Tan & Shahrill 303

Students’ learning of college level calculus

Jordan Peng Horng Tan, Universiti Brunei Darussalam, Brunei
Darussalam
Masitah Shahrill, Universiti Brunei Darussalam, Brunei Darussalam

Introduction
Mathematics (subject code 9709) is one of the popular subjects that is widely chosen
among Year 12 and 13 students in Brunei Darussalam and the syllabus follows Brunei
Cambridge GCE A-Level (BC GCE A Level). Table 1 below shows the statistical data
of students enrolling in Mathematics in a pre-university (Sixth form) colleges in Brunei
Darussalam for three consecutive years.

Table 1. Student population in one of the pre-university colleges in Brunei Darussalam

Referring to Table 1, the number of students studying Mathematics has increased

significantly each year. This has also led to a dramatic change in students‟ performance
as the number of students achieving grade U had increased drastically from 18.2% (63
students) in 2011 to 20.9% (109 students) in 2013 (data obtained from the respective
college). One of the few possible contributing factors is the students‟ level of
understanding in representing logical reasoning in A-Level Mathematics.

Sutherland (2007) and Idris (2009) stated that mathematics learning is a complex
and dynamic process. Therefore, teachers teach with the aim of delivering contextual
learning such that students understand the concepts and principles on what they have
been taught and why learning takes place. But what exactly do we mean by
understanding? Idris (2009) stated that understanding in mathematics is not simply
memorising the formulae or being able to follow procedure in solving a question but to
include the process of applying the knowledge understood. She further elaborated that
learning no longer emphasises the correctness of the final answer but rather, has shifted
to emphasizing the process, content and understanding. Therefore, it is important to
identify students‟ level of understanding.

Skemp (1979) categorised understanding into three categories, namely instrumental,

relational and logical understanding. Kastberg (2002, p. 13) summarised these three
categories as:
1. Instrumental understanding is the ability to apply an appropriate remembered rule
to the solution of a problem without knowing why the rule works. In other words,
we know “how” but we do not know “why”.
2. Relational understanding is the ability to deduce specific rules or procedures from
more general mathematical relationships. In short, one knows both “how” and
“why”.

______________________________________________________________________________
7th ICMI-East Asia Regional Conference on Mathematics Education
11-15 May 2015, Cebu City, Philippines
304 Students’ learning of college level calculus

3. Logical understanding is the ability to connect mathematical symbolism and

notation with relevant mathematical ideas and to combine these ideas into chains
of logical reasoning.

Statement of the problem

Calculus is one of the core topics in A-level pure mathematics (paper 1 and paper 3)
where students often encountered difficulties. It also carries a high weight in A-level
examinations, which include the application of differentiation, integration and solving
differential equations that require students to apply their skills learnt from both
differentiation and integration. As mentioned above, the number of students obtaining
grade U had increased from year to year, raising the question of how the level of
students‟ understanding can be attributed to this quality of students‟ achievement.

Purpose of the study

The purpose of this study is to investigate students‟ level of understanding in Calculus
based on Skemp‟s theory of understanding and to identify the type of understanding that
might affect students‟ performances. This study may be able to provide a good view on
the correlations between students‟ achievement and their level of understanding.

Scope and limitation of the study

As the scope of the study is limited to two school terms (January to June) and only
involved two upper 6 classes (Year 13) from a Pre-University college in Brunei-Muara
District, this sample will not be a representative sample of the overall Pre-University A-
Level students‟ population in Brunei Darussalam. It specifically focused on Calculus in
Pure Mathematics.

Literature review
A local researcher, Chan (2009), found that factors such as gender, confidence level,
attitude, parents‟ and teachers‟ perception, social economic status, home environment,
motivation and learning styles had significant influences on students‟ achievement in
solving A-level paper 3 (Pure) mathematics. An insight into Chan‟s research findings
further raised the interest in exploring the level of understanding in students that may be
affecting their academic performance. Luitel (2003) also highlighted “Are
understanding and learning related to each other?” in his article on „Developing and
Probing Understanding in Mathematics‟ (p.1). This further attracted the interest in
investigating students‟ level of understanding in calculus. Hiebert and Carpenter (1992)
stated that any mathematical idea or procedure or fact is understood if it is a part of a
person‟s internal mental network. However, Luitel (2003) contradicted this by
elaborating that understanding is an outcome of learning whereas learning is a process
and learning of some skills does not require developing understanding. These sparked
an interest in exploring students‟ level of understanding. Furthermore, Skemp (1976)
marked the beginning of the study of understanding in mathematical education research
by identifying three kinds of understanding; instrumental, relational and logical
understanding. His article “Relational and Instrumental Understanding” defined and
described these two types of understanding and explained why many teachers felt that
instrumental understanding was a type of understanding, through knowing how but not
knowing why. We thought about the level of students‟ mathematical understanding in
Brunei Darussalam and decided to analyse it in one of the pre-universities.

7th ICMI-East Asia Regional Conference on Mathematics Education

11-15 May 2015, Cebu City, Philippines
 Tan & Shahrill 305

According to Chan (2009), calculus is one of the most difficult topics in Paper 3
(Pure) Mathematics for teachers to teach and for students to learn. This had been
researched by Tall (1992) in which he claimed that whichever way calculus is
approached; there were difficult concepts which seemed to cause problems no matter
how they are taught. Chan (2009) further elaborated that the concepts were causing
cognitive difficulties not only to students but also teachers. In relation to the main study,
students in Brunei Darussalam may be facing these cognitive difficulties.

Idris (2009) claimed that many students have used the wrong methods in
Mathematics learning process as they think memorising rules and substituting numerals
into the chosen formula is the right way. Borasi and Rose (1989) further elaborated that
only a few students hope for meaningful Mathematics learning and only a small number
of them see Mathematics as requiring creative thinking. It is therefore no longer unusual
to find students who use procedures without understanding the situation. There is a
quote from Franklin in 1750 cited from the Northern Illinois University, “Tell me and I
forget, teach me and I may remember, involve me and I learn.” Learning a particular
topic in Mathematics by listening and imitating the workings of the teacher may not lead
to understanding. Is it true that only a small number of students in Brunei Darussalam
hope for meaningful Mathematical learning and see Mathematics as requiring creative
thinking?

Horvath (2008) conducted a study on students‟ understanding from inside-out

mainly focusing on the relationship between the chain rule and function composition
where he claimed that chain rule is one of the hardest ideas to convey to students in
calculus due to the difficultness in expressing the symbols and in representing the ideas
into words. Therefore, students have difficulties in remembering it and hence, they are
unable to apply (Gordon, 2005 as cited in Horvath, 2008). His finding showed that even
though students have written down the correct answer, they may not be thinking of the
right answers and there may still be subtle misunderstandings hidden in the notation.
Could this be due to the lack of relational or logical understanding?

As far as we know, there has not been much research conducted on students‟ level of
understanding in Calculus in Brunei Darussalam. However, this study could serve as a
continuation of Chan‟s research, which partly focuses on students‟ understanding of
Integration in Brunei Darussalam. Therefore, a study on exploring students‟
understanding in pre-university calculus was conducted, which specifically focused on
students‟ level of understanding based on Skemp‟s theory of understanding and also the
possible correlation between students‟ achievement and understanding.

Research methodology
Research design
This study involved a two cycled action research and the combination of quantitative
and qualitative methods to gather and analyse the data. The second cycle of action
research was carried out after the pre-test, which mainly focused on students‟
difficulties shown in the test. Furthermore, the topics were chosen from among students
in existing classes in order not to disturb the school routine.

______________________________________________________________________________
7th ICMI-East Asia Regional Conference on Mathematics Education
11-15 May 2015, Cebu City, Philippines
306 Students’ learning of college level calculus

Sample of the study

Two Year 13 (mean age 17.8) classes from one of the pre-university colleges in Brunei-
Muara district were selected for this study. These students are categorised by ability
based on their Mathematics syllabus D and additional mathematics performance in GCE
O-level examination. The sample consists of 19 students from the high ability class
(hereafter referred to as Class A) and 21 students from the medium ability class
(hereafter referred to as Class B) who all studied paper 3 (Pure) Mathematics and had
learnt paper 1 (pure Mathematics) in the previous year.

Instrumentation of the study

The diagnostic test, which comprised of two questions (simple indefinite integral and
integration using trigonometric identities) were given to students before the lessons,
followed by a pre-test and post-test of seven questions each. The time frame between
the pre-test and post-test is one week. A questionnaire consisting of a Likert scale with
23 questions was given to the students after the pre-test to identify their learning styles
in calculus.

Validity and reliability

All the three tests (diagnostic, pre and post-test) were face-validated by three teachers
from the pre-university college who have more than 22 years of experience in teaching
Mathematics to ensure that the tests are appropriate in assessing the students‟
knowledge and understanding.

The SPSS software was used to calculate the reliability of the data. The correlations
between students‟ results from the pre-test and post-test were generated based on
Pearson product-moment correlation coefficient, which gave a value of 0.889 (two-
tailed test at 1% significant level). The questions in the pre-test and post-test were
deliberately set differently (the questions were extracted from past GCE A-level papers
covering the same topics with the same levels of difficulty, such as∫ √ ( )
and∫ ( ) ) to avoid students memorising the workings. Apart from that, the
√
internal pre-test and post-test reliability were obtained using Cronbach‟s alpha reliability
coefficient. Table 2 below gives the reliability coefficients.

Table 2. Cronbach alpha reliabilities for pre-test and post-test

The students‟ perception questionnaire was face-validated by one senior lecturer

from the University of Brunei Darussalam and one experienced teacher from the
respective school. The reliability test, Cronbach alpha, was performed and the value was
0.706, which is in the acceptable region (Kline, 1999 as cited in Field, 2005). The
qualitative data consisted of data from individual interviews with two students that were
audiotaped.

7th ICMI-East Asia Regional Conference on Mathematics Education

11-15 May 2015, Cebu City, Philippines
 Tan & Shahrill 307

Results and discussion

Students’ understanding in calculus
After the diagnostic test, lessons on integration were designed, which focused on
students‟ understanding including the ways to derive formulae and the relations between
various topics in mathematics. After the completion of these topics, the researcher gave
a pre-test to both classes A and B. Table 3 below shows students‟ pre-test results.

Table 3. Students’ pre-test results

As shown in Table 3, majority of the students have difficulties in solving question 6;

i.e. ∫ . Students in class A obtained an average score of 1.35 out of 3 whereas
none of the students in class B managed to score. A strong relational understanding in
mathematics is required in order to solve this question. They should be able to apply the
double angled formula from trigonometry which will lead them to convert ∫
into ∫ . Unfortunately, based on the sample, this question showed that
students in both classes have minimal relational understanding.

In order to support the above claim, results from the questionnaire were extracted (as
shown in Table 4). Majority (60%) of the 30 participants from both classes who
completed the questionnaire agreed that mathematics is a subject that required
memorising all the formulae and that they often have difficulties in answering calculus
questions. This shows that these students had procedural understanding with minimal
relational understanding in applying mathematical concepts to solve mathematical
problems, which caused them difficulties in answering the questions.

Interestingly, the findings from question 3 to 7 in Table 4 below show positive

responses in the students‟ thought. Majority of them think Mathematics is a subject that
required relational understanding but curiously their pre-test results showed majority of
them only applied the procedure and some even applied the wrong procedure in solving
a mathematical problem. Examples are shown in Figure 1 and Figure 2 where students
applied the formula of „integration by parts‟ without knowing the proper concepts
behind. This also showed that students in this sample think they understand when they
are able to solve a particular problem but in fact don‟t (Brandt, 1993; Gordon, 2005;
Luitel, 2003; Skemp, 1976).

______________________________________________________________________________
7th ICMI-East Asia Regional Conference on Mathematics Education
11-15 May 2015, Cebu City, Philippines
308 Students’ learning of college level calculus

Table 4. Students’ questionnaire responses (testing on relational understanding)

Figure 1. Students’ response in solving integration by parts question without knowing

the proper concepts.

Figure 2. Students’ response in solving integration by using trigonometric identities.

7th ICMI-East Asia Regional Conference on Mathematics Education

11-15 May 2015, Cebu City, Philippines
 Tan & Shahrill 309

Question 7 (i.e. Express in the form . Hence, find ∫ ) in

the pre-test has an optimal score among all the 7 questions. Students in class A obtained
an average of 91.8% and class B obtained an average score of 46.6%. Based on the
interview with 2 students each from class A and B, these students claimed that they
were able to score well due to the formula given in the questions. In other words, they
were just following the procedures taught from partial fractions in order to get all three
unknowns in the question, followed by using integration by logarithmic function (as
shown in Figure 3 where this student was able to solve the partial fraction correctly but
not the integration). Thus, this further showed that students have a strong procedural
knowledge without a proper relational understanding.

Figure 3. Students’ response in partial fraction followed by integration by logarithmic

function (question 7).

The second cycle of action research was carried out after the pre-test where the
researcher retaught concepts that mainly focused on students‟ difficulties as shown in
their pre-test. This was followed by a post-test. Table 5 below shows students‟ post-test
results.

The mean score obtained by students in both classes increased and the standard
deviation for all the 7 questions in class A decreased. This proved that class A showed
greater improvement. In addition, all students in class A managed to answer questions 1,
3 and 7 correctly. Could students‟ understanding be attributed to these improvements?
To answer this question, student Y from class A was interviewed, and he stated “…the
questions are easy and looks alike… since we already know our mistakes and with
continuous practice, I don‟t think we will be making the same mistake over and over
again…” Questions 1, 3 and 7 only required students to use procedural skills without

______________________________________________________________________________
7th ICMI-East Asia Regional Conference on Mathematics Education
11-15 May 2015, Cebu City, Philippines
310 Students’ learning of college level calculus

any need of relational understanding. Thus, there is no evidence to suggest that students
have improved in their relational and logical understanding, which is derived from the
development of relational understanding.

Table 5. Student’s post-test results

Table 6 below shows the results of students‟ pre-test and post-test and the histogram
in Figure 4 was drawn to illustrate this finding. The results of the analysis (generated
through simple descriptive statistics) indicated that students in both classes showed
significant improvement in the calculus achievement test. Based on the given sample,
the findings show that instrumental understanding affects students‟ performance in as
much as instrumental understanding is the fundamental knowledge that one should have
in order to take this understanding to the next level (relational understanding). As stated
by Kim (2005), “there is a need to revisit earlier understandings and view them from a
different perspective in order to develop the next level of understanding” (p. 2).

Table 6. Summary of students’ results from the pre and post-test

100
80
60
Pre-test
40
Post-test
20
0
Class A Class B
Figure 4. Students’ pre-test and post-test results

7th ICMI-East Asia Regional Conference on Mathematics Education

11-15 May 2015, Cebu City, Philippines
 Tan & Shahrill 311

Consequently, students in both classes were not able to obtain full marks in
questions 2, 4, 5 and 6 because those are the questions that required not only procedural
skills but also students‟ understanding in representing relational and logical reasoning.
This could mean that students have insufficient understanding in applying fundamental
procedures in solving mathematical questions. Although some students scored well
using instrumental understanding, they could only answer the lower level questions,
suggesting that this type of understanding has limited effectiveness and application.

Conclusion
This research shows that students in this study have limited relational thinking at the
pre-university level. Although there was an improvement in conceptual knowledge, only
an insignificant increase was recorded. Perhaps, the learning of mathematics required
students to exhibit relational thinking rather than procedural knowledge but nonetheless,
learning of procedural knowledge should not be avoided. For example, in a simple
addition of 25 + 15, students will exhibit procedural skill of obtaining an answer of 40
but this does not imply students have limited relational and logical understanding. They
may have attained some understanding of the concepts of addition and why adding
always resulted in a solution that has a value more than the given and simply needed to
apply procedural understanding to obtain the correct solution. This is further supported
by a quote from Charles (1998), “full concept development appears to evolve over a
period of years” (p. 143). Thus, relational and logical thinking require students to build
from the fundamental knowledge (i.e. instrumental knowledge). Since only a sample of
two classes was taken, the findings could not be used to generalise the whole population
in Brunei Darussalam. Another common view held by students in this study was that
students categorised remembering a certain formula as understanding or part of
understanding. Further studies involving more concepts and larger populations from
different pre-university colleges of differing abilities are to be conducted in order to
have an insight on pre-university students‟ understanding throughout the nation.

References
Borasi, R. & Rose, B. J. (1989). Journal writing and mathematics instructions.
Educational Studies in Mathematics, 20, 347-365.
Brandt, R. (1993, April 7). On Teaching for Understanding: A Conversation with
Howard Gardner. Retrieved July 4, 2014, from ASCD:
http://www.ascd.org/publications/educational-leadership/apr93/vol50/num07/On-
Teaching-for-Understanding@-A-Conversation-with-Howard-Gardner.aspx.
Chan, S. (2009). The Effectiveness between Streaming and Non-Streaming 'A' - Level
Pure Mathematics (Paper 3) of Sixth Form Colleges in Brunei Darussalam.
Unpublished master‟s dissertation. Universiti Brunei Darussalam, Brunei
Darussalam.
Charles, M. P. (1998). A cross sectional investigation of the development of the
function concept. CBMS issues in mathematics education, 7, 114-162.
Field, A. P. (2005). Discovering statistics using SPSS. London: Sage.
Gordon, S. (2005). Discovering the chain rule graphically. Mathematics and Computer
Education, 39(3), 195-197.
Hiebert, J. & Carpenter, T. P. (1992). Learning and teaching with understanding. In D.
Grouws, Handbook for research in mathematics teaching and learning (pp. 65-97).
New York: MacMillian.

______________________________________________________________________________
7th ICMI-East Asia Regional Conference on Mathematics Education
11-15 May 2015, Cebu City, Philippines
312 Students’ learning of college level calculus

Horvath, A. (2008). Looking at calculus students' understanding from the inside-out:

The relationship between the chain rule and functional composition. FAST (Future
Academic Scholars in Teaching) Fellowship Program.
Idris, N. (2009). Enhancing students' understanding in calculus through writing.
International Electronic Journal of Mathematics Education, 4(1), 36-55.
Kastberg, S. E. (2002). Understanding Mathematical Concepts: The case of the
logarithmic function. University of Georgia. Retrieved March 2, 2014, from
http://jwilson.coe.uga.edu/pers/kastberg_signe_e_200205_phd.pdf
Kim, B. (2005). Psychophysical magic: Rendering the visible „invisible‟. TRENDS in
Cognitive Sciences, 9, 381-388.
Luitel, B. C. (2003). Developing and probing understanding in mathematics. Retrieved
Feb 10, 2014, from http://www.geocities.ws/bcluitel/understanding.pdf
Northern Illinois University. (n.d.). Experiential learning. DeKalb, Illinois, United
States. Retrieved March 2, 2014, from
http://www.niu.edu/facdev/resources/guide/strategies/experiential_learning.pdf
Skemp, R. R. (1979). Intelligence, learning, and action. New York: Wiley.
Sutherland, R. (2007). Teaching, learning & mathematics. In R. Sutherland, Teaching
for learning Mathematics (p. 10). Maidenhead, Berkshire, England: McGraw-Hill
Education.
Tall, D. (1992). Students' difficulties in calculus. ICME-7, (pp. 13-28). Quebec: Canada.

_______________________
Jordan Peng Horng Tan
Sultan Hassanal Bolkiah Institute of Education, Universiti
Brunei Darussalam, Jalan Tungku Link, Gadong, BE 1410,
Bandar Seri Begawan, Brunei Darussalam.
jordantph@hotmail.com

Masitah Shahrill
Sultan Hassanal Bolkiah Institute of Education, Universiti
Brunei Darussalam, Jalan Tungku Link, Gadong, BE 1410,
Bandar Seri Begawan, Brunei Darussalam.
masitah.shahrill@ubd.edu.bn

7th ICMI-East Asia Regional Conference on Mathematics Education

11-15 May 2015, Cebu City, Philippines