AN INVESTIGATION OF THE EFFECTIVENESS
OF PHYSICS FIRST IN MAINE
By
Michael James O’Brien
B.S. Lyndon State College, 1990
B.S. University of Maine at Farmington, 1998
A THESIS
Submitted in Partial Fulfillment of the
Requirements for the Degree of
Master of Science in Teaching
The Graduate School
The University of Maine
May, 2006
Advisory Committee:
John R. Thompson, Assistant Professor of Physics, Cooperating Assistant
Professor of Education; Advisor
Michael C. Wittmann, Assistant Professor of Physics, Cooperating Assistant
Professor of Education
Molly Schauffler, Assistant Professor of Geological Sciences
LIBRARY RIGHTS STATEMENT
In presenting this thesis in partial fulfillment of the requirements for an
advanced degree at The University of Maine, I agree that the Library shall make it
freely available for inspection. I further agree that permission for “fair use”
copying of this thesis for scholarly purposes may be granted by the Librarian. It
is understood that any copying or publication of this thesis for financial gain shall
not be allowed without my written permission.
ABSTRACT
Data from three high schools that teach physics in ninth grade and three
that teach physics in twelfth grade were used to make comparisons between these
classes. Research tools include written pre- and post-tests of kinematics and
mechanics concepts, a written physics attitudes and expectations survey, and
individual student interviews. Portions of these tools were excerpted from wellknown and thoroughly tested instruments. The normalized gains on the
conceptual survey were compared, and analyzed to determine which kinematics
and mechanics concepts ninth- and twelfth-graders appear to learn differently.
Students’ perceptions of physics from the ninth- and twelfth-grade viewpoints are
also compared. Results suggest that while the populations are similar affectively,
they have some significant differences in conceptual understanding, and this
difference is amplified by different instructional approaches.
1
Chapter 1
Introduction
Currently, in more than 99% of American high schools, physics is a
twelfth-grade course (Sheppard and Robbins, 2005). In this traditional model,
physics is the culminating high school science class for most students, and is rich
in math, problem solving and critical thinking skills. However, there is a growing
movement called “Physics First” whose supporters advocate teaching physics to
ninth graders rather than twelfth graders. In this new model, physics serves as the
first course in the sequence of high school science courses rather than the last.
This report describes the investigation of and comparison of the experiences of a
sample of ninth graders and a sample of twelfth graders, each of whom are taking
physics for the first time, with the goal of learning more about the advantages and
disadvantages of teaching physics at each age level.
1.1 History
In order to fully understand why the current sequence of high school
courses is being challenged, it is necessary to look at the history of high school
science education in America. Prior to 1892, chemistry and physics were not
taught in any specific order, and general biology did not exist as a course,
although courses in botany, physiology and zoology did exist. In 1892, the
National Educational Association (NEA) organized the “Committee of Ten.”
This was a group of scientists and college-level educators whose goal was to
determine what should be taught in high school so students entering college
would have a more uniform preparation (Sheppard and Robbins, 2005).
The Committee of Ten recommended that physics be taught before
chemistry so all students would be exposed to physics in high school, since many
high schools of the day only required one year of science. This recommendation
was implemented by the NEA in 1896, and was the preferred sequence of high
school science courses in the early 1900’s. However, the physics-chemistry
sequence did not last for two reasons. The first reason is that between 1896 and
1920, two new high school science courses were created – general science and
general biology. These courses were generally descriptive in nature, and
therefore, most schools taught them in the early years of high school before the
more abstract subjects of chemistry and physics (Sheppard and Robbins, 2005).
The second reason that the physics-chemistry sequence did not last is that the
Committee of Ten had also recommended that physics become more demanding
and mathematically sophisticated (Myers, 1987). This made physics more
suitable as twelfth grade course.
In 1920, the NEA commissioned the Committee on the Reorganization of
Science in Secondary Schools to look at how these new courses fit into the
sequence of high school science courses. The recommendation of this committee
was that general biology should be taught before chemistry and physics, but no
recommendation regarding the order of chemistry and physics was made.
However, because physics courses had become more rich in math, problem
solving and critical thinking, many schools began teaching chemistry before
physics. In 1908, approximately 55% of the high schools in the United States
2
taught physics before chemistry, while approximately 35% of high schools taught
physics after chemistry, and the remaining 10% did not offer separate courses in
both chemistry and physics. By 1941, only 42% of the schools taught physics
before chemistry, while 57% taught physics after chemistry, and only 1% did not
offer separate courses in both physics and chemistry (Sheppard and Robbins,
2005).
In 1957, the Soviet Union launched Sputnik, and The American High
School Today Committee was formed to recommend reforms necessary to help
the United States catch up to the Soviets in the space race. In 1959, this
committee recommended that physics be taught after chemistry. This
recommendation was made mainly because the problem solving and critical
thinking skills that were prevalent in physics courses were judged to be more
appropriate for older students. The sequence of biology-chemistry-physics has
been the most widely accepted sequence since that time. The most common
sequence of courses, which currently exists in more than 99% of American high
schools, is: 9th grade earth science or physical science, 10th grade biology, 11th
grade chemistry, and 12th grade physics (Myers, 1987).
1.2 Rationale for Physics First
There is a growing movement, started by educators, called “Physics First”
that advocates flipping this sequence around and teaching physics during the first
year of high school rather than the last (Sund and Trowbridge, 1971; HaberSchaim, 1984; Myers, 1987; Bardeen and Lederman, 1998). Advocates of the
Physics First movement argue that the sequence should be flipped because
science curricula have changed since the current “physics last” model was
recommended in 1959.
Biology curricula have seen the most change since 1959. This is largely
because of the important discoveries in biology that have occurred during this
time. When general biology was first developed as a course in the early 1900’s, it
consisted largely of botany, zoology and health topics. While the 19th century
discoveries of Charles Darwin and Gregor Mendel ushered in a new era of
modern biology at that time, it was Watson and Crick’s discovery of the structure
of DNA molecules in 1953 that ushered in the current era of modern biology.
Since the work of Watson and Crick, many important discoveries have
been made in molecular biology, biochemistry and genetics. A selection of these
discoveries includes: the development of the chain termination method of DNA
sequencing by Fred Sanger (Sanger et al, 1977); the development of the fluid
mosaic model of cell membranes in 1972 by Singer and Nicholson (Singer, S.J.,
Nicholson, G.L., 1972); Chang and Cohen showing that recombinant DNA can be
maintained and replicated in E. coli cells in 1973 (Chang, A.C.Y., Cohen, S.N.,
1974); the founding of the first genetic engineering company, Genetech, in 1977;
the discovery of human oncogenes in 1981 (Schwab et al, 1985); Kerry Mullis
inventing the polymerase chain reaction procedure in 1985 that is used to replicate
specific sequences of DNA (Mullis, K., 1986); the cloning of Dolly the Sheep in
1996; and the publication of the first drafts of the complete human genome in
2001. Clearly, biology has become a much more sophisticated subject since the
days of Darwin and Mendel.
3
As a result, modern biology courses emphasize molecular methods,
genetics and biochemistry. This is very different from the biology classes of 1959
that emphasized the classification of life forms. Chemistry courses have
undergone a similar evolution because of the discoveries in atomic structure by
John Dalton, Joseph John Thomson, Ernest Rutherford, Niels Bohr, and Erwin
Schrödinger during the early 20th century. As a result, modern chemistry
emphasizes atomic structure, atomic energy levels, and even some quantum
theory. Physics first advocates argue that a good understanding of modern
biology requires a chemistry background, and a good understanding of modern
chemistry requires a physics background. This is the foundation of Physics First.
Beyond these changes that have occurred in the content of high school
science courses, advocates for Physics First cite other advantages to teaching
physics to 9th graders rather than 12th graders. (Bardeen and Lederman, 1998)
1. If 9th graders take physics at the same time they are taking Algebra I,
they can see an application for the algebra they are learning. This will
help them learn algebra.
2. Most high schools only require 2-3 years of science, so currently only
about 35% of high school students take physics, and approximately 25%
of high school students take both chemistry and physics. Teaching
physics in the 9th grade will ensure that more students take physics in high
school.
3. Currently, if top students wish to take an advanced or elective science
courses, they need to “double up” (take two science courses
simultaneously) to do so. In the physics first model, earth science is
integrated into the other classes (physics, chemistry, biology). This allows
students to take advanced or elective science courses, during 12th grade
without having to “double up.”
4. Ninth-grade physics emphasizes the basics of science (forces, motion,
energy, experimental design, and data analysis). This is a better
foundation of science skills than earth science.
5. In the current sequence, there is little integration of topics between
biology, chemistry and physics. This is because very little of what is
covered in chemistry requires a biology background, and very little of
what is covered in physics requires either a biology or chemistry
background.
1.3 Resistance to Physics First
In 1959, the American High School Today Committee recommended that
physics be taught to 12th graders because of the abstract thinking, problem solving
and math skills that are a prerequisite for 12th grade physics courses. Some
educators are worried that if this course is moved to 9th grade, those students
either won’t have the skills necessary to be successful, or the course will have to
be so watered down that it will no longer teach the necessary basics of physics.
Furthermore, some educators also argue that there may be more appropriate ways
to ensure that a majority of high school students will have some experience in the
basic concepts of physics and chemistry. One method is teaching an integrated
science curriculum. In this model, students are exposed to biology, chemistry,
4
earth science and physics every year that they take science, and the different
sciences are not treated as separate and unrelated topics. This approach addresses
the criticism of the traditional sequence made by Physics First advocates that the
traditional sequence of high school science courses does not integrate the subjects
well.
Another method is to teach physical science to ninth graders. Although
there is no exact curriculum that describes physical science, a typical physical
science course covers at least the basic concepts of physics (i.e. Newton’s Laws,
conservation of energy) and chemistry (i.e. atomic structure, density, states of
matter). Generally, physical science courses are less mathematically rigorous
than a typical physics or chemistry class. Students who take physical science in
the ninth grade will have some background in physics and chemistry when they
enter the tenth grade biology classroom.
1.4 The Maine Learning Results and Physical Science vs. Physics
The Maine Learning Results is a document that describes the skills each
high school student in Maine needs to acquire in order to meet graduation
requirements. There are thirteen categories of skills in science. These categories
are: Classifying Life Forms, Ecology, Cells, Continuity and Change, Structure of
Matter, The Earth, The Universe, Energy, Motion, Inquiry and Problem Solving,
Scientific Reasoning, Communication, and Implications of Science and
Technology. Each of these categories includes several skills that students must
acquire to prove that they have met the requirements for that given category (State
of Maine Learning Results, 1997,
http://www.maine.gov/education/lres/homepage.htm). The skills that are
described in the Maine Learning Results dictate the topics that are covered in the
high school curriculum. Because of the large number of skills and the complexity
of some of these skills, developing a curriculum that allows all students to meet
all the requirements is a challenge for the high schools in Maine.
Many high schools currently require their student to pass only two years of
science for graduation. This means that some students in these schools take
science only in the ninth and tenth grades. Therefore, they must meet all the
requirements of the Maine Learning Results in their ninth and tenth grade science
classes. This is usually done by teaching physical science to ninth graders and
biology to tenth graders.
Any school that wishes to implement a Physics First curriculum must
require at least three years of science in order for students to make it to the third
course in this physics-chemistry-biology sequence. While increasing the science
requirement from two to three years is beneficial because it will increase
enrollment in the sciences, it may also require extra expenditures for the school
district to pay for more teachers, books, and supplies to meet the demands of this
increased enrollment. This increased enrollment in the sciences can also create
more demands on science classroom and lab space, and in fact, the school may
not have enough space to meet these demands.
1.5 The Transition Years
So that all students continue to have the opportunity to take the full range
of science courses, any school that makes the decision to change the sequence of
5
science courses it offers must go through at least a three-year transition period. If
the school switches from an earth science-biology-chemistry-physics sequence to
a physics-chemistry-biology sequence, there are several adjustments that must be
made during this transition. During the first year of this transition, all ninth
graders will be enrolled in physics as they are the first students to make the
transition. The rest of the students will still follow the old sequence. This means
all tenth graders will be enrolled in biology, eleventh graders will typically enroll
in chemistry, and twelfth graders will typically enroll in physics. During this
year, the number of students taking physics will be more than double the typical
enrollment levels since students from both the ninth and twelfth grades will be
taking physics simultaneously. This requires a shuffling of teaching assignments,
and may mean that some teachers (particularly the teachers that taught earth
science in the old sequence) are teaching outside their area of expertise.
During the second year, this transition becomes even more challenging for
the school. During this year, the ninth graders and the tenth graders will follow
the new sequence while the eleventh and twelfth graders are following the old
sequence. This means that all ninth graders are enrolled in physics, all tenth
graders are enrolled in chemistry, eleventh graders typically enroll in chemistry,
and twelfth graders typically enroll in physics. During the second year of the
transition, not only will physics enrollments be more than double their normal
levels, chemistry enrollments will also more than double during this year since
both tenth and eleventh graders are taking chemistry simultaneously. This means
that more teachers (particularly the earth science and biology teachers) may be
teaching outside their areas of expertise. This large number of students taking
chemistry at the same time can also add to the challenge by putting extra demands
on the need for chemistry lab space.
One way to deal with this incredible challenge is to stagger the transition
such that some of the tenth graders who took physics as ninth graders take biology
instead of the preferred tenth-grade course of chemistry. Those students will
actually follow a physics-biology-chemistry sequence. While this sequence is not
the preferred physics-chemistry-biology sequence, it may need to be done this
way during this second transition year with some students in order to make the
transition possible. (No reports of this sequence have been published.)
The third year of the transition will be much like the first year. Ninth and
twelfth graders are enrolled in physics simultaneously. Tenth graders are enrolled
in chemistry. Eleventh graders are enrolled in biology. If the school elects to
stagger the transition during the second year, some eleventh graders will be taking
chemistry during this year. Once the school makes it through this transition
period of three years, it can begin to offer more electives to its twelfth graders
who have finished the physics-chemistry-biology sequence.
6
1.6 Development of Research Questions
In order to gain insight into the experiences of 9th graders taking physics
th
and 12 graders taking physics for the first time, the following questions are
posed:
1. Is there a difference in the performance of 9th graders and 12th graders on a
survey of kinematics and mechanics concepts?
2. How do the attitudes toward and expectations of physics of 9th graders and
12th graders differ?
The students surveyed in this project come from several different high
schools in the state of Maine, and each school may have one or more teachers
teaching physics. For this reason, some of the students participating in this study
will have different experiences in physics. Some of these differences can be
caused by, and are not limited to: variations in weekly physics class time due to
the different schedules of different schools; differences in curricula between
schools and teachers; variations in class size between schools, and within schools;
variations in the styles of different teachers; and variations in the amount of
technology available at different schools.
The intended study population is typical high school students in the state
of Maine. Seven high schools in Maine have participated in this study. Three of
the schools teach physics to ninth graders, and three teach physics to twelfth
graders. One of the participating schools teaches physics to ninth graders and also
has a course for twelfth graders who did not take physics in an earlier grade. The
participating schools are schools that responded to a request sent out on the Maine
Science Listserv, and volunteered to participate. The participating schools,
teachers, and students received no compensation for their participation.
1.7 Context for Research
Twelfth-grade physics students from four different schools and four
different teachers, and ninth-grade physics students from three different schools
and six different teachers participated in this project. Each of the teachers was
asked to fill out a survey (Appendix A) in which they report the amount of time
spent on specific topics and the different teaching styles they regularly use in their
physics classroom. This was done in order to get a sense of the differences that
exist between the curricula and teaching styles of the different schools and
teachers. The results of these surveys are summarized below.
1.7.1 Self-Reported Time Spent on Specific Topics
In general, the teachers from school #2 reported spending more time on
these topics than the teachers from other schools. This is mainly because these
teachers used a method of instruction called “Modeling.” Modeling requires more
time because it is a student-centered approach which involves the students
performing experiments, collecting data, graphing the data, and creating
mathematical models to develop the rules and equations that will be used in that
class. This process takes longer than the traditional approach of the teacher
lecturing to the students about the rules and laws that they will be using. The only
school to not use modeling and spend more than a total of eleven weeks on these
topics is school #4.
7
School
1
2
2
Graphing
KineNewton’s
Total
Teacher
Skills
matics
Laws
(weeks)
(weeks)
(weeks)
(weeks)
1
0.5
3
6
9.5
2
4
8
8
20
3
4
8
8
20
Table 1: Instructional time breakdown for
ninth-grade college preparatory level classes
Number
of
Students
83
18
11
Graphing
Newton’s
Number
Kinematics
Total
School Teacher
Skills
Laws
of
(weeks)
(weeks)
(weeks)
(weeks)
Students
2
2
2
8
8
18
13
2
4
1
8
8
17
63
3
5
2
3
6
11
28
Table 2: Instructional time breakdown for ninth-grade honors level classes
Graphing
Newton’s
Number
Kinematics
Total
School Teacher
Skills
Laws
of
(weeks)
(weeks)
(weeks)
(weeks)
Students
3
5
2
2
6
10
9
4
6
2
8
6
16
54
5
7
1
3
3
7
2
6
8
1
3
6
10
40
Table 3: Instructional time breakdown for
twelfth-grade college preparatory level classes
1.7.2 Self-Reported Time Spent Using Specific Teaching Methods
The teachers that use Modeling as a method of instruction generally do not
report spending any time lecturing. This is because Modeling emphasizes a
“student-centered” approach rather than a “teacher-centered” approach. For the
rest of the teachers, their self-reported percentage of time spent lecturing varies
from 14-50%. For all schools the percentage of time on labs, small-group work,
and inquiry varies from 13-38%, 13-33%, and 0-33% respectively.
8
Small
Total
Group Inquiry Modeling
Class
Work
(% of
(% of
Time
(% of
total
total
(hours
total
class
class
per
class
time)
time)
week)
time)
1
1
29
29
14
29
0
3.5
2
2
0
(33)
(33)
(33)
100
3
2
3
0
(33)
(33)
(33)
100
3
Table 4: Instructional methods for ninth-grade college preparatory level
classes
The values that appear in parentheses represent components of the modeling
method of instruction.
Lecture
(% of
School Teacher
total
class
time)
Lab
(% of
total
class
time)
Small
InTotal
Group
quiry Modeling Class
Work
(% of
(% of
Time
School Teacher
(% of
total total class (hours
total
class
time)
per
class
time)
week)
time)
2
2
0
(33)
(33)
(33)
100
3
2
4
33
(33)
(16)
(16)
67
3
3
5
38
38
13
13
0
4
Table 5: Instructional methods for ninth-grade honors level classes
The values that appear in parentheses represent components of the modeling
method of instruction.
Lecture
(% of
total
class
time)
Lab
(% of
total
class
time)
Small
InTotal
Lecture
Lab
Group
quiry Modeling Class
(% of
(% of
Work
(% of
Teac
(% of
Time
School
total
total
(% of
her
total total class (hours
class
class
total
time)
class
per
time)
time)
class
time)
week)
time)
3
5
50
25
13
13
0
4
4
6
50
25
25
0
0
4
5
7
50
13
25
13
0
4
6
8
14
14
29
14
0
3.5
Table 6: Instructional methods for twelfth-grade college preparatory level
classes
9
Chapter 2
Literature Review
2.1 Previous Research into the Effectiveness of Physics First
In order for the Physics First movement to gain credibility in both the
physics and education communities, it needs to be able to cite quantitative data
that support its claims that teaching physics to 9th graders can be successful.
While many educators have reported success in teaching physics first, these
reports have been mostly anecdotal and lacking quantitative data. However, there
have been several small-scale studies of the effectiveness of teaching physics to
underclassmen (9th and 10th graders) that have been published since the Physics
First movement’s infancy in the late 1960’s.
2.1.1 Rome Public High School, NY: In 1966, Joseph Palombi, chair of the
science department at Rome Public High School, NY was considering switching
physics from a 12th grade course to a 10th grade course, because “when students
take biology during the sophomore year, it often becomes the last science course
attempted because the concepts are so sophisticated and complex, and the students
become discouraged.” (Palombi, 1971)
In order to determine if physics was suitable for the high school
sophomore, Rome Public High School taught one section of physics to 10th
graders during the 1966-67 school year. The 10th grade students that were asked
to participate in this course were those that achieved scores of 80% or higher on
the 1966 New York Regents Exam in algebra. In the June 1967, both the
sophomores from this physics course, and the seniors from the typical physics
classes took the New York State Regents physics exam. The 10th graders’
average score was 81%, and the 12th graders average score 82%. Palombi saw
this as an indication that 10th graders could be successful in physics. However,
these results are limited because not all the 10th graders were enrolled in physics
courses, and only those that performed above a certain level on the Regents
algebra exam the previous year were part of the study. (Palombi, 1971)
2.1.2 Illinois State Physics Project: A separate study was conducted in the late
1960’s by the Illinois State Physics Project to determine if the high school
sophomore could be successful in physics, because “biology itself has continued
to become a more sophisticated subject, demanding a prerequisite knowledge of
physics and chemistry.” (Hamilton, 1970) In this study, a group of 18 high school
physics teachers was each assigned a volunteer student who had just completed
ninth grade. During the summer of 1969, which was the summer between 9th
grade and 10th grade for the volunteer students, each teacher each worked one-onone with a student in order to teach the concepts of physics. Each week, the
teacher would present a different activity to his/her student using the
“phenomenological approach,” which Hamilton defines as “placing emphasis on
the observation and investigation of physics phenomena that can be readily
explored by the student…without requiring a great deal of mathematical
sophistication.” The results of this study were qualitative.
Hamilton lists two conclusions about the sophomore student: “the
sophomore student had little difficulty mastering the concepts presented,” and
“the sophomore student was highly motivated by the phenomenological
10
approach.” (Hamilton, 1970) While this project was successful, and supports the
hypothesis that the 10th graders are capable of learning physics, the conditions of
this project are so different from a typical high school classroom that it provides
limited conclusive evidence regarding the appropriateness of teaching physics to
all students at the 10th grade level.
2.1.3 Project ARISE: Project ARISE (American Renaissance in Science
Education) was founded in September, 1995, with the goal of developing a new
framework of science education. One of the key components of that framework is
reversing the order of the typical high school science sequence so that physics is
taught first, followed by chemistry and biology. Dr. Leon Lederman, recipient of
a Nobel Prize in physics in 1988, is the most well known participant of this
project.
In 2001, Pasero released the results of a study that he conducted titled,
“The State of Physics First Programs.” Only schools that teach physics first
participated in this study, which consisted of interviews with thirteen teachers,
and two case studies. Pasero found that the schools in this study were “generally
enthusiastic about their curricula.” He also found that none of the schools were
collecting quantitative data on the success of the physics first program in their
schools. Pasero concludes, “Finding that schools are satisfied with their change in
curriculum is nice, but it can hardly be considered a solid research base.
Quantitative data is needed for further study of the effectiveness of the physicsfirst programs.” (Pasero, 2001)
2.1.4 Germantown Friends School, PA: Another study was done by Howard
Glasser, a physics teacher at Germantown Friends School, a private school in
Philadelphia. This school switched to a physics first sequence in 1999. Glasser
looked at the effect of this switch on students’ scores on the math section of the
math section of the PSAT by comparing the scores of the students who took the
PSAT exam before the school switched to physics first to the scores of students
who took the PSAT exam after the switch. He looked at students who took the
PSAT exam for the three years prior to the school switching to physics first, and
those students who took the PSAT exam for the three years immediately after the
switch (table 7).
Glasser (Glasser, 2004) found that the students who took physics as
freshmen performed significantly better on the math section of the PSAT than
those who did not (table 8). This is true even though there were no significant
differences in the performance of the six classes on the Quantitative Ability subtest of the Comprehensive Testing Program III (CTP III) taken in the eighth
grade. This standardized test is published and distributed by the Educational
Records Bureau and is composed of eight sections that assess performance in
verbal and quantitative areas (Glasser, 2004).
11
The p-values in table 8 indicate the level of confidence in which we can
report the differences in the means that are being compared are due to random
effects and not measurable effects. p-values lower than 0.05 indicate that the
level of confidence that the differences in the means are due to random effects is
lower than 5%. These p-values are considered significant because they indicate
that the level of confidence that the differences in the means are caused by
measurable effects and not random effects is greater than 95%.
Year of
20002003 2004 2005
Graduation 2002
Mean PSAT
Mathematics’ 67.3 71.6 75.3 75.7
Percentile
Sample size
154
66
68
65
(n)
Table 7: PSAT performance of pre-inversion classes (2000-2002),
and each post-inversion class (2003-2005) (Glasser, 2004)
Year of
2003
2004
2005
Graduation
p-value
0.0935 0.0035 0.0036
Table 8: p-values for comparison of post-inversion PSAT
scores for each class with pre-inversion scores (Glasser, 2004)
2.1.5 Unnamed High School in Maine, 2003-2005: This high school in Maine
switched to physics first at the beginning of the 2002-2003 school year. Each
year, each student enrolled in physics takes the Force Concept Inventory (FCI)
(Hestenes et al, 1992) as both a pretest and post-test. These data were collected
by the teachers of the school for their own assessment and research purposes, and
not for public information. Because the teachers did not seek parental permission
for these data to be published outside of the school, the school cannot be named.
Each grade consists of approximately 150 students with approximately 60% of the
students taking honors level physics, and 40% taking college preparatory (CP)
level physics. Because this high school was in transition from a traditional
sequence to a physics-first sequence during these years, it is possible to compare
the results of 9th graders taking physics to 12th graders taking physics for the first
time within the same school. The tests were analyzed by finding the mean
normalized gains of each academic level (honors vs. college preparatory) for each
grade. The normalized gain is determined by dividing the difference of the posttest and pretest scores by the difference of a perfect score and the pretest score.
Normalized Gain <g> = (post-pre) / (perfect-pre). These normalized gains are
summarized in table 9.
12
At this school, the honors level 9th graders outperformed each of the other
sub-groups in the school. This is a strong indication that students in this
population can be very successful in physics. It should be noted that these
students were taught using the modeling method of instruction, while the rest of
the school was taught using more traditional methods. Therefore, these data are
also an indication of the efficacy of modeling over traditional methods.
The college preparatory students at each grade level achieved
approximately the same normalized gains. This is another indication that 9th
graders can be equally successful in understanding Newtonian physics concepts as
12th graders. These data, and their relevance to this study will be discussed in
more detail in the Results and Discussion section.
College
Sub-group
Honors
Preparatory
9th grade
24%
52%
th
12 grade
25%
40%
Table 9: Normalized gains on
Force Concept Inventory
for Maine high school
switching to Physics First
2.2
Previous Results of Widely Used Conceptual Diagnostics
Since the early 1990’s, several different diagnostics have been developed
and used to assess student learning in introductory physics courses at both the
college and high school level. Three of these diagnostics were used to develop
the diagnostic used in this study – the Test of Understanding of Graphs in
Kinematics, the Force Concept Inventory and the Force and Motion Conceptual
Evaluation.
2.2.1 The Test of Understanding Graphs in Kinematics
The Test for Understanding Graphs in Kinematics (TUG-K) was
developed by Robert Beichner of North Carolina State University in order to
assess seven different areas of students’ understanding of kinematics graphs
(Beichner, 1994). These seven objectives are summarized in Table 10. The third
column displays the percent of correct answers for each objective when the test
was administered to 524 college and high school students in introductory physics
courses after instruction in kinematics. The high school students in this study
were seniors and juniors. No ninth-grade physics students participated in this
study. (Beichner, 1994)
13
The student
Percent
will:
Correct
PositionDetermine
51
Time Graph
Velocity
VelocityDetermine
40
Time Graph
Acceleration
VelocityDetermine
49
Time Graph Displacement
Determine
AccelerationChange in
23
Time Graph
Velocity
Select
Kinematics
Another
38
Graph
Corresponding
Graph
Kinematics Select Textual
39
Graph
Description
Textual
Select
Motion
Corresponding
43
Description
Graph
Table 10: Objectives of the TUG-K, and data from university
and twelfth-grade students. (Beichner, 1994)
Given:
The overall mean score on the test was 40%. Beichner lists six difficulties
that the students encountered (Beichner, 1994):
“Graph as Picture” Errors
The student sees the graph as a photograph of the motion rather than an
abstract mathematical representation of the event.
Slope/Height Confusion
Students read values directly off the y-axis when they should find the
slope of the line.
Variable Confusion
Students do not distinguish between position, velocity, and acceleration.
They believe that graphs of these variables should be identical.
Non-origin Slope Errors
Students successfully find the slope of lines that pass through the origin,
but have difficulty if the line does not go through the origin.
Area Ignorance
Students do not recognize the meaning of the area under a kinematics
graph.
14
Area/Slope/Height Confusion
Students often perform slope calculations or read values off the y-axis
when area calculations are required.
2.2.2
The Force Concept Inventory
The Force Concept Inventory (FCI) was developed by Arizona State
University physics professors Ibrahim Halloun and David Hestenes, and high
school physics teachers Malcolm Wells and Gregg Swackhamer (Hestenes et al.,
1992). The FCI was developed as an assessment tool that can be used to
determine how well students understand the Newtonian concept of force. It is a
29-question multiple-choice survey whose distracters represent common
misconceptions of force. It has been administered to tens of thousands of
introductory physics students at the high school and college level. The FCI data
can be analyzed two different ways. The first method is to look at the total
percentage of correct answers on the post-test. Post-test scores greater than 60%
are accepted as the entry threshold to Newtonian thinking, and post-test scores
greater than 85% are accepted as the mastery level of Newtonian thinking
(Hestenes and Halloun, 1995). The second method is to determine the average
normalized gain, <g>, for a group of students (Hake, 1998). This is useful to
determine the percentage of the maximum possible gain that has been achieved.
Hestenes, Wells, and Swackhamer administered the FCI to hundreds of
high school and college students in Arizona. The data summarized in Table 11
are from those students that were enrolled in classes that used “traditional”
methods of teaching. The high school classes were typically 12th-grade classes.
No data were collected from 9th-grade physics classes as part of this study.
Hestenes, Wells and Swackhamer did not report the normalized gains for these
students, but they have been calculated here from the pre- and post-test scores.
<g> = (post%-pre%) / (100%-pre%).
Class
Pretest
Score
(%)
Post-test
Score
(%)
<g> (%)
N
High
School
27
48
29
612
NonHonors
High
School
33
56
34
118
Honors
High
School
41
57
27
33
AP
University
52
63
23
139
Table 11: FCI Results Using Traditional Teaching Methods
(Hestenes et al., 1992)
15
Hestenes, Wells, and Swackhamer also collected data from the classes
taught by Wells and Swackhamer. Their teaching methods were more innovative,
i.e. more interactive and student-centered. These data are summarized in Table
12. Hestenes, Wells and Swackhamer do not report testing these data for
statistical significance, and the sample sizes for the innovative classes may be too
small to show any statistical significance. However, there seems to be a
correlation between the teaching method and the normalized gains. The
normalized gains are in the 20-35% range for the students that received more
traditional instruction, and in the 40-60% range for students that were involved in
more innovative methods of instruction (as defined by Hestenes, Wells, and
Swackhamer).
Class
Pretest
Score
(%)
Wells NonHonors
Wells
Honors
Swackhamer
AP
Wells
University
Table 12:
Post-test
Score
(%)
<g> (%)
N
28
64
50
18
42
78
62
30
73
85
44
11
36
68
50
44
FCI Results Innovative Teaching Methods
(Hestenes et al, 1992)
A separate larger study that offers more evidence for a correlation between
teaching method and performance on the FCI was done by done by Richard Hake
of Indiana University and was published in the American Journal of Physics in
1998 (Hake, 1998). Hake analyzed FCI data from 6542 students enrolled in sixtytwo different introductory physics classes both at the college and twelfth-grade
level, and it is in this study that Hake defined normalized gain and found it to be a
measure of merit. The FCI was not administered to any 9th graders as part of this
study. According to these data, the average normalized gains were in the 20-30%
range for 12th graders taught by traditional methods, and in the 50-70% range for
12th graders taught by “interactive engagement” methods. The data from college
and university were very similar to the high school data. The average normalized
gains for college and university students were in the 10-30% range for traditional
classes, and in the 30-70% range for interactive engagement classes (Hake, 1998).
16
Hake defines interactive engagement (IE) methods as, “those designed at
least in part to promote conceptual understanding through interactive
engagement of students in heads-on (always) and hands-on (usually) activities
which yield immediate feedback through discussion with peers and/or
instructors” (Hake, 1998). This definition is similar to the definition Hestenes,
Wells, and Swackhamer use for innovative methods. Hake defines traditional
courses as “those that make little or no use of IE methods, relying primarily on
passive-student lectures, recipe labs, and algorithmic-problem exams” (Hake,
1998).
2.2.3 The Force and Motion Conceptual Evaluation
The Force and Motion Conceptual Evaluation (FMCE) was developed by
Ronald Thornton of Tufts University, and David Sokoloff of the University of
Oregon (Thornton and Sokoloff, 1998). The FMCE is another research-based
diagnostic used to evaluate students’ understanding of Newtonian mechanics.
Like the TUG-K and the FCI, it a multiple-choice test whose distracters are
appealing because they represent common student preconceptions of motion. The
data from the administration of the FMCE to hundreds of college students
enrolled in introductory physics is similar to the data collected from the TUG-K
and the FCI. Students typically answer 10-20% of the questions correctly before
instruction, and how well they do after instruction depends highly on the method
of instruction. The post-test scores of students enrolled in traditional instruction
classes do not improve nearly as much as those of students enrolled in “active
learning” classes (Thornton and Sokoloff, 1998).
A unique feature of the FMCE is the clustering of subsets of questions that
deal with specific topics. For example, one cluster of seven questions asks about
the force required to cause specific motions of a sled on a frictionless surface.
Another cluster of three questions asks about the force acting on a coin that has
been tossed straight up. The questions ask about the force on the coin on its way
up, its way down, and at the apex of its trajectory. Typically, student answers to
questions within clusters follow a predictable pattern given their level conceptual
understanding of Newtonian motion. The patterns of student answers to subsets
of questions can be analyzed to determine their level of conceptual understanding
of Newtonian motion (Thornton and Sokoloff, 1998).
2.3
The Maryland Physics Expectations Survey
The Maryland Physics Expectations (MPEX) Survey was developed at the
University of Maryland (Redish et al., 1998). It is a 34-item survey that probes
students’ attitudes and beliefs about physics courses. The survey contains
statements about physics and the learning of physics in which the students are
asked to respond to the statements using a Likert scale (agree-disagree). The
purpose of the survey is to probe students’ understanding of the “process of
learning physics” (Redish et al., 1998). The MPEX probes student beliefs in three
main categories: independence, coherence, and concepts. The independence
questions probe whether the students see learning physics as receiving
information or as an active process of reconstructing one’s own understanding.
The coherence questions probe students’ views of physics either as a collection of
unrelated facts or as a single coherent system of describing the physical world.
17
The concepts questions probe the relative importance students place on the
formulas and the concepts that support the formulas.
The responses to the statements are categorized as favorable if they agree
with the expert view, and unfavorable if they do not. The survey data are
analyzed by determining the ratio of favorable to unfavorable responses for each
of the three main categories of independence, coherence, and concepts. The
survey was administered to more than 1500 students enrolled in introductory
calculus-based physics courses at six different colleges and universities in the mid
1990’s at the beginning of the semester and again at the end of the semester.
When taken at the end of the semester, the average overall percentage of
favorable responses for each institution falls in the 45-60% range, the average
overall percentage of unfavorable responses falls in the 20-30% range, and the
overall percentage of neutral responses falls in the 20-30% range as well (Redish
et al., 1998). For the coherence cluster, the average overall percentages of
favorable, unfavorable, and neutral responses are in the 45-65%, 20-30%, and 1530% ranges respectively. For the concepts cluster, the average overall
percentages of favorable, unfavorable, and neutral responses are in the 35-60%,
25-40%, and 20-30% ranges respectively (Redish et al., 1998).
2.4
Relevance to this study
Excerpts of each of these four surveys will be used to evaluate the
students’ understanding of mechanics concepts, and their attitudes and
expectations toward physics. Because each of these surveys has been widely used
in these previous studies, the data that have been collected in these studies can be
used as a standard. The reliability of the data collected in this study will be
evaluated through comparison to this standard. However, it needs to be kept in
mind that most of the data has been collected in previous studies involves
university students and high school seniors. There are very few empirical data
available involving 9th graders.
18
Chapter 3
Methods and Instruments
Two important components of successful instruction are investigated in
this study – student understanding of mechanics concepts, and student attitudes
and expectations toward physics. Three different instruments were used to assess
student understanding and student attitudes and expectations: a “mechanics
concepts” survey, an “attitudes and expectations” survey, and a set of interview
questions. Because Physics First is a relatively new paradigm, and one that has
not been tested empirically, there are no instruments designed specifically to
assess 9th graders. For this reason, excerpts from commercially available
instruments were used to create instruments judged to be appropriate for assessing
9th graders.
3.1 Mechanics Concepts Survey
A 27-question multiple-choice survey of mechanics concepts (Appendix
B) was developed by excerpting and collating questions from the TUG-K, FCI,
and FMCE. This mechanics concepts survey was given to students as a pretest
during September, 2005, and as a post-test sometime after instruction in
mechanics was completed. Depending on the duration of the mechanics
instruction of the different classes, the post-test was administered sometime
between December, 2005 and March, 2006.
The TUG-K, FCI, and FMCE are each valid and reliable diagnostics that
have been widely used to assess student understanding of introductory physics
concepts. Each of these diagnostics was designed for assessment of college-level
students enrolled in introductory physics courses. Since each diagnostic contains
some questions that are too advanced for the typical ninth-grade student, it was
necessary to pick questions from each diagnostic that were judged to be
appropriate for ninth-grade students. The end result is a diagnostic that covers the
topics of kinematics graphs and Newton’s Laws, and is understandable to the
average ninth-grade student. This is important to limit the amount of frustration
that the ninth graders will experience and maximize their participation level.
The TUG-K consists of 21 questions that assess students’ understanding of
kinematics graphs. The basic kinematics concepts that are assessed are those that
involve interpreting position vs. time, velocity vs. time, and acceleration vs. time
graphs by either determining the slope of the graphs or finding the area under the
curve. Many of the graphs consist of multi-stage curves, i.e. the slope of the
curve is different for different time intervals. Because this can be confusing to
younger students, questions that contain graphs with either a constant slope or
only two different stages were chosen. These simpler graphs were judged to be
best at assessing students’ understanding of the meaning of slope. Students who
understand slope may not answer the more complicated graph questions correctly
because they get confused processing all the different bits of information that
exist in the multiple-stage graphs.
The FCI consists of 29 questions that assess student understanding of
Newtonian force concepts. Each multiple-choice question consists of five
choices. Each of the choices represents a different way that students with specific
19
misconceptions would answer the questions. Often times these different
misconceptions are only slightly different from one another. Because many of the
choices represent specific misconceptions, they may contain fairly long
explanations of the reasoning behind the misconceptions. Therefore, the students
must read these choices carefully. Because this high level of analytical reading
may be overwhelming for the typical ninth grader, the questions that were chosen
from the FCI are generally those that do not demand the highest level of analytical
reading skills.
The FMCE consists of 47 questions that assess student understanding of
Newtonian force concepts, kinematics graphs, and the Law of Conservation of
Energy. None of the questions that assess student understanding of kinematics
graphs were chosen from the FMCE because these were judged to be more
complex than the graphs that were chosen from the TUG-K. None of the
questions that assess student understanding of the Law of Conservation of Energy
were chosen since that is not one of the goals of this project. Only questions that
assess student understanding of Newtonian force concepts were chosen from the
FMCE.
Many of these questions require the same analytical reading skills as the
FCI, but subsets of the FMCE questions are arranged in clusters. This clustering
of questions allows students to focus on one topic while answering several
successive questions about that topic. This makes it easier for students to focus
because there are fewer transitions from topic to topic as the students move from
question to question. This clustering of questions also allows for assessment of
any patterns of thinking that students may follow when thinking about slightly
different aspects of specific topics.
The mechanics concepts survey that was used a pre- and post-test consists
of six questions from the TUG-K, five questions from the FCI, fifteen questions
from the FMCE, and one that was created by the investigator. These questions
assess five different topics: kinematics graphs, free-fall, and each of Newton’s
three laws. In addition, there are two subsets of questions that are used to assess
whether students have the “impetus” misconception that is common among
introductory students (Clement, 1985). Students with the impetus misconception
believe that a force is required to keep an object in motion, and this force can
exist even when the object is not in contact with the agent supplying the force.
For example, the correct Newtonian view is that a sled on a frictionless surface
that is given a short push will continue to move at a constant velocity after the
push until acted upon by an another outside force. The impetus view is that the
sled will slow down and stop after the push because the applied force diminishes
with time, and in order for the sled to move at a constant velocity, there must be a
constant force applied to the sled.
This survey was administered to students as a pretest at the beginning of
their physics courses in September, 2005. It was also administered as a post-test
some time after the students received instruction in kinematics and mechanics. In
order to evaluate student learning, the normalized gains, <g>, of each student
20
were calculated. The normalized gain is the absolute gain divided by the
maximum possible gain:
<g> = (post-test score – pretest score) / (perfect score – pretest score).
3.2 Attitudes and Expectations Survey
A 14-question survey of student attitudes and expectations (Appendix C)
was created by using questions from the 34-question MPEX survey. This survey
was administered once at the same time as the post-test of mechanics concepts.
The MPEX survey was written for evaluation of the attitudes and expectations of
(university) students in introductory physics courses. For this reason, some of the
questions may not be applicable to the ninth graders’ experiences in their physics
courses, and some of the questions may be written at a level above the typical
ninth grader’s reading level.
In order to adapt the MPEX survey to the needs of this project, it was
necessary to eliminate these questions, and make the survey shorter so it is more
appropriate for ninth graders. From the fourteen questions that were chosen, seven
belong to the coherence subgroup, five belong to the concepts subgroup, and two
belong to the independence subgroup. Like the MPEX survey, each of the
fourteen questions used in this study consist of statements about how physics is
best taught and learned. The students are asked to choose their degree of
agreement/disagreement with each statement using a five-point Likert scale. This
scale consists of “Strongly Disagree,” “Disagree,” “Neutral,” “Agree,” and
“Strongly Agree.”
This survey is used to evaluate students’ attitudes and expectations toward
their physics courses by determining the percentages of students that answer each
question favorably, unfavorably and neutrally. A favorable response is one that is
the same as an expert would answer. Redish et al. define an expert as
“experienced physics instructors who have a high concern for educational issues
and a high sensitivity to students” (Redish et al., 1998). A favorable response on
the MPEX results from agreeing with the statement when an expert would agree,
and disagreeing when an expert would disagree. An unfavorable response results
from disagreeing when an expert would agree, or agreeing when an expert would
disagree.
3.3 Interview Questions
The purpose of the interviews (interview protocols are in Appendix D)
was to probe more deeply into the thoughts of some students than could be done
with multiple-choice surveys. The interviews were conducted at some point
during mechanics instruction. This varied slightly depending on the pace of
instruction at different schools, and the timing of the interviews. The interviews
were mostly conducted during students’ study hall times with those students who
volunteered to participate. In order to maximize the participation of students, the
interviews were designed to be approximately five minutes long. There are two
reasons that shorter interviews were thought to maximize participation: students
would be more wiling to give up a small amount of their study hall time than a
large amount of time, and more interviews can be done in a study hall period
when the interviews are shorter.
21
Some of the interview questions focused on affective aspects of the
students’ experiences in physics, i.e., student enjoyment of their physics course,
and students’ perception of the difficulty of their physics course. The other
interview questions focused on whether students held the impetus misconception
by asking about student understanding of two different situations – a ball thrown
straight up into the air, and a sled on a frictionless surface. In order to keep the
interviews to their prescribed length, not every question was asked to every
interviewee. This practice allowed more students to be interviewed, but resulted
in some questions being asked more often than others.
3.4 Analysis of Surveys and Interview Questions
Students used bubble sheets to answer each of the multiple-choice
questions of the pretest and post-test versions of the mechanics concepts survey.
Student names were removed from the answer sheets, and replaced by their
assigned identification codes. The answer sheets were scored at the University of
Maine’s computer testing center. Responses to each question, and the raw score
for each student were recorded into an Excel spreadsheet. The normalized gains
for each student that completed both a pre- and post-test were calculated. Only
matched were used in the final analysis of normalized gains.
A separate answer sheet was not used for the attitudes and expectations
survey. The spaces for students to circle their responses were on the survey itself.
The student responses were transferred to and recorded in an Excel spreadsheet,
and the percentages of favorable, unfavorable, and neutral responses were
calculated and recorded in the Excel spreadsheet. Student names were removed
from these surveys and replaced with their assigned identification codes.
Some students were interviewed during study hall times, and some were
interviewed during physics class time. This was determined by the different
agreements that were reached with the different physics teachers who volunteered
to participate in this study. Each interview was conducted in a private office space
within the school. The interviews were audio-recorded and transcribed, and
identified only by the students’ assigned identification codes on the transcription.
In order to evaluate and compare how the different populations answered the
different interview questions, all the answers to each question were categorized
into groups of like answers. The percentages of each group of like answers for
each question were recorded in an Excel spreadsheet.
22
Chapter 4
Results and Discussion
4.1 Pretest Results and Discussion
On the pretest, both the 9th and 12th graders did only slightly better than
random guessing. This indicates the students from both grade levels had very
little background in physics at the beginning of their physics courses. While
random guessing would result in answering 4.4 questions out of 27 correctly
(16%), the ninth graders answered 5.3 questions correctly (20%), and the twelfth
graders answered 6.0 questions correctly (22%) (Table 13a). These results were
analyzed using ANOVA, and this difference in scores between the grade levels
was found to be significant (p=0.02).
Analysis of the pretest by topic (Table 13b) indicates that the students
have similar levels of understanding of the different topics, as the scores in each
topic are similar. Only, the differences in the scores of the free fall and Newton’s
First Law categories between the 9th and 12th graders were significantly different
according to ANOVA (p<0.05). This indicates that possibly at some point in
either their academic careers or personal experience these twelfth graders who
have never taken physics have acquired understanding of these topics that the
ninth graders have not. It is also possible that these twelfth-grade students were
able to answer these questions better because of their more advanced reading level
or test-taking skills.
9th Grade
12th Grade
N
216
105
Mean Score
5.3
6.0
(out of 27)
% Correct
20
22
A: Overall Pretest Scores
23
9th
Grade
12th
Grade
p-value
(9th vs.
12th)
0.001
0.031
Free Fall*
40
54
st
1 Law*
37
44
Kinematics
27
27
0.489
Graphs
3rd Law
14
13
0.874
nd
2 Law
9
11
0.181
Impetus
7
7
0.781
B: Pretest Scores by Topic (% Correct)
* indicates topics whose means are significantly
different for the different grade levels
Table 13: Pretest Scores
A: Overall Pretest Scores
B: Pretest Scores by Topic
4.2 Interview Results and Discussion
A total of fifty-four 9th graders and forty-eight 12th graders were
interviewed at some point during mechanics instruction. Because the protocols
were rotated, most questions were answered by approximately five to twenty
students from each grade. The most remarkable result of the interviews is that the
students in the different grade levels generally answered the questions very
similarly. Analysis of the responses can be made manageable by categorizing the
questions into three general categories: student enjoyment of physics, student
difficulties with physics, and understanding of selected physics concepts.
4.2.1 Student Enjoyment Category
In the student enjoyment category (tables 14a-e), approximately 50% of
students in each grade level reported that what they liked most about their physics
class was either the hands-on activities, or that what they are learning has
practical, real-world applications (table 14a). Also, approximately 70% of
students from each grade level reported that physics is either more enjoyable than
or equally enjoyable as the other classes they are presently taking, while
approximately 15-20% reported that physics is less enjoyable than the other
classes they are presently taking (table 14d). A majority of students from each
grade level reported that physics is more enjoyable than their past science classes
(table 14b). This number is slightly higher for the 9th graders than the 12th
graders. The attributes of the physics courses that are most commonly reported
by the students are the subject is more interesting than other courses, the course is
easier than other courses, and it is more hands-on than other courses (tables 14c
and 14 e). These data seem to suggest that 9th graders are enjoying their physics
classes at least as much as 12th graders are.
24
Question 1:
What do you
like most
about your
physics class?
Hands on/experiments/labs
Practical/real world applications
Math/equations
Figuring out how stuff works
It is interesting
Others
9th
37
13
10
8
8
24
N=38
12th
26
17
9
9
9
31
N=23
A: Student responses to question 1(by percent)
Table 14: Student Enjoyment Category Interview Results
A: Student Responses to Question 1
Question 2: Compared to other science
9th
classes you have taken in the past, is
More
71
physics more enjoyable,
Less
6
less enjoyable, or about the same?
Same
23
N=31
12th
53
13
33
N=15
B: Student responses to
question 2 (by percent)
Question 2A: If you
find physics more enjoyable than
other science classes you
have taken in the
past, what makes it
more enjoyable?
It is more
interesting
It is more
hands on
I like the
teacher
It is easier
9th
12th
91
75
27
25
18
13
14
13
N=22 N=8
C: Student responses
to question 2A
(by percent)
Question 3: Compared
to the other classes you
are taking this year, is physics
more enjoyable,
less enjoyable, or about
the same?
More
Less
About the
same
9th
54
14
32
12th
50
20
30
N=28
N=10
D: Student responses to
question 3 (by percent)
25
Table 14 Continued
B: Student Responses to Question 2
C: Student Responses to Question 2A
D: Student Responses to Question 3
Question 3A: If you find physics
more enjoyable than the other
classes you are taking this year,
what makes it more enjoyable?
It is more
interesting
It is easier
I like
science
It is hands
on
I like the
teacher
I like
math
9th
12th
27
60
13
40
20
0
13
0
7
20
7
0
N=15
N=5
E: Student responses
to question 3A
(by percent)
Table 14 Continued
E: Student Responses to Question 3A
4.2.2 Physics Difficulty Category
In the difficulty category, there are also similarities between the grade
levels, as well as some differences (tables 15a-g). Approximately 2/3 of students
from each grade level reported that physics is more difficult than the other classes
they are presently taking (table 15d), with a majority of students from each grade
level citing either the involvement of math or non-intuitive concepts as the factors
that most contribute to their difficulties in physics class (table 15a). However, the
students from the different grade levels did not agree on the difficulty of physics
relative to past science courses. Sixty-three percent of 9th graders reported that
physics is more difficult than past science courses, and only 31% of 12th graders
reported this (table 15b). This may not be surprising considering the last science
course the 9th graders took previous to physics was 8th grade science, and the last
26
science course for most 12th graders previous to physics was chemistry. This is
likely to be especially true since the physics courses cover the more tangible
subject of mechanics at the beginning of the school year when the interviews were
done. However, a majority of students from each different grade level report that
the level of math in physics is what makes it more difficult than other science
courses (table 15c).
Another similarity and difference that was evident between the grade
levels in the difficulty category was the level of counter-intuitiveness that students
perceive in the concepts of physics, and how they dealt with these counterintuitive concepts. In both grade levels, a large majority (70-80%) of students
who were interviewed reported that the concepts they are learning in physics are
sometimes counter-intuitive (table 15e). However, students from the different
grade levels dealt with these concepts differently. When interview subjects were
asked, “When you need to learn a concept in physics class that is counterintuitive, how do you help yourself understand it?”, 41% of the 9th graders
reported that they ask the teacher for help, while only 13% of the 12th graders
reported asking the teacher for help. Eighty percent of the 12th graders who were
interviewed reported that they help themselves learn the counter-intuitive
concepts by thinking about them on their own, while only 49% of 9th graders who
were interviewed reported doing this. Furthermore, 20% of the 12th graders
interviewed reported that they accept what the teacher says even if it doesn’t
make sense, while no 9th graders reported this (table 15f). These results seem to
suggest that the 12th graders are less likely than the 9th graders to ask for help
from their teacher, and more likely to try and figure things on their own when
they are having difficulty understanding a concept. This increased level of
independence in the 12th graders could be simply a function of their age and
maturity.
Question 4: What
9th
12th
is most difficult
Math/formulas/problem
74
69
about learning physics?
solving
It is not intuitive/it is
18
19
hard
Others
8
12
N=39 N=16
A: Student responses to
question 4 (by percent)
Question 5: Compared to
other science classes you have
taken in the past, is physics
more difficult, less difficult, or
about the same?
More
Less
About the same
9th
12th
63
35
22
35
15
30
N=27 N=17
B: Student responses to
question 5 (by percent)
27
Question 5A: Compared to other
science classes you have taken in the
past, is physics more difficult, less
difficult, or about the same?
There is
more math
It is more in
depth
The pace is
faster
It is not
intuitive
9th
12th
59
67
35
0
6
0
0
33
N=17 N=6
C: Student responses to
question 5A (by percent)
Table 15: Physics Difficulty Category Interview Results
A: Student Responses to Question 4
B: Student Responses to Question 5
C: Student Responses to Question 5A
Question 6: Compared to
the other classes you are
taking this year, is physics more
difficult, less difficult, or
about the same?
More
Less
About the
same
9th
63
10
12th
64
27
27
9
N=30 N=11
D: Student responses to
question 6 (by percent)
Question 6A: If physics is more
difficult than your other classes,
what makes it more difficult?
The math is
hard
It is not
intuitive
Others
9th
12th
36
50
36
50
27
0
N=19 N=6
E: Student responses to
question 6A (by percent)
28
Question 7: Are the concepts of physics
ever counter-intuitive?
9th
12th
Yes
71
83
No
29
17
N=45 N=18
F: Student responses to
question 7 (by percent)
Table 15 Continued
D: Student Responses to Question 6
E: Student Responses to Question 6A
F: Student Responses to Question 7
Question 7A: When
you are confronted
with a concept that
is counter-intuitive,
what do you do to
help yourself understand it?
Ask teacher
for help
Accept what
teachers
says even if
doesn’t
make sense
Think about
it
Read text
9th
12th
41
13
0
20
49
80
6
0
N=32 N=15
G: Student responses to
question 7A (by percent)
Table15 Continued
G: Student Reponses to Question 7A
4.2.3 Physics Knowledge Category
In the physics knowledge category, a majority of students from the
different grade levels answered most of the questions very similarly – sometimes
correctly, and sometimes incorrectly (tables 16a-i). For example, when asked
about the acceleration of a vertically launched projectile at the top of its path,
close to 100% of students from each grade level incorrectly stated that the
acceleration would be zero at that point (table 16b). This is also true when the
same question was asked on the written post-test where 82% of the students
answered this question incorrectly. However, a majority of students from both
grade levels answered most of the physics knowledge questions interview
29
correctly and similarly. This is true even if they did not answer the same
questions correctly on the written post-test.
When asked about the forces acting on a projectile, approximately 70% of
the students from each grade level correctly did not mention the initial applied
force (table 16a). However, a majority of students (85%) answered similar
questions on the written post test as though they believe there is an impetus force.
When asked about the motion of an object that is on a frictionless surface and
given a push, a majority of students (10 of 14) from each grade level correctly
stated that the object would travel at a constant velocity until something caused it
to stop (table 16c).
This question was asked in a different form on the written post-test.
Instead of asking what happens to a sled on a frictionless surface that is given a
push, the question asks to choose which horizontal force is necessary to keep a
sled on a frictionless surface moving at a constant velocity toward the right.
Eighty-three of 105 students chose, “the force is to the right and is of constant
strength.” Only 7 of 105 chose the correct answer, “no applied force is needed.”
These drastically different results seem to indicate that those students who
responded during the interviews that the sled would travel at a constant velocity
until acted upon by an outside force were really reciting Newton’s First Law more
than they were demonstrating a complete understanding of Newtonian motion.
Furthermore, in the interviews, only a small percentage of students from
each grade level (25% of 9th graders and 11% of 12th graders), were able to
correctly identify the forces acting on this sled that is traveling at a constant
velocity on a frictionless surface (table 16d). This indicates that even though they
are able to recognize that the sled’s velocity will not change in the absence of an
unbalanced force, they are not as able to recognize the actual forces that are acting
on the sled. When asked why an object sliding across a table will quickly come to
a stop, a majority of students from each grade level correctly stated that there is an
outside force of friction acting on the bottle (table 16e).
When asked about letting go of two bowling balls that are identical except
for mass from the same height above the Earth’s surface at the same time, 75% of
students from each grade level correctly stated that both balls would hit the
ground at approximately the same time (table 16f). This compares with 71% of
students answering this question correctly on the written post-test. Even though
75 % of students from each grade level answered this interview question
correctly, few of the students from the different levels were able to explain the
physics behind this occurrence correctly. Only one 9th-grade student out of the
fifteen (7%), and three of the twelve (25%) 12th graders, who answered this
question correctly, when asked why the heavier ball doesn’t hit first, stated the
reason (a=F/m) correctly. Only an additional three of the fifteen (20%) 9th
graders, and four of the twelve (33%) 12th graders, knew that it had to do with
“some kind of ratio.” Nine of the fifteen (60%) 9th-grade students, and five of the
twelve (42%) of the 12th graders, who answered that the balls will land at the
same time gave answers that indicated a complete lack of understanding of how
Newton’s Second Law applies to this situation (such as “the balls have the same
30
inertia,” “the normal force acting on both balls is the same,” or “I don’t know”)
(table 16g).
When asked about this same scenario on the moon, students from each
grade level again answered the question very similarly, and the results indicate
students’ common misconception of the effects of gravity on the moon. Thirtysix percent of 9th graders and 53% of twelfth graders correctly reported that the
balls will hit the ground at the same time (table 16h). This compares with 59% of
students answering this question correctly on the written post-test. Thirty-six
percent of the 9th graders and 40% of the 12th graders interviewed reported that
the balls do not fall because there is no gravity on the moon (table 16i). The
interviewer did not probe more deeply into students reasoning behind this answer,
but it is worth noting that 35-40% of the students who were interviewed reported
believing that there is no gravity on the moon.
Table
Question 8: Suppose
9th
12th
you throw a tennis ball
Gravity and air resistance
75
55
straight up into the air.
Gravity
0
11
After the ball leaves
Applied
force,
gravity,
air
resist
17
22
your hand, and while it
Applied force, gravity
8
11
is on the way up, what
N=12
N=9
forces are acting on the
ball?
A: Student responses to question 8 (by percent)
Question 8A: What is the
acceleration of the ball at the top of its
path?
m/s2
9.8
Zero
9th
11
89
N=9
B: Student responses to
question 8A (by percent)
31
12th
36
64
N=14
Table 16: Physics Difficulty Interview Results
A: Student Responses to Question 8
B: Student Responses to Question 8A
Question 9: A sled on a frictionless
surface is given a push. After the push,
Constant speed
what will the sled do if there is no
forever
friction or air resistance?
Slow down and
stop
Continue to speed
up
9th
12th
100
71
0
21
0
7
N=8
N=14
C: Student responses to
question 9 (by percent)
Question 9A: What forces are
acting on the sled after it is
pushed?
Gravity only
Don’t know
Gravity and normal force
Gravity and applied force
None
9th
0
0
25
25
50
N=8
12th
44
22
11
11
11
N=9
D: Student responses to
question 9A (by percent)
Question 9B: When an
object is slid across the
table, why does it stop?
There is an outside force of friction
There is an outside force of gravity
The applied force is gone
The object is in contact with the table
9th
12th
57
67
21
0
7
33
7
0
N=14 N=6
E: Student responses to question 9B
(by percent)
Table 16 Continued
C: Student Responses to Question 9
D: Student Responses to Question 9A
E: Student Responses to Question 9B
32
Question 10: You hold
two bowling balls in your
hands that are identical
except one is heavier than
the other. If you hold
them at the exact same
height, and let go of them
at the exact same time,
what will happen?
Question 10A: If
the balls land at
the same time,
why?
They both land at the same
time
The heavier ball lands first
9th
12th
71
29
N=21
75
25
N=16
F: Student responses to question 10
(by percent)
a=F/m
Newton’s Second Law
They are dropped from a small height
Some kind of ratio
Don’t know
Others
9th
6
6
6
19
19
44
N=16
12th
25
0
0
33
8
33
N=12
G: Student responses to question 10A (by percent)
Question 10B: If the heavier
ball hits first, why?
The heavier ball has more
gravitational force
The heavier ball is less
affected by air resistance
9th
12th
60
50
40
N=5
50
N=4
H: Student responses to question 10B
(by percent)
Table 16 Continued
F: Student Responses to Question 10
G: Student Responses to Question 10A
H: Student Responses to Question 10B
33
Question 10C:
What if the same
two bowling balls
were dropped by
a person
standing on the
surface of the
moon?
They both land at the same
time
Neither balls falls/there is no
gravity
The heavier ball lands first
The lighter ball lands first
Don’t know
9th
12th
40
53
33
13
7
7
N=15
40
7
0
0
N=15
I: Student responses to question 10C (by percent)
Table 16 Continued
I: Student Responses to Question 10C
4.3 Attitudes and Expectations Survey Results and Discussion
During the course of this project the physics teachers who volunteered to
let their classes participate in this study were asked to fill out surveys for each of
the classes that they taught to determine if any measurable differences exist in
student population, curriculum, or teaching method. It was discovered that all the
12th grade classes were non-honors level, but the ninth grade classes were a
mixture of honors level and non-honors level. Honors level is reserved for the
most highly achieving students, and the non-honors level consists of everyone
else. The non-honors level is typically called the “College Preparatory” (CP)
level.
All the percentages from each group are within the range of percentages
that were obtained when the MPEX survey was administered by its creators at the
University of Maryland to approximately 1500 college students from five
different universities who were enrolled in introductory physics courses. (Redish
et al., 1998) As measured by these questions, the high school students in this
study share similar attitudes and expectations toward physics as the college
students in the University of Maryland study.
Analysis of these results, in Table 17, shows that the percentages of
favorable, unfavorable, and neutral responses are almost identical for the 9th
graders at the honor’s level, and the non-honors level 12th graders. Those
percentages for the 9th graders at the non-honors level are only slightly different,
with a slightly lower percentage of favorable responses and slightly higher
percentages of unfavorable and neutral responses.
34
Sub-Group
%
Favorable
% Unfavorable
%
Neutral
University of
Maryland Study
45-60
20-30
20-30
9th Grade
Non-Honors
N=112
45
29
26
9th Grade
Honors
N=104
53
23
24
12thGrade
N=105
52
24
24
Table 17: Overall Results of Attitudes and Expectations Survey
Although the overall results show that approximately 50% of the
responses were favorable for each group of students, the results are different when
the responses are split into the coherence and concepts clusters. Favorable
responses to the coherence questions indicate the student agrees with the expert
view that physics is a connected, consistent framework of knowledge, while
unfavorable responses indicate the student sees physics as a bunch of unrelated
facts. Favorable responses to the concepts questions indicate the student sees the
concepts as more important than the equations, while unfavorable responses
indicate the student sees the equations as more important than the underlying
concepts.
The attitudes and expectations survey data in this study was broken down
in this manner (Table 18). In general, the percentage of favorable responses to the
coherence questions was above 50%, while the percentage of favorable responses
to the concepts questions was below 50%. This indicates the students in this
study generally see the different topics of physics as connected and related, but
they generally do not think the concepts underlying the equations as more
important than the equations themselves. This is generally consistent with the
MPEX survey results obtained by the researchers at the University of Maryland.
However, all of the high school students in this study answered the concepts
questions less favorably than the college students in the University of Maryland
study. The non-honors level 9th graders especially do not find the concepts
underlying the equations as more important, as only 29% of those students
answered the concepts questions favorably and 40% answered unfavorably. This
is the only category for any group of students in which the unfavorable responses
outnumber the favorable responses.
35
Sub-Group
Coherence
Concepts
University of Maryland
53 / 25 / 22
47 / 32 / 21
Study
9th Grade
Non-Honors
53 / 23 / 24
29 / 40 / 31
N=112
9th Grade
Honors
59 / 19 / 22
39 / 34 / 27
N=104
12th Grade
59 / 20 / 21
40 / 34 / 26
N=105
Table 18: Attitudes and Expectations Survey Results by Category
(% Favorable / % Unfavorable / % Neutral)
4.4
Physics Concepts Post-Test Results and Discussion
From the teacher survey results, it was discovered that all the 12th grade
classes were similar in that they were all traditionally taught (a mixture of lecture,
lab, small group activities, and problem solving), non-honors level classes.
However, two large differences were discovered in the 9th grade classes. They
were a mixture of honors and non-honors levels, as well as a mixture of two
different instructional methods. Some of the classes were taught with a traditional
method of instruction and some were taught with a modeling method of
instruction. The modeling method of instruction is a student-centered, inquirybased method in which the students perform experiments and analyze the data to
create mathematical models that are used to create physical rules and equations
(Wells et al., 1995). These differences in instruction required that both the nonhonors and honors level 9th grade populations be divided into sub-groups of those
that received modeling-based instruction and those that did not.
4.4.1
Our Data and Results
The honors level 9th graders had the highest post-test scores and <g>’s of
any of the sub-groups. The non-honors level 9th graders had the lowest post-test
scores and <g>’s of any of the sub-groups. All the <g>’s (i.e., post-test score vs.
pretest score) are significant except for those of the 9th graders in the non-honors
level classes that did not receive modeling-based instruction. For this sub-group
only, the post-test scores are not significantly different from the pretest scores.
The results for each sub-group are displayed in Table 19.
Modeling appeared to have a large effect on the non-honors level ninthgrade students’ performance on the post-test and their normalized gains. These
results are not inconsistent with data that have been collected on the efficacy of
modeling in previous studies. Hake (Hake, 1998), Wells (Wells, et al., 1995) and
Hestenes (Hestenes, 2000) found that the normalized gains of students in courses
36
that employ modeling are approximately twice as large (40-60%) as those of
students in traditionally taught courses (20-35%). In Hake’s, Wells’ and
Hestenes’ results, this was generally true for both honors and non-honors level
students. However, in this study, the use of modeling appeared to have no effect
on the honors level ninth-grade students’ performance on the post-test or their
normalized gains compared to traditional instruction.
Sub-group
N
Pretest
Score
(out of
27)
Post-test
Score
(out of
27)
<g>
p-value
(post-test
vs.
pretest)
9th Grade
Non-honors
80
5.6
6.3
.03
0.072
Non-modeling
9th Grade
Non-honors
32
5.0
8.9
.18
0.000
Modeling
9th Grade
Honors
28
5.5
13.0
.35
0.000
Non-modeling
9th Grade
Honors
76
4.9
12.5
.35
0.000
Modeling
12th Grade
Non-honors
105
6.0
10.9
.23
0.000
Non-modeling
Table 19: Overall Post-Test Scores and <g>
Because the results of the two 9th grade honors sub-groups were so similar,
a comparison of their means was performed. This was done to determine the
level of statistical significance of the difference between their means. It was
found that the p-value for comparison of the post-test means was 0.977. This
means we can report with 97.7% confidence that the difference between these
means is caused by random effects, and therefore not measurable effects. In other
words, we can report with only 2.3% confidence that this difference in means is
caused by measurable effects and not random effects. The p-value for the
normalized gain means is 1.000. This means we can report with 100% confidence
that any differences in the normalized gains in these sub-groups are caused by
random effects and not measurable effects. Given these high p-values, all 9th
grade honors level students were placed in one group for comparison with the
other groups.
37
The post-test scores (Table 20) and normalized gains (Table 21) for each
sub-group are significantly different from each other, except for the differences in
both the post-test scores (p=0.056) and normalized gains (p=0.239) between the
non-honors level 9th graders who received modeling instruction and the 12th
graders. Given the large difference in <g> between the 9th-grade non-honors
students who were taught with the modeling method of instruction and those who
were not, the effectiveness of modeling seems to be evident for this sub-group
regardless of whether or not those students achieve the same post-test scores and
gains as the 12th-grade students. For descriptive statistics on the pretest scores,
post-test scores and normalized gains of each sub-group, see Appendix E.
9th Grade
All
12th Grade
9th Grade
th
Non-honors
9 Grade
Non-honors
Sub-group
Non-honors
NonHonors
NonModeling
modeling
modeling
9th Grade
Non-honors
1.000
Nonmodeling
9th Grade
Non-honors
0.010
1.000
Modeling
All
th
9 Grade
0.000
0.000
1.000
Honors
12th Grade
Non-honors
0.000
0.056
0.008
1.000
Nonmodeling
Table 20: p-values for physics concept survey post-test score comparisons
38
Sub-group
9th Grade
Non-honors
Nonmodeling
9th Grade
Non-honors
Modeling
All
th
9 Grade
Honors
9th Grade
Non-honors
Nonmodeling
9th Grade
Non-honors
Modeling
All
Grade
Honors
9th
12th Grade
Non-honors
Nonmodeling
1.000
0.002
1.000
0.000
0.000
1.000
12th Grade
Non-honors
0.000
0.239
0.000
1.000
Nonmodeling
Table 21: p-values for <g> comparisons for physics concepts survey
When the results are broken down by topic, as shown in Table 22, we see
that each sub-group had the most difficulty with the topics of impetus and
Newton’s Second Law. Based on these results, it appears that a large majority of
students in each sub-group have the impetus (force required for motion)
misconception even after instruction, and therefore, it is logical that the gains for
Newton’s Second Law are also small.
In general, even though the non-modeling, non-honors level 9th graders
normalized gains are smaller than those of the other sub-groups, the trends of
improvement for each sub-group are similar. Each sub-group achieved its highest
gains in free fall, Newton’s First Law, and Newton’s Third Law. Impetus and
Newton’s Second Law were the lowest scores for each sub-group. These results
indicate that while all the non-honors 9th graders had difficulty with impetus and
Newton’s Second Law, the modeling method of instruction can help these
students achieve roughly the same gains as the 12th graders on each of the other
topics.
39
Subgroup
9th Grade
Nonhonors
Nonmodeling
9th Grade
Nonhonors
Modeling
All
th
9 Grade
Honors
12th
Grade
Nonhonors
Nonmodeling
Free
fall
Impetus
Kinematics
Graphs
1st
Law
2nd
Law
3rd
Law
0
1
0
9
1
6
42
1
23
33
0
50
47
17
41
39
9
65
40
14
15
30
5
45
Table 22: <g> by Topic (%) on physics concepts survey
4.4.2 The Impetus Misconception
The impetus questions ask about the net force and acceleration on an
object that is thrown vertically into the air as it is traveling upward, at the top of
its path, and while it is traveling downward. The correct Newtonian answers are
that the net force and the acceleration are both constant and in the downward
direction for the entire time that the object is in the air.
On the pretest, only 8% of the responses in this category were correct.
Sixty-four percent of the responses indicated that the net force and acceleration
are both upward and decreasing as the object is traveling upward; both are zero at
the peak of the object’s path; and both are downward and increasing as the object
travels downward. This is consistent with previous research (Clement, 1982,
Hestenes, et al, 1992) in which students either confuse the net force and the
acceleration with the velocity, or they think that there must be a net force in the
direction of travel in order for motion to be possible. Students with this impetus
misconception believe that the force applied to the object by the hand that is
throwing it stays with the object and dies away with time. They also believe that
this force that stays with the object must be larger than the gravitational force as
the object travels upward. When the magnitude of the applied force is the same as
the gravitational force, the object stops traveling upward. When the gravitational
force is larger than the applied force, it travels downward. Prior to instruction in
physics, 64% of the students in this study seem to have this impetus
misconception.
40
On the post-test, the results are only slightly different from the pretest.
Eighteen percent of the responses to these questions are correct, and 55% are
consistent with the impetus misconception.
4.4.3 Comparison of physics concepts survey results with previous largescale studies
Because the assessment of 9th graders taking physics has not been done
previously, there are no data available for comparison with the results of this
study. Therefore, the data from this study must be compared with previous
studies on university students and 12th graders.
In this study, the data for <g> generally correlate with published results
from previous studies. Beichner (1994) found that university students generally
answered 40% of the questions on the entire TUG-K post-test correctly. In this
study, these high school students collectively answered 41% of these selected
TUG-K questions correctly on the post-test. Hestenes, Wells and Swackhamer
(1992) found that the normalized gain for non-honors level 12th graders on the
entire FCI post-test was in the 20-30% range. In this study, the free fall, and
Newton’s First Law questions were those that were taken from the FCI. The
normalized gains on these questions were generally at or slightly above the 30%
level. Thornton and Sokoloff (1995) found very little improvement on the FMCE
post-test compared to the FMCE pretest, i.e. <g>’s between 5% and 20%, when
students were taught with traditional methods. In this study, the impetus,
Newton’s Second Law, and Newton’s Third Law questions were taken from the
FMCE. The normalized gains for impetus and Newton’s Second Law were
generally within this 5-20% range. However, in general the normalized gains for
Newton’s Third Law in this study are well above this 5-20% range. These results,
in general, indicate that the performance of the students in this study is similar to
the performance of students in previous studies (table 23).
While these comparisons indicate a general correlation of the results of
previous studies and this study, it should be noted that the questions that were
chosen for this study were, in many cases, thought to be the least difficult
questions of the surveys. These less difficult questions were chosen so the 9th
graders could understand and interpret them more easily. This could account for
the students in this study answering approximately 65% of the FCI questions
correctly on the post-test when only about 45-50% would be expected. This could
also account for the higher than expected normalized gains on some of the FMCE
questions. Because the FMCE questions were used for three different topics
(Impetus, Newton’s Second Law, Newton’s Third Law) in the mechanics
concepts survey used in this study, there is a wide range of <g>’s (4-40%) for
FMCE excerpts. This is a reflection of the low gains on Newton’s Second Law
questions (4%), and the higher gains on Newton’s Third Law questions (40%).
For a question-by-question comparison of the results in this study with previous
studies, see Appendices F and G.
41
Sub-group
TUG-K
FCI
FMCE
Previous Results for
Traditional Methods
~40%1
45-50%2 10-20%3
of Instruction
Previous Results for
Interactive
Engagement
N/A*
60-80%2 60-80%3
Methods of
Instruction
12th graders in this
38%
66%
29%
study
Honors level 9th
55%
63%
36%
graders in this study
CP level 9th graders
with Modeling
41%
55%
21%
method of
instruction
CP level 9th graders
with traditional
28%
46%
12%
method of
instruction
Table 23: Comparison of Post-test Scores from Previous Studies to Post-test
Scores from Physics Concepts Survey Questions in This Study
1. Beichner, 1994
2. Hestenes et al, 1992
3. Thornton and Sokoloff, 1995
* Beichner only reported results for traditional instruction
42
Sub-group
TUG-K
FCI
FMCE
Previous Results
for Traditional
N/A*
20-30%2
5-20%3
Methods of
Instruction
Previous Results
for Interactive
Engagement
N/A*
40-60%2 40-60%3
Methods of
Instruction
th
12 graders in this
15%
35%
21%
study
Honors level 9th
graders in this
41%
43%
30%
study
CP level 9th
graders with
23%
38%
17%
Modeling method
of instruction
CP level 9th
graders with
0%
5%
3%
traditional method
of instruction
Table 24: Comparison of <g> from Previous Studies
to <g> for Physics Concepts Survey Questions in
This Study
* Beichner did not report pretest scores on the TUG-K
2. Hestenes et al, 1992
3. Thornton and Sokoloff, 1995
4.4.4 Comparison of FCI Results to a Small-Scale Study in Maine
In section 2.1.5, the results of an unnamed high school’s administration of
the FCI to its students were reported. This high school was in the process of
transitioning from a physics last sequence to a Physics First sequence during the
2002-2003 to 2004-2005 school years. During these years, both their 9th and 12th
grade students were taking physics for the first time. They administered the FCI
to all the students taking physics in their school. Their results showed that the
non-honors level students in both 9th and 12th grades achieved <g>’s of 24-25%.
This correlates with the 12th grade <g>’s of 23% in this study, but not with the 7%
<g>’s for the population of non-honors-level 9th graders in this study. Their
honor’s level 9th graders achieved the highest <g>’s (52%) of any sub-group in
their school, which is consistent with the results of this study.
4.4.5 Student Achievement of Newtonian Thinking
Hestenes and Halloun (1995) define the entry threshold, and the mastery
level of Newtonian thinking as achieving a score of 60 % and 85%, respectively,
43
on the FCI. While the complete FCI was not used as the instrument in this study,
these values are a good starting place to determine who achieved these levels of
Newtonian thinking in this study. As a population, none of the sub-groups in this
study achieved even the entry threshold of Newtonian thinking. However, there
were some individuals who did achieve this. In the non-honors, non-modeling 9th
grade sub-group, 1 of 76 students (1%) achieved the entry threshold into
Newtonian thinking. In the non-honors, modeling 9th grade sub-group, 1 of 28
students (4%) achieved the entry threshold. In the honors 9th grade subgroup, 24
of 105 students (23%) achieved the entry threshold, and 3 of those students (3%
total) achieved the mastery level. In the 12th grade sub-group, 18 of 104 students
(17%) achieved the entry threshold, and 2 of those students (2% total) achieved
the mastery level (table 25). Using these standards, a large majority of students
from each sub-group are below the entry threshold into Newtonian thinking.
Sub-group
Percentage of
Percentage of
Students
Students
Achieving at Least Achieving Mastery
Entry Level
Level
Non-honors, Non1
0
modeling 9th Grade
Non-honors,
Modeling
3
0
9th Grade
Honors Level
23
3
9th Grade
12th Grade
17
2
Table 25: Percentage of Students Achieving Entry
and Mastery Level of Newtonian Thinking
defined by Halloun and Hestenes (1995)
4.4.6 Effects of Other Variables
In the surveys that were distributed to teachers, the teachers reported the
total amount of time (in weeks) they spent in mechanics instruction. They also
reported the percentage of class time each week they spent each week engaged in
different types of instruction (lecture, lab, small group work, inquiry and
modeling). The effects of modeling vs. non-modeling and honors vs. non-honors
on the 9th grade sub-group have already been demonstrated in the analysis of the
data. ANOVA was performed on the data in order to determine which of the
other variables may also affect the normalized gains.
For the 9th grade non-honors population, any differences in these variables
were tied directly to the method of instruction (modeling vs. non-modeling). The
classes at this level who used modeling as a method of instruction spent
considerably more time on mechanics instruction (20 weeks for modeling vs. 9.5
weeks for non-modeling). However, any differences in the normalized gains of
44
these sub-groups are more likely due to the modeling method as opposed to the
amount of time spent on the topic.
At the 9th grade honors level, the gains of the different sub-groups
(modeling vs. non-modeling) are identical, so none of the variables had any effect
on the gains of this population.
The data from the 12th grade population was the only data that did not
include any effects of modeling as none of the teachers in this population used
modeling. The effects of each of the variables (number of weeks spent on
mechanics, percentage of class time engaged in lecture, percentage of class time
engaged in labs, percentage of class time engaged in small group work, and
percentage of class time engaged in inquiry) on the normalized gain were tested
using ANOVA. For this population, none of these variables had a significant
effect (table 26). Therefore, it can be concluded that any differences in <g>
between sub-groups are more likely to be the result of ability level (honors vs.
non-honors) or method of instruction (modeling vs. non-modeling).
p-value for variable’s effect on
Variable
<g>
Number of Weeks
0.092
of Mechanics Instruction
Percentage of Class Time Spent
0.114
On Lecture-based Activities
Percentage of Class Time Spent
0.271
On Lab-based Activities
Percentage of Class Time Spent
0.087
On Small Group Work
Percentage of Class Time Spent
0.129
On Inquiry-based Activities
Table 26: Effects of different variables on <g>
for the 12th grade population
The activities of student-centered and interactive methods of instruction
such as modeling require more instructional time. This increase in instructional
time introduces this new variable when these methods are included in a study. In
order to try and control for this variable, Redish and colleagues at the Universtiy
of Maryland (1997) conducted a study in which a physics instructor taught and
tested two different sections of the same introductory physics class. One section
used microcomputer-based labs (MBL), which require an increase in instructional
time. The same instructor taught the second section using “lecture demonstrations
with the same MBL apparatus with much student interaction and discussion.”
These lectures lasted the same amount of time as the MBL labs in which
the students interacted with the MBL apparatus. On a post-test consisting of five
questions, the students in the lecture section (n=100) answered 36% of the
questions incorrectly, while the students in the lab section (n=161) averaged 14%
incorrect (Redish et al., 1997). This is an indication that the interactive method of
instruction, and not the amount of instructional time, that most affects student
learning. This supports the finding of our study in which the use of modeling as a
45
method of instruction had a significant effect on the post-test scores and
normalized gains, but the amount of time spent using traditional methods of
instruction did not.
4.5 Summary and Implications
The data collected suggest that the students in both grade levels enter their
respective physics courses with similar backgrounds in and understanding of
physics as indicated by their pretest scores. These pretest scores are generally
similar to those of 12th graders on the complete versions of the well known and
thoroughly tested instruments from which our pretest was assembled. In general,
pretest scores on the TUG-K, FMCE and FCI are similar to those that would be
achieved by random guessing. (Thornton and Sokoloff, 1998; Beichner, 1994;
Hake, 1998) Student performance on the survey used in this study was also
slightly better than random guessing.
The students also appear similar affectively, reporting similar attitudes
toward and expectations of their physics courses in interviews with the
investigator, and on the written survey of attitudes and expectations.
The students’ performance on this survey is also similar to the performance of
university students in introductory physics courses on the MPEX. Results from
the MPEX show these students generally answer 45-60% of the questions
favorably, 20-30% of the questions unfavorably, and 20-30% of the questions
neutrally. In this study, the students from each population (9th grade non-honors,
9th grade honors, and 12th grade non-honors) were within these ranges. However,
the ninth-grade students at the non-honors level were generally on the lower end
(45%) of this favorable range, and the higher end (29%) of this unfavorable range.
The apparent difference in attitudes and expectations of these students is most
noticeable when the survey is separated into its constituent categories of
coherence and concepts. Slightly more than 50% of students in each population
responded favorably to questions regarding the coherence of physics. However,
approximately 40% of students in this study responded favorably to questions
regarding the relative importance of physics concepts and physics equations. This
40% rate of favorable responses for this category is lower than the 50% rate of
favorable responses to the coherence questions. It is also lower than the 47% of
university students who answered favorably to this category in the University of
Maryland study (Redish, et al., 1998).
Favorable responses to the concepts questions indicate that students agree
with the expert view that the physics concepts are more important than the
physics equations. The lower favorable response rate of these high school
students may indicate that high school students are more easily overwhelmed by
the use of math and equations in their physics classes than the university students
are. This is corroborated by the interview results in this study, in which a large
majority of students who were interviewed reported that the mathematical
component and the use of equations in their physics classes were the most
46
difficult aspects of physics for them. Furthermore, the rate of favorable responses
in this category for non-honors level 9th graders is lower than the rest of the
students in this study. Only 29% of 9th grade non-honors students responded
favorably to these questions, compared to approximately 40% of the rest of the
students in this study. This is an interesting result because the non-honors, ninthgrade physics courses are thought to be generally more conceptual than the
twelfth-grade courses. This indicates that either the non-honors, ninth-grade
courses in this study are not as conceptual as they are assumed to be, or that the
typical non-honors, ninth-grade student is more easily overwhelmed by even the
small emphasis on equations that exists in the traditional ninth-grade physics
course.
The differences reported by the non-honors, ninth-grade students may be
an indication of the significant differences in conceptual understanding of
kinematics and mechanics concepts for this population that were revealed in the
post-test data. The students in this sub-group achieved the lowest post-test scores
and the lowest normalized gains, and these low scores and gains are amplified
when the method of instruction these students received is considered. The posttest scores of the students from this population whose teachers used traditional
lecture- and lab-based instruction were not significantly higher than their pretest
scores. The post-test scores and normalized gains of the students from this
population whose teachers use a modeling-based method of instruction were
significantly higher, and approached the level of the 12th grade students – all of
whom were non-honors level and received traditional instruction.
Finally, students in each grade level, regardless of instruction type, had the
most difficulty with the impetus misconception, and consequently Newton’s
Second Law. The pretest scores and the normalized gains for these topics were
the lowest of any topic for each population of student. This is an indication that
the intuition of an impetus force is deeply seated with students, and it is hard to
help these students understand the correct relationship between net force,
acceleration and motion. Therefore, it is easy to see why the students in this study
also did not do well with the Newton’s Second Law questions. These students
with the impetus misconception see a relationship between net force and motion
rather than the correct Newtonian relationship between net force and acceleration.
47
Chapter 5
Conclusions
The pretest and attitudes and expectations results indicate that, in general,
students in both 9th and 12th grade are equally ready to learn physics and have
about equal understanding of the content entering a physics course. The post-test
results indicate that, in general, ninth graders are capable of understanding
kinematics and mechanics concepts as well as twelfth graders do. However, the
level of the ninth-grade student and the teaching method used in a 9th grade course
can have very significant effects on student learning in the class. Honors-level
ninth graders scored significantly higher on the post-test than both their nonhonors-level counterparts in both the ninth and twelfth grades.
The relative success of the non-honors-level ninth graders whose teachers
employed a modeling-based method of instruction compared to their counterparts
whose teachers did not use modeling indicates that schools and teachers
considering teaching physics to these students need to carefully consider how the
course will be taught. These results suggest that in order for these students to be
able to understand the basic kinematics and mechanics concepts that are typically
taught in an introductory high school physics course, teachers need to employ a
more student-centered approach rather than the traditional lecture-based approach
that is employed in most twelfth-grade courses.
Furthermore, the non-honors level 9th graders responded favorably to a
relatively low percentage of the statements in the concepts portion of the attitudes
and expectations survey. This indicates that these students may be more easily
intimidated by an equations-based approach to physics, and that a more careful
approach to mathematical learning is necessary at this level.
It is important to note that the schools, teachers, and students who
participated in this study were all volunteers who received no compensation for
their participation. This may result in a self-selection issue that could potentially
skew the results, i.e. the schools, teachers and students who were motivated to
participate may have been more confident in their performance on the surveys,
and those that were not confident did not volunteer. It is also important to note
that there are many variables that cannot be controlled when conducting an
experiment of this type. Most noticeable are the differences in curricula, student
populations, and teaching styles that exist from classroom to classroom and from
school to school. Another factor is that all the schools in this study that teach
physics to ninth graders, teach physics to all ninth graders, while physics is an
elective course in the schools that teach physics to twelfth graders.
The differences (honors vs. non-honors, modeling vs. non-modeling) that
were represented in the population of ninth-grade students were not represented in
the population of twelfth-grade students. All of the twelfth-grade students were
non-honors level whose teachers employed a traditional method of instruction. It
would be helpful to include sub-groups of twelfth graders students at the honors
level and sub-groups who received modeling-based instruction. The exclusion of
these sub-groups leaves some important questions unanswered, such as:
 How would honors-level 12th graders compare to the other subgroups?
48

What effect would a modeling-based method of instruction have
on 12th graders?
The self-selection issue that may exist in this study may also exist in the
other sources of information on this topic. Some teachers who teach Physics First
frequently contribute presentations at physics education conferences, and articles
to physics education journals. They generally share their positive experiences
with the Physics First sequence, but rarely provide any empirical data on its
effectiveness. It is unlikely that schools and teachers who have not had success
teaching Physics First would volunteer to share their experiences by speaking at a
conference or submitting an article to a journal.
Other questions that would be beneficial to answer are:

Are any schools that have made the switch to Physics First
struggling with this new sequence?

Are there any schools who have considered switching to Physics
First that decided not to make switch?

If so, what factors influenced their decision not to switch?

Are there any schools that made the switch to Physics First and
switched back to the traditional sequence because it was
unsuccessful?

If so, what made it unsuccessful?
As logical or illogical as teaching physics to ninth graders may seem to
different people, until more empirical data is collected, arguments for and against
this sequence will remain largely opinion-based rather than fact-based. Therefore,
there are some additional important questions that need to be answered when
investigating the effectiveness of Physics First that were not addressed in this
study. Advocates of Physics First argue that modern biology courses have
become more sophisticated, and a physics and chemistry background is necessary
to understand the biochemical concepts that now dominate these courses. While
this is certainly a logical argument, there is little to no empirical data in existence
that either supports or refutes this argument. Studies should be done on the effect
of Physics First on the performance of students in their subsequent chemistry and
biology courses. A related study would examine the effect of a chemistry course
on performance in a subsequent biology course, especially in microbiological
concepts.
Advocates of Physics First also point to other benefits of this sequence
such as helping students achieve a better understanding of algebra by teaching
physics during the same year that Algebra I is taught, increasing the numbers of
12th graders taking science since 12th grade will be freed up for more interesting
electives, and ultimately creating more science majors at the university level by
creating more interest in the sciences in high school. While these arguments may
also seem logical, additional studies need to be performed to determine whether
or not they are true. Without these data, advocates of Physics First may face an
uphill battle when trying to convince the physics (and chemistry and biology) and
education communities of the benefits of this sequence.
Since the majority of 9th graders taking physics are non-honors level
students, the results for these students are the most important. In general, the
49
honors level students are more self-motivated, and their level of success is less
dependent on the classroom situation. Any complete study should consider the
success of honors level students, but because the success of the non-honors level
student is more dependent upon the curriculum, enthusiasm of the teacher,
teaching method, etc., it is these students who will be the standard-bearers of the
success for Physics First. If there are factors that improve the likelihood of
success with these students, these factors need to be identified. If Physics First is
not successful with these students, it cannot be considered successful. As Art
Hobson says in the November, 2005 edition of The Physics Teacher, “Physics
First will succeed or fail depending on the way it is implemented. If all it does is
offer a math-based first course focusing on classical physics, similar to many first
physics courses now offered in the 11th or 12th grade, it will fail for the same
reason that those courses fail” (Hobson, 2005).
50
REFERENCES
AAPT Statement on Physics First, 2002.
http://www.aapt.org/policy/physicsfirst.cfm
Bardeen, M.G., and Lederman, L., “Coherence in Science Education,” Science
281, 178-179 (1998).
Beichner, R.J., “Testing Student Interpretation of Kinematics Graphs,” American
Journal of Physics 62 (8), 750-762 (1994).
Chung, A.C.Y., and Cohen, S.N., “Genome Construction between bacterial
species in vitro: Replication and expression of Straphylococcus plasmid genes in
Esherichia coli,” The Proceedings of the National Academy of Sciences 71, 10301034 (1974).
Clement, J., “Students’ Preconceptions in Introductory Mechanics,” American
Journal of Physics 50(1), 66-71(1982).
Ewald, G., Bond Hickman, J., Hickman, P., Myers, F., “Physics First: The RightSide-Up Science Sequence,” The Physics Teacher 43, 319-320 (2005).
Glasser, H., “The Effect of Ninth-Grade Physics in One Private School on
Students’ Performance on the Mathematics Section of the PSAT” (1999),
http://www-ed.fnal.gov/arise.
Haber-Schaim, U., “High School Physics should be Taught before Chemistry and
Biology,” The Physics Teacher 22, 330-332 (1984).
Hake, R.R., “Interactive engagement vs. traditional methods: A six-thousandstudent survey of mechanics test data for introductory physics courses,” American
Journal of Physics 66, 64-74 (1998).
Hamilton, D.K., “Physics and the High School Sophomore,” The Physics Teacher
8, 457-458 (1970).
Hehn, J., Newschatz, M., “Physics for All? A Million and Counting!” Physics
Today 106, 37-43 (2006).
Hestenes, D., Halloun, I., “Interpreting the Force Concept Inventory,” The
Physics Teacher 33, 502-506 (1995).
Hestenes, D., Wells, M., Swackhamer, G., “Force Concept Inventory,” The
Physics Teacher 30, 141-166 (1992).
51
Hestenes, D., “Findings of the Modeling Workshop Project” (2000),
http://modeling.asu.edu/r&e/research.html.
Hickman, P., “Freshman Physics?” The Science Teacher 57 (3), 45-47 (1990).
Hobson A., “Considering Physics First,” The Physics Teacher 43, 485 (2005).
Lederman, L.M., “Getting High School Science in Order,” Technology Review,
61-62 (1996).
Lederman, L.M., “A Science Way of Thinking,” Education Week XVIII, 1-3
(1999).
Lederman, L.M., “Revolution in Science Education: Put Physics First!” Physics
Today 101, 11-12 (2001).
Lederman, L.M., “Physics First?” The Physics Teacher 43, 6-7 (2005).
McLaughlin, G., “Effect of Modeling Instruction on Development of Proportional
Reasoning I: an empirical study of high school freshmen” (2003),
http://modeling.asu.edu/modeling-hs.html.
McLaughlin, G., “Effect of Modeling Instruction on Development of Proportional
Reasoning II: theoretical background” (2003), http://modeling.asu.edu/modelinghs.html.
Mervis, J., “U.S. Tries Variations on High School Curriculum,” Science 281, 161162 (1998).
Meyers, F., “A Case for a Better High School Science Sequence in the 21st
Century,” The Physics Teacher 25, 78-81 (1987).
Mullis, K., et al., “Specific enzymatic amplification of DNA in vitro: the
polymerase chain reaction,” Cold Spring Harbor Symposia on Quantitative
Biology 51, 263-273 (1986).
National Commission on Excellence in Education (NCEE), A Nation at Risk: The
Imperative for Education Reform (NCEE, Washington, DC, 1983)
Pallrand, G., Lindenfeld, P., “The Physics Classroom Revisited: Have We
Learned Our Lesson?” Physics Today 85, 46-52 (1985).
Palombi, J., “The Illogic of Teaching Biology before Chemistry and Physics,”
The Physics Teacher 9, 39-40 (1971).
52
Pasero, S., “The State of Physics-First Programs,” Project ARISE, June 2001.
Physics First Website, http://members.aol.com/physicsfirst.
Project ARISE final report (revised 2003),
http://fnalpubs.fnal.gov/archive/2001/pub/pub-01-206.pdf.
Project ARISE website, http://www-ed.fnal.gov/arise
Redish, E.F., Saul, J.M., Steinberg, R.N., “On the Effectiveness of Activeengagement Microcomputer-based Laboratories,” American Journal of Physics
65, 45-54 (1997).
Redish, E.F., Saul, J.M., Steinberg, R.N., “Student Expectations in Introductory
Physics,” American Journal of Physics 66, 212-224 (1998).
Sanger, F., Nicklen, S., Carlson, A.R., “DNA Sequencing with ChainTerminating Inhibitors,” Proceedings of the National Academy of Sciences 74,
5463-5467 (1977).
Scott, E., “Comparing NAEP, TIMSS, and PISA in Mathematics and Science,”
National Center for Educational Statistics, December 2004.
Schwab, M., Bishop, J.M., Varnus, H.E., “Human N-myc gene contributes to
neoplastic transformation of mammalian cells in culture,” Nature 316, 160-162
(1985).
Sheppard, K., Robbins, D.M., “Chemistry, the Central Science? The History of
the High School Science Sequence,” Journal of Chemical Education 82, 561-566
(2005).
Sheppard, K., Robbins, D.M., “Lessons from the Committee of Ten,” The Physics
Teacher 40, 426-431 (2002).
Singer S.J., Nicholson, G.L., “The fluid mosaic model of the structure of cell
membranes,” Science 175, 720-731 (1972).
Sousa, D.A., “Are We Teaching High School Science Backward?” NASSP
Bulletin 80 (577), 9-15 (1996).
State of Maine Learning Results
http://www.maine.gov/education/lres/homepage.htm
Swartz, C., “Let’s Do Away with High School Courses in Earth Science, Biology,
Chemistry, and Physics,” The Physics Teacher 9, 65-66 (1971).
53
Thornton, R.K., “Conceptual Dynamics: Changing Student Views of Force and
Motion,” in Thinking Physics for Teaching (Bernadini, Tarsitani, Vicenti, 1995).
Thornton, R.K., Sokoloff, D.R., “Assessing Student Learning of Newton’s Laws:
The Force and Motion Conceptual Evaluation and Evaluation of Active Learning
Laboratory and Lecture Curricula,” American Journal of Physics 66, 338-352
(1998).
Washton, N.S., Teaching Science Creatively in the Secondary Schools. (W.B.
Saunders Company, Philadelphia, 1967), pp. 78-79.
Wells, M., Hestenes D., Swackhamer, G., “A Modeling Method for High School
Physics Instruction,” American Journal of Physics 63 (7), 606-619 (1995).
54
Appendix A
Physics Teacher Survey
Please answer the following questions to help provide a context for the
explanation of the student survey data. If you had classes from different grade
levels or academic levels participate in this study, please complete a different
survey for each class.
Your name: ___________________________ Your school: ________________
Textbook used: ________________________ Grade of your physics class: ____
Level of your physics class:
Honor College Prep
General
Heterogeneous
Use the following numbers to indicate how much time was spent on the following
topics:
0 = no time 1 = less than one week 2 = 1-2 weeks 3 = more than 2 weeks
____ Kinematics
____ Graphing Skills
____ Newton’s First Law
____ Newton’s Second Law
____ Newton’s Third Law
____ Energy and Momentum
Please indicate the approximate numbers of hours per week for each style of
teaching that you used:
____ Lecture
____ Lab
____ Whiteboarding
____ Small group work
____ Modeling (As described by Arizona State University, students perform
experiments, and mathematically model the data they collect to develop physics
rules and equations.)
____ Inquiry (Students perform experiments, and/or participate in tutorials and
questioning strategies that allow them to use inductive reasoning to make
discoveries about physics concepts.)
_____ Other (please specify)
How often is probeware used in your class? (please circle)
Daily A couple times a week Weekly A couple times a month
month or less
Once a
Other comments about your classroom/students/school/curriculum that may be
helpful:
55
Appendix B has been intentionally omitted. It consists of sample questions.
Appendix C
Physics Expectations Survey
Here are 14 statements which may or may not describe your beliefs about this
course. You are asked to rate each statement by circling the correct letters, where
the letters mean the following:
SD: Strongly Disagree D: Disagree N: Neutral A: Agree SA: Strongly
Agree
Answer the questions by circling the letters that best expresses your feeling.
Work quickly. Don’t overanalyze the meaning of each statement. They are
meant to be taken as straightforward and simple. If you don’t understand a
statement, leave it blank. If you understand, but have no strong opinion, circle N.
If an item combines two statements, and you disagree with either one, choose SD
or D.
1
2
3
4
5
6
7
8
9
Physical laws have little relation to what I
experience in the real world.
Knowledge in physics consists of many
pieces of information each of which applies
primarily to a specific situation.
Only very few specially qualified people are
capable of really understanding physics.
In this course, adept use of formulas is the
main thing needed to solve physics
problems effectively.
Learning physics helps me understand
situations in my everyday life.
The results of an exam don’t give me any
useful guidance to improve my
understanding of the course material.
The main skill I get out of this course is
learning how to solve physics word
problems.
Understanding physics means being able to
recall something you’ve read or been
shown.
“Problem solving” in physics basically
means matching problems with facts or
equations and then substituting values to get
a number.
56
SD
D N
A SA
SD
D N
A SA
SD
D N
A SA
SD
D N
A SA
SD
D N
A SA
SD
D N
A SA
SD
D N
A SA
SD
D N
A SA
SD
D N
A SA
10
11
12
13
14
A significant problem in this course is being
able to memorize all the information I need
to know.
The main skill I get out of this course is to
learn how to reason logically about the
physical world.
To really help us learn physics, teachers
should show us how to solve lots of
problems, instead of spending so much time
on concepts, and one or two problems.
I use mistakes I make on the homework and
the exam problems as clues to what I need
to do to understand the material better.
In this physics course, I do not expect to
understand equations in an intuitive sense.
They just need to be taken as givens.
57
SD
D N
A SA
SD
D N
A SA
SD
D N
A SA
SD
D N
A SA
SD
D N
A SA
Appendix D
Interview Protocols
I. Ball Tossed Straight Up Protocol
Consider a ball that is tossed straight up into the air by a person standing on the
ground.
1. After the ball leaves the person’s hand, and as it is going up, describe the
ball’s motion.
2. What forces are acting on the ball as it is going up? In which directions are
each of these forces acting? Is one force larger than the other? Does this
change as the ball is going up?
3. As the ball is coming back down, describe the ball’s motion.
4. What forces are acting on the ball as it is going down? In which directions are
each of these forces acting? Is one force larger than the other? Does this
change as the ball is going down?
5. Describe the ball’s motion when the ball is at its highest point.
6. What forces are acting on the ball when it is at its highest point? In which
directions are each of these forces acting? Is one force larger than the other?
II. Sled on Ice Protocol
Consider a sled on a smooth icy surface. The surface is so smooth it can be
considered frictionless. A person standing on the ice is wearing spiked shoes so
she can move on the ice.
1.
The person gives the sled a short push to the right, and stops pushing.
Describe what happens to the sled a) while it is being pushed, and b) after
the person stops pushing.
2.
What forces are acting on the sled after the person stops pushing?
3.
If the person continues to push the sled to the right instead of letting go
after a short push, describe what happens to the sled.
4.
Suppose the sled were traveling to the right at a constant speed, describe
how the person could make the sled slow down to a stop.
III. Enjoyment Question Protocol
58
1.
What do you like most about physics class?
2.
What do you like least about physics class?
3.
In the past, did you consider science one of your favorite subjects? If so,
what did you like about science in those classes?
4.
In the past did you consider science one of your least favorite subjects? If
so, what did you dislike about science in those classes?
5.
Is physics more or less enjoyable than other science classes you have
taken? What makes it more/less enjoyable?
IV. Difficulty Question Protocol
1.
What is most difficult about physics?
2.
What is least difficult about physics?
3.
In the past, was science a difficult class for you? If so, what made it
difficult?
4.
In the past, was science an easy class for you? If so, what made it easy?
5.
Is physics more or less difficult than other science classes you have taken?
What makes it more difficult?
59
Appendix E
Descriptive Statistics
9th Grade Non-Honors Non-Modeling
Mean
Range
Pretest
Post-test
5.6
6.3
<g>
0.03
0 – 11
2 – 18
-0.33 –
0.53
95%
Standard
Confidence
Deviation
Interval
5.1 – 6.1
2.1
5.7 – 6.9
2.8
-0.005 –
0.14
0.059
30
20
0.2
10
0.1
0
0
5
10
Pretest Scores
15
0.0
20
30
20
0.2
10
0.1
0
0
5
10
15
Post-test Scores
60
0.0
20
Proportion per Bar
Frequency
0.3
Proportion per Bar
Frequency
0.3
9th Grade Non-Honors Modeling
Mean
Range
Pretest
Post-test
5.2
8.9
<g>
0.18
2 – 10
2 – 18
-0.28 –
0.57
20
95%
Standard
Confidence
Deviation
Interval
4.4 – 6.0
2.2
7.6 – 10.1
3.4
0.14 – 0.22
0.6
0.5
Frequency
0.4
10
0.3
0.2
5
Proportion per Bar
15
0.1
0
0
5
10
Pretest Score
15
20
0.0
20
0.6
0.5
Frequency
0.4
10
0.3
0.2
5
0.1
0
0
5
10
15
Post-test Score
0.0
20
61
Proportion per Bar
15
0.17
9th Grade Honors
Mean
Range
Pretest
Post-test
5.0
12.6
<g>
0.35
1 – 14
5 – 26
-0.24 –
0.94
95%
Standard
Confidence
Deviation
Interval
4.5 – 5.5
2.5
11.8 – 13.4
4.2
0.31 – 0.38
40
0.3
Frequency
20
0.2
10
0.1
0
0
10
20
Pretest Score
Proportion per Bar
30
0.0
30
40
0.3
Frequency
20
0.2
10
0.1
0
0
Proportion per Bar
30
0.0
30
10
20
Post-test Score
12th Grade
62
0.19
Mean
Range
Pretest
Post-test
6.0
10.9
<g>
0.23
0 – 17
3 – 24
-0.21 –
0.72
95%
Standard
Confidence
Deviation
Interval
5.5 – 6.5
2.8
10.0 – 11.8
4.5
0.19 – 0.27
0.20
30
0.3
20
0.2
10
0.1
0
0
Proportion per Bar
Frequency
40
0.0
30
10
20
Pretest Score
30
0.3
20
0.2
10
0.1
0
0
Proportion per Bar
Frequency
40
0.0
30
10
20
Post-test Score
Appendix F
Item Analysis, by Survey Question Number,
63
and Comparison to Results of Previous Studies
Newton’s First Law
Previous Results (Hestenes et al., 1992)
Sub-Group
1
6
27
Total
12th Grade
63
26
42
44
% Pretest
12th Grade
88
47
72
69
% Post-test
12th Grade
0.41
0.28
0.52
0.45
<g>
Results of this Study
Sub-Group
12th Grade
% Pretest
12th Grade
% Post-test
12th Grade
<g>
th
9 Grade
Honors
% Pretest
9th Grade
Honors
% Post-test
9th Grade
Honors
<g>
9th Grade
CP Model
% Pretest
9th Grade
CP Model
% Post-test
9th Grade
CP Model
<g>
1
6
27
Total
68
20
52
44
83
38
69
61
0.47
0.23
0.35
0.30
56
12
35
34
88
22
70
59
0.64
0.11
0.54
0.39
57
7
57
40
84
50
47
50
0.63
0.46
-0.23
0.33
64
Sub-Group
1
6
27
Total
9th Grade
CP
% Pretest
58
17
45
39
71
27
39
44
0.31
0.12
-0.11
0.09
60
15
45
40
82
31
60
58
0.55
0.19
0.27
0.33
9th Grade
CP
% Post-test
9th Grade
CP
<g>
All Students
% Pretest
All Students
% Post-test
All Students
<g>
Newton’s Second Law
Previous Results (Thornton and Sokoloff, 1998)
Sub-Group
15
16
17
18
Total
th
12 Grade
19
17
18
16
18
% Pretest
12th Grade
26
20
24
30
25
% Post-test
12th Grade
0.09
0.04
0.07
0.17
0.08
<g>
Sub-Group
12th Grade
% Pretest
12th Grade
% Post-test
12th Grade
<g>
Results of This Study
15
16
17
18
Total
16
2
16
15
11
19
7
20
21
16
0.04
0.05
0.05
0.07
0.05
65
Sub-Group
9th Grade
Honors
% Pretest
9th Grade
Honors
% Post-test
9th Grade
Honors
<g>
9th Grade
CP Model
% Pretest
9th Grade
CP Model
% Post-test
9th Grade
CP Model
<g>
th
9 Grade
CP
% Pretest
9th Grade
CP
% Post-test
9th Grade
CP
<g>
All Students
% Pretest
All Students
% Post-test
All Students
<g>
15
16
17
18
Total
9
1
5
9
6
14
11
21
13
15
0.05
0.10
0.16
0.04
0.09
20
7
0
20
11
9
6
6
19
11
-0.10
0.00
0.06
0.00
0.00
17
1
8
17
11
18
5
6
20
12
0.00
0.04
-0.02
0.04
0.01
14
2
9
14
10
16
7
16
18
14
0.02
0.05
0.08
0.05
0.04
Newton’s Third Law
Sub-Group
12th Grade
% Pretest
12th Grade
% Post-test
12th Grade
<g>
19
Previous Results
20
21
22
23
Total
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
66
Sub-Group
12th Grade
% Pretest
12th Grade
% Post-test
12th Grade
<g>
th
9 Grade
Honors
% Pretest
9th Grade
Honors
% Post-test
9th Grade
Honors
<g>
9th Grade
CP Model
% Pretest
9th Grade
CP Model
% Post-test
9th Grade
CP Model
<g>
th
9 Grade
CP
% Pretest
9th Grade
CP
% Post-test
9th Grade
CP
<g>
All Students
% Pretest
All Students
% Post-test
All Students
<g>
19
Results of this Study
20
21
22
23
Total
26
22
10
9
6
13
63
58
46
48
51
52
0.50
0.46
0.40
0.43
0.48
0.45
20
20
4
14
8
13
78
72
67
64
65
69
0.73
0.65
0.48
0.66
0.62
0.65
17
23
20
10
13
17
61
31
28
47
44
43
0.53
0.10
0.10
0.41
0.36
0.50
27
20
9
5
8
14
28
26
14
13
16
19
0.01
0.08
0.05
0.08
0.09
0.06
23
21
9
10
8
14
59
50
43
44
46
48
0.46
0.37
0.38
0.38
0.42
0.40
67
Free Fall
Previous Results (Hestenes et al, 1992)
Sub-Group
2
3*
14
Total
th
12 Grade
33
N/A
45
39
% Pretest
12th Grade
67
N/A
75
71
% Post-test
12th Grade
0.50
N/A
0.55
0.52
<g>
* Question #3 is an original question
Sub-Group
12th Grade
% Pretest
12th Grade
% Post-test
12th Grade
<g>
th
9 Grade
Honors
% Pretest
9th Grade
Honors
% Post-test
9th Grade
Honors
<g>
9th Grade
CP Model
% Pretest
9th Grade
CP Model
% Post-test
9th Grade
CP Model
<g>
Results of this Study
2
3
14
Total
45
52
72
54
76
69
74
72
0.56
0.35
0.07
0.40
40
29
45
37
58
63
76
66
0.30
0.48
0.56
0.47
47
23
23
31
63
41
75
60
0.30
0.23
0.68
0.42
68
Sub-Group
9th Grade
CP
% Pretest
9th Grade
CP
% Post-test
9th Grade
CP
<g>
All Students
% Pretest
All Students
% Post-test
All Students
<g>
2
3
14
Total
38
49
58
48
41
46
57
48
0.05
-0.06
-0.02
0.00
47
41
55
47
60
59
71
63
0.25
0.30
0.35
0.30
Impetus
Previous Results (Thornton and Sokoloff, 1998)
Sub-Group
8
9
10
24
25
26
th
12 Grade
N/A
N/A
N/A
N/A
N/A
N/A
% Pretest
12th Grade
N/A
N/A
N/A
N/A
N/A
N/A
% Post-test
12th Grade
N/A
N/A
N/A
N/A
N/A
N/A
<g>
Sub-Group
12th Grade
% Pretest
12th Grade
% Post-test
12th Grade
<g>
th
9 Grade
Honors
% Pretest
9th Grade
Honors
% Post-test
9th Grade
Honors
<g>
Total
10
20
0.11
8
Results of This Study
9
10
24
25
26
Total
10
6
12
9
4
8
7
18
24
31
13
15
21
20
0.09
0.19
0.22
0.04
0.11
0.14
0.14
7
11
15
6
4
8
8
28
27
28
21
17
25
24
0.23
0.18
0.15
0.16
0.14
0.18
0.17
69
Sub-Group
9th Grade
CP Model
% Pretest
9th Grade
CP Model
% Post-test
9th Grade
CP Model
<g>
9th Grade
CP
% Pretest
9th Grade
CP
% Post-test
9th Grade
CP
<g>
All Students
% Pretest
All Students
% Post-test
All Students
<g>
8
9
10
24
25
26
Total
10
10
7
10
3
10
8
3
3
13
10
3
13
9
-0.08
-0.08
0.06
0.00
0.00
0.03
0.01
1
1
7
7
3
16
4
4
3
13
4
4
8
6
0.03
0.01
0.07
-0.03
0.01
-0.09
0.01
7
7
11
8
4
10
8
18
18
24
13
12
18
17
0.14
0.11
0.14
0.06
0.09
0.07
0.10
Kinematics Graphs
Results of Previous Study (Beichner, 1992)
Sub-Group
4
5
7
11
12
13
12th Grade
N/A
N/A
N/A
N/A
N/A
N/A
% Pretest
12th Grade
62
73
37
67
21
37
% Post-test
12th Grade
N/A
N/A
N/A
N/A
N/A
N/A
<g>
70
Total
N/A
50
N/A
Sub-Group
12th Grade
% Pretest
12th Grade
% Post-test
12th Grade
<g>
th
9 Grade
Honors
% Pretest
9th Grade
Honors
% Post-test
9th Grade
Honors
<g>
9th Grade
CP Model
% Pretest
9th Grade
CP Model
% Post-test
9th Grade
CP Model
<g>
th
9 Grade
CP
% Pretest
9th Grade
CP
% Post-test
9th Grade
CP
<g>
All Students
% Pretest
All Students
% Post-test
All Students
<g>
4
Results of this Study
5
7
11
12
13
Total
16
48
23
37
19
9
27
54
54
26
58
17
26
38
0.45
0.12
0.04
0.33
-0.02
0.19
0.15
43
32
17
29
16
8
24
73
59
44
80
27
53
55
0.53
0.40
0.33
0.72
0.13
0.49
0.41
60
17
17
30
10
14
21
63
28
28
56
16
47
41
0.01
0.13
0.13
0.37
0.07
0.38
0.23
66
46
22
21
22
13
28
73
37
19
18
10
13
28
0.21
-0.14
-0.04
-0.05
-0.03
0.00
0.00
54
38
16
28
17
10
27
66
49
30
55
18
33
42
0.26
0.17
0.17
0.37
0.01
0.26
0.21
71
APPENDIX G
Pretest vs. Post-test
Distribution of Answers by Percent and
Comparison to Previous Studies with Available Data
Newton’s First Law
Previous Study (Hestenes et al., 1992)
12th
Grade
N=612
A
B
C
D
E
1 Pre
1 Post
6 Pre
6 Post
27
Pre
27
Post
27
63
8
1
1
9
88
3
1
0
40
21
13
26
0
26
15
11
47
0
13
42
16
10
19
4
72
6
3
15
This Study
12th
1 Pre
1 Post
6 Pre
6 Post
27
Pre
27
Post
28
68
3
1
0
13
84
3
0
0
48
20
11
20
1
35
18
9
38
0
30
52
4
6
6
14
69
4
3
10
1 Pre
1 Post
6 Pre
6 Post
27
Pre
27
Post
34
56
8
1
0
9
88
4
0
0
55
29
5
12
0
36
38
5
22
0
41
35
7
2
15
14
68
7
0
8
9th CP
Model
N=28
1 Pre
1 Post
6 Pre
6 Post
27
Pre
27
Post
A
B
C
D
E
39
57
4
0
0
11
89
0
0
0
64
14
14
7
0
18
7
7
54
0
25
57
57
4
11
18
46
46
4
18
Grade
N=105
A
B
C
D
E
9th
Honors
N=104
A
B
C
D
E
Bold=Correct Answer
72
Newton’s First Law
(Continued)
9th CP
N=76
A
B
C
D
E
1 Pre
1 Post
6 Pre
6 Post
34
59
7
0
1
24
72
3
0
0
43
21
14
17
3
37
24
12
26
0
Bold = Correct Answer
73
27
Pre
30
41
3
5
12
27
Post
45
38
1
5
11
Newton’s Second Law
This Study
12th
Grade
N=105
A
B
C
D
E
F
G
9th
Grade
Honors
N=104
A
B
C
D
E
F
G
9th
Grade
CP
Model
N=28
A
B
C
D
E
F
G
15
Pre
15
Post
16
Pre
16
Post
17
Pre
17
Post
18
Pre
18
Post
80
16
1
0
0
2
1
73
19
5
2
0
0
0
7
83
2
2
2
2
0
5
79
4
7
2
0
1
0
0
67
4
6
16
6
5
4
49
5
7
20
10
8
3
0
1
3
14
67
4
4
1
5
3
21
62
15
Pre
15
Post
16
Pre
16
Post
17
Pre
17
Post
18
Pre
18
Post
86
9
2
0
0
2
1
81
14
2
1
1
0
0
4
90
4
0
0
2
0
1
86
2
10
0
0
1
0
1
75
5
5
6
6
1
0
57
4
4
20
11
2
0
0
2
9
9
78
2
0
1
1
5
14
77
15
Pre
15
Post
16
Pre
16
Post
17
Pre
17
Post
18
Pre
18
Post
61
18
11
0
4
0
4
89
7
0
4
0
0
0
11
75
7
7
0
0
0
0
96
0
4
0
0
0
4
4
75
14
0
0
4
0
0
79
0
11
4
4
0
11
4
0
11
18
57
0
0
0
0
0
21
79
Bold = Correct Answer
74
Newton’s Second Law
(Continued)
9th
Grade
CP
N=76
A
B
C
D
E
F
G
15
Pre
15
Post
16
Pre
16
Post
17
Pre
17
Post
18
Pre
18
Post
75
17
1
1
1
3
1
76
18
4
0
0
1
1
8
75
8
1
4
4
0
13
76
3
5
3
1
0
5
6
64
5
9
8
1
1
3
73
3
8
6
6
4
3
0
4
13
17
60
6
1
5
0
10
20
57
Bold = Correct Answer
75
Newton’s Third Law
This Study
12th
Grade
n=105
A
B
C
D
E
F
J
9th
Grade
Honors
n=104
A
B
C
D
E
F
J
9th
Grade
CP
Model
N=28
A
B
C
D
E
F
J
19
Pre
19
Post
20
Pre
20
Post
21
Pre
21
Post
22
Pre
22
Post
23
Pre
23
Post
63
2
3
4
26
1
1
27
2
5
2
63
2
0
13
45
4
2
22
10
3
6
22
3
0
57
10
0
3
66
4
8
10
10
0
10
35
5
0
46
3
0
7
75
3
4
9
1
1
2
42
2
4
47
2
0
0
77
8
3
7
7
0
5
33
3
5
50
3
0
19
Pre
19
Post
20
Pre
20
Post
21
Pre
21
Post
22
Pre
22
Post
23
Pre
23
Post
65
2
4
7
20
1
1
21
0
1
0
78
0
0
12
48
5
4
20
7
4
8
16
0
0
72
4
0
10
58
12
7
4
4
5
4
25
0
2
67
1
0
4
78
2
1
14
0
1
2
32
1
0
64
0
0
5
62
9
5
8
3
6
2
32
0
0
65
0
0
19
Pre
19
Post
20
Pre
20
Post
21
Pre
21
Post
22
Pre
22
Post
23
Pre
23
Post
73
0
0
7
17
3
0
35
0
4
0
61
0
0
10
40
17
3
23
7
0
15
39
6
3
30
3
0
13
37
3
10
20
13
0
24
33
3
3
27
6
0
13
60
13
0
10
3
0
3
36
0
9
45
3
0
7
57
7
7
14
7
3
0
42
3
0
42
9
0
Bold = Correct Answer
76
Newton’s Third Law (Continued)
9th
Grade
CP
n=76
A
B
C
D
E
F
J
19
Pre
19
Post
20
Pre
20
Post
21
Pre
21
Post
22
Pre
22
Post
23
Pre
23
Post
44
7
9
1
27
9
1
56
3
5
3
28
4
3
12
37
4
7
20
11
7
15
37
9
3
25
10
0
11
51
9
2
9
11
7
11
54
5
4
14
11
0
15
58
9
3
4
7
1
6
66
3
8
13
4
1
3
52
10
9
8
6
10
8
57
4
5
16
9
0
Bold = Correct Answer
77
Free Fall
Previous Study (Hestenes et al., 1992)
12th
Grade
n=612
A
B
C
D
E
3
14
14
Post
Pre Post
*
1
1
N/A N/A
11
7
N/A N/A
8
4
33
67
58
29
N/A N/A
45
75
8
3
N/A N/A
27
11
0
0
N/A N/A
9
3
* Question #3 is an original question
2
Pre
2
Post
3
Pre*
This Study
12th
Grade
n=105
A
B
C
D
E
9th
Grade
Honors
n=104
A
B
C
D
E
9th
Grade
CP
Model
n=28
A
B
C
D
E
2
Pre
2
Post
3
Pre
3
Post
14
Pre
14
Post
0
44
40
6
8
0
76
20
4
0
32
52
0
11
5
13
69
0
14
3
7
4
72
15
2
9
6
75
9
1
2
Pre
2
Post
3
Pre
3
Post
14
Pre
14
Post
0
40
45
10
4
0
58
34
5
2
47
29
3
15
5
28
63
0
7
2
17
7
45
22
7
8
6
75
9
2
2
Pre
2
Post
3
Pre
3
Post
14
Pre
14
Post
0
47
50
0
3
0
62
38
0
0
63
23
3
17
0
50
40
0
3
7
17
13
23
33
13
7
3
75
13
3
Bold=Correct Answer
78
Free Fall (Continued)
9th
Grade
CP
N=76
A
B
C
D
E
2 Pre
2 Post
3 Pre
3 Post
14
Pre
14
Post
3
38
43
5
12
1
41
35
15
8
39
49
3
6
3
46
46
0
7
1
8
9
58
18
5
18
11
57
8
6
Bold = Correct Answer
79
Impetus
This Study
12th
8
8
9
9
10
10
24
24
25
25
26
26
Grade
Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post
n=105
A
10
22
6
24
12
32
9
13
4
16
8
22
B
11
15
6
3
64
52
6
10
3
7
56
34
C
4
2
2
4
6
5
9
17
5
0
14
9
D
1
2
81
64
1
2
1
5
85
64
0
3
E
5
5
2
5
2
4
8
11
2
10
5
10
F
15
14
0
0
14
5
24
10
0
1
16
16
G
54
40
3
0
0
1
43
34
1
2
1
6
9th
Grade
8
8
9
9
10
10
24
24
25
25
26
26
Honors Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post
n=104
A
7
28
11
27
14
28
6
21
4
17
8
25
B
5
13
9
5
70
55
5
14
10
7
71
59
C
6
3
2
1
7
6
7
9
2
1
7
4
D
1
2
67
59
1
1
0
0
78
73
1
0
E
4
2
4
3
2
1
8
6
1
1
3
4
F
10
10
0
0
4
6
12
11
1
0
7
8
G
67
41
5
5
1
2
62
37
3
0
1
0
9th
Grade
CP
Model
n=28
A
B
C
D
E
F
G
8
8
9
9
10
10
24
24
25
25
26
26
Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post
10
10
3
3
0
30
43
3
0
13
0
6
23
55
10
13
0
53
13
0
10
3
9
0
75
0
0
9
7
63
23
0
10
3
0
12
69
6
0
0
6
3
10
3
17
0
0
23
47
Bold = Correct Answer
80
10
0
17
60
3
37
37
3
10
7
90
7
3
10
3
0
3
3
0
0
3
10
53
17
3
0
17
0
13
61
7
3
0
17
0
Impetus (Continued)
9th
Grade 8
8
9
9
10
10
24
24
25
25
26
CP
Pre Post Pre Post Pre Post Pre Post Pre Post Pre
n=76
A
1
4
1
3
6
10
7
5
3
5
5
B
3
5
1
10
71
56
5
8
3
5
45
C
1
1
3
3
17
22
9
14
6
5
23
D
1
2
74
72
0
5
9
5
71
78
0
E
11
10
9
3
1
14
12
14
8
3
3
F
25
27
3
5
1
22
22
22
3
10
10
G
58
50
6
1
3
35
31
35
7
1
1
Bold = Correct Answer
81
26
Post
8
47
18
3
5
20
0
Kinematics Graphs
Previous Study (Beichner, 1992)
12th
Grade
and
College
n=895
A
B
C
D
E
4
Pre
Post
5
Pre
Post
7
Pre
Post
11
Pre
Post
12
Pre
Post
13
Pre
Post
N/A
N/A
N/A
N/A
N/A
8
0
20
62
10
N/A
N/A
N/A
N/A
N/A
3
2
73
18
4
N/A
N/A
N/A
N/A
N/A
11
11
37
37
5
N/A
N/A
N/A
N/A
N/A
14
67
8
2
9
N/A
N/A
N/A
N/A
N/A
21
46
8
7
19
N/A
N/A
N/A
N/A
N/A
19
9
37
12
23
4
5
7
11
12
13
This Study
12th
4
4
5
5
7
7
11
11
12
12
13
13
Grade
Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post
n=105
A
8
23
3
4
45
29
34
25
29
19
19
17
B
0
0
2
4
16
23
18
26
5
11
36
59
C
16
14
24
17
4
3
4
1
46
55
9
26
D
48
37
4
4
7
11
14
8
53
55
10
26
E
23
8
1
1
5
5
21
9
50
45
43
36
9th
Grade
4
4
5
5
7
7
11
11
12
12
13
13
Honors Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post
n=104
A
28
16
1
5
27
11
38
8
31
8
15
27
B
0
0
5
6
33
25
16
32
5
9
29
79
C
14
7
13
17
14
8
6
5
32
59
8
52
D
59
26
4
0
5
4
15
12
43
73
16
45
E
15
4
2
4
10
3
15
5
55
33
39
19
Bold=Correct Answer
82
Kinematics Graphs
(Continued)
9th
Grade
CP
Model
n=28
A
B
C
D
E
9th
Grade
CP
n=76
A
B
C
D
E
4
Pre
4
Post
5
Pre
5
Post
7
Pre
7
Post
11
Pre
11
Post
12
Pre
12
Post
13
Pre
13
Post
20
0
7
60
10
21
0
9
60
6
3
3
17
77
0
3
9
27
54
3
27
23
20
17
13
18
27
15
27
9
37
30
20
7
7
21
54
18
0
3
10
0
0
10
80
15
9
9
3
60
17
7
13
7
53
6
12
45
6
27
4
Pre
4
Post
5
Pre
5
Post
7
Pre
7
Post
11
Pre
11
Post
12
Pre
12
Post
13
Pre
13
Post
30
3
0
66
3
21
0
1
73
3
55
21
16
3
5
58
18
10
9
5
22
23
8
9
38
10
19
4
9
58
33
7
7
13
40
31
6
12
18
31
5
1
21
17
1
0
49
47
5
11
45
36
42
44
22
19
7
19
3
5
Bold = Correct Answer
83
APPENDIX H
HUMAN SUBJECTS RESEARCH PROPOSAL
An Investigation of the Effectiveness of Physics First in Maine
General Topic with Rationale:
I would like to research the “physics-first” model of high school science
instruction, in which physics in a ninth-grade course. Traditionally, high school
science instruction has followed a “physics-last” model, in which physics is a
twelfth-grade course. There is some literature that suggests that physics-last is
not the best model, such as Dr. Leon Lederman’s article in the September, 2001
of the “Physics Teacher”. The physics-first model is gaining more attention. A
growing number of high schools have already switched to this model of
instruction, and many more are considering it. I would like to find a population of
ninth graders in Maine who are taking physics, and investigate their progress in
this course.
General Methodology
My goal is to find out what ninth graders learn compared to what twelfth
graders learn in physics. I will do this by developing an instrument to assess
students’ gains in physics, and administer it to a population of ninth graders and
to a population of twelfth graders that are both currently taking physics. I will
then compare the results from each grade level to each other, and look for
differences in normalized gains in each grade level. I will do this through the use
of pre- and post- tests.
I will also conduct student interviews throughout the course to find out more
about the students that I can’t learn through the pre- and post-tests. I will choose
students randomly and ask them questions about the concepts being taught, and
about their attitudes toward the physics course they are taking. By conducting
interviews, I can ask questions about the physics concepts being taught that
require them to give explanations, and I can ask different follow up questions
depending on the answers they give me.
Personnel
Michael O’Brien is a former high school science teacher and current
student in the Master of Science in Teaching program.
Subject Recruitment
High schools in the state of Maine that teach physics to ninth graders will
be asked to participate in this study. The subject population will be composed of
those students at participating high schools who are taking physics.
Informed Consent
84
Students will not participate without permission of a parent or guardian.
Parents or guardians must submit a signed permission slip in order for students to
participate.
Confidentiality
Student names will be removed from surveys, and replaced with id codes.
The surveys will be kept in the investigator’s locked office indefinitely. Only the
investigator, and his faculty advisors will have access to the data.
Risk to Subjects
There is no risk to the subjects beyond their time and inconvenience.
Benefits
Subjects will receive no monetary benefits.
85
Informed Consent Form
My name is Michael O’Brien. I am a graduate student at the University of Maine
in the Master of Science in Teaching program. As part of my master’s work, I am doing
research on the effectiveness of teaching physics to ninth graders. Your child’s school
has agreed to participate in this research project. Traditionally, physics has been taught
to twelfth graders. However, more and more schools are changing the sequence of
science courses, and teaching physics to ninth graders. It is important to do research in
this area to help schools better make this transition.
Since your child is taking physics, I am requesting your permission to allow your
child to participate in this study. Their participation is totally voluntary. They will be
asked to take three surveys in total. One survey will be administered at the beginning of
their physics course, and the second survey will be administered later in the year. These
two surveys will be used to determine student understanding of important physics
concepts at the beginning of the course, and after the concepts are taught. The third
survey will be administered at the end of the course to determine the students’ attitudes
toward physics.
Results of these surveys will be compiled and used to assess the effectiveness of
teaching physics to ninth graders. Each test will take about 30-40 minutes to complete,
and will be done during physics class time. Your child’s name will not be used in any
publication. As soon as survey information is received, your child’s name will be
removed and an ID code will be assigned. The survey data (without your child’s name)
will become a permanent part of the research data maintained by the University of Maine.
These surveys will not be used in any way to determine your child’s grade in physics
class.
Additionally, I will be interviewing students periodically throughout the school
year to gain insight into their attitudes toward physics. This part of the research, like the
survey, is totally voluntary and will not take place without your permission. The
interviews will take place during study hall times. Some interviews will be audio taped
for better record keeping. These audiotapes will not include your child’s name, and will
become a permanent part of the research data maintained by the University of Maine.
Again, your child’s name will not be used in any publication.
Please complete and sign the attached permission slip indicating whether or not
you want your child to participate. Thank you for your time and consideration in this
matter.
Sincerely,
Faculty Sponsor:
Michael O’Brien
John Thompson
5709 Bennett Hall
5709 Bennett Hall
University of Maine
University of Maine
Orono, ME 04469
Orono, ME 04469
(207)581-1031
(207)581-1030
michael_obrien@umit.maine.edu
john_thompson@umit.maine.edu
86
If you have any questions regarding your rights as a research participant, please contact:
Gayle Anderson, Assistant to the University of Maine’s Protection of Human Subjects
Review Board, 5717 Corbett Hall, Room 443, University of Maine, Orono, ME 04469
(207)581-1498
gayle_anderson@umit.maine.edu
Sample Physics Concepts Survey Questions
Sample Attitude Survey Questions
Students will be asked to read and determine their level of agreement with statements like
the following:
A significant problem in this course is being able to memorize all the information I need
to know.
The main skill I get out of this course is to learn how to reason logically about the
physical world.
Sample Interview Questions
Is physics class more or less interesting than other science classes you have taken?
Is physics class more or less difficult than other science classes you have taken?
What do you like most about physics?
What do you like least about physics?
Parental Permission Form
Permission Slip
I (please circle) (do / do not) give my child, ___________________ ,permission to
participate in this study by taking two surveys in physics class– one at the beginning of
the course, and one at the end of the course.
____________________
(parent’s signature)
I (please circle) (do / do not) give my child, ____________________, permission to
participate in this study by participating in interviews during study hall time with the
researcher, Michael O’Brien.
_________________________
(parent’s signature)
BIOGRAPHY OF THE AUTHOR
Michael J. O’Brien was born in Westborough, Massachusetts on December 27,
1966. He graduated from Westborough High School in 1984, and from Lyndon
State College in Vermont with a Bachelor of Science (in Meteorology) in 1990.
Michael taught high school math and science in Maine for seven years. He is a
candidate for the Master of Science in Teaching degree from the University of
Maine in May, 2006.
87