Conceptual models and individual cognitive learning styles
in teaching recursion to novices
Abstract of Doctoral Thesis
The University of Texas at Austin, 1993
Supervisors: Ralph W. Cain and Nell B. Dale
This study investigated how different types of conceptual models and
cognitive learning styles influence novice programmers when learning recursion.
A pretest-posttest, 2 X 2 (conceptual models X learning styles) factorial
experimental design was implemented in order to study the problem. Two hundred
thirty-seven students enrolled in an introductory computer science course
at a major southwest research university served as the subjects for this
study. Subjects were randomly assigned to either an abstract model group
or a concrete model group and the groups were of approximately equal size.
Different conceptual models (abstract or concrete) were used to present
recursion to the two model groups. Within each model group, subjects were
identified as either an abstract learner or a concrete learner based on
their scores on the scrambled Kolb's Learning-Style Inventory 1985. A posttest
and two retention tests were administered after the treatment to compare
students' performance in different groups. A pretest administered prior
to the treatment was used to equate the variance caused by students' prior
knowledge in the statistical analysis. The statistical procedure of two-way
ANCOVA was employed to analyze all of the performance data.
The findings of this study are: Concrete conceptual models were better
than abstract conceptual models in teaching recursion to novice programmers.
However, the teaching effects weakened several weeks after classroom instruction.
Novice programmers with abstract learning styles performed better than those
with concrete learning styles when learning recursion. Finally, abstract
learners did not necessarily benefit more from abstract conceptual models,
and concrete learners did not necessarily benefit more from concrete conceptual
A replication study with a longer treatment period that covers more aspects
of recursive programming is recommended for future research. Additional
research needs to be conducted to better understand students' mental models
of recursion. Furthermore, future research should investigate how the other
dimension of Kolb's learning styles (i.e., active-reflective) relates to
the instructional methods provided. It is also recommended that the relationship
between the characteristic of learning tasks (or domains) and the matching
of learning styles with conceptual models be investigated
Copyright (c) 1993 by Cheng-Chih Wu, Ph.D. Presentation
of this material by the Department of Computer Sciences at the University
of Texas at Austin was made possible under a limited license grant from
the author, who has retained all copyrights in the works.