PHYSICAL REVIEW SPECIAL TOPICS—PHYSICS EDUCATION RESEARCH 11, 020129 (2015)

Students’ difficulties with vector calculus in electrodynamics

Laurens Bollen,1,* Paul van Kampen,2,† and Mieke De Cock1,‡
1
Department of Physics and Astronomy & LESEC, KU Leuven,
Celestijnenlaan 200c, 3001 Leuven, Belgium
2
Centre for the Advancement of Science and Mathematics Teaching
and Learning & School of Physical Sciences, Dublin City University,
Glasnevin, Dublin 9, Ireland
(Received 27 January 2015; published 2 November 2015)
Understanding Maxwell’s equations in differential form is of great importance when studying the
electrodynamic phenomena discussed in advanced electromagnetism courses. It is therefore necessary that
students master the use of vector calculus in physical situations. In this light we investigated the difficulties
second year students at KU Leuven encounter with the divergence and curl of a vector field in mathematical
and physical contexts. We have found that they are quite skilled at doing calculations, but struggle with
interpreting graphical representations of vector fields and applying vector calculus to physical situations.
We have found strong indications that traditional instruction is not sufficient for our students to fully
understand the meaning and power of Maxwell’s equations in electrodynamics.

DOI: 10.1103/PhysRevSTPER.11.020129 PACS numbers: 01.40.Fk, 01.40.gb, 01.40.Di

I. INTRODUCTION Maxwell’s equations can be formulated in differential

It is difficult to overestimate the importance of or in integral form. In differential form, the four laws are
Maxwell’s equations in the study of electricity and magnet- written in the language of vector calculus that includes
ism. Together with the Lorentz force law, these equations the differential operators divergence and curl. These are
provide the foundations of classical electrodynamics. They typically expressed using the nabla ∇ symbol to denote
can be used to solve problems concerning electromagnetic the del operator. The divergence of a vector field (∇ · A)
phenomena, including those occurring in electric circuits is a scalar quantity that measures the magnitude of a
and wave optics. Moreover, Maxwell’s equations are the source or sink of the field at a given point. The curl of a
first example of a gauge theory (which is commonly used in vector field (∇ × A) results in a vector field that describes
particle physics) and are the starting point for Einstein’s the infinitesimal rotation at any point in the field. Both
theory of special relativity, both historically and in many quantities are defined locally: they only describe the
curricula. It is therefore desirable that students have a characteristics of a vector field at a single point. This is
profound understanding of these equations. the most important distinction from Maxwell’s equations
For physics education researchers, an electrodynamics in integral form, which describe the electromagnetic
course is an ideal “laboratory” to explore the link between field in a region of space. For physics majors, the two
mathematics and physics since students have acquired formulations are equally important since they have both
knowledge of the physical concepts in an introductory certain advantages and limitations in specific contexts.
electricity and magnetism course, and have learned the The research described in this paper focuses on students’
necessary mathematical techniques during instruction on understanding of Maxwell’s equations in differential form
calculus. To describe more complex electromagnetic phe- and includes their knowledge and skills concerning vector
nomena, they will need to apply this advanced mathematics calculus.
in the description of the physical reality. This study has In Sec. II we provide an overview of the related
charted some problems our students encounter with physics literature, including work on the link between mathematics
and mathematics when using Maxwell’s equations. and physics: research on electromagnetism, vector calcu-
lus, and the combination of the two. This leads to our
research goals, which are described in Sec. III. To formulate
*
[email protected] our research questions, we distinguish four kinds of skills
†
[email protected] that need to be acquired by the students:
‡
[email protected] (1) Structural understanding [1,2] of the concepts of
gradient, divergence. and curl.
Published by the American Physical Society under the terms of
the Creative Commons Attribution 3.0 License. Further distri- (2) The interpretation of these operators in the context of
bution of this work must maintain attribution to the author(s) and a graphical representation of the field.
the published article’s title, journal citation, and DOI. (3) Doing calculations that involve vector operators.

1554-9178=15=11(2)=020129(14) 020129-1 Published by the American Physical Society

BOLLEN, VAN KAMPEN, AND DE COCK PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

(4) Conceptual understanding of Maxwell’s equations [46,51–53]. However, little is known about situations where
in differential form. the del operator is applied to scalar or vector fields. Gire and
The educational context and methodology are described in Price discussed the option to use graphical representations
Sec. IV, followed by the results of our study in Sec. V. This when teaching about vector fields and vector calculus.
section contains a discussion of our students’ skills and Based on their experience with different types of in-class
difficulties concerning the four topics that are listed above. activities, they argued that algebraic representations are
The most important findings and possible implications for useful since they can easily be manipulated, but students
teaching are summarized in Sec. VI. gain more insight into the differences between components
and coordinates when using a graphical approach.
II. RELATED LITERATURE Moreover, they expect that students will benefit from being
able to translate one representation to another [54]. Singh
The majority of physics education research (PER) at the and Maries report that about half of their graduate students
university level to date concerns introductory courses before instruction, and one out of three after instruction, are
(examples can be found in the summary of McDermott unable to determine where the divergence or curl is (non)
and Redish [3]). These studies have, among other things, zero when provided with a graphical representation of a
yielded an extensive inventory of conceptual problems in vector field. They argue that physics courses are often a
physics and the finding that many students struggle with the missed learning opportunity because they strongly focus on
application of their mathematical knowledge in a physical mathematics but fail to develop a functional understanding
context [4–6]. The role of mathematics in physics educa- of the underlying concepts [55].
tion has been an important topic in recent PER projects While there is quite some physics and mathematics
[6–12]. Manogue and Dray went so far as to state that education research on vector calculus, the amount of
physicists and mathematicians speak a different language, research on vector calculus in the context of electrody-
but use the same vocabulary [13]. A tendency for students namics is limited. The educational setting in this context,
to focus on equations and calculations rather than on the however, is different. Manogue and Dray pointed out that in
physical meaning behind the symbols has been identified as mathematics the gradient, divergence, and curl are used in a
a recurring issue [14–16]. Another source of difficulties is general and abstract way, while in physics they are mostly
the use of multiple representations in physics and math- used in certain symmetries (Cartesian, cylindrical, or
ematics. Students have severe difficulties combining the spherical) [56]. Research at the University of Colorado
information in texts, equations, symbols, graphs, and showed that problems arise when asking students to
figures into a single unambiguous story [17–22]. Our work determine where the divergence of an electric field vanishes
adds to the investigation of these issues in the context of an for a given charge distribution. This type of question can be
intermediate electrodynamics course. solved with the differential form of Gauss’s law in a
Introductory electricity and magnetism courses are a straightforward way. However, they report that only 26%
popular setting to learn about student misconceptions of their students were able to give a correct answer [57].
[23–35]. A fair amount of research has been carried out on Baily and Astolfi found that what students from St.
the use of integrals in electricity and magnetism [36–44]. This Andrews learned about the divergence in one context
research informs the work we present here, since we have (Gauss’s law) often did not translate to their understanding
adopted some of the ideas and methodologies in these papers in other contexts (e.g., the continuity equation) [58].
as a starting point for our own research. At Dublin City In summary, the literature reviewed here shows that
University students’ ideas about integrals were investigated students struggle when they have to use their knowledge
using an approach based on the idea of the concept image, i.e., from mathematics in a physical context. Clearly, this also
all the mental processes activated when students encounter a applies to the specific case of applying vector calculus in
certain concept (e.g., an integral of a vector operator) [36]. It is electrodynamics. However, there is still a lot of research to be
unique for every person, and therefore differs from the done on the subject. In the next section, the contribution of our
(formal) concept definition, which is a description that is study is formulated in terms of goals and research questions.
accepted by the wider community [45]. One aspect of our
study concerns our students’ concept image of vector
III. RESEARCH DESIGN
operators, which we relate to the difficulties they encounter
when applying Maxwell’s equations in differential form. This paper gives an account of an exploratory study of
In advanced courses (often called electricity and magnet- students’ strengths and weaknesses in using vector calculus
ism2 or electrodynamics) vector calculus plays an important in mathematical and physical contexts. The research
role. It is known that operations with vectors and vector extends the previous findings mentioned in Sec. II by
fields (e.g., vector addition and the dot product) provide adopting a broader approach to ascertain the knowledge,
students with many problems [46–51]. Furthermore, stu- skills, and understanding our students have acquired. The
dents struggle with the use of vectors in different coordinate goal of this study is twofold: it aims to provide both
systems and the application of appropriate unit vectors researchers and teachers with insights into the learning

020129-2
STUDENTS’ DIFFICULTIES WITH VECTOR … PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

results of traditional instruction of electrodynamics, and it traditional thirteen week intermediate electrodynamics
is the first stage of a large scale investigation of students’ course. The students major in physics or mathematics at
understanding of Maxwell’s equations in differential form. the KU Leuven. They use Griffiths’ textbook [59] and are
In this first stage of the investigation we aim to gain instructed in one two-hour lecture and one two-hour
insight into the difficulties students encounter with vector problem solving session per week in which they discuss
calculus in a purely mathematical or physical context. To typical end of chapter problems from the textbook. The
this end, we distinguish four different kinds of skills and students have already completed an introductory electro-
competencies students need to acquire: structural under- magnetism course using the textbook of Serway and Jewett
standing [1,2] of divergence and curl, graphical interpre- [60] that leads up to Maxwell’s equations in integral form,
tation of vector fields, calculation of divergence and curl, and at least two calculus courses [61] that include a chapter
and conceptual understanding of Maxwell’s equations in on vector calculus. Therefore, they have encountered the
differential form. Based on their instruction prior to the necessary mathematical tools and physical situations pre-
electrodynamics course and results from the literature, we sented in the electrodynamics course.
expect our students to have reasonable facility with the To identify the prior knowledge of our students they
mathematical techniques needed to carry out calculations were given a pretest before the first lecture in the advanced
while lacking experience with interpreting graphical rep- electromagnetism course based on Griffiths’ textbook [59].
resentations of vector fields. We have investigated our Since these students had encountered vector calculus
students’ attainment at the start of the electrodynamics mostly in a mathematical setting, the questions on the
course and have tried to establish to what extent their pretest do not contain any physical context. To encourage
understanding of Maxwell’s equations in differential form students to write down their reasoning, calculations, and
changes while taking the course. We have focused on the thinking process, all questions were open ended. A post-
following research questions: test was given after instruction on chapters 1–7 of Griffiths’
• Did our students acquire a structural understanding textbook [59], during a lecture about halfway through the
[1,2] of gradient, divergence, and curl from their semester. The post-test assignments are mostly similar to
introductory and intermediate mathematics courses? those on the pretest; however, a physical context is
(i) What is their concept image [45] of gradient, introduced in some cases to investigate whether informa-
divergence, and curl? tion on the physical situation affects the students’ ability to
(ii) How do they describe the meaning of the vector interpret the divergence and curl of the (electromagnetic)
operators? vector fields. The post-test also comprises questions that
• Can students interpret a graphical representation of a evaluate students’ understanding of Maxwell’s equations.
vector field in terms of its divergence and curl? There were no time constraints for the students to complete
(i) Can they deduce where the divergence and the pre- and post-test.
curl of vector fields are (non)zero in a purely The analysis focuses on the solution method and think-
mathematical context? ing process rather than the result. To describe and explain
(ii) Can they deduce where the divergence and curl the variation in students’ conceptions, ideas of phenom-
of electromagnetic fields are (non)zero? enography are used. Phenomenography is an empirical
(iii) What strategies do they use to interpret these approach that aims to identify and categorize the different
representations? qualitative ways in which different people perceive and
• Did our students acquire the necessary mathematical understand phenomena [28,63]. The categories used in the
techniques to perform calculations involving vector
analysis of our data were established in a bottom-up
operators with and without a physical context?
approach where one of us proposed a set of categories
(i) What technical difficulties do they encounter?
based on the answers students gave, the strategies they
(ii) Do different kinds of coordinate systems
used, and the mistakes they made. After an elaborate
(Cartesian, cylindrical, spherical) present differ-
discussion with the other collaborators about some specific
ent challenges?
student answers, we refined our classification and decided
• Do students conceptually understand Maxwell’s equa-
on a final set of categories. To confirm that our categories
tions in differential form?
are well defined, we evaluated the interrater reliability
(i) Are they able to correctly deduce whether the
by calculating Cohen’s kappa (κ). For individual questions,
divergence and curl of an electromagnetic field
are zero or nonzero in a given situation? Cohen’s kappa ranged from 0.76 to 1.00, indicating a
substantial to almost perfect agreement. Since the number
of students is limited (N ¼ 30 on the pretest and N ¼ 19 on
IV. EDUCATIONAL CONTEXT
the post-test), the percentages should be generalized with
AND METHODOLOGY
care. Nevertheless, they should give a clear view of the
To answer the research questions, we gave written paper- limitations in our students’ understanding of vector calcu-
and-pencil questions to second year university students in a lus in electrodynamics.

020129-3
BOLLEN, VAN KAMPEN, AND DE COCK PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

V. RESULTS AND DISCUSSION We do not suggest that students do not know something
they did not write, but we do think the question reveals
In this section the results of the pre- and post-test are
what is cued first and foremost.
presented and discussed. The questions can be found in
Only a few students gave a description of the operators
Appendixes A and B.
we deemed conceptual. Some provided a more or less
correct description that resembles the concept definition:
A. Pretest
The pretest shown in Appendix A was given to all 30 “The curl tells you how strong and which way the vector
students at the beginning of the course to probe their field A rotates.”
knowledge and understanding acquired in previous
This student did not make a statement about the local
courses. The first part of the pretest identifies students’
character of the curl, but does seem to have a rather good
concept images [45] of the operators grad, div, and curl. In
idea about what the curl represents. Others had very
the second part, the students’ calculational skills and their
incomplete or incorrect conceptual ideas:
ability to interpret graphical representations of vector fields
are tested. For this part only, some useful formulas were
“The divergence is a measure for how the field is
attached to the questions (Appendix C).
changing.”
The three questions on the pretest correspond to the first
three research questions that were discussed above. Since
our students only studied Maxwell’s equations in integral “The gradient of A is the vector normal to the plane.”
form during their introductory course, we did not include a
question that assesses their understanding of the differ-
ential form. We also observed that some students misidentified the
vector or scalar character of the expression. More than half
of the students mentioned which operations result in a
1. Concept image of grad, div, and curl vector field, and which produce a scalar field. Sometimes
The concept image question serves to get a better they explicitly wrote it down, in other cases it could be
understanding of what students associate with the gradient, derived from their notation. The notation in this student’s
divergence, and curl in a very general sense. Expressions answer, for example, shows he thinks the divergence of a
for grad, div, and curl are given to the students, and they vector field is a vector:
are asked to write down everything they think of. From
this, we can make some statements about the students’ “In three dimensions this is the divergence and, there-
concept image [36,45,64] of these operators. The students’ ~ ·A
fore, ∇ ~ ¼ ∂ A=∂x
~ ~
þ ∂ A=∂y ~
þ ∂ A=∂z”
responses to the questions are described qualitatively
in Table I. We distinguished three important emerging One of our students wrote that the gradient of a scalar is
categories in the students’ answers: information about again a scalar. Two students also seemed to think the
the structural meaning of the vector operators, the scalar divergence of a vector field is again a vector field. We did
or vector character of the expression, and the name and not observe a single misidentification of the vector char-
symbolic expression that students wrote down. Obviously, acter of the curl of a vector field. This corroborates the
a student can give more than one interpretation and, findings of Barniol and Zavala, who showed that students
therefore, the percentages sum to more than 100%. have significantly more problems with the vector or scalar
nature of the dot product than of the vector product
[51]. Since the students correctly described A as being a
scalar (field) and A as being a vector (field), we have no
TABLE I. Categorization of students’ interpretation of the indications that there was a problem with the notation in the
expressions ∇A, ∇ · A, and ∇ × A. question. About one out of three students did not make any
Category ðN ¼ 30Þ ∇A ∇·A ∇×A
statement concerning the vector or scalar character in their
answer (e.g., they just named ∇ · A “divergence,” without
Correct concept 10% 0% 10% any explanation).
Incorrect or incomplete concept 23% 10% 7% The category “Naming” contains all students who wrote
Scalar 3% 63% 0% down the correct name of the expression. No students
Vector 60% 7% 53% remembered names incorrectly or mixed up the terms
Naming 70% 50% 53% gradient, divergence, and curl. Nevertheless, half of the
Formula 63% 50% 37% students did not explicitly identify ∇ · A as the divergence
Other 3% 3% 7% and ∇ × A as the curl of a vector field. Only five of our
students wrote that ∇ is the nabla symbol, and one student
No answer 3% 10% 10%
called it the del symbol. About one out of four, however,

020129-4
STUDENTS’ DIFFICULTIES WITH VECTOR … PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

called this symbol the gradient or the Laplacian. In some that they could not decide whether the divergence was zero,
cases it was not possible to determine whether students had we deemed their answer correct.
the concepts confused, or the names, or both: First the students’ answers were checked for correctness.
Figure 1 shows that our students have severe difficulties
“∇ × A is the vector product between the gradient with these graphical representations. For Fig. 4, half of the
and A.” students gave a correct answer, but for the other three vector
fields less than one out of four figured out correctly where
Furthermore, about half of the students wrote down a the divergence and curl are nonzero. In Fig. 7, just two
formula from memory, some incorrectly; all of these are students could determine that the divergence is nonzero
counted in the category “formula.” In the category “other” only at the center of the field. About 30% of the students
there were some correct statements that explained the link made at least one statement that pointed toward the typical
with Stokes’ law, the divergence theorem, or conservative error [57] of confusing the derivative of the field with its
fields. These students seem to have made connections with value (e.g., the derivative is zero when the field is zero). Our
the integral form of Maxwell’s equations. students were easily misdirected, and very inconsistent in
On the whole, students seemed to feel these vector
their reasoning. Moreover, a significant number of students
operators are a tool to evaluate something. A similar focus
did not answer the question (about 20%–30% for the
on evaluation was also seen in students’ concept image of
divergence and 30%–40% for the curl). It is likely that
integration. In that particular case, students rather tried to
these students did not know how to solve these problems,
evaluate an integral that was impossible to calculate than
since they did answer the other pretest questions.
to describe it as an area under a curve or a sum of
Second, we looked at the strategy students used to obtain
infinitesimal parts [36]. This particular question does not
their answer. The prevalence of the strategies is shown in the
elicit a structural understanding [1,2] of the gradient,
second column of Table II. We distinguish five categories:
divergence, and curl. The second part of the pretest is
• Concept based strategy: This category includes ex-
designed to investigate the students’ ability to interpret
planations from students that show a good under-
graphical representations of vector fields and their skill at
standing of the underlying concepts. Their answers are
doing calculations in vector calculus.
based on drawings together with the definition and
potentially some derived formulas that link the differ-
2. Graphical interpretation of vector fields ential and integral form. Typically the change in flux
We gave our students a two-dimensional representation per unit area is determined by drawing a small box
of four different vector fields and asked them to indicate around a point: if there is no net flux in the area
where the divergence and curl are (non)zero. The assign- bounded by the box, the divergence is zero as there is
ments can be found in Appendix A. The divergence is no source or sink in this area. To obtain the curl, a
nonzero everywhere in Fig. 4 and the curl is nonzero virtual paddle wheel is placed in the field. If it rotates,
everywhere in Figs. 5 and 6. The curl in Fig. 7 clearly is the curl is nonzero at that particular location.
zero everywhere, but determining the divergence is less • Formula based strategy: The student mostly relies on
straightforward. The field we sketched has 1=s depend- “the formula” for divergence and curl, and uses the
ence, so that the divergence is nonzero only at the center derivatives of x and y to get an answer. Some students
of the field. This could, for example, be the electric field of even (try to) obtain an algebraic expression for the
a charged wire pointing in the z direction. However, if vector field and then apply the definition of the
students saw an unspecified dependence on s and stated operator (typically in Cartesian coordinates).

FIG. 1. Results for the question about graphical representations of vector fields on the pretest (N ¼ 30).

020129-5
BOLLEN, VAN KAMPEN, AND DE COCK PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

TABLE II. The prevalence and success rate of strategies students In the last four columns of Table II, the success rate
used to determine the divergence and curl of vector fields based on for certain strategies is given. The student who used the
graphical representations used in the pretest ðN ¼ 30Þ. concept based technique was very successful in determin-
ing the divergence and curl of the fields. The formula based
Number Success rate
Divergence technique is useful if the vector function can be found and
of students Figure 4 Figure 5 Figure 6 Figure 7
calculations are carried out correctly, which may cause
Concept based 3% 100% 100% 100% 100% problems for complex cases [e.g., the fourth vector field
Formula based 23% 71% 57% 57% 0% (Fig. 7)]. Students who used a description based strategy or
Description 23% 71% 14% 14% 0% give little or no explanation seem to have a low chance of
based being successful in determining the divergence and curl of a
Unclear 30% 56% 44% 33% 11% graphical representation of a vector field. There are some
No answer 20%     exceptions to these generalizations, like a student who
determined the divergence and curl correctly for every
Number Success rate field, but gave no explanation whatsoever. However, his
Curl
of students Figure 4 Figure 5 Figure 6 Figure 7 answers were probably well considered, as he gave a fairly
Concept based 3% 100% 100% 100% 100% accurate description of the divergence and curl in the first
Formula based 23% 100% 57% 71% 57% part of the pretest:
Description 20% 100% 17% 33% 67%
based “The divergence of A is a scalar field that tells you how
Unclear 23% 100% 29% 14% 71% much is added to the vector field A”
No answer 30%    

“The curl of A is a vector field that tells you how strong

• Description based strategy: The students use a and in which way the field A turns”
(correct-incorrect-incomplete) qualitative description This leads us to believe that this particular student used his
of the divergence and curl to obtain an answer. conceptual understanding of divergence and curl to tackle
Typically the student relies on the common English problems concerning the graphical representation of vector
definition of the words “divergence” and “curl” and fields.
links this in naive way to the graphical representation
of a vector field. This is illustrated by some examples 3. Calculation of divergence and curl
for the third vector field (Fig. 6):
(i) False descriptions: In the last set of questions on the pretest we asked our
students to calculate the divergence and curl of three vector
“∇ · A ≠ 0 because the length of the arrows fields (see Appendix A). Two fields were given in Cartesian
increases.” coordinates, the third in cylindrical coordinates. For each of
the 6 calculations, we split answers into four categories:
complete and correct calculations, calculations with minor
“∇ × A ¼ 0 because the field is not rotating.” mistakes or omissions (e.g., a forgotten minus sign or an
expression is left unsimplified), calculations with major
(ii) More or less correct descriptions: mistakes (e.g., an error in the use of the formula for div or
curl, an error when taking the derivative or inappropriate
“∇ · A ¼ 0 since nothing is added to the field use of unit vectors), or no answers. The results are
anywhere.” presented in Table III.
Exercise (a) required students to make a quite straight-
“∇ × A ≠ 0 because the field is rotating forward calculation. Nevertheless, only 60% of the students
locally.” were able to calculate the divergence correctly, and,
allowing for minor errors, about three-quarters calculated
It is, of course, possible that students who wrote down the curl correctly. One student calculated the Laplacian
these descriptions have some conceptual insights as instead of the curl. He did not give an answer for the
well, but their answers provided no evidence for this. other parts.
• Unclear: The reasoning is not explained or it is Exercise (b) was more difficult than the first one, since
unclear. some challenging algebra is required to evaluate the
• No answer: The student did not answer the question. expression in Cartesian coordinates. This explains the
Students generally used just one of these strategies to solve higher number of students who make minor errors.
the question. However, some used a different approach to When allowing for minor errors, about 60% of students
determine the divergence and the curl of a field. gave correct answers. These are mostly students that could

020129-6
STUDENTS’ DIFFICULTIES WITH VECTOR … PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

TABLE III. Categorization of students’ calculations of the in the concept image of the students: most of our students
divergence and curl of three vector fields in the pretest (N ¼ 30). lack a conceptual understanding of the divergence and curl,
and focus on evaluation.
Exercise (a) Exercise (b) Exercise (c)
Divergence Curl Divergence Curl Divergence Curl B. Post-test
Correct 60% 53% 23% 43% 60% 57% After instruction up to Chap. 7 in Griffiths’ textbook
Minor error 0% 20% 33% 13% 7% 3% [59], we gave the students a post-test questionnaire (see
Major error 33% 16% 33% 3% 7% 7% Appendix B). It comprises two questions concerning
No answer 7% 10% 10% 40% 27% 33%
graphical interpretation of vector fields (one with and
one without physics context), two questions where students
have to calculate the divergence and curl after imposing a
also correctly calculate the divergence and curl in the first condition, and two conceptual questions in which they had
exercise. Two students converted the equation to polar to use the differential form of Maxwell’s equations to
coordinates and then calculated the divergence and curl. interpret a series of physical situations. Therefore, the
One of them knew the formula for the divergence in two questions on the post-test correspond to the last three
dimensions [65] (only the formulas for three dimensions research questions in Sec. III. We did not include a question
were given), the other student made a mistake at this point. that aims to examine the concept image of the divergence,
Both of them noticed that the curl is zero without doing any curl, and gradient because we wanted to exclude the
calculation at all. Two other students did not calculate the possibility of a retest effect and to limit the workload for
curl in this part, because they argued that a vector product is the students. Since the number of attendants in the non-
only defined in three dimensions. It may not have occurred mandatory lecture dropped over the semester, only 19
to them that the vector field could be considered three- students filled in the post-test. All of these participants also
dimensional with zero z component. While almost all took the pretest. Based on the pretest data, the population of
students attempted to calculate the divergence, over one- students that took the post-test is equivalent to the pop-
third of the students did not attempt to calculate the curl. ulation of students that took the pretest.
They may have been discouraged by difficulties they had
when calculating the divergence. 1. Graphical interpretation of (EM) vector fields
The calculation required in exercise (c) is as straightfor- Vector fields without physical context.—This question is
ward as that of exercise (a), and a similar fraction of similar to the question 1 of part 2 on the pretest. Figures 8
students calculated the divergence and curl correctly. and 9 on the post-test are analogous to Figs. 6 and 4 on the
Again, this is more or less the same group of students pretest, respectively (however, the “view” is changed a bit).
that could do the calculations in the first two exercises. The results are summarized in Fig. 2 and are compared to
However, many more students did not give an answer at the answers on the pretest.
all. It is unlikely that they did not know how to calculate It seems that students did better with the first vector field
the divergence and curl in cylindrical coordinates, since of the post-test, but made a more or less equal number of
expressions were given to them. mistakes when interpreting the second vector field. This
In general, we can see that approximately 60% of the was analyzed more profoundly by looking at how many
students are able to calculate the divergence and the curl of students’ answers improved (incorrect at pretest; correct
given vector fields, independent of the level of difficulty in post-test), disimproved (correct at pretest; incorrect at
and the coordinate system used, if we allow minor errors. post-test), and stayed the same. For the first field, three
The major errors can be classified in three subcategories: 10 students could correctly determine divergence and curl in
times an error was made when taking the derivative, 11 the post-test, but not in the pretest. Not a single student
students used the expressions incorrectly, and in 10 cases made a “new” mistake. For the second field, four students
unit vectors were used inappropriately (e.g., unit vectors improved their answers, but two went from answering
were appended to terms in the divergence of a vector field). correctly to answering incorrectly. This means that most
Of course a single student could make multiple errors students stick to their answers: there is a slight increase in
during one calculation. Concerning the use of unit vectors, correct answers, but still only about 50% can determine the
it was striking to see that students used them very divergence and curl from a graphical representation of a
inconsistently in the pretest. simple vector field. Note that these percentages are similar
When we compare the results of the calculations to the to the results that Singh and Maries found when testing
number of correct answers in the graphical representation their graduate students (before instruction) [55].
question, the prior knowledge of these students clearly However, four students made incorrect statements like
shows. They seem to have some difficulties calculating the
divergence or curl, but struggle much more with exercises “The divergence is zero in the x direction, but not in the
that ask for more insight. This confirms what we observed y direction.”

020129-7
BOLLEN, VAN KAMPEN, AND DE COCK PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

FIG. 2. Results for the context-free graphical representation of vector fields on the post-test ðN ¼ 19Þ compared to the results on the
pretest ðN ¼ 30Þ. The first vector field on the post-test (Fig. 8) should be compared to Fig. 6 on the pretest, and Fig. 9 to Fig. 4.

This kind of reasoning was not seen in the graphical pretest which may explain that formula based reasoning is more
questions, though it did emerge in the calculational pretest popular and effective in the post-test.
questions. We intend to explore this issue further in the
future; it illustrates in any case that many students still Electromagnetic fields.—The post-test questions on
struggle with divergence. graphical representations of electromagnetic fields ask
Looking at the strategies the students used to determine similar questions in context. Furthermore, the fields are
the divergence and curl (Table IV) we see an increase in a bit more “difficult” in the sense that they have a
concept based reasoning and formula based reasoning. The cylindrical instead of a Cartesian symmetry. The number
number of unclear answers decreased and every student at of correct answers is very small: only a few students could
least tried to give an answer this time. This effect might be correctly determine both the divergence and the curl
due to instruction: in lectures they were told that it is (Fig. 3).
possible to use a paddle wheel to determine the curl, for To solve this question, students could use the same
example. However, students were not asked to use this idea strategies as before, but could also use Maxwell’s equations
in any exercises, which may explain the many errors they (physics based reasoning). Because some students used
made. The students did a lot of calculations during the such an argument to confirm their answer based on another
tutorial sessions and the fields are fairly straightforward, strategy, it is now possible that they are entered in multiple
categories. The answers are summarized in Table V. The
first part shows the percentage of students that used a
TABLE IV. The prevalence and success rate of strategies certain approach. Some students used a generic approach
students used to determine the divergence and curl of vector together with a physics based strategy, so the total exceeds
fields based on graphical representations used in the post-test 100%. The second part of the Table shows the success rate
ðN ¼ 19Þ. when students use a certain strategy.
Success rate
Divergence Number of students
Figure 8 Figure 9
Concept based 16% 67% 67%
Formula based 37% 100% 100%
Description based 26% 0% 20%
Unclear 21% 0% 0%
No answer 0%  

Success rate
Curl Number of students
Figure 8 Figure 9
Concept based 21% 50% 50%
Formula based 32% 100% 100%
Description based 16% 0% 33%
Unclear 32% 17% 17%
FIG. 3. Percentage of correct answers for the graphical repre-
No answer 0%  
sentation of electromagnetic fields in the post-test ðN ¼ 19Þ.

020129-8
STUDENTS’ DIFFICULTIES WITH VECTOR … PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

TABLE V. Strategies students used to interpret the divergence try to set up the equation of the field in Cartesian
and curl of graphical representations of electromagnetic fields in coordinates.
the post-test ðN ¼ 19Þ.

Figure 10 Figure 11 2. Calculation of divergence and curl of EM fields

Prevalence
Divergence Curl Divergence Curl These questions are intended to check if students can do
Concept based 5% 16% 0% 0% calculations in an electromagnetism context, which takes
Physics based 42% 11% 21% 26% the form of imposing a simple condition. To determine
Formula based 16% 11% 5% 0% which field could be a magnetic field, students should
Description based 16% 16% 11% 16% check if ∇ · B ¼ 0 applies; to check whether a field could
Unclear 26% 42% 53% 58% be an electrostatic field, they need to verify that ∇ × E ¼ 0.
No answer 0% 5% 11% 5% Recognizing and imposing this condition proved to be
problematic for our students: 74% (14) students were able
Figure 10 Figure 11 to do this for the magnetic field and only 53% (10) for the
Success rate
Divergence Curl Divergence Curl electric field. One student did not give answers to any of
Concept based 100% 0%  
these questions. Others used qualitative reasoning that
Physics based 100% 50% 50% 100% contained something about the radial or z dependence:
Formula based 67% 0% 0% 
Description based 67% 0% 0% 100% “A magnetic field spreads radially outward from its
Unclear 17% 0% 10% 91% point of origin. The first one doesn’t do that because it is
in Cartesian coordinates.”
A few students calculated the divergence of the potentially
Students tend to change their strategy between questions electrostatic fields, but then struggled to interpret the result.
(this is why both questions are treated separately) and often Almost all students who obtained the correct condition
fail to correctly apply the strategies they use. They seem calculated the divergence and curl correctly. Some even did
especially unsure about the use of Maxwell’s equations. To not need to calculate a full expression, but could determine
illustrate this, we analyze the answer of a student for the whether the divergence and curl were (non)zero by sight.
magnetic field question (Fig. 10): Despite the observation that some students still made
errors, a slight progression could be noticed concerning
Concerning the divergence, this student states correctly the ability to perform calculations. This may be explained
it is zero: “∇ · B ¼ 0 → there is no magnetic monop- by the huge emphasis on calculations in Griffiths’textbook
ole”, which is indeed always true. When looking at the [59] and the exercise sessions.
curl, he writes “∇ × B ¼ μ0 I”, but then crosses out
the right-hand side, and simply writes “nonzero”. In the
end, it is unclear what argument he used to obtain this 3. Conceptual understanding of Maxwell’s equations
incorrect result. The last set of post-test questions intends to investigate
students’ conceptual understanding of and insight into
Many students did similar things: they tried to use Ampère’s
Maxwell’s equations in differential form. To this end they
law, but failed to apply it correctly. They did not appear to
had to determine whether the curl and divergence are zero
understand that the curl of a magnetic field is only nonzero
where a current flows, and that it varies from point to point. or not in four electric and five magnetic fields [66]. When
Furthermore, they misinterpreted the use of a paddle wheel: Maxwell’s equations in differential form are applied cor-
they seemed to think it rotates everywhere, but it does so rectly in every situation, one obtains the following answers:
only in the center of the field (everywhere else it translates in ∇ · E ¼ 0 always except for the first situation, ∇ × E ¼ 0
a circle around the current carrying wire). Similar mistakes in situations a, c, and d, ∇ · B ¼ 0 always, and ∇ × B ¼ 0
were made in the case of the electric field (Fig. 11): no more in the last two situations. We did not give the students a list
than three students understood that the divergence is nonzero of Maxwell’s equations to avoid pointing students in a
only where charges are present. The most occurring mistake particular direction. In our opinion students who under-
(37%) was that students thought the divergence is nonzero stand these laws will be able to reproduce at least the
everywhere because of the appearance of the field: causality between the fields and sources. However, as
shown in Table VI, students encountered tremendous
“All arrows point towards a certain point, so the difficulties in answering this question.
divergence is zero nowhere.” Only one student did not make a single mistake, while all
other 18 students made at least three errors. We could not
Formula based reasoning is less effective here, because the find a correlation between the errors, but some patterns did
students struggle with the use of cylindrical coordinates or emerge:

020129-9
BOLLEN, VAN KAMPEN, AND DE COCK PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

TABLE VI. The percentage of students who could correctly incomplete and contained incorrect information. Some
determine whether the divergence and curl of the electric students were confused about the vector or scalar character
(magnetic) field described in a situation is zero or not ðN ¼ 19Þ. of the operators, and used incorrect terminology.
Electric field ∇·E ∇×E
Interpreting graphical representations of vector fields is a
difficult exercise for students. Even after instruction, only
Situation (a) 47% 95% half of the students were able to determine where the
Situation (b) 16% 53% divergence and curl of a simple vector field are (non)zero.
Situation (c) 26% 74%
When more complex and realistic electromagnetic fields
Situation (d) 95% 95%
had to be considered, only a few students succeeded in
Magnetic field ∇·B ∇×B
solving the question correctly. Moreover, many students
used various strategies inconsistently. This suggests they
Situation (a) 63% 79% lack a structural understanding of the mathematical con-
Situation (b) 74% 79%
cepts, and on top of this they are unable to use their
Situation (c) 74% 42%
Situation (d) 84% 74%
acquired skills in a physical context.
Situation (e) 84% 68% Since we (and many others [54,55,57]) believe that these
graphical representations are helpful when trying to con-
ceptualize the abstract mathematical structures in vector
• Gauss’ law, which states that the divergence of the calculus, we think it would be advisable to put more effort
electric field is nonzero only where charge is present, into this kind of exercise in both physics and mathematics
or more conceptually still, that the source (sink) of an instruction. In our opinion it would help students to
electric field is a positive (negative) electric charge, understand the physical meaning of Maxwell’s equations,
elicited most errors. Only one student could correctly which has applications beyond electricity and magnetism—
determine where the divergence is nonzero for all five e.g., in subsequent problems concerning electromagnetic
situations. This confirms some of the findings from radiation, gauge theory, and the introduction to special
the University of Colorado: students have diffi- relativity.
culties applying the divergence in an electromagnetic In Griffiths’ textbook [59] a lot of exercises focus on
context [57]. complex calculations. We found that students are reason-
• At least one mistake was made in the application of ably comfortable with the required algebra, but have
Faraday’s law by 53% (10) of students. Nine students problems when they need to interpret the context of a
did not appear to know that the curl of an electric field calculation. One out of four students was unable to come up
is nonzero when there is a changing magnetic field, as with the condition a vector field should satisfy in order to
stated explicitly in situation (b). be a realistic magnetic field, and only half of them knew
• Despite the elaborate discussion on the nonexistence this condition for an electrostatic field. Since understanding
of magnetic monopoles during instruction, 53% (10)
and explaining electrodynamic phenomena is one of the
of students did not check every box under ∇ · B. This
main objectives of this course, we suggest more attention
is alarming, since it is a truly elementary law that is
should be paid to the interpretation of the equations and
easy to apply.
setting up the problem at the expense of doing algebraic
• Only 21% (4) of students were able to correctly
manipulations.
evaluate Ampère-Maxwell’s law in every situation.
When we investigated the competencies students show
Strikingly, fewer than half of the students could
when they encounter situations that can be solved using
interpret situation (c), which is the classic textbook
Maxwell’s equations in differential form, we observed that
example to show Maxwell’s correction to Ampère’s
law. It was discussed both during the lectures and in students had tremendous difficulties with the application of
Griffiths’ textbook [59]. all four laws. This calls for an instruction that puts more
These results show that our students did not profoundly effort in linking mathematics and physics and uses a more
understand Maxwell’s equations in differential form. qualitative approach. Some great ideas can be found in
Huang et al. [67], although we think even more graphical
and conceptual examples are needed in order to fully show
VI. CONCLUSIONS AND IMPLICATIONS
FOR TEACHING students the power and usefulness of Maxwell’s equations
in differential form.
We have investigated students’ understanding of diver- In future work we are planning to conduct student
gence and curl in mathematical and physical contexts. interviews to gain more insight in the thinking process
Concerning their initial concept image, we found that they of students when they solve problems linked to the differ-
focused on evaluation, and appeared to pay little attention ential form of Maxwell’s equations. This will help us
to the conceptual meaning of the vector operators. understand how graphical representation and a better
Furthermore, their conceptual descriptions often were structural understanding of the mathematical concepts

020129-10
STUDENTS’ DIFFICULTIES WITH VECTOR … PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

can help students to apply their skills in a physical context.

At a later stage the results of these interviews will be used
to create new or improved questions on the pre- and post-
test, and to iteratively design a tutorial that aims to help

y
students understanding Maxwell’s equations in differen-
tial form.

ACKNOWLEDGMENTS x

We gratefully acknowledge fruitful discussions with FIG. 5. Pretest field 2.

Charles Baily and thank him for his valuable comments
on the paper. We greatly appreciate the cooperation with
Wojciech De Roeck and the students who made this study
possible.

APPENDIX A: PRETEST QUESTIONS

y
The first part was given to the students without any
expressions for div, grad, and curl. After they finished and
turned in Part 1, they were given Part 2 which contained
x
the expressions given in Appendix C. Some space was left
blank for the students to answer after each question. Figures FIG. 6. Pretest field 3.
are displayed smaller than in the actual test.

1. Part 1
(1) Interpret (i.e., write down everything you think of
when you see) the following operations.
(a) ∇A
(b) ∇ · A
y

(c) ∇ × A

2. Part 2 x

(1) Indicate where the divergence and/or curl is (non)

zero for the next vector fields in the x; y plane. The z FIG. 7. Pretest field 4.
component is zero everywhere. Explain and show
your work (Figs. 4–7).
(2) Calculate the divergence and curl of the following APPENDIX B: POST-TEST QUESTIONS
vector fields.
(a) va ¼ x2 êx þ xêy − 2xzêz The expressions of Appendix C were appended to these
xê þyê questions. Some space was left blank for the students to
(b) vb ¼ ðx2 þy
x y
2 Þ3=2 answer after each question. Figures are displayed smaller
(c) vc ¼ðr=2;rθ;−zÞ (Hint: cylindrical coordinates) than in the actual test.
y

x x

FIG. 4. Pretest field 1. FIG. 8. Post-test field 1.

020129-11
BOLLEN, VAN KAMPEN, AND DE COCK PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

(4) Which of these equations could represent a realistic

static electric field (E0 is a constant with the
appropriate units)? Explain.
(a) Ea ¼ E0 ½ðz − xÞx̂ þ ðz þ xÞŷ þ xẑ

y
(b) Eb ¼ E0 ½sð2 þ sin2 ϕÞŝ þ s sin ϕ cos ϕϕ̂ þ 3zẑ
(5) Check the box(es) if ∇ · E and/or ∇ × E are equal
to zero.
x
∇·E ¼ 0 ∇×E ¼ 0
FIG. 9. Post-test field 2. (a) The electric field at a distance r<R □ □
from the center of a uniformly
charged sphere with radius R.
(b) The electric field generated by a □ □
changing magnetic field.
(c) The electric field at a distance r □ □
from a pure electric dipole.
(d) The electric field inside a charged □ □
y

conductor.

(6) Check the box(es) if ∇ · B and/or ∇ × B are equal

to zero.
x
∇·B ¼ 0 ∇×B ¼ 0
FIG. 10. Post-test field 3.
(a) The magnetic field generated by a □ □
changing electric field
(b) The magnetic field at a distance □ □
r < R from the axis of a
cylindrical conductor with radius
R carrying a steady current.
(c) The magnetic field between the □ □
y

plates of a charging capacitor.

(d) The magnetic field inside a □ □
solenoid with a steady current
passing through it.
x (e) The magnetic field at a distance r of □ □
a large conducting plate carrying a
FIG. 11. Post-test field 4. steady surface current density K.

(1) Indicate where the divergence and/or curl is (non)

zero for the following vector fields in the ðx; yÞ APPENDIX C: FORMULAS
plane. The z component is zero everywhere. Explain This section contains the formulas that were handed to
and show your work (Figs. 8 and 9). the students with the pretest (part 2) and post-test. In the
(2) For the following physical situations, explain where pretest, these equations were given using the notation the
the divergence and/or curl of the field are (non)zero. students learned in their calculus courses. In the post-test
The z component of the fields is zero everywhere. we used Griffiths’ notation [59]. Only the latter are
Show your work (Figs. 10 and 11). presented here.
∂vy ∂vz
(a) The magnetic field of an infinite current carrying • ∇ · v ¼ ∂v
∂x þ ∂y þ ∂z
x

wire along the z axis. ∂vy ∂v

(b) The electric field of a charged infinitely long • ∇ × v ¼ ð∂v
∂y −
z
∂z Þx̂ þ ð∂v ∂vz ∂vx
∂z − ∂x Þŷ þ ð ∂x − ∂y Þẑ
x y

cylinder with radius R. In the figure, the cross ∂vϕ

section in the x; y plane is given. • ∇ · v ¼ r12 ∂r
∂ 1 ∂
ðr2 vr Þ þ r sin 1
θ ∂θ ðsin θvθ Þ þ r sin θ ∂ϕ
(3) Which of these equations could represent a realistic 1 ∂ ∂vθ 1 1 ∂vr ∂
• ∇ × v ¼ r sin θ ½∂θ ðsin θvϕ Þ − ∂ϕ r̂ þ r ½sin θ ∂ϕ − ∂r ðrvϕ Þ
magnetic field (B0 is a constant with the appropriate θ̂ þ 1r ½∂r
∂ ∂v
ðrvθ Þ − ∂θr ϕ̂
units)? Explain. ∂vϕ ∂vz
(a) Ba ¼ B0 ½4xyx̂ − y2 ŷ þ ðx − 2yzÞẑ • ∇ · v ¼ 1s ∂s
∂
ðsvs Þ þ 1s∂ϕ þ ∂z
∂vϕ
(b) Bb ¼ B0 ½r̂ þ 4r2 θ̂ − 2 sinðθÞϕ̂ • ∇×v ¼½1s ∂v ∂vs ∂vz 1 ∂ ∂vs
∂ϕ − ∂z ŝþ½ ∂z − ∂s ϕ̂þ s ½∂s ðsvϕ Þ− ∂ϕ ẑ
z

020129-12
STUDENTS’ DIFFICULTIES WITH VECTOR … PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

[1] A. Sfard, On the dual nature of mathematical conceptions: [21] M. De Cock, Representation use and strategy choice in
Reflections on processes and objects as different sides of physics problem solving, Phys. Rev. ST Phys. Educ. Res.
the same coin, Educ. Stud. Math. 22, 1 (1991). 8, 020117 (2012).
[2] J. Tuminaro, Ph.D. thesis, University of Maryland, 2004. [22] J. F. Wagner, C. A. Manogue, J. R. Thompson, N. S.
[3] L. C. McDermott and E. F. Redish, Resource Letter: Rebello, P. V. Engelhardt, and C. Singh, Representation
PER-1: Physics Education Research, Am. J. Phys. 67, issues: Using mathematics in upper-division physics, AIP
755 (1999). Conf. Proc. 1413, 89 (2012).
[4] M. Artigue, J. Menigaux, and L. Viennot, Some aspects of [23] D. P. Maloney, T. L. OKuma, C. J. Hieggelke, and A. Van
students’ conceptions and difficulties about differentials, Heuvelen, Surveying students conceptual knowledge of
Eur. J. Phys. 11, 262 (1990). electricity and magnetism, Am. J. Phys. 69, S12 (2001).
[5] F. R. Yeatts, Calculus and physics: Challenges at the [24] D. P. Maloney, Charged poles?, Phys. Educ. 20, 310 (1985).
interface, Am. J. Phys. 60, 716 (1992). [25] B. W. Adrian and R. G. Fuller, A qualitative investigation
[6] E. F. Redish, Problem Solving and the Use of Math in of college students conceptions of electric fields, report
Physics Courses, presented at the World View on Physics presented at the Meeting of the Arkansas-Oklahoma-
Education in 2005, Delhi, 2005 (2006), p. 1, http://www Kansas and Nebraska AAPT Sections (1997).
.compadre.org/per/items/detail.cfm?ID=3706. [26] V. Albe, P. Venturini, and J. Lascours, Electromagnetic
[7] J. Tuminaro and E. F. Redish, Understanding students’ concepts in mathematical representation of physics, J. Sci.
poor performance on mathematical problem solving in Educ. Technol. 10, 197 (2001).
physics, AIP Conf. Proc. 720, 113 (2004). [27] M. G. M. Ferguson-Hessler and T. de Jong, On the quality
[8] T. J. Bing and E. F. Redish, The cognitive blending of of knowledge in the field of electricity and magnetism, Am.
mathematics and physics knowledge, AIP Conf. Proc. 883, J. Phys. 55, 492 (1987).
26 (2007). [28] J. Guisasola, J. M. Almudí, and J. L. Zubimendi, Difficul-
[9] J. Tuminaro and E. F. Redish, Elements of a cognitive ties in learning the introductory magnetic field theory in the
model of physics problem solving: Epistemic games, Phys. first years of university, Sci. Educ. 88, 443 (2004).
Rev. ST Phys. Educ. Res. 3, 020101 (2007). [29] M. H. P. Kesonen, M. A. Asikainen, and P. E. Hirvonen,
[10] T. J. Bing and E. F. Redish, Analyzing problem solving University students conceptions of the electric and mag-
using math in physics: Epistemological framing via war- netic fields and their interrelationships, Eur. J. Phys. 32,
rants, Phys. Rev. ST Phys. Educ. Res. 5, 020108 (2009). 521 (2011).
[11] B. R. Wilcox, M. D. Caballero, D. A. Rehn, and S. J. [30] C. Smaill and G. Rowe, Electromagnetics: How well
Pollock, Analytic framework for students use of math- is it understood by first- and second- year electrical-
ematics in upper-division physics, Phys. Rev. ST Phys. engineering students?, in Proceedings of the ASEE Annual
Educ. Res. 9, 020119 (2013). Conference, San Antonio, Texas, 2012 (American Society
[12] R. Karam, Framing the structural role of mathematics in for Engineering Education (ASEE), San Antonio, Texas,
physics lectures: A case study on electromagnetism, Phys. 2012), p. 13.
Rev. ST Phys. Educ. Res. 10, 010119 (2014). [31] I. Garzon, M. De Cock, K. Zuza, P. van Kampen, and J.
[13] C. A. Manogue and T. Dray, Bridging the Gap between Guisasola, Probing university students’ understanding of
Mathematics and the Physical Sciences, http://www.math electromotive force in electricity, Am. J. Phys. 82, 72
.oregonstate.edu/bridge/papers/bridge.pdf. (2014).
[14] J. Larkin, J. McDermott, D. P. Simon, and H. A. Simon, [32] J. Guisasola, J. L. Zubimendi, and K. Zuza, How much
Expert and novice performance in solving physics prob- have students learned? Research-based teaching on elec-
lems, Science 208, 1335 (1980). trical capacitance, Phys. Rev. ST Phys. Educ. Res. 6,
[15] B. L. Sherin, How Students Understand Physics Equations, 020102 (2010).
Cognit. Instr. 19, 479 (2001). [33] N. W. Preyer, Surface charges and fields of simple circuits,
[16] E. Kuo, M. M. Hull, A. Gupta, and A. Elby, How students Am. J. Phys. 68, 1002 (2000).
blend conceptual and formal mathematical reasoning in [34] B. A. Sherwood and R. W. Chabay, A unified treatment of
solving physics problems, Sci. Educ. 97, 32 (2013). electrostatics and circuits, URL: http//cil.andrew.C.edu/…
[17] P. Kohl and N. Finkelstein, Student representational (2009).
competence and self-assessment when solving physics [35] A. Sihvola, J. Leppävirta, and H. Kettunen, Signs, curls,
problems, Phys. Rev. ST Phys. Educ. Res. 1, 010104 and time variations: Learning to appreciate Faraday’s law,
(2005). Adv. Electromagn. 1, 1 (2012).
[18] P. Kohl and N. Finkelstein, Effects of representation on [36] L. Doughty, E. McLoughlin, and P. van Kampen, What
students solving physics problems: A fine-grained char- integration cues, and what cues integration in intermediate
acterization, Phys. Rev. ST Phys. Educ. Res. 2, 010106 electromagnetism, Am. J. Phys. 82, 1093 (2014).
(2006). [37] J. Guisasola, J. M. Almudí, J. Salinas, K. Zuza, and M.
[19] P. Kohl and N. Finkelstein, Patterns of multiple represen- Ceberio, The Gauss and Ampere laws: different laws but
tation use by experts and novices during physics problem similar difficulties for student learning, Eur. J. Phys. 29,
solving, Phys. Rev. ST Phys. Educ. Res. 4, 010111 (2008). 1005 (2008).
[20] D. Nguyen and N. S. Rebello, Students’ difficulties in [38] C. A. Manogue, K. Browne, T. Dray, and B. Edwards, Why
transfer of problem solving across representations, AIP is Ampères law so hard? A look at middle-division physics,
Conf. Proc. 1179, 221 (2009). Am. J. Phys. 74, 344 (2006).

020129-13
BOLLEN, VAN KAMPEN, AND DE COCK PHYS. REV. ST PHYS. EDUC. RES 11, 020129 (2015)

[39] R. E. Pepper and S. V. Chasteen, Our best juniors still With Spherical Unit Vectors in Intermediate E&M, AIP
struggle with Gauss’s Law: Characterizing their difficul- Conf. Proc. 1289, 173 (2010).
ties, AIP Conf. Proc. 1289, 245 (2010). [54] E. Gire and E. Price, Graphical representations of vector
[40] C. Singh, Student understanding of Symmetry and Gausss functions in upper-division E&M, AIP Conf. Proc. 1413,
law, AIP Conf. Proc. 790, 65 (2005). 27 (2012).
[41] A. L. Traxler, K. E. Black, and J. R. Thompson, Students’ [55] C. Singh and A. Maries, Core graduate courses: A missed
use of symmetry with Gauss’s law, AIP Conf. Proc. 883, learning opportunity? AIP Conf. Proc. 1513, 382 (2013).
173 (2007). [56] C. A. Manogue and T. Dray, THE VECTOR CALCULUS
[42] C. S. Wallace and S. V. Chasteen, Upper-division students GAP: Mathematics Physics, PRIMUS 9, 21 (1999).
difficulties with Ampères law, Phys. Rev. ST Phys. Educ. [57] R. E. Pepper, S. V. Chasteen, S. J. Pollock, and K. K.
Res. 6, 020115 (2010). Perkins, Observations on student difficulties with math-
[43] D. Nguyen and N. S. Rebello, Students difficulties with ematics in upper-division electricity and magnetism, Phys.
integration in electricity, Phys. Rev. ST Phys. Educ. Res. 7, Rev. ST Phys. Educ. Res. 8, 010111 (2012).
010113 (2011). [58] C. Baily and C. Astolfi, Student Reasoning About the
[44] D. Hu and N. S. Rebello, Understanding student use of Divergence of a Vector Field, PERC Proceedings 2014,
differentials in physics integration problems, Phys. Rev. ST (AIP, Minneapolis, MN, 2014).
Phys. Educ. Res. 9, 020108 (2013). [59] D. J. Griffiths, Introduction to Electrodynamics, 4th Ed.
[45] D. Tall and S. Vinner, Concept image and concept (Addison-Wesley, Reading, MA, 2012).
definition in mathematics with particular reference to [60] R. Serway and J. Jewett, Physics for Scientists and
limits and continuity, Educ. Stud. Math. 12, 151 Engineers with Modern Physics (Cengage Learning,
(1981). Boston, 2009), p. 1552.
[46] P. Barniol, G. Zavala, N. S. Rebello, P. V. Engelhardt, and [61] An introductory calculus course based on the textbook
C. Singh, Students’ difficulties with unit vectors and scalar by Adams and Essex [62], and a course on differential
multiplication of a vector, AIP Conf. Proc. 1413, 115 equations including some paragraphs, examples, and
(2012). exercises about gradient, divergence, and curl.
[47] L. Doughty, Designing, Implementing and Assessing [62] R. A. Adams and C. Essex, Calculus: A Complete Course,
Guided-Inquiry based Tutorials in Introductory Physics, 7th Ed. (Pearson Education Canada, Ontario, 2009),
Ph.D. thesis, Dublin City University, 2013. p. 1152.
[48] R. D. Knight, The vector knowledge of beginning physics [63] F. Marton, Phenomenography—describing conceptions of
students, Phys. Teach. 33, 74 (1995). the world around us, Instr. Sci. 10, 177 (1981).
[49] N. Nguyen and D. E. Meltzer, Initial understanding of [64] S. Vinner and T. Dreyfus, Images and definitions for the
vector concepts among students in introductory physics concept of function, J. Res. Math. Educ. 20, 356 (1989).
courses, Am. J. Phys. 71, 630 (2003). [65] This actually is the formula for the divergence in
[50] J. Van Deventer, Comparing student performance on cylindrical coordinates in three dimensions, but with a z
isomorphic math and physics vector representations, component equal to zero.
Ph.D. thesis, University of Maine, 2008. [66] After doing the analysis and having some fruitful dis-
[51] P. Barniol and G. Zavala, Test of understanding of vectors: cussions, we decided that a fifth electric field was described
A reliable multiple-choice vector concept test, Phys. Rev. in an ambiguous way, and the physics was more involved
ST Phys. Educ. Res. 10, 010121 (2014). than intended. Therefore, we left this situation out of the
[52] T. Dray and C. A. Manogue, Conventions for spherical analysis.
coordinates, Coll. Math. J. 34, 168 (2003). [67] H. Huang, J. Wang, C. Chen, and X. Zhang, Teaching
[53] B. E. Hinrichs, C. Singh, M. Sabella, and N. S. Rebello, divergence and curl in an Electromagnetic Field course, Int.
Writing Position Vectors in 3-d Space: A Student Difficulty J. Electr. Eng. Educ. 50, 351 (2013).

020129-14