Samuel Lau | Instructor-Centered Design of Tools to Support Teaching Programming and Data Science At Scale

Abstract

Instructors of technical subjects like programming and data science use a wide array of software tools that enable them to create sophisticated and engaging lessons at scale. Although there are many such tools available, instructors often find themselves repurposing software originally designed for other people, like professional software engineers. This mismatch of intent adds extra logistical complexity to the already-challenging task of designing and delivering effective learning content.

To address these issues, this dissertation takes an instructor-centered approach. It surfaces previously unmet needs through studies of instructors, their goals, and their software tools. The key findings are that instructors constantly seek to update their learning materials, yet encounter heavy logistical challenges in doing so because the tools they use to help design their lessons were not intended for instructional use.

This dissertation also contributes novel interactive systems that directly support teaching by designing for instructor needs. In particular, this dissertation contributes program visualization tools that enable instructors to show how code transforms data: TweakIt helps learners work with unfamiliar code snippets, and the Pandas/Tidy Data/SQL Tutors automatically visualize code that manipulates data tables step-by-step. Together, this dissertation provides the first evidence that the insights gathered from an instructor-centered approach can lead to tools that better support the work of instruction.

Committee members:

Philip J. Guo, Chair, Cognitive Science

James D. Hollan, Cognitive Science

Ranjit Jhala, Computer Science and Engineering

Bradley Voytek, Cognitive Science

Haijun Xia, Cognitive Science

.

Date & Time : Tuesday, June 6th, 1:00pm – 2:00pm

Location : Design and Innovation Building, Room 406 or Join with Zoom