Unhealthy Comparisons to Promote Healthy Behavior? Exploring the Impact of Social Comparison Strategies on Well-Being within Health &Wellness Applications.

Abstract

Social comparison methods are frequently used in health & wellness applications despite concerns about their impact on users' well-being. These comparisons can lead to unfavorable self-perceptions and expectations, resulting in lower self-confidence. Research primarily focuses on the direct effects on motivation, user engagement, and behavior change, often neglecting other feelings. Therefore, my thesis aimed to investigate the effect of social comparison strategies in health & wellness apps. Since existing literature did not clearly depict how social comparisons are implemented, I reviewed commercial apps (n = 42) to create an overview. Metrics related to exercise, sleep, nutrition, anthropometric health, physiological health, and cognitive health are mainly integrated into social comparison strategies. The strategies include sharing data with others, insights into others' data, comparisons to combined data, competitive comparisons, and reminders to compare.

Interviews with users (n = 12) revealed that while social comparisons often lead to motivation, they can also result in negative feelings like inferiority, insecurity, and disappointment. In some cases, this led to longer-term effects such as preoccupations, overtraining, and injuries, reducing participants' well-being. The experiences depend on multiple factors, including personal aspects and comparison elements. The strategy, topic, and the persons with whom users are compared play a role. A vignette study (n = 192) further explored the effect of different comparison elements on positive and negative affect using the Positive And Negative Affect Schedule (PANAS). The study tested differences between comparison topics (step count and body fat), comparison strategy implementations (placement in a distribution, sharing and insights into feed messages, and placement on a leaderboard), and the comparison others (better than the user, similar to the user, or worse than the user).

The results show that not every combination of social comparison elements is suitable in health & wellness applications. Some health metrics are more sensitive within the context of social comparisons than others. Body fat comparisons consistently caused more negative feelings than step count comparisons, especially when comparing with others who were similar to or better than the user. The comparison strategy also influences this effect, with leaderboards having the most significant impact on users' feelings.

This work demonstrates the need for careful implementation of social comparisons in health & wellness applications. I have formulated design guidelines for app developers to support them in this task. However, further research is needed to determine the suitability and sensitivity of various metrics; the influence of personal characteristics remains largely unexplored. This exploratory work underscores the necessity for further research in this area.