Socio-technical Journeys: Using Digital Narratives for Managerial Decision-Making 

This project explored how major life-transition journeys shared on socio-technical systems, such as blogs, are valuable first-hand, rich data for understanding users' behaviors, emotions, and opinions throughout their journeys, and how systematic analysis of these crowdsourced data at scale can promote community support and inform data-driven management decision-making.

I introduced the notion of socio-technical journeys and led a multi-phase, mixed-methods workstream to enhance sustainable managerial decision-making within the long-distance hiking community.

Background and Goals

We have some key goals in order to guide our user-centered research.

Background

1. Individuals undergoing major life transitions often share their emotional experiences through socio-technical systems, such as community-centered blogs, for various reasons.

2. Emotional well-being plays a central role in shaping how individuals navigate life’s changes. It helps them maintain resilience, stay focused on their goals, and cope with uncertainty.

3. Appalachian Trail management requires ​the systematic integration of digital technologies and hikers' experiences to support sustainable management.

4. Trail management staff's digital literacy and objectives influence their technology adoption to provide community support.

Generative Research Goals:

Research Methods and Process

1. How can amorphous data from socio-technical systems be refined to craft a dataset useful for understanding long-distance hiking as a socio-technical journey?

2. How can we characterize factors that contribute to individuals' emotional states in a dynamic socio-technical journey, such as long-distance hiking, at the community scale?

3. How do trail management objectives and perspectives on the LD hiking experience shape the ways in which crowd-sourced hikers' information, shared in socio-technical systems, is integrated into management practices?

4. How can technology-facilitated records of LD hikers’ experiences support cooperative and sustainable trail management decision-making?​​​

​

Data Collection

• I developed and deployed a cross-platform dataset of 240K+ anonymized long-distance hiking records, combining user-generated journals, location data, timestamps, and mileage with weather metadata (temperature, humidity, rainfall) to support large-scale behavioral analysis.​​​

​​

Multi-phased Mixed-methods Research

1. Context-Understanding Workshops

   • Conducted participatory workshops with the Appalachian Long Distance Hikers Association (ALDHA) to understand the research context and build trust with the hiking community. Sessions included Collaborative App Design workshops (N=17, 2022) and Digital Trail Data discussions (N=6, 2023) to explore how hikers generate and interpret trail data shared on socio-technical systems.

   • Methods: Workshop, Journey Mapping, Storyboards, participatory design, quantitative data analysis, emotional modeling

   • My Role: Co-facilitated workshops and presented research insights while collaborating with both research and executive team members.

2. Emotional Journey Modeling​

   • Applied machine learning, topic modeling, and emotion analysis to examine hikers’ journal entries, extracting contextual insights to identify themes and emotional patterns throughout long-distance journeys. I also analyzed the relationship between emotions and weather conditions to understand how environmental factors affect hikers’ experiences and identify stages at which hikers may need additional community support. I also provided design recommendations for technology-mediated reflective tools.

   • Methods and Tools: Machine learning, topic modeling, natural language processing, emotional modeling, statistical analysis, Thematic Analysis

   • My Role: Led the data modeling and analysis to uncover patterns in hikers’ emotional journeys.

3. Technology Probe Development​

   • Developed a Digital Analytics Dashboard as a technology probe to transform crowdsourced hikers' data into actionable insights for trail management stakeholders. The dashboard visualized hiking activity and combined emotional insights with journey-related factors such as communal experiences, overnight stays, and hikers’ movement patterns, enabling multi-aspect analysis. It was used to facilitate discussions with stakeholders, helping them explore how digital trail data could support research goals and inform trail management decisions.

   • Tech: Python, Streamlit, Dashboard, Visual Analytics, Classification models, Spatio-temporal analysis 

   • My Role: Tech lead for the project and mentor to a team of five undergraduate computer science students, guiding the design and development of the dashboard. 

4. Stakeholder Interviews​​

   • Conducted 19 semi-structured interviews with trail and visitor-use management professionals, including trail managers, stewards, and visitor center managers. Participants were recruited through snowball sampling across professional networks to capture diverse perspectives on trail management and the use of digital trail data.  I then synthesized the interviews using thematic analysis to generate actionable insights.

     • The technology probe (Digital Analytics Dashboard) was used during interviews to prompt discussion, gather feedback, and explore how data-driven tools could support stakeholder decision-making.

   • Methods: Semi-structured interviews, usability testing, concept testing, mixed-methods research

   • My Role: Led the research process end-to-end, including participant recruitment, interview facilitation, engaging stakeholders using the technology probe, and synthesizing insights.

​

Findings and Crucuial Insights

• Identified key factors shaping hikers’ emotions throughout the long-distance journey
  Analysis of hikers’ journals revealed recurring themes across six core emotional states (happiness, sadness, anger, disgust, fear, and surprise), showing that long-distance hiking is an emotional rollercoaster!

​​

• Discovered environmental, cultural, and social factors shaping hikers’ emotions
  Combining hikers’ narratives with contextual and environmental data (e.g., weather, social interactions) revealed types of relationships among these factors and emotional experiences along the trail.

​

• Showed that crowdsourced user-generated data can support management decisions
  Combining emotional insights from hikers with movement patterns and environmental data demonstrated how digital narratives can help managers better understand hikers’ experiences and needs.​

​

• Validated the usefulness of visual analytics tools for non-profit organizations
  The digital analytics dashboard helped trail managers interpret unstructured hikers' digital data more easily and facilitated discussions about how data-driven insights could inform trail management and visitor support strategies.

​

• Revealed gaps between hikers’ lived experiences and trail management frameworks
  Trail management policies often rely on zoning plans and standardized frameworks, but interviews showed that hikers’ experiences vary widely across trail sections and conditions. Managers must balance maintaining trail sustainability with supporting diverse hiking experiences.

​

• Identified barriers to integrating crowdsourced hiker data into management decisions
  Trail managers face challenges using crowdsourced data due to fragmented data ecosystems, weak data-sharing pipelines, and concerns about reliability and representativeness. As a result, much of the data currently used for decisions still comes from formal surveys rather than social media platforms.

​

• Demonstrated how visual analytics can support collaborative, data-driven trail management
  The analytics dashboard helped managers interpret complex hiking data and explore patterns in hikers’ experiences. Stakeholders noted that such tools could support planning, communication, collaboration, and negotiation when making trail management decisions.