Summary

This project investigates how digital mental health tools exclude blind and low-vision users and what inclusive alternatives should look like. As lead researcher, I ran an explanatory sequential mixed-methods study — a survey of 93 BLV respondents (analyzed with Kruskal-Wallis tests) to establish where experiences diverged, followed by 10 semi-structured interviews to explain why — and turned the combined evidence into recommendations teams can use to design accessible DMH products.

Objective

Digital mental health apps (for mood tracking, meditation, journaling, and self-care) are increasingly common, but many rely on visual interfaces, charts, and interactions that are inaccessible to blind and low-vision users. As one participant put it, "sighted people have their pick of the litter." The goal was to define the problem space and research questions so that design and product teams could build more inclusive DMH tracking services.

Research questions

1. What do BLV individuals need from digital mental health tracking services?
2. What barriers do they face with existing DMH tools, and how can research and design better include them?

Work

I ran the study with and for the blind community in two connected phases:

1. Quantitative phase. A survey of 93 BLV respondents about their use of and barriers with DMH tracking tools. I used Kruskal-Wallis tests to identify where responses differed significantly across groups, which pinpointed the barriers worth explaining in depth.
2. Qualitative phase. 10 semi-structured interviews, sampled to follow up on the survey signals, to capture lived experiences and the "why" behind the numbers, analyzed with reflexive thematic analysis.

Methods: Explanatory sequential mixed methods; survey design and non-parametric analysis (Kruskal-Wallis); semi-structured interviews and reflexive thematic analysis; integration of quant and qual strands into evidence-based recommendations for product and design. Collaboration with academic advisor and community stakeholders throughout.

End result

Insights & recommendations

The survey established statistically distinct patterns in how BLV users experience DMH tools, and the interviews explained them — pointing to concrete design priorities like screen reader support, non-visual feedback for tracking and progress, and reduced reliance on charts. Findings were published as a CHI EA '25 late-breaking work and a fuller ASSETS 2026 paper.

Impact

Pairing quantitative signal with qualitative explanation gives product and design teams both the "how many / how much" and the "why" they need to prioritize accessibility work in an underserved domain. The study also demonstrates independent mixed-methods capability — survey design, non-parametric statistics, and qualitative integration — not qualitative research alone.

Reflection

Leading this project underscored how much product teams assume a sighted user by default. If I were to do it again, I would run a short diary study alongside interviews to capture in-the-moment barriers and pair the work with a lightweight design sprint to turn top recommendations into concrete design concepts for stakeholders.

Research skills demonstrated

• Explanatory sequential mixed-methods study design
• Survey design and quantitative analysis (non-parametric / Kruskal-Wallis)
• Semi-structured interviews and reflexive thematic analysis
• Integration of quantitative and qualitative strands into evidence-based recommendations
• Stakeholder collaboration throughout

Resources

• Full bibliography: [P3] CHI EA 2025 · [C6] ASSETS 2026 (to appear)
• CHI EA '25: Unpacking the need for DMH tracking services with and for the blind community [DOI]
• arXiv preprint

Methods

Explanatory sequential mixed methods: survey design and non-parametric analysis (Kruskal-Wallis), semi-structured interviews and reflexive thematic analysis, integration of quantitative and qualitative findings into product and design recommendations, stakeholder collaboration.