Mixed methods research, Participant recruitment, UX Research principles and Experimental study designReporting and presentation.
Mixed methods research, Participant recruitment, UX Research principles and Experimental study designReporting and presentation.
Mixed methods research, Participant recruitment, UX Research principles and Experimental study designReporting and presentation.
UI DESIGN: THE FINAL DELIVERABLE
The chat screens.
UI DESIGN: THE FINAL DELIVERABLE
Home screen refinement.
UI DESIGN: THE FINAL DELIVERABLE
After all the design iterations, these are the final designs.
UI DESIGN: THE FINAL DELIVERABLE
After all the design iterations, these are the final designs.
UI DESIGN: THE FINAL DELIVERABLE
After all the design iterations, these are the final designs.
UI DESIGN: THE FINAL DELIVERABLE
After all the design iterations, these are the final designs.
UI DESIGN: THE FINAL DELIVERABLE
Home screen, Main pages.
UI DESIGN: THE FINAL DELIVERABLE
Home screen, Main pages.
UI DESIGN: THE FINAL DELIVERABLE
Home screen, Main pages.
Role of Data visualizations and user control over health data in fostering user' trust in health information delivered by health chatbots
Role of Data visualizations and user control over health data in fostering user' trust in health information delivered by health chatbots
Role of Data visualizations and user control over health data in fostering user' trust in health information delivered by health chatbots
Role of Data visualizations and user control over health data in fostering user' trust in health information delivered by health chatbots
Role of Data visualizations and user control over health data in fostering user' trust in health information delivered by health chatbots
Miro Link: You can find the coding and data analysis here.
Miro Link: You can find the coding and data analysis here.
Miro Link: You can find the coding and data analysis here.
Notion file Link: You can find detailed information here.
Notion file Link: You can find detailed information here.
Notion file Link: You can find detailed information here.
UI DESIGN: VERSION 1
The insights
gathered from the participants and the Nursing stakeholders were useful in creating more intuitive designs for the users helping us understand the needs better and making the required design iterations.
This project explores whether visualizations of health information can build trust in chatbots for delivering medical recommendations. By comparing visual vs text-only representations, I aim to identify which approach fosters more confidence. This study addresses the trust gap in healthcare chatbots.
This project explores whether visualizations of health information can build trust in chatbots for delivering medical recommendations. By comparing visual vs text-only representations, I aim to identify which approach fosters more confidence. This study addresses the trust gap in healthcare chatbots.
This project explores whether visualizations of health information can build trust in chatbots for delivering medical recommendations. By comparing visual vs text-only representations, I aim to identify which approach fosters more confidence. This study addresses the trust gap in healthcare chatbots.
THE TEAM
UX Designer / UX Researcher (me)
Dr. Hidy Kong
Nursing Faculty (University of Washington)
4 Developers
THE TEAM
UX Designer / UX Researcher (me)
Dr. Hidy Kong
Nursing Faculty (University of Washington)
4 Developers
THE TEAM
UX Designer / UX Researcher (me)
Dr. Hidy Kong
Nursing Faculty (University of Washington)
4 Developers
MY ROLE
Preliminary Research, Literature survey, Prototype redesign, Recruitment, Experimental study, Usability testing, Qualitative and Quantitative analysis.
MY ROLE
Preliminary Research, Literature survey, Prototype redesign, Recruitment, Experimental study, Usability testing, Qualitative and Quantitative analysis.
MY ROLE
Preliminary Research, Literature survey, Prototype redesign, Recruitment, Experimental study, Usability testing, Qualitative and Quantitative analysis.
TIMELINE
August 2023 - Present
TIMELINE
August 2023 - Present
TIMELINE
August 2023 - Present
THE PROBLEM STATEMENT
Many users are reluctant to trust chatbots for medical recommendations due to concerns about their trustworthiness. This project aims to explore whether visualizing health information can help build trust in users, particularly caregivers, and improve the acceptance of chatbot-delivered medical advice.
THE PROBLEM STATEMENT
Many users are reluctant to trust chatbots for medical recommendations due to concerns about their trustworthiness. This project aims to explore whether visualizing health information can help build trust in users, particularly caregivers, and improve the acceptance of chatbot-delivered medical advice.
THE PROBLEM STATEMENT
Many users are reluctant to trust chatbots for medical recommendations due to concerns about their trustworthiness. This project aims to explore whether visualizing health information can help build trust in users, particularly caregivers, and improve the acceptance of chatbot-delivered medical advice.
RESEARCH QUESTIONS TO BE ANSWERED THROUGH THIS STUDY ARE
RQ1. How does the presence of visualization of health information in a chatbot affect users' trust in information in the chatbot?
RQ2. How can user control over the visualization of personal data help build trust in health information in health chatbots?
GOALS AND MOTIVATION
COCO is a healthcare app that offers caregivers personalized medical support based on user input and real-time data from wearables like Fitbit. The chatbot uses visualizations to enhance communication, transparency, and comprehension. This project aims to examine how visualizations and user control over data impact trust in the chatbot's health recommendations. The goal is to explore if visualizing information improves user trust and acceptance of health chatbots in healthcare.
RESEARCH OBJECTIVES
The main aim of this research is to find out if there is a possible relationship between visualized health data and user trust. It will also focus on providing user controls over personal health information in visualizations and it will focus on how providing user controls can affect trust in chatbots. This is done to improve the usability of chatbots for health recommendations.
RESEARCH PROCESS
To kickstart the research process, I
Read and analyzed 100+ research papers, industry reports and trends.
Skimmed through prior research in the field that helped in understanding existing knowledge.
Identified gaps, trends, and informed research questions.
STAGES OF TESTING
User Interface Design: Wireframing and Prototyping.
Test Plan
Recruitment
Pilot testing
Usability studies
Data Analysis
STAGES OF TESTING
User Interface Design: Wireframing and Prototyping.
Test Plan
Recruitment
Pilot testing
Usability studies
Data Analysis
STAGES OF TESTING: STEP 1 : USER INTERFACE DESIGN
Along with the design team at the University of Washington, we designed the user interface of the COCO app. Considering all the needs of the users the app was designed to be more user friendly and intuitive.
STAGES OF TESTING: STEP 2: TEST PLAN
Along with my advisor, I came up with a study plan to test out the theory that visualizations help in increasing trust in health chatbots.
To begin with, the study took into account the visualization literacy of the participants to keep check of the bias. To test this, I took the help of Mini-VLAT which is a short-form Visualization Literacy Assessment Test.
Following this, the participants had to fill out a pre-study survey which included questions about their demographic information and their concerns regarding data visualizations.
The pre-study was followed by the participants interacting with the COCO prototype. The participants were divided into 2 groups. One group was tested with Text first, then Visualization and the other was presented with visualization first, and text later.
The last part of the study was a follow-up Interview. This interview helped in gathering the user's perspective about the UI, the text, and the visual representations of their health data (See Notion file attached at the end of the project for more details)
STAGES OF TESTING: STEP 3: RECRUITMENT
For the participant recruitment, the Nursing faculty and I sent out emails inviting people to participate in our study (more details are included in the notion file attached at the end of the project)
STAGES OF TESTING: STEP 4: PILOT STUDY
Before diving into the testing of the chatbot with the caregivers, I conducted 3 Pilot tests to time the test sessions and they came up to 50 minutes on average.
STAGES OF TESTING: STEP 5: USABILITY STUDIES
I conducted 8 user testing sessions. The participants were divided into 2 equal groups. Group 1 was Text first, Visualization second, and the rest Visualization first and then Text. This helped me understand how the perspective of the participants changed when they saw visualizations first and text later and vice versa.
STAGES OF TESTING: STEP 2: TEST PLAN: THE DATA VISUALIZATIONS USED IN THE PROTOTYPE.
These are the visualizations used in the test. One of them is the sleep graph and the other is a symptom graph.
UI DESIGN: VERSION 1
Here is how the initial version of the app looked like. We tested the first version and got feedback from the participants as well as stakeholders.
UI DESIGN: THE FINAL DELIVERABLE
Taking into account all the insights, we iterated on the designs.
QUALITATIVE DATA AND ANALYSIS
The interviews were audio recorded. I transcribed the recordings and cleaned them to extract quotes and code the data. This stage is crucial and helped in getting insights from the interviews.
UI DESIGN: VERSION 1
The insights
gathered from the participants and the Nursing stakeholders were useful in creating more intuitive designs for the users helping us understand the needs better and making the required design iterations.
IN PROGRESS:
This study served as a pilot study to test with more participants. The next study will be done with 40 caregivers and it will be a longitudinal study where the app will be given to the users to use for 2 weeks.
IN PROGRESS:
This study served as a pilot study to test with more participants. The next study will be done with 40 caregivers and it will be a longitudinal study where the app will be given to the users to use for 2 weeks.
IN PROGRESS:
This study served as a pilot study to test with more participants. The next study will be done with 40 caregivers and it will be a longitudinal study where the app will be given to the users to use for 2 weeks.
IN PROGRESS:
This study served as a pilot study to test with more participants. The next study will be done with 40 caregivers and it will be a longitudinal study where the app will be given to the users to use for 2 weeks.
PIGEONS WORK, BUT
THIS WAY IT'S QUICKER
Here are 5 reasons why I’m your perfect next hire.
#designedwithLoveandGoodVibes
PIGEONS WORK, BUT
THIS WAY IT'S QUICKER
Here are 5 reasons why I'm your Pefect next hire.
#designedwithLoveandGoodVibes
PIGEONS WORK, BUT THIS WAY IT'S QUICKER
Here are 5 reasons why I'm your perfect next hire.
#designedwithLoveandGoodVibes
PIGEONS WORK, BUT
THIS WAY IT'S QUICKER
Here are 5 reasons why I'm your Pefect next hire.
#designedwithLoveandGoodVibes
PIGEONS WORK, BUT THIS WAY IT'S QUICKER
Here are 5 reasons why I'm your perfect next hire.
#designedwithLoveandGoodVibes
PIGEONS WORK, BUT THIS WAY IT'S QUICKER
Here are 5 reasons why I'm your perfect next hire.
#designedwithLoveandGoodVibes
