Stress Eating Study, Part 2: From Learning to Doing

Welcome to phase two of the semester-long journey where I document my team’s research and design process as part of our core classes at Georgia Tech’s MS-HCI program. Our goal for the semester is to create a solution that allows college students to manage stress-eating.

To learn more about how we identified the problem space in phase 1, please read: https://medium.com/georgia-tech-mshci/stress-eating-study-part-1-the-cortisol-induced-food-frenzy-96e4e9e2b66a

The literature reviews are done. The problem space has been identified. Next stop: the enigmatic world of data collection.

Data Collection

Having pored over existing literature about stress among college students and its effects on eating behavior, we needed to gather information about our target population (college students) to better understand their needs, goals, and pain points when it came to the problem of stress eating.

To start our data collection, we considered observations, focus groups, surveys, and interviews.

Observations were ruled out as we could not observe people’s eating habits in person due to the pandemic. Focus groups were not feasible because people are usually conscious about their eating habits and may not reveal their eating habits authentically in a group setting. We then considered interviews but decided that those would be optimal after we had a firmer grasp on trends about stressors and eating behavior among college students. Ultimately, in the competition for the best data collection method to begin our data collection, surveys came out on top.

Surveys

First, surveys would provide us with clear, quantitative data about students’ eating habits and coping strategies, allowing us to identify trends and patterns clearly. Second, due to the relative ease of distributing and analyzing surveys, we would be able to collect data from a larger number of target users, giving us a set of data that is reliable. To stay consistent with our target group, only undergraduate and graduate students were eligible to participate in the survey.

Through the survey, we wanted to gather answers to the following research questions:

Question 1: Does stress change students’ eating habits in any way?

Question 2: In which settings do students stress-eat the most?

Question 3: How do students relieve stress?

Survey results

Finding 1: 20 participants out of the 50 who responded to this question said that stress definitely changes their eating habits while 25 participants said that stress maybe changes their eating habits.

Finding 2: An overwhelming majority of the participants (60 out of 74) responded to experiencing academic stress.

Finding 3: Students relieve stress in many different ways, some popular ones being working out, calling a loved one, and streaming movies or TV shows.

Interestingly, another popular response was “eating comfort foods”. The emergence of this category as a popular stress-relief activity validated the scope and importance of our problem space and the need to create a solution.

Definition taken from the Nielsen Norman Group:https://www.nngroup.com/articles/diary-studies/

Diary Study

Why we decided to do a diary study:

Though we were able to gather a lot of data from our survey, we wanted to delve deeper into the daily habits, stressors, and stress relief methods of our target users. Since we wanted highly contextual data that varies from day-to-day, we decided to conduct a diary study. We aimed to follow participants’ eating habits, mood, and stress levels over the course of seven days. Our research questions for the diary study were:

Question 1: Are participants snacking due to stress?

Question 2: Are participants willing to perform stress-alleviation activities to relieve their stress?

To design the diary study, we studied the initial survey data to determine which activities users reported as being most effective for alleviating stress. Through the initial survey data, the activities that people reported as being most effective for stress-alleviation were physical activity, talking to a friend or loved one, a short guided meditation session, playing a game, and watching TV.

Our diary study protocol then assigned the aforementioned activities to each participant for a duration of two days per task and randomized the order in which they received the activity. We also asked participants about their meals and snacks for each day, and for each snack, we asked participants why they consumed that snack. This part of the study was inspired by a 2017 study where researchers conducted a diary study to understand emotional eating (Vandewalle et al., 2017).

Overall, we were able to collect seven days worth of data from six participants, five days' worth of data from two participants, and one participant did not end up filling out the diary study at all.

Diary Study Results

Finding 1: Participants snack for different reasons, one of which is stress; however, participants cannot always accurately identify when they are eating due to stress.

Throughout our diary study, participants cited multiple reasons for consuming a particular snack. While most participants somewhat agreed that they consumed a snack due to stress, many had difficulty accurately identifying why they ate a snack as they couldn’t always pinpoint whether it was due to hunger, loneliness, stress, or boredom. We found that for approximately 64% of the snacks consumed, the same participant chose multiple reasons for consuming the same snack.

Since participants cannot accurately identify if they are eating due to stress, we realized the need for an external physiological measure for stress-detection.

Finding 2: Participants are willing to perform stress-alleviation activities if there is a chance that they may help relieve stress.

All eight participants who were part of the diary study expressed a willingness to perform the interventions when asked about in the screening survey. Out of the six participants who completed all seven days of the study, on average, participants successfully performed the intervention that was assigned to them.

Reflection

While our research methods yielded adequate findings that allowed us to answer all of our research questions, our chosen methods were not without their limitations. While the diary study was designed and distributed with the intention of capturing contextual data about students’ stress-eating behavior, the fact that we had few participants and limited resources to carry out the diary study reduces our confidence in our findings.

1. We realized that people have complex eating patterns which may not be reflected accurately in a week’s food tracking.
2. We realized that people are often not aware of why they ate a particular food item, and they often attributed eating the same snack to many different mood states. This made drawing clear inferences from the data extremely challenging.
3. Finally, due to the pandemic, many of our participants reported changes to their lifestyle. Some participants reported moving back in with their parents, which may have caused changes to their usual eating behaviors.

Look out for Part 3 where our team establishs design requirements and comes up with divergent design ideas!