Tailored Education

Measuring the Return on Infrastructural Investment in Student Learning

About TailorEd

Throughout the 1990s and 2000s, active learning began to attract increasing attention in STEM fields in response to demands from industry to produce more STEM graduates by - among other things - improving recruitment and retention in STEM undergraduate programs. Funded by large grants from the NSF and other agencies, STEM faculty began large controlled experiments testing the impact of specific active learning methods like demonstration predictions, think-pair-share, minute papers, strip sequences, student-generated test questions, et cetera in entry-level courses in disciplines like physics, biology, and math, which flip the classroom by replacing traditional lectures with student activities which test and apply students' growing understanding of course content.

This study used automated aerial photography to document the instructional formats used in 74 courses taught in humanities, social sciences, engineering, and business disciplines in 12 active learning and 2 traditional (control) classrooms. We used the machine learning tools available in Matlab to analyze the photos to identify instances of lecture format, large group discussion, small group discussion, approximating different active learning formats.

My Role

While this project was collecting data to feed into the machine learning image recognition algorithm, an online tool was needed so that professors could access this data.

For this project, I took on the role of UX Researcher and UX Designer . I conducted user research using the professors at our school while looking at other similar projects that were available at other colleges and universities. I also defined the product with deliverables such as a user personas, while ideating designs, conducting usability tests, and applying revisions.

Research

As part of the competitor or landscape analysis, we researched similar projects and sites to understand where TailorEd stands in the market and to see the strengths and weaknesses of our competitors. We looked at a variety of factors including the purpose of the tool, cost to use, target audience and more.

	UBC Grades	Atlas	Berkeleytime
Cost	Free	Free for Students, Staff & Faculty	Free
Target Audience	Professors	Professors & Students	Students
Purpose of Tool	To view and visualize historic grade distribution of courses at the University of British Columbia Vancouver and Okanagan, dsplaying grade reports in a more useable manner than what is officially provided by UBC and reporting older grade data that has been removed from public access	To display historic data about curriculum to guide University of Michigan students, instructors, and staff in decision making	To help students view and compare grade distributions for each course and semester and to track accurate, real-time course enrollment trends and history

User Interviewing

We interviewed a total of ten professors at SCU to better understand how they design their course curriculum and how useful they would find a tool like ours.

With these objectives in mind, we aimed to answer the aimed to discover how professors would answer the following questions:

How often do they make changes to your teaching methods and what motivates or prompts you to implement these changes?
What are the most common challenges that they experience when developing their course activities?
How do they aim to build and maintain an inclusive academic environment for all their students

Below are a few highlight quotes from the interviews that we conducted.

"I'm interested in understanding student socioeconomic backgrounds and how this affects student learning (how they learn and whether they choose to learn)."

"I prefer to make minor changes every quarter rather than a massive reconstruction of my course."

"Professors ideally try to mix [classroom] activities, it could be hard to implement because some classes have fixed seating arrangements."

User Personas

Using the information from both the competitive landscape analysis and the user interviews, I needed to build a few user personas to get a better grasp of the type of user that we are designing for and their needs and frustrations.

Based on the research findings, I created John Miller and Cynthia Tucker - with goals and characteristics that represent the needs of the larger group. With their goals, needs, frustrations, and fears in mind, I have a better understanding of who we will be designing for.

Prototypes

With the research that we had conducted in mind, we wanted to make sure that we covered the following questions:

How could we make the information accessible to processors of different majors and departments?
How can we present our results in a way that would be helpful for professors who are building their course curriculum?
How can professors manipulate their search results so that it would be representative of the students that they were teaching?

Using those guiding questions, we began to build our initial low-fidelity prototypes of the tool: