Towards scalable, diverse, and secure assessment in college stem education

Chen, Binglin

Permalink

https://hdl.handle.net/2142/116207 Copy

Description

Title

Towards scalable, diverse, and secure assessment in college stem education

Author(s)

Chen, Binglin

Issue Date

2022-07-12

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

Zilles, Craig

Doctoral Committee Chair(s)

Zilles, Craig

Committee Member(s)

• West, Matthew
• Forsyth, David
• Zhai, Chengxiang
• Fox, Armando

Department of Study

Computer Science

Discipline

Computer Science

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

Assessment

Abstract

Assessment is a key component of education. Routine grading of students’ work, however, is time consuming. Automating the grading process allows instructors to spend more of their time helping their students learn and engaging their students with more open-ended, creative activities. One way to automate grading is through computer-based assessment, where questions can be randomly parameterized and deterministic algorithms can be implemented to grade students’ responses to these questions. Computer-based assessment is particularly well suited for courses in STEM, where many of the building block skills can be formulated as algorithmically-scored tasks. This thesis focuses on computer-based assessment in three directions. The first direction concerns the adoption of computer-based assessments in a summative context, where I demonstrate that randomization of questions and a centralized proctored facility result in secure asynchronous computer-based summative assessments. In the second direction, I attempt to extract patterns from computer-based assessment data that allow instructors to adjust the assessments for the better. Lastly, I present an attempt to expand the set of autograded questions to include “Explain in Plain English” questions where students are asked to give natural language descriptions of what a piece of code does. I discuss the lessons learned from the attempt and propose how automated short answer grading should be evaluated in general.

Graduation Semester

2022-08

Type of Resource

Thesis

Copyright and License Information

Copyright 2022 Binglin Chen

Owning Collections