positive, negative, and apathetic emotions.

First-year college students experiencing negative emotions may be less motivated to make healthy food choices and more likely to make poor food choices [10, 18, 19, 26, 27]. Eating behaviors are not solely determined by physiological needs. Theoretical frameworks of eating behaviors include the affect regulation model (e.g., coping with negative affect by consuming and sometimes overeating, unhealthy foods [43‐45] or consuming healthy foods when experiencing higher, positive affect) [46] and the ego depletion model (e.g., in stressed situations, one is unable to exert self-control, thus consuming, and sometimes overeating, unhealthy foods) [47]. In the context of the current findings, more research is needed to understand the psychological contexts that predict eating choices. For example, the current study found that on occasions when participants reported negative emotions, they were more likely to report consuming meats/proteins. Food is often used as a distraction to draw one’s attention away from their negative emotions [8]. Given that meats/proteins tend to be high in fat and tend to be easily overconsumed [10], our finding provides a new ecological understanding of why first-year college students may be making unhealthy food choices during their first year on campus. These data can be used for future studies to examine how food choices and emotions are related to the development of overweight and obesity that occurs during the first year of college [5, 48‐50].

Our study found participants who reported positive emotions relatively more often than their peers were more likely to consume meats/proteins. Participants were also more likely to report the consumption of sweets, and less likely to consume pizza/fast food, on occasions when they reported positive emotions. The association of lower consumption of pizza/fast food is consistent with one previous retrospective study which found that more “healthful” foods and less “junk” foods were consumed during positive emotions [25]. However, the increased consumption of sweets was also positively associated with positive emotions which is not consistent with the retrospective study findings [25]. In focus groups, college students have said that making healthy food choices can be difficult because of time constraints, reliance on pre-cooked meals and unhealthy foods served on campus, and not knowing how to prepare healthy foods for themselves [51, 52]. Given that emotions are rarely examined in the context of healthy eating, more research is needed to replicate the current findings. If indeed emotions are related to healthy food choices, helping students to be more mindful of their emotions may promote overall greater nutritional status.

Apathetic emotions were not related to first-year college students’ food choices. Previous studies have reported that boredom may be a frequent reason for snacking in college students [53, 54]. Snacks often consumed by college students consist of chips, crackers, and nuts [53], which are generally salty and/or fried. Given that snacking contributes significantly to weight problems in college students [54] and because apathetic emotions and foods choices have never been assessed among this population using mEMA methodology, it will be imperative that we continue to assess apathetic emotions and food choices to confirm these novel findings, and use this new information to create interventions that will educate first-year college students on how to avoid and/or make better use of their apathetic emotional experiences. For example, because this study found that positive emotions are positively associated with choosing sweet foods, future research may consider investigating whether strategies to temporarily regulate affect to more neutral states would sometimes be beneficial to overall eating behaviors.

This study is the first to examine within and between person associations of positive, negative, and apathetic emotions and food choices. Interestingly, there were few similarities in statistically significant findings across the within and between person findings. We observe a few more within-person findings than between person findings, suggesting that the effects are acute, short-term, and transitory in nature (play out on a micro-timescale), particularly when reporting positive emotions in relation to food choices. To date, most studies examining emotions and eating behaviors have focused on differences between individuals [8‐23]. While these current findings need to be replicated, future research may consider the examination of why we see more variation at the intra-individual level. Additionally, these findings indicate why it is important to examine relationships at varying levels of analysis.

There are a number of strengths that should be taken into account when considering the findings of this study. This study is the first to examine the associations among negative, positive, and apathetic emotions and a relatively large variety of food choices. Using mEMA instead of traditional retrospective surveys allowed for multiple, repeated, real-time assessments within the participant’s natural environment. This reduced the risk of retrospective recall bias and ultimately provided ecologically-valid data for this study. The app was developed in both android and iPhone operating systems so that participants used their own mobile phones to complete the mEMAs. Additionally, the large sample size and racial/ethnic diversity of the sample was another major strength, enhancing the generalizability of the findings.

The observational study design used here allows for examination of how variables covary together over time, it does not allow for causal inferences. Additionally, the data came from a convenience sample of first-year college students at a single university; therefore, the findings may not generalize to other emerging adults who are not first-year college students or to students in other universities. Eating choices were examined as groups of food based on previous formative work and were designed to keep the assessments brief [39]. However, the specific foods and the quantity of food within each group that were consumed is unknown; as such, we are unable to examine under- and overeating behaviors as they are associated with food choices.

Additionally, it is possible that with certain response sets (i.e., individual response patterns), some participants may have been more diligent and willing to complete the entire food and emotion item set (thus reporting more food choices and emotions experienced), while other participants may have rushed through the assessments (thus reporting fewer food choices and emotions experienced). Given that positive emotions were associated with both healthy and unhealthy food choices, individual response patterns may account for the observed findings. The measures in this study were self-reported, which could result in recall bias; however, mEMA methodology may have minimized this bias.

Very little is known about how emotions are associated with food choices, especially in first-year college students. The findings of this secondary data analysis show that experiencing positive emotions more frequently than others is associated with an increase in consumption of meats/proteins between individuals. When participants experienced negative emotions, they were also significantly more likely report the consumption of meats/proteins; however, these results are significant for the within-person analyses only. When participants experienced positive emotions, they were significantly more likely to be associated with the report of the consumption of sweets and less likely to be associated with the report of the consumption of pizza/fast food in within person. Apathetic emotions were not associated with any of the food choices. The findings from this mEMA study add to the small body of literature that exists regarding emotions and food choices, providing a better understanding of the food choices that first-year college students make. Future research should continue to assess more specific food choices as well as how positive, negative, and apathetic emotions affect quantity of foods consumed. mEMA methodology provides a unique opportunity to examine these associations within- and between-people, providing insights for individual and population-level interventions. These findings can be used to develop and test interventions that encourage healthy food choices among first-year college students and ultimately reduce the risk of weight gain.