Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 133

Journal of Research in Education Sciences

2014, 59(3), 133-160
doi:10.6209/JORIES.2014.59(3).05

Developing an Instrument for Capturing High

School Teachers’ Mathematics Knowledge
for Teaching: An Exploratory Study
Yung-Chi Lin Chien Chin
Graduate Institute of Science Education, Department of Mathematics,
National Changhua University of Education National Taiwan Normal University

Abstract
This paper addresses an effort to design and test a set of items for assessing the mathematics
knowledge for teaching. The paper begins by describing how a representative item was developed
based on three experienced high school teachers’ analyzing classroom instruction. Subsequently,
item refinement by two experts is documented. Finally, the initial results of a pilot-test of a focus
group of 11 high school mathematics teachers are reported. This study revealed three useful methods
for creating items for assessment: (1) using mathematics problems containing certain key
mathematical ideas; (2) using nonstandard approaches of students that are unfamiliar to the teachers;
and (3) using the errors of teachers.

Keywords: high school mathematics, instrument, item development, mathematics knowledge for
teaching (MKT), teacher knowledge

Corresponding Author: Yung-Chi Lin, E-mail: yclin@cc.ncue.edu.tw

Manuscript received: Aug. 27, 2013; Revised: Oct. 21, 2013, Feb. 18, 2014; Accepted: May 27, 2014.
 134 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

Introduction
Although MKT and its domains have been the subject of much research and discussion, and of
many educational reform efforts in recent years (e.g., Ball, Thames, & Phelps, 2008; Bell, Wilson,
Higgins, & McCoach, 2010; Hill, 2010; Kersting, Givvin, Sotelo, & Stigler, 2010), little is known
about the general conception and its implementation at the high school level. Most studies of MKT
have focused on elementary mathematics instruction. Furthermore, it seems unlikely that high school
teachers would lack common content knowledge (CCK). As a result, it may seem more suitable for
researchers to investigate other kinds of knowledge (i.e., MKT) required for effective teaching at the
high school level.
There is currently no well-described method to measure a high school teacher’s MKT and few
research has been done to create items for assessing high school teachers’ MKT. This paper details
an initial effort to develop several items that capture high school mathematics teachers’ MKT. In
particular, the report documents how the representative item was developed from beginning to end.
An account of this work should be helpful and useful to those who want to conceptualize, identify
and measure MKT at the high school level. Furthermore, this work is an important precursor to the
design and implementation of large-scale studies to assess high school mathematics teachers’ MKT.
Finally, the long-term intention for this work is to help high school mathematics teachers to improve
their MKT, which will benefit in students’ learning.
The purpose of this study is to describe the process of developing the items to access high
school teachers’ MKT. It should be noted that our interest in this study is not about how high school
teachers perform on the MKT items but in reporting the process (method) to create items to measure
high school teachers’ MKT. We asked:
1. What is the process of developing high school MKT items?
2. What principles are useful in creating high school MKT items?
3. What items are usable in capturing high school teachers’ MKT?

Literature Review
Mathematics Knowledge for Teaching
The notion of MKT refers to the mathematics knowledge needed to carry out the work of
teaching mathematics beyond what can be seen in the tables of contents of school textbooks or in
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 135

curriculum frameworks (Ball et al., 2008). MKT has traditionally been divided into six domains.
The first domain, CCK, refers to mathematics knowledge and the skills used in an applied,
rather than a teaching, setting. CCK includes simply calculating an answer, knowing the material to
be taught, or, more generally, correctly solving a problem (Ball et al., 2009). Examples of questions
that can be used to measure teacher CCK include “What is a number that lies between 1.1 and 1.11?”,
“Is a square a rectangle?” “Is 0/7 equal to 0?” and “Are diagonals of a parallelogram necessarily
perpendicular?” (Ball et al., 2008). The second domain, specialized content knowledge (SCK), is
specialized to the work of teaching and only teachers need to know it (Hill, Sleep, Lewis, & Ball,
2007). SCK is the mathematics knowledge and skill unique to teaching and requires knowledge
beyond that being taught to students. Herbst and Kosko (2012) defined SCK as mathematics
knowledge used particularly to accomplish the tasks of teaching. Such knowledge encompasses
things that teachers routinely do in the classroom, such as “presenting mathematical ideas,”
“responding to students’ why questions” or “finding an example to make a specific mathematical
point” (Ball et al., 2008). Horizon content knowledge (HCK) is an understanding of the broader set
of mathematics ideas to which a particular idea connects (Ball et al., 2008). HCK is the sort of
understanding that gives teachers peripheral vision for where they are and where their pupils are
headed, so they are conscious of the consequences of how ideas are represented or of the later
development that is enabled (Ball et al., 2009). Elaborating on Ball et al.’s (2008) idea of HCK,
Jakobsen, Thames, Ribeiro, and Delaney (2012, p. 4642) further characterize HCK as enabling
teachers “to make judgments about the importance of particular ideas or questions, and to treat the
discipline with integrity, all resources for balancing the fundamental task of connecting learners to a
vast and highly developed field”. In this study, we define HCK as a body of knowledge that allows a
teacher to recognize most core (fundamental) concepts in teaching problems.
The other domains of MKT represent strands associated with PCK and contain knowledge of
content and students (KCS), knowledge of content and teaching (KCT) and knowledge of content
and curriculum (KCC). These comprise the special amalgams that are deeply embedded in the work
of teaching what makes a topic difficult for pupils; the ways in which learners tend to develop
understanding of a particular idea; and, ways to sequence and structure the development of a
mathematics topic, including representations likely to help pupils learn (Ball et al., 2009; Ball et al.,
2008; Hill, Ball, & Schilling, 2008).
KCS is very closely related to student thinking. It is knowledge of common student conceptions
and misconceptions about particular mathematical concept (Ball et al., 2008). For example, “when
assigning a task, teachers need to anticipate what students are likely to do with it and whether they
 136 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

will find it easy or hard” (Ball et al., 2008, p. 401). KCT is how teachers sequence particular content
when teaching a concept. For example, “during a classroom discussion, a teacher must decide when
to pause for more clarification, when to use a student’s remark to make a mathematical point, and
when to ask a new question or pose a new task to further students’ learning” (Ball et al., 2008, p.
401). KCC was excluded from the present study of high school MKT domains, because the
definition of KCC is not yet clear in the literature.

An Example of High School Mathematics Knowledge for Teaching

The present study offers an example based on Heron’s (shortest distance) problem to better
explain what is meant by high school MKT (Lin & Chin, in press):
Find a point on a given line L, the sum of whose distances to two given points, A and B, on the
same side of the line is the minimum.

Figure 1. Heron’s Shortest Distance Problem

Most high school teachers will know how to find the point A’ that is symmetric to A with
respect to the line L. Thus, the line joining B and A’ intersects L at P, which is the point that is the
solution (Figure 1). This is mathematics knowledge that most teachers commonly hold (CCK).
However, being able to solve this problem is not sufficient to teach it. How might a teacher explain
the meaning of “finding the point that is symmetric to A with respect to the line L” to students? One
way that most teachers use is to prove it. For any other point P’ on L, it can be proved that
AP ' B > APB (Figure 1) by referring to the triangle inequality for △ A’BP’ such that
AP + PB = A ' P + PB = A ' B < A ' P ' + P ' B = AP ' + P ' B . This shows the SCK in teaching this
problem. Furthermore, if the teacher knows that the problem could be extended to the following case
(Figure 2):
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 137


Figure 2. Extended Heron’s Problem ( AB and L Do Not Intersect with Each Other)

Find a point on a line L “in 3-D space”, the sum of whose distances to two given points, A and
B, is the minimum.
A teacher with this kind of awareness which reveals one aspect of HCK might focus on
reducing the previous Heron’s problem to the consideration of joining two points by a straight line,
since the extended problem would also be solved by emphasis on the same approach. Moreover, if a
teacher knows that most students memorize the procedure for solving Heron’s problem without
conceptual understanding, that teacher is equipped with KCS. Finally, KCT is “knowing different,
instructionally viable modes for teaching Heron’s problem and providing two different instructional
approaches that could help students understand why their method did not work.”

Sample Items of the Existing MKT Instrument

A set of MKT items that was developed by Ball and her colleagues (Ball et al., 2008; Hill, 2010;
Hill, Schilling, & Ball, 2004; Hill et al., 2007) was used as a reference for this study. These items
were mainly designed in the multiple-choice format for use in large-scale studies, and each of them
corresponded to one domain of MKT.
In contrast to Ball and colleagues’ multiple-choice design, Bell et al. (2010) developed a set of
open-ended MKT items. These items included several questions, and a question might assess more
than one domains of MKT. For example, the following item (Bell et al., 2010, p. 492) assessed
teachers’ SCK (identity and explain students’ errors), SCK and KCS (understand what students know)
and KCT (help students understand why their method did not work) for questions a, b and c,
respectively.
Imagine you are a fourth-grade teacher. At the beginning of the year several of your students are
making the same error in multiplication problems:

24×36=20×30+4×6=600+24=624
 138 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

16×18=10×10+6×8=100+48=148

a. Explain what the students might have been thinking. Why does this method give the wrong
answer?
b. List up to four things these students seem to know about solving multiplication problems of
this type.
c. As their teacher, what are two approaches you might take to help students understand why
this method doesn’t work? (include examples to clarify what each approach entails).
Kersting et al. (2010) developed an online measure that gauges teacher knowledge by
examining responses to 13 short video clips of teaching (each about three to five minutes). Although
the “classroom video analysis” does not measure exactly the same construct as the MKT instrument,
it does measure a closely related one (Kersting, 2008). Video clips of classroom instruction, which
teachers were asked to view and respond to in writing, served as item prompts to elicit their
knowledge of teaching.
Osana, Lacroix, Tucker, and Desrosiers (2006) measured teachers’ ability to use the
Mathematical Tasks Framework to sort 32 mathematics problems into four different levels by the
cognitive complexity of the problems. Although teachers’ ability to classify problems is not clearly
stated in the current MKT model (Ball et al., 2008), we believe that it is a type of MKT. This is
because Osana et al. assert that the ability to classify problems is related to teachers’ knowledge of
content as well as their knowledge of students and curriculum. The ability to classify problems will
therefore influence teachers’ effectiveness.
In a more recent study, Chazan, Herbst, and Sela (2011) used animations and comic strips as a
medium to facilitate pre-service teachers’ development. They designed a standard method and a
nonstandard method for teaching algebra word problems in their animated story. Teachers who used
the nonstandard method did not show students how to solve the problem. Instead, the teacher
focused on the answers that students obtained and asked students to determine if the answers are
reasonable given the circumstances described in the problem. Teachers who used the standard
method focused first “on the correct answer and then moved swiftly to understanding how students
got their answers” (Chazan et al., 2011, p. 26). Although Chazan et al.’s study did not directly
measure teachers’ MKT, their design inspired us to use teachers’ non-standard behaviors or students
non-standard reactions as prompts to elicit respondents’ MKT.
The present study followed up previously existing sample MKT items in two ways. First, the
intent was to construct items using an open-ended format. This is much like Bell et al.’s (2010)
design. Hill et al. (2008) suggested that further item development might invest in open-ended items
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 139

to prevent teachers from using mathematical reasoning, rather than MKT, to arrive at the correct
answer. Furthermore, high school MKT is a new research field; the open-ended items could help us
better understand high school teachers’ MKT. Second, transcribed excerpts from the three
experienced high school mathematics teachers’ classroom instruction were used as item prompts to
elicit respondents’ MKT. In this approach, teachers might be helped to analyze classroom events
outside of an online environment, by directly using paper-and-pencil assessments.

Method
This study used a qualitative research method. We chose “vectors and 3-D geometry” as our
focus because the topics were very important to high school students and difficulties existed in the
teaching and learning of these topics.
Based on Ball and colleagues’ approach for the development of MKT measures (e.g., Hill et al.,
2008; Hill et al., 2004), a qualitative method was undertaken in an attempt to create a group of items.
It involved: 1. analyzing three experienced high school mathematics teachers’ classroom instruction
video-records to construct initial items; 2. reviewing and revising these items; 3. pilot-testing on a
focus-group of 11 high school mathematics teachers; 4. conducting follow-up interviews with four
selected teachers from the focus-group survey; and 5. analyzing the pilot-test and interview results to
produce the final items.

Participants
Informant Teachers
There were three experienced in-service high school mathematics teachers enrolled in this study.
Teacher Chang, Weng and Hu (pseudonyms) served as informant teachers to offer information about
how teachers usually teach mathematical topics at high schools. All the informant teachers had
taught mathematics at high schools at less for 16 years and they all taught at top-ranked public high
schools in the local district.

Research Group
The research group in this study was comprised of a retired and experienced high school
mathematics teacher (Chiu), a tertiary mathematics teacher educator (Chin), a postdoctoral fellow
(Lin), two Ph.D. students and two graduate students. In this group, one of the Ph.D. students and one
of the graduate students were current high school mathematics teachers with roughly eight years of
teaching experience. One purpose of the research group was to analyze the informant teachers’
 140 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

instructions to create the MKT assessment items. Chiu had taught for 39 years in the top-ranking
high school in Taiwan. He showed a high level of MKT, in particular, in the domain of HCK, as
reported by the Lin and Chin (in press). Other members of the research group all received rigorous
training in mathematics and mathematics teaching, and had research experience in the field of
mathematics education.

Pilot-test Participants
For the pilot-test of initial items, a focus group of 11 high school mathematics teachers from a
high school department of mathematics was selected. These teachers varied with respect to
educational degrees and number of years of teaching experience. Their mathematics teaching
experience ranged from 1 year to 29 years, with a mean of 12.8 years. About two-thirds of the
7
teachers ( ) had master’s degrees. The majority of the teachers were mathematics majors except
11
one statistics major and one engineering major. The non-master’s-degree teachers all held college
degrees in mathematics. Presumably, the teachers might represent different levels of MKT due to
their differing levels of education and differing levels of experience.
In order to understand whether the items developed by the research group actually measured
these teachers’ MKT, four of the eleven teachers (two high and two low performances in answering
these items) were selected for interviews to further explore and clarify their responses. These four
teachers were asked to describe their answers and reasoning in more detail. These interviews were
audio-recorded and transcribed verbatim, then analyzed by the research group.

Research Procedures
Item development was carried out in three phases. The first phase was “Initial Item
Construction”. The purpose was to collect possible MKT items by analyzing the classroom practices
of the three informants (Chang, Weng, and Hu).
The next phase was “Item Refinement” in which each research group members individually
reviewed the initially constructed items and provided suggestions for revising the items, eliminating
ambiguous items, or adding useful items. After this, the entire research group met to discuss the
proposed changes. Through consensus-based discussion, the research group reached decisions on
what changes would be made to the MKT items. If consensus was not reached on a particular item,
the change was debated further until either a consensus was reached or the matter was recorded and
dropped.
Reaching a consensus was not always easy. For example, when we reviewed Mr. Chang’s
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 141

teaching in the smallest volume problem (see item 7 in the Appendix), there was a debate on whether
we should let students realize that “P is the centroid” in this problem. Four of the group members did
not agree that students should be taught that “P is the centroid” (center property) because they
thought that students could find out the coordinates of P without knowing “P is the centroid.” If
students do not know the center property, they still can get the answer correctly. However, the other
members, including Mr. Chiu, had a strong sense that knowing the center property is very important
for both earlier and later learning. The group then posed the question of whether or not the center
property is, in fact, a key mathematical concept that students should master. Mr. Chin conjectured
that the center property will still hold when the space is extended to more than three dimensions. Mr.
Chiu responded agreement to Mr. Chin’s conjecture. However, Mr. Chiu could not give a concrete
example immediately. The research group finally decided to postpone the discussion in the next
meeting because Mr. Chin wanted to give the team members time to study if any mathematical
example is relevant to teaching the center property. In the next meeting, Mr. Chiu proposed the
center property will hold from 2 to n dimensions and also for axes that are not perpendicular. He also
gave the formal proofs to support his argument. After reviewing Mr. Chiu’s proofs and discussing
with him, the research group finally had a consensus that teaching the center property is very
important to students and was able to give more examples.
The final phase was the “Pilot Test” which included two separate steps. First, the instrument
was distributed to the 11 focus-group teachers. Second, in order to determine whether the items
successfully measured MKT and to learn more about the knowledge used to answer the items,
in-depth, face-to-face interviews were conducted. Later, these items were further discussed and
refined by the research group members. The following section describes each of these phases in
detail.

Trustworthiness of This Study

As Hill et al. (2008) and Manizade and Mason (2011) suggested, four methods were useful to
ensure the instrument possessed acceptable trustworthiness: 1. the research group’s review,
discussion and revision; 2. triangulation through all phases of data collection and analysis; 3.
detailed description and reporting of the research process; and 4. the interviews.
First, to ensure that the content of the MKT items was consistent in what was being measured,
the items were reviewed, discussed and then revised by the research group members. Each of the
research group members had prior high school teaching experience, and two of them had experience
in designing instruments. The research group also judged the appropriateness of the instrument for
high school mathematics teachers by spending a substantial amount of time reading and analyzing
 142 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

every word of the items.

Second, two different types of triangulation were conducted. In data triangulation, the MKT
instrument was constructed based on literature review, analysis of the three teachers’ classroom
instruction, the research group’s opinions about the items’ appropriateness, and comprehension and
pilot-test results. In investigator triangulation, the ideas and suggestions generated by the research
group members were considered.
Third, the detailed description and reporting of the research process was a much more
rudimentary form of validity.
Finally, the interviews were used to determine whether MKT was actually measured by the
items and to learn more about the knowledge used to answer the items. We also checked the face
validity from the interview results. Every pilot-test participant reported that they could understand
the items well. Additionally, some participants particularly liked the items and indicated that the
items were much better at measuring teachers’ practical knowledge as compared to a standard test of
teachers’ content knowledge.

Pilot Test Scoring

Teacher responses in each question in the pilot-test were scored on a 3-point scare. Table 1
shows detailed information about scoring criteria (for the pilot-test questions, please see Figure 7).

Table 1
The Scoring Criteria for the Pilot Test Questions
Score Definition Example Response
Q1 0 The response was not critical to teaching distance formula
Heron’s problem
1 The response was related to how to prove triangle inequality
that PA + PB is the shortest path
2 The response included that the shortest the shortest distance between two point is
distance between two points is a straight line a straight line
Q2 0 The response did not include any comments. agree with Teacher A’s teaching
1 The response included one comment, but the let students observe and discuss it first
comment is related to a general teaching
method
2 The response included a comment and the using another way to explain why to do
comment is specific to the teaching of the reflection
Heron’s problem
(continue)
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 143

Table 1
The Scoring Criteria for the Pilot Test Questions (continue)
Score Definition
Q3 0 The response was similar to Teacher A
1 The response was included proving that
PA + PB is the shortest path
2 The response was related to letting students
appreciate that the shortest distance between
two points is a straight line
Q4 0 The response was related to memorizing the
solving process
1 The response was related to proving that
PA + PB is the shortest path.
2 The response was related to the fact that the
shortest distance between two points is a
straight line.
Q5 0 The response solved the extended problem
incorrectly
1 The response solved the extended problem
partially successfully
2 The response solved the extended problem
successfully
Example Response
the same as Teacher A
prove that PA + PB is the shortest path.
teach point A and B on different sides first
memorize it
prove that PA + PB is the shortest path.
consider A and B on different sides of a
plane
Respondents leave a blank
Respondents try to find the reflection of
point B but cannot find where should be
the exactly position of B’.
Respondents can solve the problems by
rotating B to B’ (see Figure 6)

Two coders independently coded all the responses. The interrater reliability (Cohen’s kappa
values) across these five questions ranged from .78 to .85. All disagreement was reconciled in the
following research group meeting by all group members. According to their average scores in the
pilot-test, the 11 participants were further split into high performance and low performance. The
participants who scored in the top 23% were categorized as high-performing; the participants who
scored in the bottom 23% were categorized as low-performing.

Findings
From the analysis of their classroom instructions, 13 initial items were formulated that
contained a total of 35 questions (an item may have three or more questions attached, and others may
 144 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

have just one). In the end, 8 final items remained (a total of 19 questions; for sample items see
Appendix).

Initial Item Construction

In order to better describe the development process, the present study provides a representative
example (Heron’s problem item) of how to build an item from the very beginning (early idea stage)
to the end (reviewed, revised, critiqued, polished, pilot-tested and analyzed). We selected Heron’s
problem (also known as the shortest distance problem) as a representative item for several reasons: 1.
This problem is an interesting and famous ancient problem which implies that light always takes the
shortest path (Heath, 1921). 2. The problem is familiar to Taiwanese high school mathematics
teachers. Teachers are likely to give rich responses to teaching this problem. 3. Solving the problem
is not straightforward to students. In particular, students may not find it easy to understand why the
reflection is necessary (Figueiras & Deulofeu, 2005). 4. Teaching this problem can reveal teachers’
HCK and HCK is a relatively new idea in the field of teacher knowledge.
The mathematical problem contained in this item was relevant to Heron’s problem (Figure 3
was supplemented by the authors. Thus, the students would not know the positions of A and B if they
could not make a judgment or the teacher did not teach it):

Given A (4, 3, 1), B (5, 2, 6) and a plane E: x+2y+2z=3, a point P on the plane E such that
PA + PB is the minimum. Find the coordinate of P and the sum of PA + PB .

‧B (5, 2, 6)
‧A (4, 3, 1)

‧P=? E: x+2y+2z=3

Figure 3. Heron’s Problem Via a Plane

While teaching this problem, the informant teacher Weng, first, asked students to recall their
experiences in solving Heron’s problem (Figure 1, see previous section). After thinking of the
analogous problem, Weng then solved the given problem by finding the reflection of A over the
plane E. After this, he continued to provide another method for solving this problem. He drew AA ' ,
BB ' , AM and A ' B ' on the board (Figure 4) and then used the Pythagorean theorem to find the
length of A ' B . He said this method was more effective for finding the shortest distance of
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 145

B
3 3
3
A M 6
3 2
3
Plane E:
P
3 3

A’ B’

Figure 4. Another Method for Solving Heron’s Problem Via a Plane

PA + PB and the coordinate of P could also be obtained by using “P both on the plane E and
A ' B .”
In the discussion of the Weng’s instruction (above) by the research group, Chiu indicated that
Weng did not teach the following key mathematical idea for solving the problem: “the shortest
distance between two points is a straight line”. Depending on this approach, said Chiu “Mr. Weng
should start to teach the problem from two points on different sides of a given line L.” Excerpts of
the research group’s discussion (Miss Cheng, Ph.D. student with pseudonyms) follow:

Lin: I agree with Mr. Chiu’s idea that a teacher needs to illustrate the key idea before
solving the problem. But I am wondering whether it is really a key idea for this
problem? How about the concept of “congruence of triangles” or “triangular
inequality”?
Cheng: They are also quite important for this problem, because you need to prove P is
unique. However, “the shortest distance between two points is a straight line”
might be more fundamental in solving this problem.
Chin: I thought “straight line” was very primitive in this case. This problem is an
application of the famous Heron’s shortest distance problem. With realization that
“the shortest distance between two points is a straight line” in this problem, it
makes for a clearer explanation as to why one given point was reflected across the
line or over the plane.

The other two informant teachers taught a problem similar to this. Teacher Hu’s teaching was
found to be roughly the same as Weng. It was noteworthy that teacher Chang did not teach the same
problem, but, instead, taught an extended Heron’s problem (Figure 2, see previous section). This
 146 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

problem was more complex than Heron’s problem, since it extended the 2-D shortest distance
problem to a 3-D space. In teaching the more complex problem, Mr. Chang let students know that by
rotating B about L, the diagram will become a circle whose center is the projection of B on L and the
radius is the perpendicular distance of B from L. He then let E be a plane containing L and point A
and cutting this circle. Now B’ is on plane E (A and B’ are on opposite sides of L). The intersection
point P of AB ' with L is the solution (Figure 5).

Figure 5. An Extended Heron’s Problem to 3-D

In our analysis, first it was noted that Chang did not see the relation between the Heron’s and
the extended problem. Instead of connecting these two problems with each other, he had decided to
teach the extended problem alone. Chiu suggested that solving this series of problems should stay
with the basic idea of “the shortest distance between two points is a straight line.” Thus a solver
could more easily comprehend the use of a rotation. Chiu’s comments collected from the research
group discussion follow:

Chiu: This is what I said about the key idea for solving the problem. If we hold the basic
principle: “the shortest distance between two points is a straight line”, it would be
more sensible to rotate B to B’ and connect A to B’. In fact, when an attempt was
made to find the symmetric point about L in the earlier (attempt to solve) example
of Heron’s problem, this symmetry is one kind of rotation. It rotated in 3-D space
through 180° about L.

At the end of this discussion, the relevant knowledge for teaching the shortest distance problem
had emerged: 1. teachers should guide students to understand that the shortest distance between two
points is the most fundamental idea to solve the problem; 2. teachers might first examine whether
points A and B are on the same side or on different sides of a given plane, and teach the problem
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 147

containing points on different sides first; 3. it is necessary to prove the uniqueness of P through the
use of congruent triangles and triangular inequality; and 4. the problem could be extended further,
into 3-D space (see Figure 2).
From the analysis above, the initial item was written below (Figure 6). It consists of two
different perspectives on teaching this problem. Teacher B’s teaching seemed to be broader than
teacher A’s. This difference was used to elicit respondents’ MKT related to this problem.

Problem: Given A (4, 3, 1), B (5, 2, 6) and a plane E: x+2y+2z=3, and a point P on the plane E such that
PA + PB is the minimum. Find the coordinate of P and the sum of PA + PB .
Teacher A’s main teaching process is shown below:
(a) Reviewed the similar, but easier, problem of “Find a point on a line L, the sum of whose distances
to two given points, A and B on the same side of L, is the minimum.” (Heron’s problem)
(b) Reminded students of how Heron’s problem was solved previously: reflecting A across L to obtain
point A’ and the point of the intersection of L with A ' B is the answer.
(c) Solved the problem by analogy with the solution in (b).
(d) Taught another effective method for obtaining PA + PB and P:

B
3 3
3
A M 6
3 2
3
Plane E:
P
3 3

A’ B’

Teacher B’s main teaching process is shown below:

(a) Taught the fundamental idea for solving this problem: “the shortest distance between two points is a
straight line”
(b) Reviewed the problem first “Find a point on a line L, the sum of whose distances to two given points,
A and B on different sides from L, is the minimum?” Then further discussed the situation of points A
and B lying on the same side of L.
(c) Found the reflection of A over E (named A’) and then substituted the parametric equations of
A ' B into the equation of the plane to obtain the point of intersection, P.
(d) Proved PA + PB is minimal by using congruent triangles and triangular inequality.
(e) Extended the problem to 3-dimensions case: “Where a line L in space is given and two points, A (4,
3, 1) and B (5, 2, 6), not in the same plane with L, find a point P on L in order to minimize PA + PB ”
Please comment on teacher A’s and teacher B’s teaching processes.
Figure 6. The First Version of Heron’s Problem Item
 148 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

Item Refinement
When these MKT items were formulated, they were then reviewed by two experts (Chin, a
teacher educator and Chiu, a retired teacher). The educator’s job was to make sure these items were
accurate and relevant to the examination of mathematics teachers’ MKT, and retired teacher’s work
was to check the appropriateness to high school mathematics, and revealing high school teaching
context, across the items. Thus, they deleted some inappropriate items and reworded others. In
addition, some new questions were also added to an item because of their putative relationship to
MKT.
In the review of the shortest distance item, for example, Chin suggested that teacher B’s
teaching process should be removed and substituted with some concrete questions: 1. “What are the
key mathematics ideas for solving the problem?”; 2. “What is your comment about teacher A’s
teaching?”; and 3. “How do you teach the problem in your classes?” He changed the original
commentary question due to concerns that the question might be too open and respondents might not
know how to give an adequate comment, or might give irrelevant information that this study did not
aim to focus on (e.g., issues about beliefs). On the other hand, Chiu suggested to remove “another
effective method” from teacher A’s teaching process (i.e., (d)), because he thought this was not the
most common way used to teach the problem. After these exchanges, the entire research group
worked once more to compare these suggested revisions and finally come to an agreement about the
wording, revision and organization of the instrument.
Afterward, two more questions for this item arose (Q4 and Q5, see Figure 7). Q4 was based on
a discussion about providing mathematical explanations for students (which belongs to the domain
of SCK). To reflect A across E is not so trivial, and it reduces the problem to the consideration of
joining two points by a straight line. Those who did this had the insight to apply their knowledge that
the shortest distance between two points is a straight line, and other knowledge related to the
preservation of distance under reflection must come into play. Q5 is the extended problem. It
assesses teachers’ capacity to see how the current problem would be extended to, and interact with,
later problems. This question seems to measure teachers’ HCK. The second revision of this item is
shown in Figure 7 (changes underlined).

The Pilot Test

After the second revision of the test items, the pilot test was implemented to ensure that the
items adequately captured high school mathematics teachers’ MKT, to determine whether the
instrument does in fact differentiate well among individuals, and to perform validation work. Table 2
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 149

Problem: Given A (4, 3, 1), B (5, 2, 6) and a plane E: x+2y+2z=3, a point P on the plane E such that
PA + PB is the minimum. Find the coordinate of P and the sum of PA + PB .
Teacher A’s main teaching process is shown below:
(a) Reviewed the similar, but easier, problem of “Find a point on a line L, the sum of whose distances
to two given points, A and B on the same side of L, is the minimum.” (Heron’s problem)
(b) Reminded students of how Heron’s problem was solved previously: reflecting A across L to obtain
point A’ and the point of the intersection of L with A' B is the answer.
(c) Solved the problem by analogy with the solution in (b).
Q1. What are the key mathematical ideas needed to solve the problem? (to assess HCK)
Q2. What is your comment about teacher A’s teaching? (to assess SCK & KCT)
Q3. How do you teach the problem in your classroom? (to assess SCK & KCT)
Q4. If a student asks you: “how do you know that to reflect point A across E leads to the solution?”
What would be your responses? (to assess SCK)
Q5. If a student asks you: “Find a point on a line L in space, the sum of whose distances to two given
points A and B is the minimum?” ( AB and L did not intersect with each other) What would be your
response? (to assess HCK)
Figure 7. The Second Version of Heron’s Problem Item

summarizes the 11 focus-group teachers’ responses for each question in the Heron’s problem item.
Through research group discussion, their responses were classified into several categories based on
the similarities of each question. For example, in Q1, if the respondents’ answers to the relevant
concepts used to prove that PA + PB is the shortest path, such as the congruence of a triangle,
triangle inequality, and so forth, then they were classified into a category.
As Table 2 shows, the most frequent response was “prove that PA + PB is the shortest path.”
High percentages were found on Q1 (36%), Q3 (36%) and Q4 (64%). This result seems to suggest
that “prove that PA + PB is the shortest path” was most valued in solving this problem by these
respondents. Nevertheless, according to previously analysis, the “shortest distance between two
points” is the most key (fundamental) concept in solving this problem. Proof is also very important
but it is used to verify the conjecture (i.e., to reflect point A across E) in solving the problem. This is
because, once Heron’s problem is extended (Q5), a solver should solve the problem by equipped
“the shortest distance between two points is a straight line.”
Another noteworthy response might be the physics approach to teaching this problem (Q3 and
Q4). Although this was not the highest frequency (9% on Q3 and 18% on Q4), it demonstrated an
unusual appearance of SCK. And it seemed to illuminate the problem:
 150 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

Table 2
Teachers’ Responses to Heron’s Problem Item
Responses N* (%)
1. concepts about the shortest distance: the shortest distance between two points is a 2 (18)
straight line, light travels along the shortest path
2. concepts used to prove that PA + PB is the shortest path: congruence of triangle, 4 (36)
triangle inequality, perpendicular bisector theorem
Q1
3. concepts about the relationship between points and plane: examining A and B’s 3 (27)
locations, reflection of a point (symmetric image of a point)
4. prior concepts: distance formula, a normal vector of a plane, coordinates of the 2 (18)
midpoint
1. comments: explain the thinking process about how to solve it in more detail, let 7 (64)
students observe and discuss first, solve it by another method (differential), trail
Q2
and error (trying different points of P).
2. agreement with his teaching 4 (36)
1. different than teacher A: teach A and B on different sides first (18%), prove that 7 (64)
Q3 PA + PB is the shortest path (36%), teach mirror image (9%)

2. the same as teacher A 4 (36)

1. physics perspectives: assume that the plane was a plane mirror, optical property 2 (18)
2. proof perspective: prove that PA + PB is the shortest path. 7 (64)
Q4
3. geometry perspective: consider A and B on different sides of a plane 1 (9)
4. rule perspective: memorize it 1 (9)
1. successful in solving the problem 8 (73)
Q5
2. unsuccessful in solving the problem 3 (27)
Note. * The total number of the participants is 11.

As long as we assume this plane is a mirror plane, the problem becomes one of
determining the path from A to B via a mirror which has a minimum optical length. Thus,
it makes more sense for students to find the mirror image of A. (said an interviewee with
high-performing in the pilot-test)

Results from Table 2 also show the discrimination among teachers of different knowledge
levels. For example, compared with “prior concepts” in Q1, a teacher who emphasized the
importance of “shortest distance between two points” might make more sense to students trying to
comprehend the problem. In Q3, a more knowledgeable teacher could offer more opinions about
teaching this problem, rather than simply agreeing with teacher A’s teaching.
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 151

In addition, the results from Table 2 also offer three clues about revisions of this item. First, Q1
should restrict teachers to writing “the most” key idea (see Q1’ in Figure 8), because most of them
wrote too many ideas that did not seem critical. The initial intention was not to restrict respondents
to only one answer, because doing this might narrow their expressions. The consideration was that
allowing more ideas might help them truly show their thoughts. However, the decision was made not
to do this in the final version. Some teachers totally ignored the word “key”; they just wrote as many
ideas as they could. One teacher even wrote seven ideas, and it was not apparent, without an
interview, which of them was the most important to him.

(Problem’s description and teacher A’s main teaching process were omitted, see Figure 7; changes
underlined)
Q1' What is the most important key mathematical idea needed to solve this problem? (to assess HCK)
Q2' Compared with teacher A, your teaching of this problem is: □Exactly the same □Not exactly the
same but not totally different □Totally different.
Explain: (to assess SCK & KCT)
Q3' If a student asks you: “How do you know that to reflect point A across E leads to the solution?”
What would be your responses? (to assess SCK)
Q4' What might be the mathematical extensions for the problem; what might be an extended problem
after solving the problem? And how does the extended problem influence you in teaching the
problem? (to assess HCK)
Figure 8. The Final Version of Heron’s Problem Item

Second, Q2 and Q3 from this item should be combined in the final version (as Q2’, in Figure 8)
because the responses were so closely related. While these respondents wrote about their teaching in
Q2, most of them also gave comments based on their teaching when answering Q3. Another reason
is that some respondents, despite the fact that they reported that they taught in another way, still
expressed opinions that teacher A’s teaching was acceptable. Thus, integrating these two questions
might help respondents to demonstrate their actual MKT in teaching this problem.
Finally, Q5 which was used to examine teachers’ HCK in the second version was changed to
Q4’ in the final version (Figure 8). From the interview results, teachers who were able to solve Q5
did not necessarily have HCK if they did not see these three related problems (Heron’s problem,
Heron’s problem via a plane, and the extended problem) as a whole. That is the reason why Q5 was
changed to Q4’.
 152 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

Discussion and Conclusion

The results of the responses to Q2 suggest that commenting on authentic instructional practice
“only” might not be particularly helpful in uncovering high school teachers’ MKT. This differs from
Kersting et al.’s (2010) results when using authentic video clips of classroom instruction to elicit
teachers’ MKT. Although “video clips” and “texts” were not compared with each other, it remains to
be seen whether texts could work. This could be examined if detailed classroom instruction
transcripts were provided and if teachers were asked not only to comment on them but also to
analyze them. However, the simplicity of the original intention might be lost with such an
arrangement. From this consideration, it seems worthwhile to attempt a “comic strips” design
(Chazan et al., 2011). On the other hand, Taiwanese culture is unlikely to convey criticism of the
others’ teaching, since it would be regarded as impolite. For example, many teachers in the
interviews assumed that teacher A must be the interviewer (Lin) and therefore feel that it is
inappropriate to criticize the teaching scenarios. For this reason we believed the comic
strip/animation design to be a superior method for uncovering teacher MKT.
Writing items and pilot testing them help illuminate and refine the content and structure of the
MKT conceptual framework. For example, SCK should be categorized further into two types. First,
teachers’ justification of what is the most important key concept in solving the problem might lead to
different levels of SCK (e.g., teachers who thought “shortest distance between two points” is the key
concept, may use the example of points on different sides first while teaching the problem). Second,
teachers’ broader view of mathematics (HCK) might influence the formation of SCK (e.g., the
physics perspective). However, these propositions still need to be further examined.
From the item development experience, three principles were useful in creating high school
MKT assessment items. The first concerns the use of mathematics problems in terms of certain key
(primitive) mathematics concepts, such as Heron’s problem item and item 7 (see Appendix). In item
7, the most key concept in solving this problem is “P is the centroid of △ABC.” This result will still
be true even in 4 to n dimensions. Having this sort of knowledge could help in providing the vision
useful in seeing connections to much later mathematics ideas. The second is to use and formulate
students’ nonstandard approaches that are unfamiliar to the teacher. For instance, in item 2 (see
Appendix), a student asserted that he would only need two points to establish a plane: “the
perpendicular bisecting plane of the line segment joining points A and B.” How might a teacher
explain and justify his mathematical idea? The third is to use and formulate the teacher’s errors. For
instance, in item 10 (see Appendix), the teacher drew an incorrect diagram to represent the problem
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 153


( v should be in the inverse direction), although he still solved the problem successfully. It is
interested to know whether the respondents could identify it. We therefore assume that the high
school mathematics teachers’ MKT might be captured and understood more insightfully from their
debates on these three different types of items.
In the Heron’s problem item (Figure 8), we mainly adapted the first principle “a mathematics
problem has some key (core) concepts.” This principle agrees with Ball and Bass’ (2009) definition
of HCK. They defined HCK as “a kind of elementary perspective on advanced knowledge” (Ball &
Bass, 2009, p. 10). We tended to explain that the “elementary perspective” consists in the ability to
identify primitive core concepts and then use (emphasize) these concepts in one’s teaching. Since we
could use the item to examine whether teachers have this kind of ability, we thought the principle
was useful. That is, we found that some teachers could really specify the key concept, but others
could not. In addition, when teachers could specify the key concept of “straight line” in the Heron’s
problem, they usually could provide more meaningful comments on Teacher A’s teaching. We
provided two examples (Heron’s problem item and the smallest volume problem item) in this study.
However, additional research analyzing teaching practice is needed to find more key concepts in
order to develop more similar items.
The second principle related to teachers’ understanding about students thinking and came from
Hill et al.’s (2008) definition of KCS. In the classroom teaching, students may sometimes propose
solution methods that seem incorrect even though a teacher cannot immediately figure out the
mistake. We believe that this is a good source for us to develop a KCS question because some
teachers (without KCS) may not understand how to respond to students. This is also an echo of
Chazan et al.’s (2011) “non-standard” design.
The third principle is related to using mathematics precisely. Sometimes teachers unconsciously
make some errors in teaching. We collected these incidents as item prompts because using errors is
thought to be a good way to elicit one’s knowledge (Borasi, 1996)
Results from our construction and analysis of MKT items indicate that this approach might
work for capturing knowledge that is distinct from pure content or pedagogical content knowledge.
However, the item instrumentation is still an ongoing process. In the next phase, a more fully
developed set of items will be administered to a larger group of high school mathematics teachers to
check its psychometric validity and reliability.
The questions proposed to assess high school mathematics teacher’s MKT were designed to
situate professional knowledge in the context of its practice, but how such knowledge was actually
used in the classroom needs to be further investigated. Much work remains to be done to reach a
 154 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

greater understanding of the feature of, and the relevance among, those domains of high school
mathematics teacher’s MKT, such as uncovering more concrete examples of HCK, or examining the
possible relationships between SCK and HCK.
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 155

References
Ball, D. L., & Bass, H. (2009, March) With an eye on the mathematical horizon: Knowing
mathematics for teaching to learners’ mathematical futures. Paper presented at the 43rd
Jahrestagung der Gesellschaft für Didaktik der Mathematik, Oldenburg, Germany.
Ball, D. L., Thames, M. H., Bass, H., Sleep, L., Lewis, J., & Phelps, G. (2009). A practice-based
theory of mathematical knowledge for teaching. In M. Tzekaki, M. Kaldrimidou, & H.
Sakonidis (Eds.), Proceedings of the 33rd conference of the international group for the
psychology of mathematics education (Vol. 1; pp. 95-98). Thessaloniki, Greece: PME.
Ball, D. L., Thames, M. H., & Phelps, G. (2008). Content knowledge for teaching: What makes it
special? Journal of Teacher Education, 59(5), 389-407. doi:10.1177/0022487108324554
Bell, C. A., Wilson, S. M., Higgins, T., & McCoach, D. B. (2010). Measuring the effects of
professional development on teacher knowledge: The case of developing mathematical ideas.
Journal for Research in Mathematics Education, 41(5), 479-512.
Borasi, R. (1996). Reconceiving mathematics instruction: A focus on errors. Santa Barbara, CA:
Greenwood Publishing Group.
Chazan, D., Herbst, P., & Sela., H (2011). Instructional alternatives via a virtual setting: Rich media
supports for teacher development. In O. Zaslavsky & P. Sullivan (Eds.), Constructing
knowledge for teaching secondary mathematics: Tasks to enhance prospective and practicing
teacher learning (pp. 23-37). New York, NY: Springer. doi:10.1007/978-0-387-09812-8_2
Figueiras, L., & Deulofeu, J. (2005). Visualising and conjecturing solutions for Heron’s problem. In
M. Bosch (Ed.), Proceedings of the CERME 4 international conference (pp. 420-427). Sant
Feliu de Guixols, Spain: CERME.
Heath, T. L. (1921). A history of Greek mathematics (Vols. I&II). New York, NY: Dover.
Herbst, P., & Kosko, K. (2012, November). Mathematical knowledge for teaching high school
geometry. Paper presented at the Proceedings of the 34th annual meeting of the North American
Chapter of the International Group for the Psychology of Mathematics Education, Kalamazoo,
MI.
Hill, H. C. (2010). The nature and predictors of elementary teachers’ mathematical knowledge for
teaching. Journal for Research in Mathematics Education, 41(5), 513-545.
Hill, H. C., Ball, D. L., & Schilling, S. G. (2008). Unpacking “pedagogical content knowledge”:
Conceptualizing and measuring teachers’ topic-specific knowledge of students. Journal for
 156 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

Research in Mathematics Education, 39(4), 372-400.

Hill, H. C., Schilling, S. G., & Ball, D. L. (2004). Developing measures of teachers’ mathematics
knowledge for teaching. The Elementary School Journal, 105(1), 11-30. doi:10.1086/428763
Hill, H. C., Sleep, L., Lewis, J., & Ball, D. L. (2007). Assessing teachers’ mathematical knowledge:
What knowledge matters and what evidence counts. In F. K. Lester (Ed.), Second handbook of
research on mathematics teaching and learning (pp. 111-156). Charlotte, NC: Information Age.
Jakobsen, A., Thames, M. H., Ribeiro, C. M., & Delaney, S. (2012, July). Using practice to define
and distinguish horizon content knowledge. Paper presented at 12th International Congress on
Mathematics Education, Seoul, South Korea.
Kersting, N. (2008). Using video clips as item prompts to measure teachers’ knowledge of teaching
mathematics. Educational and Psychological Measurement, 68(5), 845-861. doi:10.1177/
001316 4407313369
Kersting, N. B., Givvin, K. B., Sotelo, F. L., & Stigler, J. W. (2010). Teachers’ analyses of classroom
video predict student learning: Further explorations of a novel measure of teacher knowledge.
Journal of Teacher Education, 61(1-2), 172-181. doi:10.1177/0022487109347875
Lin, Y.-C., & Chin, C. (in press). Extending the conception of Horizon Content Knowledge:
Fundamental mathematical knowledge and a sample assessment item. Chinese Journal of
Science Education.
Manizade, A. G., & Mason, M. M. (2011). Using Delphi methodology to design assessments of
teachers’ pedagogical content knowledge. Educational Studies in Mathematics, 76(2), 183-207.
doi:10.1007/s10649-010-9276-z
Osana, H. P., Lacroix, G. L., Tucker, B. J., & Desrosiers, C. (2006). The role of content knowledge
and problem features on preservice teachers’ appraisal of elementary mathematics tasks.
Journal of Mathematics Teacher Education, 9(4), 347-380. doi:10.1007/s10857-006-4084-1
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 157

Appendix: Sample items

2. When teaching the topic of “specifying a plane with three non-collinear points,” a student drew a
diagram and asserted that he would only need two points (P and Q) to establish a plane.

P Q

Q1. Explain what the students might have been thinking. Is it legitimate to argue this? (to assess
SCK & KCS)
Q2. As his teacher, what might you do to help him understand this argument? (to assess KCT)
7. Problem: The plane E passes through the point P (2, 1, 3) and cuts off the smallest volume in the
first octant (the plane cuts the x-axis at A (a, 0, 0), the y-axis at B (0, b, 0), and the z-axis at C (0,
0, c)).
z
C

‧P(2, 1, 3)
y
O B
A
x

(i) Find the smallest volume of the tetrahedron cut by the plane.
(ii) Find the equation of the plane.
Teacher A’s teaching process is shown below:
x y z 2 1 3
(a) Let the plane E be + + = 1 ; since (2, 1, 3) lies on the plane E, we have + + =1.
a b c a b c
 158 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

2 1 3
+ +
2 1 3
(b) The inequality of arithmetic and geometric means gives a b c ≥ 3 × × ; therefore
3 a b c

3
⎛ ⎞
1 ⎜ 3 ⎟
V(a, b, c)= abc ≧ ⎜ ⎟ ≧27
6 ⎜⎜ 2 + 1 + 3 ⎟⎟
⎝a b c⎠

(c) Now we notice that (a, b, c)=(6, 3, 9) gives equality in this inequality; thus the equation of

x y z
the plane E is + + = 1.
6 3 9

(d) Next let us replace 2, 1 and 3 with x0, y0 and z0; by the same procedure, we get (a, b, c)=(3x0,

x y z
3y0, 3z0) and the plane is + + = 1.
3x0 3 y0 3z0

(e) Actually, P is the centroid of △ABC.

Q1. What is the most “key” mathematical idea for solving the problem? (to assess SCK)
Q2. Compared with teacher A, your teaching of this problem is: □Exactly the same □Not
exactly the same but not totally different □Totally different.
Explain: (to assess SCK & KCT)
Q3. Will the result of “P is the centroid of △ABC” still hold in the affine coordinate system
(two non colinear axes not necessarily perpendicular to each other)?
Explain: (to assess HCK)
Q4. Will the result of “P is the centroid of △ABC” still hold, if we extend the whole problem to
4-dimensions?
Explain: (to assess HCK)

10. Problem: Consider △ABC in which A is (2, 5), B is (5, 1) and C is (3, 7). P is on BC and the
  3 4
projection vector of AP on AB is (- , ) . Find the coordinates of P.
5 5
Teacher A drew the diagram and taught the problem as follows:
Yung-Chi Lin & Chien Chin Developing an Instrument for Capturing teachers’ MKT 159

C(3, 7)
P(x, y)

A(2, 5)

v
B(5, 1)

    3 4
(a) Let v be the projection vector of AP on AB ; therefore v = (- , ) .
5 5
   
   AP ⋅ AB  1 AP ⋅ AB
(b) We know AB = (3, -4) and v = t AB = (  2 ) AB ; therefore t = - and (  2 )
AB 5 AB

1
=-
5
 
(c) Getting a parametric equation of BC and P=(5 - 2t, 1+6t); we have AP = (3 - 2t, -4+6t) and
 
AP ⋅ AB 1 (3 - 2t , - 4 + 6t ) (3, - 4) 1
(  2 ) = - . Therefore =-
( )
2
AB 5 5
32 + (-4) 2

(d) t=1, P=(3, 7)

Q1. Compared with teacher A, your teaching of this problem is: □Exactly the same □Not
exactly the same but not totally different □Totally different.
Explain: (to assess SCK & KCT)
 160 Developing an Instrument for Capturing teachers’ MKT Yung-Chi Lin & Chien Chin

教育科學研究期刊 第五十九卷第三期
2014 年，59（3），133-160
doi:10.6209/JORIES.2014.59(3).05

發展高中教師教數學所需的知識
之問卷：探索性研究
林勇吉* 金鈐
國立彰化師範大學 國立臺灣師範大學
科學教育研究所 數學系

摘要

本研究旨在描述如何發展高中教師「教數學所需的知識」
（MKT）問卷。研究結果以一個
具代表性的問題為例。呈現這個代表性問題如何發展與不斷精緻，最後報導本問卷的前導測
驗結果。本研究發現三個有效的創造問卷的策略：一、使用具有「關鍵」教學知識的數學問
題（教導這個數學問題，需要一個基礎、重要且不能忽略的核心知識）
。二、使用學生非例行
性解法。三、使用教師在實際教學上的錯誤。本研究希望提供相關領域研究者，如何設計 MKT
問卷的參考依據。

關鍵字：高中數學、評量工具、問卷發展、教數學所需的知識、教師知識

通訊作者：林勇吉，E-mail: yclin@cc.ncue.edu.tw
收稿日期：2013/08/27；修正日期：2013/10/21、2014/02/18；接受日期：2014/05/27。
 © 2014. This work is published under
https://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”).
Notwithstanding the ProQuest Terms and Conditions, you may use this
content in accordance with the terms of the License.