Overview

The Problem

For new students, faculty, and visitors, navigating an unfamiliar, complex campus environment can be a challenging experience. Even for existing and long-time students and staff, navigation may be hindered by temporary conditions caused by factors such as construction on campus, elevator outages, parking lot closures, etc. These temporary conditions present a more significant barrier for people with disabilities or temporary mobility impairments. The goal of this project was to design an accessible campus navigation system that provided necessary information for all pedestrians on the Georgia Tech campus.

Role

Lead UX Researcher, UX Designer

Duration

Aug 2022 - Dec 2022

Tools

Qualtrics, Figma, Miro

Team

Prachi Chopde, Karla Gil, Yuwei Li

The Process

This work was a part of a semester long project in one of my core courses at Georgia Tech titled ‘Research Methods for HCI.’ The goal of our process was to develop a user-centered, evidence-based design solution.

The Solution

Based on findings extrapolated from our research, we devised TechNAV. TechNav is a mobile pedestrian navigation app that improves pedestrian navigation for students, faculty, and visitors on the Georgia Tech campus by providing real-time data. TechNav provides the critical information needed to improve navigation at Georgia Tech, including information on path blockages, amenity locations, and indoor navigation. TechNAV emphasizes accessibility, allowing users to customize settings, such as step count and topographical information, to better suit their needs.

research

The goal with our research for this project was to gain a strong contextual understanding of our problem space, and understand the needs and pain points of our target user group.

The initial research for this project was divided into two phases. We began with a preliminary survey sent out to students at Georgia Tech to understand the scope and significance of the problem space. Our second phase of research consisted of to a variety of research methods aimed at gathering data at various granularities about our users and the problem domain. The research methods used are as follows:

Following the collection of data using the above methods, we consolidated our data into an affinity map, from which we were able to extrapolate several key research findings and insights. These research findings informed our creation of storyboards, journey maps, empathy maps, and design concepts.

Preliminary Survey

To understand the navigation barriers faced by pedestrians on campus, we distributed a survey across the Georgia Tech campus for preliminary research on the problem space. In this initial survey, we received 48 responses from Georgia Tech students. of 48 Georgia Tech students.

Based on the results of the survey, we identified the following problems:

The lack of real-time information about the accessibility of campus pathways, given the impact of temporary conditions such as construction, weather, crowd surges, events, etc. is a major concern as it obstructs travel around campus. These conditions present a larger barrier for people with disabilities or temporary mobility impairments, as research has shown that traveling, particularly in unfamiliar environments, is an arduous task for wheelchair users due to both interpersonal and environmental barriers
There is a lack of information about building access including details about entrance ramps, stairs, elevator outages, and student ID access requirements.

The results from our survey allowed us to identify key issues within the campus navigation problem space. Through our preliminary survey, we also came to recognize that problems faced by students at Georgia Tech, are also faced by other pedestrians on campus. We were thus able to identify our target user group.

Competitive Analysis

Based on our preliminary survey, we found that pedestrians on the Georgia Tech campus were most often relying on Google Maps, Apple Maps, the Georgia Tech Interactive Map, and Transloc (a bus tracking application) to get across campus. We, therefore, chose to conduct a competitive analysis with these four navigation tools to understand their strengths and weaknesses, and identify opportunities for improvement. From our analysis we learned:

Users need appropriate feedback on route progress and accurate travel times
Visual representations of buildings or landmarks are important for orientation during outdoor navigation
Voice assistance is a key accessibility feature needed to support navigation for all users
These applications lack real-time information for pedestrians, and are unable to account for temporary route blockages
Navigation through a college campus frequently involves in-building navigation. Current applications lack indoor navigation functionality.

Observation

We conducted an observation session on the Georgia Tech campus as one of our initial research methods. The general purpose was to observe the barriers students had indicated they face on campus in the preliminary survey. This included experiences with physical map kiosks as well as the lack of physical signage throughout campus. We aimed to use this knowledge to inform our question and theme construction for user interviews. From this session, we made three key observations.

Survey

The primary purpose of our survey was to garner a high-level understanding of our users’ navigational needs when traversing through the Georgia Tech campus. We used the data from our preliminary survey to our development of an additional survey, based on our understanding of our user’s pain points within the problem space. Before constructing the survey, we identified three main areas we wanted to cover and tailored our questions accordingly.

Our final survey in Qualtrics consisted of 15 questions. We distributed the survey across Georgia Tech channels/platforms, including Slack, Facebook groups, and Georgia Tech Reddit channels. We received a total of 148 responses from current students, faculty, alumni, and visitors of Georgia Tech.

Findings

51.3% of respondents rely on walking as their primary mode of transportation across campus
57.34% of respondents reported using Google Maps/Apple Maps on campus as their current choice of navigation tools.
50% of respondents reported being neutral or dissatisfied with their current navigation tool of choice
Respondents indicated they would largely use an on-campus navigation tool to find a location by name (e.g. 'HR Office'), find a parking lot, find a classroom within a building, and find meeting and collaboration rooms.
In an attempt to understand what navigation features people on campus value most, we asked participants to rate certain features on a usefulness scale. The findings are as follow:

"Sometimes I can't find the sign that denotes what building it is and sometimes when it's busy, I feel like it's hard to see signs and stuff because I'm short and people who are taller often block the signs."

"I walk everywhere, and the map app doesn't always show cut-throughs or sidewalks (not super pedestrian-friendly). There's also randomly construction blocking off some paths."

Contextual Inquiry

Following our survey, we chose to conduct a contextual inquiry with a current student at Georgia Tech. Through a contextual inquiry, we aimed to gain a clearer understanding of our users’ motivations, habits, and behaviors by observing a user in their natural context. We had our user walk us through a typical day navigating the Georgia Tech campus, and then provided a task-based scenario, asking the user to find a specific room in a building they were not familiar with. The questions we asked and observations we made aimed to understand the following:

What is a user’s thought process when navigating campus? More specifically, how does a user ultimately decide which route to take to reach their destination?
What navigation tools do users prefer? How do these preferences form?
How well do current navigation tools work in supporting a user’s navigation goal?
What are the strengths and weaknesses of these navigation tools?
How does a user overcome navigational barriers when confronted with them?
How might past experiences impact the route decisions a user currently makes?

Semi-Structured Interviews

Following the collection of qualitative and quantitative data from our survey, we chose to conduct semi-structured interviews to dive deeper into our users' attitudes, beliefs, and emotions towards navigation at Georgia Tech. By interviewing students of diverse backgrounds (i.e. undergraduates, graduate students, and visitors of varying abilities) we aimed to gain insights from the differences in opinions and behaviors.

In total, we recruited 4 participants - two graduate students, an undergraduate student, and a visitor - for 30 minute interviews. We asked questions in a semi-structured format, allowing for flexibility in asking follow-up questions as needed. Our questions surrounded individual users’ navigation preferences in terms of modes of transportation, preferences for navigation tools and their features, and behaviors and emotions towards campus navigation.

Data Analysis

To organize and understand the data collected from both user interviews and the contextual inquiry, we chose to create an affinity map. By visualizing the information hierarchy, affinity maps facilitate the generation of new insights and highlight emerging patterns and themes in the data. The process of collaboratively immersing ourselves in the data makes affinity modeling a strong tool to build design ideas off of each other and understand overarching themes in our data.

‍

With the 4 user interviews and 1 contextual inquiry we conducted, we were able to transcribe audio recordings and utilize notes to generate 108 interpretation notes. We chose to use Miro for this process. Over several hours, we then worked to group our interpretation notes based on the general topics and behaviors.

Based on the groupings in our affinity map and data collected from our survey, we generated the following insights.

Personas, Empathy Maps, and Storyboards

Based on the outcome of our research analysis, we synthesized and organized our findings into personas, empathy maps, and storyboards. We used these these tools as a way to empathize with and tell a story about our users, capturing their key characteristics, pain points, and expectations. In doing so, we gained a better sense of who we were designing for and how we can address their needs.

Personas

Empathy Maps

Storyboard

We then created a storyboard, choosing a research-backed user scenario to express users' pain points and needs in the form of storytelling. The protagonist of the story is a freshman at Georgia Tech who faces difficulty finding a particular meeting room on campus.

DESIGN

Our design process consisted of several iterations, including sketched concepts, wireframes, and our high-fidelity prototype. Following each phase, we conducted evaluation sessions with several real users to identify the strengths and weaknesses of each design and continuously improve our product.

Concept Sketches

To begin the ideation process, we translated our research findings into design implications. We then generated two preliminary sketched design concepts that aimed to address the problem statement by incorporating these implications. We conducted 45-minute concept sketch feedback sessions with four Georgia Tech graduate students and a Georgia Tech faculty member, specialized in accessibility. Our sessions aimed to address three goals:

Understand the perceived value of our designs and gauge our users' reactions towards our high-level features
Identify concerns about our designs and information users found was lacking
Gauge how accessible our designs would be for people of all ages and abilities.

Based on our feedback and findings from the concept sketch sessions, we chose to implement the mobile navigation app. It not only provided users with a method of receiving more accurate real-time data but also better addressed additional user needs such as indoor navigation and amenity location. We found that the mobile navigation app provided us with the flexibility needed to address the concerns and suggestions raised by our users, particularly those related to accessibility. The portability of a mobile application also made the navigation application more practical for users than a stationary kiosk.

Wireframes

With the concept sketch feedback in mind, we identified the features described as most useful by users across both sketches. These features were further supported by data collected and analyzed in our surveys and interviews.

Real-time information regarding navigation obstacles
Indoor navigation
Location of amenities across campus, including printers, microwaves, and restrooms
Images of destination

We also included additional accessibility features to better support navigation for students, faculty, and visitors of all ages and abilities. These features would also differentiate our application from current navigation tools on campus.

Step counts to support users with vision impairments
Conversational interface
Information regarding campus topography (slopes, uphill, downhill)

Following the development of our wireframes, we conducted wireframe feedback sessions with three users - an undergraduate student, faculty, and visitor. Our feedback sessions consisted of a task-based portion, where users were asked to guide us through how they would complete a series of tasks, followed by a semi-structured interview. Feedback received during our wireframe sessions informed the design of our high-fidelity prototype.‍

High-Fidelity Prototype

To develop our high-fidelity prototype, we reviewed the comments and suggestions from the user feedback sessions. Since we were building a navigation application for Georgia Tech, we wanted to keep the color scheme consistent with the Georgia Tech brand. We developed a design language that included colors inspired by the Georgia Tech brand, icons, and design elements to ensure consistency during the prototype development phase.

Evaluation

To evaluate the efficacy of our design and identify any usability issues, we conducted heuristic evaluations with 3 expert evaluators, and moderated usability tests with 5 users. Our users consisted of three students, a faculty member, and a visitor.

Heuristic Evaluation

The team chose to conduct heuristic evaluations as a way to detect usability issues when users interact with the product and identify ways to resolve them. As part of the heuristic evaluation process, experts tested the application prototype against Nielsen's 10 Usability Heuristics.

‍

As part of the evaluation, we had experts independently interact and familiarize themselves with the prototype using a think-aloud method. The team then asked experts to quantify their feedback based on the 10 heuristic principles on a 1-5 scale of severity.

1 = I don't agree that this is a usability problem at all

2 = Cosmetic problem only: need not be fixed unless extra time is available on project

3 = Minor usability problem: fixing this should be given low priority

‍ 4 = Major usability problem: important to fix, so should be given high priority

5 = Usability catastrophe: imperative to fix this before the product can be released

Our findings based on feedback provided during the think-aloud and the heuristic ratings, we found usability issues with the following heuristics:

‍

Moderated Usability Tests

Following our expert evaluations, we conducted usability tests with 5 users to identify challenges in use and opportunities for improvement. More specifically, we aimed to identify:

How easily users were able to complete a set of tasks relevant to user needs and goals identified in our research phase
If the application addressed our users’ needs
Design inconsistencies that made application use more challenging
Features users found most valuable or invaluable

Our usability test consisted of two phases: a task-completion portion, where users were asked to complete four tasks, followed by a semi-structured interview. During the interview we asked participants about their general experience with the application, including any challenges they faced or suggestions for improvement. As part of the task completion portion of the session, we recorded the number of errors users made for each task and the task completion rate.

Following the completion of each task, we asked participants to rate the difficulty level of this task on a scale of 1 to 5 where 1 is very easy, 2 is easy, 3 is neutral, 4 is difficult, and 5 is very difficult. Below are the participant ratings for each task:

Based on user feedback, we identified several themes emerging within the data. The findings represent all five users and their common reactions, suggestions for improvements, and errors in the prototype.

Design Recommendations

Based on the feedback received from our expert evaluations and user testing, we formulated several design recommendations that future iterations of our design should consider.

Change the image of the icons or adding descriptors to improve symbol recognition and establish meaning across the application
Turn 'ADA entrances' off by default to reduce the amount of information displayed on the home screen. Users will still be able to toggle this feature on in Settings.
Redesign the 'Buzzcard' access tag to remove the circle around it and change the color, as many users believed this was a button. Add descriptors or the word 'Buzz' to make it more easily recognizable.
Reduce uncertainty on the amenities page by making the icon denoting the number of amenities clearer (e.g. state '1 printer')
Change color of topography scale so it doesn't conflict with users expectations of what 'red' and 'blue' means on a mobile navigation app.
Make the 'Add' button more prominent on the reports page
Make the description field of a new report optional

Reflection

Lessons Learned

As my first research-focused project at Georgia Tech, I found the process of researching to design to be very rewarding. Along the way, I learned a couple of key lessons.

The importance of framing the problem - In the early stages of the project, our focus lay on the topic of 'Accessibility' on the Georgia Tech campus. We found that the term 'campus accessibility' was too broad, and led to confusion as a team on what our problem focus was. We found as a team it was difficult to articulate the problem we were attempting to solve with just this phrase, and struggled to find relevant literature. In reframing the problem to center around campus navigation, while still including accessibility at the forefront, we were able to gain a clear understanding of the problem space.

‍The value of diverse research methods - During this project, we were able to use 5 research methods that collected data at various granularities. In doing so, I learned the strengths and weaknesses that each method provides, and how data collected from one method can be enhanced with data rom another method. Utilizing a variety of research methods is key in gathering rich data and gaining a clear understanding of the user group.

Feedback is key - Throughout the process, we received feedback from both users and experts. Expert feedback helped us in re-framing our problem space, while user feedback helped in our iterations from concept sketches to the final prototype.