Designing for Trust: The Role of Scientific Cues in Health App UI/UX
During my UI/UX design internship at DTT, I conducted a research study to investigate how incorporating scientific cues in a mobile health app description impacts user purchase intention. The goal was to understand whether evidence-based language could enhance user trust and increase app subscriptions. Using an A/B testing approach in a controlled experimental setting, I tested two different app descriptions—one with scientific claims and one without—to measure their influence on user behaviour.
Project Type
Research
Timeline
2024
Tools
Figma, SPSS, Qualtrics
My Role
UI/UX Researcher / Designer
Problem & Research Question
PROBLEM
Many health apps struggle to convert free users into paying subscribers. Research suggests that credibility cues (such as scientific evidence) can influence user trust and decision-making, but their effectiveness in a mobile health app context was unclear.
RQ1: To what extent do scientific cues (present vs. absent) in a health app description influence users’ purchase intention?
RQ2: Does a user’s personal interest in scientific information moderate this effect?
Methodology
STUDY DESIGN
I designed a between-subjects experiment (A/B test) to compare two versions of a mobile health app description:
Version A (Scientific Cues Present): Included claims like “Listening to Sleepo white noise can help you fall asleep 47% faster, backed by scientific research.”
Version B (No Scientific Cues): Contained general statements like “Good choice to use Sleepo to aid your sleep.”
PARTICIPANTS & DATA COLLECTION
• 124 participants recruited.
• Survey conducted via Qualtrics with randomized assignment to Version A or B.
• Measured variables: Purchase intention (via Likert scale) and interest in scientific information (moderator).
Used two-way ANOVA to determine the effects of scientific cues on purchase intention and whether user interest moderated this effect.
PARTICIPANTS & DATA COLLECTION
Results
MAIN & INTERCATION EFFECTS
A two-way analysis of variance was conducted on the influence of two independent variables (presence of scientific cues, interest in scientific information) on the purchase intention of the app subscription. Presence of scientific cues included two levels (present, not present) and interest in scientific information included two levels each (high – above the mean of means, low – below the mean of means). Only the effect for the presence of scientific cues was significant at .05 significance level F (3, 120) = 8.3, p < .005, indicating a significant difference between the two conditions - presence of scientific cues (M = 4.09, SD = 1.5) and no presence of scientific cues (M = 2.73, SD = 1.7) under low-interest condition and presence of scientific cues (M = 3.26, SD = 1.8) and no presence of scientific cues (M = 2.56, SD = 1.8) under high-interest condition. The main effect for interest in scientific information yielded F ratio of F (3, 120) = 1.29, p = .257, indicating an insignificant difference between higher interest in scientific information (M = 2.56, SD = 1.8) and lower interest in scientific information (M = 2.73, SD = 1.7).
The interaction effect of interest in scientific information on the relationship between the presence of scientific cues and purchase intention of the app subscription was insignificant F (3, 120) = .44, p = 506. In this case, support was found for H1; therefore, the null hypothesis for H1 can be rejected. This indicates that the presence of scientific cues in the health app interface leads to higher purchase intention of the app subscription for mobile app consumers compared to the non-presence of scientific cues. However, no support was found for H2; therefore, the H2 null hypothesis cannot be rejected. This indicates that the main effect of scientific cues in the health app interface on the purchase intention of the app subscription did not significantly increase among the users with a high interest in scientific information compared to users with a low interest in scientific information.
Key Findings
✅ Scientific Cues Significantly Increased Purchase Intention
• Users exposed to scientific cues were more likely to subscribe compared to those who saw a non-scientific description (p < .005).
• This suggests that integrating evidence-based claims in UI text can boost perceived credibility and user trust.
❌ User Interest in Science Did NOT Moderate the Effect
• Surprisingly, users’ pre-existing interest in scientific information did not significantly influence their purchase decision.
• This suggests that all users, regardless of their interest in science, were positively influenced by scientific cues in the app UI.
Reflections
🔹 Data-Driven Design Decisions Matter
This project reinforced the importance of using research and experimentation to inform design. Instead of relying on assumptions, I applied A/B testing and statistical analysis to validate how UI copy influences user behaviour.
🔹 The Power of UX Writing & Microcopy
Small details, like the presence of scientific cues, can significantly impact user perception and decision-making. Thoughtful UX writing can boost credibility and engagement.
🔹 Balancing Trust & Persuasion in UI/UX
Designing for trust isn’t just about aesthetics—it’s also about how we communicate value. Scientific cues increased purchase intention, proving that evidence-based messaging can be a powerful UX tool.
🔹 Experimentation Enhances UX Strategy
This project showed me how testing different variations of a design can lead to more informed, strategic decisions. Leveraging research-backed insights allows me to make design choices that drive real impact.