INGI - Public Thesis Defense of Olivier Goletti
ingi |
Improving Tutors’ Practice by Integrating Explicit Instructional Programming Strategies in Introductory Programming Courses at University.
Wednesday February 12, 225 at 10:00 a.m. Auditorium A.01 - Place des Sciences 1348 Louvain-la-Neuve
Context. Often, university-level introductory programming courses (CS1)
make use of undergraduate teaching assistants (UTAs). Such UTAs gen-
erally are responsible for most of the students’ interactions with the CS1
course. UTAs usually have no pedagogical background. While some univer-
sities provide pedagogical training and research documents these trainings,
challenges and interactions, pedagogical content knowledge specific to the
teaching of programming has not been extensively researched.
Objective. In this dissertation, we explore how UTAs’ teaching practices
can be improved by training them with dedicated evidence-based explicit
instructional programming strategies. Moreover, we want to understand
and characterise how UTAs use and adapt these strategies, what hinders
and helps them.
Methods. Based on learning transfer processes and cognitive load the-
ory, we selected four strategies: explicit tracing, subgoal learning, Parsons
problems and explicit problem solving. We used a design research approach
to iterate on our interventions on the instructional design of the course’s
setup in three different years. This iterative process allowed us to gradu-
ally tailor explicit tracing and subgoal learning to the studied CS1 course.
We progressively increased the number of UTAs participating in our study,
while refining our intervention up to a full integration of the subgoal learn-
ing strategy throughout the course. We collected data on tutors’ usage of
the instructional strategies by mixing methods: semi-directed interviews,
surveys and observations. The data was analysed using thematic analysis
with both inductive and deductive coding, and the concept of fidelity of
implementation.
Findings. This dissertation reports on the results of this iterative process.
We found that integrating subgoal learning throughout the course pushed
UTAs to apply explicit instructional programming strategies in their practice.
We also report on triggers for strategy uses by UTAs, adaptations made by
UTAs and UTAs’ fidelity of implementation of explicit tracing and subgoal
learning.
Implications. Since UTAs are widely used in CS courses, this work pro-
vides research-informed actionable advice to practitioners willing to inte-
grate explicit instructional strategies in their own setup. We also discuss
the feasibility of applying these strategies to school teachers.
Thesis Committee:
Prof. Kim Mens (Supervisor) UCLouvain/ICTEAM, Belgium
Prof. Felienne Hermans (Supervisor) VU Amsterdam, The Netherlands
Prof. Mariane Frenay UCLouvain/IPSY, Belgium
Prof. Virginie März UCLouvain/IACS, Belgium
Prof. Jan Vahrenhold Universität Münster, Germany
Prof. Charles Pecheur (Chair) UCLouvain/ICTEAM, Belgium
Prof. Efthymia Aivaloglou TU Delft, The Netherlands