Mechansim design for collaborative work

Lee, Andy C

Permalink

https://hdl.handle.net/2142/120291 Copy

Description

Title

Mechansim design for collaborative work

Author(s)

Lee, Andy C

Issue Date

2023-04-20

Director of Research (if dissertation) or Advisor (if thesis)

Sundaram, Hari

Department of Study

Computer Science

Discipline

Computer Science

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• Mechanism design
• education

Abstract

Performance in educational settings has significant impacts to students after they gradu- ate. It can substantially influence placement in labor markets which has crucial downstream impacts on life satisfaction. It is important that instructors be fair and efficient. Mechanism design provides a set of tools and methodologies for creating algorithms for strategic agents which attains some socially desirable outcome. This work contributes to the growing body of research at the intersection of education and mechanism design. We focus on the problem of mechanism design for collaborative work. Collaborative work is pedagogically important as it provides students an opportunity to learn from their peers and provides experience for work outside the educational setting. Students may also be able to complete more complicated problems in groups than by themselves. Because of its importance it is critical that group work be administered fairly and efficiently. This not only maximizes student satisfaction but educational outcomes as well. We split collaborative work into three problems: group formation, task allocation, and peer grading. Group formation is the problem of matching students into project groups together. Task allocation is the problem of matching students to tasks within a project. Peer grading is the problem of assessing students evaluations of their groups once a project is complete. For each problem mechanisms solving the problem are given. These mechanisms are evaluated on their theoretical properties as well as by simulation when appropriate. For group formation it was found that a theoretically strong mechanism did not substantially outperform a baseline mechanism. For task allocation mechanisms are given which balance competing preferences students may have. For peer grading a mechanism is given which incentivizes students to report their contributions honestly.

Graduation Semester

2023-05

Type of Resource

Thesis

Copyright and License Information

Copyright 2023 Andy Lee

Owning Collections