S. L. Wong et al. (Eds.) (2010). Proceedings of the 18th International Conference on Computers in Education.

Putrajaya, Malaysia:
Asia-Pacific Society for Computers in Education.

The Impact of Learning Styles and

Instructional Methods on Students Recall and
Retention in Programming Education
Hui-Hui TIEa,b* & Irfan Naufal UMARb*
a
School of Degree Studies, SEGi College Penang, Malaysia
b
Centre for Instructional Technology & Multimedia, Universiti Sains Malaysia, Malaysia
*
[email protected]
*
[email protected]

Abstract: Learning styles are the predetermined indicator which learners respond to and use
stimuli in the context of learning. It is the preferred mode that learners perceive and process
new information at a certain pace, which in turn affects their academic performance. This
paper focuses on the visual-verbal learning styles as well as pair programming (PP) and
direct instruction (DI) methods. A group of 83 first year computing students was involved in
this experimental study. These students from two intact classes were randomly assigned
either to the experimental group that received PP, or to the control group that received DI.
During the seven-week treatment, the participants in PP group worked in pairs, and those of
DI group worked individually to solve novel programming scenarios. Three computer
programming performance tests (CPPT) a pretest, an immediate posttest and a delayed
posttest were conducted to evaluate the students recall and retention performance on
programming comprehension. The post hoc test results revealed that both the visual and
verbal students taught in the PP and DI groups performed equally good in recall
performance. However, only the verbal students in the DI group significantly outperformed
the visual students in retention. In conclusion, the diversity of learning styles influence
students engagement in understanding the programming concepts; significantly affects
information retention and later on their academic achievements.

Keywords: Learning style, direct instruction, pair programming, programming

performance

Introduction

Computer programming requires students to understand the programming process stages:

problem, design, coding and maintenance, in solving programming problems. It is a subject
which demands complex cognitive skills such as reasoning, problem-solving and planning
that students find it too difficult to comprehend and master [1]. Grand and Smith [2]
describe basic programming concepts as a set of rules, like expressions, functions, variables
declarative, statements, sequencing and conditionals which first year computing students
need to know in following any programming courses.
Various teaching methods have been applied by instructors in teaching these
concepts to undergraduate students. According to Baldwin and Kuljis [3], these methods
include changing the programming languages which is to be taught first, using different
textbooks, slowing the course delivery, lowering the standard, reducing the course
requirements, as well as switching between bottom-up and top-down approaches in teaching
basic programming concepts. However these methods did not significantly change the
students performance. Thus, some instructors opted to reduce the quality and to lower the
standard of material taught and concentrate on important topics [3].

ICCE2010 | 191
S. L. Wong et al. (Eds.) (2010). Proceedings of the 18th International Conference on Computers in Education. Putrajaya, Malaysia:
Asia-Pacific Society for Computers in Education.

Programming concepts tend to be difficult to grasp and require complex cognitive

task. Students who have adequate program-solving skills and manage to phase solution to
programming problems, could find it difficult to turn the problems into syntactically correct
program flowchart and Pseudocode [1][2]. Consequently, they tend to make the same
mistakes and become frustrated trying to understand and comprehend even the most basic
concepts [4]. Hence, those who are strong in logical reasoning skills outperform their
counterparts in the programming course, and others who do not appreciate the beauty of
programming will eventually drop out from the field.

1. Learning Styles

Learning style (LS) is a preferred learning mode in which students respond to and use
stimuli in the context of learning. For James and Gardner [5] and Kolb [6], it is the way in
which students perceive, process, store and recall attempts of learning. These styles have
been acknowledged as the prime construct that serve as relatively stable indicators of how
students respond to the learning environment [7]. However, with variation, these stable
indicators somehow may change from one learning environment to another as the students
adapt to the learning approach, progress and respond to the programming problems.
Researchers have claimed that students consistently indicate positive improvements in
performance when novel concepts are illustrated in their preferred style [8][9]. In this study,
the Felder-Solomans [10] Index of Learning Style was selected as it focused primarily on
learning styles, comprehensive, parsimonious and is relevant to science education.

2. Instructional Methods

Instructional method is an approach used by educators in course content deliveries. It is an

educational approach for revolving knowledge into learning which focusing on the how to
in delivery of training [11]. In this study, two instructional methods and two experience
lecturers identified are randomly assigned to the treatment groups. In programming
methodology, two basic constructs: (i) sequence and (ii) selection are the syllabus covered
in this study. The undergraduate semester one computing students who worked in pairs
received the PP instruction method. Those in the control group are taught in the DI method.

3. Research Methodology

The prime focus of this study is to investigate the effects of pair programming as
cooperative learning approach on the programming recall and retention performance
amongst the visualverbal learning style dimension computing students. The emphasis of
this research is on whether the learning style preference could be the moderating factors
when different instructional methods are applied in classroom environment.

3.1 Research Design

A 2 x 2 quasi-experimental design was used to examine the effects of PP and DI on the two
dependent variables (recall and retention) with one LS dimension being the moderating
variables. These dependent variables were based on the immediate and delayed posttest
scores obtained from the computer programming performance test (CPPT). A total of 83
first year undergraduate computing students (n = 83) in this intact group were randomly
assigned to the two treatment groups: the experimental group that received the PP treatment
and the control group that received the DI treatment.

ICCE2010 | 192
S. L. Wong et al. (Eds.) (2010). Proceedings of the 18th International Conference on Computers in Education. Putrajaya, Malaysia:
Asia-Pacific Society for Computers in Education.

3.2 Research Instrument

The computer programming performance tests (CPPT) instrument which consist the
immediate and delayed posttest were administered to measure the students recall and
retention performance. The tests were designed to assess the students programming
performance on the theories and practical knowledge of the sequence and selection
programming constructs. The Felder-Solomans Index of Learning Styles Questionnaire
(ILSQ), based on the Felder and Silvermans [12] learning styles model was used to access
the students learning style preference prior to the treatment. In this study, only one
dimension measuring the visual and verbal learning style (Cronbachs Coefficient alpha of
0.78 [13]) was used to determine the students learning preference. This instrument uses a
multiple choice format in presenting options. The students who responded mostly a in the
questionnaire were classified as visual learners and those who responded mostly b were
identified as verbal learners.

3.3 Data Collection Procedures

The experiment was conducted for seven-week duration to the two intact classes. These two
classes were randomly selected from the first year computing course and assigned to the two
treatment groups. The ILSQ questionnaire was administrated to these groups prior to the
treatment in order to classify the students as visual or verbal learners. During the treatment,
the students in both groups received program flowchart and Pseudocode in learning the
basic programming concepts: (i) sequence and (ii) selection constructs. In the PP group,
each pair consisted of one visual and verbal student with an assigned role of either as
driver or navigator. These roles are interchangeable at regular intervals. The pairs
worked together on the same programming task to discuss and provide constructive
criticisms in generating quality programming solutions.

4. Findings

The quantitative data collected to corroborate the research hypotheses were analyzed using
the SPSS 16.0 for Windows and computed at the 0.05 level of significant. As covariate was
used in the analysis, the adjusted posttest mean scores were reported. The MANCOVA
results clearly revealed a significant difference in recall performance between the visual and
verbal students taught in the two treatment groups (F=6.10; p=0.00; p<0.05). A significant
difference in retention was also observed between these students who received the different
treatment methods (F=4.30; p=0.01; p<0.05). Therefore, a post hoc test was administrated
to further examine the differences. This study reveals the comparisons involved only
between the visual and verbal students in both the PP and DI groups. The comparisons
between the visual students in the PP and DI groups, as well as between the verbal students
in both the treatment groups were not the scope of this discussion.

Hypothesis 1: There are no significant differences in recall between the visual and verbal
students who received the PP method and those who received the DI method.
In this study, the post hoc result indicates no significant difference between the
visual and verbal students taught in the PP group (MPP-visual=65.86; MPP-verbal=66.27; p=0.89)
as well as those in the DI group (MDI-visual=55.31; MDI-verbal=59.65; p=0.15). Thus, this
finding has accepted the first hypothesis.

ICCE2010 | 193
S. L. Wong et al. (Eds.) (2010). Proceedings of the 18th International Conference on Computers in Education. Putrajaya, Malaysia:
Asia-Pacific Society for Computers in Education.

Hypothesis 2: There are no significant differences in retention between the visual and
verbal students who received PP method and those who received DI method.
For the PP group, the post hoc test result reveals no significant difference in
retention performance between the visual and verbal students taught in this PP method
(MPP-visual= 52.03; MPP-verbal= 58.33; p=0.051; p>0.05). However, a significant difference in
retention was observed between them in the DI group (MeanDiff =6.53; p=0.037; p<0.05),
with the verbal students performed significantly better than those of visual students
(MDI-visual=46.95; MDI-verbal=53.48). Thus, the second hypothesis has been rejected.

5. Discussion

This study revealed that the preferred style has significantly correlated and influenced
students programming comprehension; in return affecting their mental process [7][8]. For
the recall performance, both the visual and verbal students taught in the PP and DI methods
performed equally good in the programming recall test. Knowing the students individual
LS allows them to create a richer learning context of the basic programming concepts. With
regards to the instructional methods, LS assists both students taught in the PP and DI
methods in that they understand their individual preference in learning programming that
create a higher level of conceptual understanding and promote logical thinking skills with
respect to programming recall performance [14]. Despite the fact that visual students learn
better with diagrams or charts, they are able to adapt and change learning approach from one
learning environment to another as they progress and respond to programming problems
that subsequently trigger the actual learning process in the long term memory for further
information recall [15].
In retention performance, only the verbal students taught in the DI method
significantly outperformed the visual students taught in the same method. Working in pairs,
both visual and verbal students provided with the opportunity to explain and discuss novel
concepts with their partners, thus promotes meaningful learning and retention [16].
Moreover, students are benefited from pairing with heterogeneous LS pair, especially with
diversity of thought that fosters learning process [9]. The verbal students in the DI method
incline to have longer retention rate on the basic programming concepts taught as the
explanation of abstract concepts are predominantly conducted in the form of lectures,
written texts and on the whiteboard in the classroom environment [17]. Thus, the verbal
students are able to process and build better schemata in long term memory as they
remember better through oral and written explanations as compared to their counterparts in
both instructional methods. In contrast, the visual students learn and remember better
through pictorial form than auditory clues. So, they are experiencing difficulties in attaining
the pedagogical goals for programming performance in retention than those verbal students
taught in both PP and DI groups [14][18].

6. Conclusion

The study has emphasized the importance of considering LS components in learning the
basic programming concepts for classroom delivery. LS is a predictable stable indicator that
influences the learning approach of students in perceiving, interacting with and responding
to programming concepts. The preferred LS of students and the instructional methods
applied in programming classes are correlated with the students recall and retention
performance. Disparity between the LS and instructional methods in comprehending the
fundamental programming concepts may cause significant impact on students overall
performance (recall and retention). Therefore, both the visual and verbal representation of
ICCE2010 | 194
S. L. Wong et al. (Eds.) (2010). Proceedings of the 18th International Conference on Computers in Education. Putrajaya, Malaysia:
Asia-Pacific Society for Computers in Education.

novel concepts need to be embraced into the learning context in order to reinforce long term
memory retention, enhance cognitive knowledge and promote logical thinking skills with
respect to programming performance.

References

[1] Miliszewska, I. & Tan, G. (2007). Befriending computer programming: A proposed approach to teaching
introductory programming. Issues in Informing Science and Information Technology, 4, 277-289.
[2] Grant, M. & Smith, S. (2005). Programming languages. Free software foundation, (1.02). Retrieved
February 10, 2009, from http://www.e-booksdirectory.com/ details.php?ebook=1139
[3] Baldwin, L. P. & Kuljin, J. (2000). Visualisation techniques for learning and teaching programming.
Journal of Computing and Information Technology ICT, 8(4), 285-291.
[4] Stamouli, I., Doyle, E. & Huggard, M. (2004). Establishing structured support for programming students.
Proceedings of the 34th ASEE/IEEE Frontiers in Education Conference, 2, F2G-5-F2G-9, Savannah,
GA., October.
[5] James, W. & Gardner, D. (1995). Learning styles: Implications for distance learning. New Directions or
Adult and Continuing Education, 67.
[6] Kolb, D. A. (1984). Experiential learning: Experience as the source of learning and development.
Englewood Cliffs, NJ: Prentice-Hall.
[7] Deborah, G. (2005). Meeting differing learning styles of non-traditional students in the second language
classroom. Journal of College Teaching and Learning, August, 2(8), 1-7.
[8] Pallapu, P. (2007). Effects of visual and verbal learning styles on learning. Institute for Learning Styles
Journal, 1, 34-39.
[9] Davis, G. A. (2006). Learning style and personality type preferences of community development
extension education. Journal of Agricultural Education. 47(1), 90-99.
[10] Felder, R. M. & Soloman, B. (2000). Learning styles and strategies. North Caroline State University,
Resources in Science and Engineering Education. Retrieved December 30, 2009, from
http://www2.ncsu.edu/unity/lockers/users/f/felder/public/ILSdir/styles.html
[11] Nguyen, F. & Clark, R. C. (2005). Efficiency in e-learning: Proven instructional methods for faster,
better, online learning. A publication of the eLearning guides, November 7, 1-8.
[12] Felder, R. M. & Silverman, L. K. (1988). Learning styles and teaching styles in engineering education.
Engineering Education, 78(7), 674-681.
[13] Zywno, M. S. (2003). A contribution to validation of score meaning for Felder-Solomans Index of
Learning Styles. Proceeding of the 2003 American Society for Engineering Education Annual
Conference & Exposition, American Society for Engineering Education.
[14] Fenrich, P. (2006). Getting practical with learning styles in Live and computer-based training settings.
Issues in Informing Science and Information Technology, 3, 233-242.
[15] Anabela, G., Alvaro, S., Lilian, C. & Mendes, A. J. (2007). Learning styles in an e-learning tool. Paper
presented at International Conference on Engineering Education ICEE 2007, Coimbra, Portugal.
[16] Chung Ruey-Gwo & Lo Chieh-Ling (2006). The study of different cooperative learning and
problem-based instructions in promoting students teamwork competences. World Transactions on
Engineering and Technology Education, 5(3).
[17] Layman, L., Cornwell, T., Williams, L. & Osborne, J. (2007). Personality profiles and learning styles of
advanced undergraduate computer science students. NC: North Caroline State University.
[18] Franzoni, A. L. & Assar, S. (2009). Student learning styles adaptation method based on teaching
strategies and electronic media. Educational Technology and Society, 12(4), 15-29.

ICCE2010 | 195