THINKING LIKE A COMPUTER: AN EXPLORATORY STUDY OF INTRODUCTORY PROGRAMMERS' LEARNING PROCESSES IN SCRATCH
Date
2020-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
As block-based programming environments are increasingly used to introduce
students to computational thinking (CT), researchers have developed various techniques
and tools for assessing students' CT skills using these visual programming languages.
Most approaches focus on analyzing students' nal code products; however,
insights into how to improve teaching also can be gained from observing students'
programming processes as they code, beyond what is learned from their products.
Unfortunately, observing each individual student's process rsthand and interviewing
students is time-intensive and non-scalable.
This thesis describes a logging and sequential pattern analysis methodology that
enables scalable data collection and analysis of students' programming processes in the
Scratch visual programming environment. As students code, we record their interactions
with the constructs and the environment itself. Using this capability, we explore
the kinds of programming behaviors that can be learned from the mined patterns of
logs recorded as students performed the same coding task in two college courses introducing
programming (37 Education and Human Development majors taking an Ed
Tech course and 44 non-CS majors taking a general introductory CS course). Our
ndings indicate that the logs and patterns can describe one's demonstration of uncertainty
with a certain construct, attempted use of blocks that are not required in an
assigned programming task, and heavy focus on speech-based constructs that do not
involve CT skills, all of which can serve as starting points of discussion concerning how
CT teaching methodologies could be improved.
Description
Keywords
computer science, scratch