Alexandria Digital Research Library

Enabling wide-scale computer science education through improved automated assessment tools

Author:
Boe, Bryce A.
Degree Grantor:
University of California, Santa Barbara. Computer Science
Degree Supervisor:
Diana Franklin
Place of Publication:
[Santa Barbara, Calif.]
Publisher:
University of California, Santa Barbara
Creation Date:
2014
Issued Date:
2014
Topics:
Education, Sciences and Computer Science
Keywords:
Assessment tools
Scratch
Static analysis
Computer science education
Genres:
Dissertations, Academic and Online resources
Dissertation:
Ph.D.--University of California, Santa Barbara, 2014
Description:

There is a proliferating demand for newly trained computer scientists as the number of computer science related jobs continues to increase. University programs will only be able to train enough new computer scientists to meet this demand when two things happen: when there are more primary and secondary school students interested in computer science, and when university departments have the resources to handle the resulting increase in enrollment. To meet these goals, significant effort is being made to both incorporate computational thinking into existing primary school education, and to support larger university computer science class sizes. We contribute to this effort through the creation and use of improved automated assessment tools.

To enable wide-scale computer science education we do two things. First, we create a framework called Hairball to support the static analysis of Scratch programs targeted for fourth, fifth, and sixth grade students. Scratch is a popular building-block language utilized to pique interest in and teach the basics of computer science. We observe that Hairball allows for rapid curriculum alterations and thus contributes to wide-scale deployment of computer science curriculum. Second, we create a real-time feedback and assessment system utilized in university computer science classes to provide better feedback to students while reducing assessment time. Insights from our analysis of student submission data show that modifications to the system configuration support the way students learn and progress through course material, making it possible for instructors to tailor assignments to optimize learning in growing computer science classes.

Physical Description:
1 online resource (148 pages)
Format:
Text
Collection(s):
UCSB electronic theses and dissertations
ARK:
ark:/48907/f3vh5m02
ISBN:
9781321349146
Catalog System Number:
990045116700203776
Rights:
Inc.icon only.dark In Copyright
Copyright Holder:
Bryce Boe
File Description
Access: Public access
Boe_ucsb_0035D_12298.pdf pdf (Portable Document Format)