RESOURCE LETTER

Roger H. Stuewer, Editor

School of Physics and Astronomy, 116 Church Street
University of Minnesota, Minneapolis, Minnesota 55455

This is one of a series of Resource Letters on different topics intended to guide college physicists,
astronomers, and other scientists to some of the literature and other teaching aids that may help
improve course content in specified fields. @The letter E after an item indicates elementary level or
material of general interest to persons becoming informed in the field. The letter I, for intermediate
level, indicates material of somewhat more specialized nature; and the letter A, indicates rather
specialized or advanced material.# No Resource letter is meant to be exhaustive and complete; in time
there may be more than one letter on some of the main subjects of interest. Comments on these
materials as well as suggestions for future topics will be welcomed. Please send such communications
to Professor Roger H. Stuewer, Editor, AAPT Resource Letters, School of Physics and Astronomy, 116
Church Street SE, University of Minnesota, Minneapolis, MN 55455.

Resource Letter: PER-1: Physics Education Research

Lillian C. McDermott
Department of Physics, University of Washington, Seattle, Washington 98195
Edward F. Redish
Department of Physics, University of Maryland, College Park, Maryland 20742
~Received 9 September 1998; accepted 11 March 1999!
The purpose of this Resource Letter is to provide an overview of research on the learning and
teaching of physics. The references have been selected to meet the needs of two groups of physicists
engaged in physics education. The first is the growing number whose field of scholarly inquiry is ~or
might become! physics education research. The second is the much larger community of physics
instructors whose primary interest is in using the results from research as a guide for improving
instruction. © 1999 American Association of Physics Teachers.

I. INTRODUCTION widely distributed, journals from outside of the United States

are not included. References to conference proceedings and
Experienced instructors recognize that in spite of their best books have been kept to a minimum.
efforts many students emerge from their study of physics In the selection of references, careful consideration has
with serious gaps in their understanding of important topics. been given not only to quality but also to breadth in objec-
In the last two decades, physicists have begun to approach tives, methods, and subject matter. The emphasis is on sys-
this problem from a scientific perspective by conducting de- tematic investigations of student learning. Thus many in-
tailed systematic studies on the learning and teaching of sightful and inspirational reflections based on teaching
physics. These investigations have included a wide variety of experience have not been included. Descriptions of the de-
populations, ranging from young children to professional velopment and implementation of innovative courses have
not been cited unless they also expand our knowledge of
physicists. This Resource Letter is not intended to provide
how students learn. Also absent are articles in which the
either a complete listing or a historical record of this re-
effectiveness of instruction is primarily assessed by the per-
search. Rather it is meant to contribute to the establishment
formance of students on traditional end-of-chapter problems,
of a research base that can serve as a resource for ongoing by their own assessment of their learning, or by how they ~or
improvement and enrichment of student learning in physics. their instructors! feel about an educational innovation.
Although some studies involving precollege students are Choices have been made among similar studies by different
included, the primary emphasis is at the university level. A investigators. When there are multiple papers by the same
major consideration in the selection of references has been authors on similar topics, only the more readily available are
their intellectual and physical accessibility to readers of the cited.
American Journal of Physics. Most of the articles cited are The references have been organized into sections. Section
from the American Journal of Physics and The Physics II contains bibliographies and conference proceedings. Read-
Teacher. Additional sources include Physics Today, Comput- ers unfamiliar with the literature might find it helpful to be-
ers in Physics, the Journal of Research in Science Teaching, gin with the reviews and overviews in Sec. III. Section IV,
Science Education, and a few other multidisciplinary jour- the core of the Resource Letter, is devoted to empirical stud-
nals on the teaching and learning of science and mathemat- ies. The references in Sec. V contain some theoretical per-
ics. Except for the International Journal of Science Educa- spectives. A few references from related fields are listed in
tion ~formerly the European Journal of Science Education! Sec. VI. In Sec. VII are examples of instructional materials
and Physics Education, which are published in English and that have been developed on the basis of findings from re-

755 Am. J. Phys. 67 ~9!, September 1999 © 1999 American Association of Physics Teachers 755
search and that also have been evaluated through docu- 10. ‘‘Research on conceptual understanding in mechanics,’’ L. C. McDer-
mented use with students. Section VIII identifies some ear- mott, Phys. Today 37 ~7!, 24–32 ~1984!. This article identifies critical
elements of research on student understanding in physics and reviews
lier Resource Letters that can provide useful background for
the early research on conceptual and reasoning difficulties in mechanics.
readers interested in conducting research in physics educa- 11. ‘‘Scientific approaches to science education,’’ F. Reif, Phys. Today 39
tion. Articles that fit into more than one category are cross- ~11!, 48–54 ~1986!. This article takes a more theoretical perspective
referenced. For the most part, references within sections and than the one above.
subsections are ordered chronologically, from earliest to lat- 12. ‘‘A view from physics,’’ L. C. McDermott, in Toward a Scientific Prac-
est. tice of Science Education, edited by M. Gardner, J. G. Greeno, F. Reif,
A. H. Schoenfeld, A. diSessa, and E. Stage ~Lawrence Erlbaum Asso-
ciates, Hillsdale, NJ, 1990!, pp. 3–30. This paper presents a broad re-
II. GENERAL REFERENCES view of research on conceptual understanding.
13. ‘‘Instructional design, cognition, and technology: Applications to the
A. Bibliographies
teaching of scientific concepts,’’ F. Reif, J. Res. Sci. Teach. 24 ~4!,
There is an extensive literature on research in science edu- 309–324 ~1987!. This article presents a good overview of how cognitive
cation. Readers interested in exploring this literature should science and educational theory can contribute to the design of effective
instruction.
consult one or more of the following bibliographies. 14. ‘‘Learning to think like a physicist: A review of research-based instruc-
1. Research on Students’ Conceptions in Science: A Bibliography, P. Car- tional strategies,’’ A. Van Heuvelen, Am. J. Phys. 59, 891–897 ~1991!.
michael et al., Centre for Studies in Science and Mathematics Education This article reviews research on student learning of physics with a focus
~University of Leeds, Leeds, UK, 1990!. This bibliography should be on general issues such as knowledge representation and concept organi-
useful to readers who are interested in learning about studies with pre- zation. Some instructional strategies are discussed.
college students. 15. ‘‘Research on problem solving: Physics,’’ D. P. Maloney, in Handbook
2. Bibliography: Students’ Alternative Frameworks and Science Education, of Research on Science Teaching and Learning, edited by D. Gabel
4th Edition, H. Pfundt and R. Duit ~IPN Reports-in-Brief, Kiel, Germany, ~MacMillan, New York, 1993!, pp. 327–354. This article provides a
1994!. This bibliography is also available on the Internet at the ftp site very detailed and comprehensive review of the extensive literature in
ftp.topgun.idbsu.edu in the directory /pub/plrnr in the files: plr11mac.bin education and cognitive science on the use of physics problems as a
and plr12mac.bin ~Word for the Macintosh!, or plr11pc.doc and context for examining cognitive processes and approaches to problem
plr12.pc.doc ~Word for Windows!. solving.
Two additional bibliographies that focus on physics are also available on 16. ‘‘Teaching physics: Figuring out what works,’’ E. F. Redish and R. N.
the same site. Steinberg, Phys. Today 52 ~1!, 24–30 ~1999!. This paper discusses re-
3. D. I. Dykstra Jr., F. Monte, and S. Schroeder, Boise State University, search on improving instruction in engineering physics. The focus is on
filename: plr03mac.bin ~Word 5.1 for the Macintosh!, plr03pc.doc ~Word
what has been learned about the teaching of concepts and about the
for Windows!.
attitudes that students bring to their study of physics.
4. D. Maloney, Indiana University–Purdue University, Fort Wayne, file-
name: plr16mac.bin ~Word 5.1 for the Macintosh!, plr16pc.doc ~Word Perspectives of research groups have appeared in pub-
for Windows!. lished versions of the Millikan Award Lectures, in the 1996
ICUPE Proceedings ~see Ref. 9!, and in Guest Comments in
B. Conference proceedings the AJP. @The Robert A. Millikan Award recognizes ‘‘no-
table and creative contributions to the teaching of physics.’’
Physics education research has been a major theme at sev- This is an annual award of the AAPT ~American Association
eral national and international conferences. Readers who of Physics Teachers!.# These also provide extensive lists of
would like to explore beyond the papers cited in this Re- references.
source Letter may wish to consult the published proceedings
17. ‘‘Millikan Lecture 1990: What we teach and what is learned—Closing
listed below.
the gap,’’ L. C. McDermott, Am. J. Phys. 59, 301–315 ~1991!.
5. Research on Physics Education, Proceedings of the First International 18. ‘‘Millikan Lecture 1994: Understanding and teaching important scien-
Workshop, La Londe Les Maures, France, edited by G. Delacôte, A. tific thought processes,’’ F. Reif, Am. J. Phys. 63, 17–32 ~1995!.
Tiberghien, and J. Schwartz ~Éditions du CNRS, Paris, France, 1983!. 19. ‘‘Bridging the gap between teaching and learning: The role of re-
6. Relating Macroscopic Phenomena to Microscopic Particles, Proceedings search,’’ L. C. McDermott, AIP Conf. Proc. 399, 139–165 ~1997!. ~See
of an International Seminar, Utrecht, The Netherlands, 23–26 October item 9.!
1989, edited by P. L. Lijnse, P. Licht, W. de Vos, and A. J. Waarlo 20. ‘‘How can we help students acquire effectively usable physics knowl-
~CD-gbs, Utrecht, NL, 1990!. edge?’’ F. Reif, AIP Conf. Proc. 399, 179–195 ~1997!. ~See item 9.!
7. Research in Physics Learning: Theoretical Issues and Empirical Studies, 21. ‘‘Guest comment: How we teach and how students learn—a mis-
Proceedings of an International Workshop, Bremen, Germany, 4–8 match?’’ L. C. McDermott, Am. J. Phys. 61, 295–298 ~1993!.
March 1991, edited by R. Duit, F. Goldberg, and H. Niedderer ~IPN, 22. ‘‘Who needs physics education research?’’ D. Hestenes, Am. J. Phys.
Kiel, Germany, 1992!. 66, 465–467 ~1998!.
8. Conference on the Introductory Physics Course on the Occasion of the
Retirement of Robert Resnick, Troy, NY, 20–23 May 1993, edited by Jack Since many conceptual and reasoning difficulties identi-
Wilson ~Wiley, New York, 1997!. fied among younger students are also common among under-
9. The Changing Role of Physics Departments in Modern Universities: Pro- graduates, familiarity with the pre-college literature is impor-
ceedings of the International Conference on Undergraduate Physics Edu-
tant for physicists who conduct research with students of any
cation (ICUPE), College Park, MD, 31 July–3 Aug. 1996, edited by E. F.
Redish and J. S. Rigden, AIP Conf. Proceedings No. 399 ~American age. The two reviews below are concerned with student
Institute of Physics, Woodbury, NY, 1997!. learning in high school.
23. ‘‘Pupils and paradigms: A review of literature related to concept devel-
opment in adolescent science students,’’ R. Driver and J. Easley, Stud.
III. REVIEWS, OVERVIEWS, AND PERSPECTIVES
Sci. Educ. 5, 61–84 ~1978!.
A number of reviews, overviews, and perspectives on re- 24. ‘‘Learning and instruction in pre-college physical science,’’ J. Mestre,
Phys. Today 44 ~9!, 56–62 ~1991!.
search in physics education have been written by physicists.
The articles below include extensive references and provide The information contained in the papers above is also use-
a good background for an initial study of the literature in this ful for faculty who teach physics or physical science to
field. K–12 teachers. An additional set of articles on the applica-

756 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 756
tion of physics education research to the preparation of identify all the relevant variables or to perform a controlled
teachers can be found in the following on-line book. experiment in which only a single variable is changed. ~For
25. Connecting Research in Physics Education with Teacher Education, example, experiments are not repeatable for individual quan-
edited by A. Tiberghien, E. L. Jossem, and J. Barojas @http:// tum events.! Yet, both in physics education and in quantum
www.physics.ohiostate.edu/;jossem/ICPE/BOOKS.html#. physics, experience demonstrates that reliable and reproduc-
ible results can be obtained.
In an idealized physics experiment, an effort is made to
IV. EMPIRICAL STUDIES ensure that the effect of a probe on the system that is being
measured is small. However, it is not always possible to find
In selecting the references for this section, we have been such a probe, especially in quantum systems. In physics edu-
guided by several criteria that can be summarized as follows: cation research, weak coupling is not always desirable. For
~1! The focus is on the phenomenon being studied, which in example, to learn what is really going on in the minds of
this case is the learning of physics by students. ~2! The re- students, the investigator often must interact strongly with
search is conducted in a systematic manner. ~3! The proce- them.
dures are described in sufficient detail so that they can be The level of quantification must be appropriate to the situ-
replicated. ation that is being studied. In traditional physics experi-
The primary consideration in all cases has been that the ments, the goal is to obtain quantitative results with the un-
investigation be focused on the student as a learner, not on certainty in the measurements well specified and as small as
the instructor or on the material covered. The authors must possible. However, a meaningful interpretation of numerical
show that they attempted to find out what students actually results requires a sound qualitative understanding of the un-
thought and explain how that information was determined. derlying physics. In studies involving students, the value of
They should provide evidence that the investigation was con- quantitative results also depends on our understanding of
ducted carefully and systematically. The authors should de- qualitative issues, which usually are much less well under-
scribe the context for the study, such as the physical setting, stood than in the case of physical systems. To be able to
time frame, and the size and characteristics of the student determine the depth of students’ knowledge and the nature of
population involved. If the response to instruction is being their difficulties, it is necessary to probe the reasoning that
probed, it is necessary to note specific features of the course, lies behind their answers. The analysis of numerical data
including length, sequence of topics, and any special charac- alone may lead to incorrect interpretations. Detailed investi-
teristics. Since in an educational framework results can be gations with a small number of students can be very useful
sensitive to environmental and contextual details, the com- for identifying conceptual or reasoning difficulties that might
pleteness of the description is of considerable importance. be missed in large-scale testing. On the other hand, if the
Enough information should be given so that, under similar population involved is too small, the results may be idiosyn-
conditions, the experiment is reproducible. For this to be cratic and important information may be missed.
possible, the report of the research should include a thorough The empirical studies in this section have been divided
description of the instrument used to assess understanding, into overlapping categories that vary considerably in scope
the degree of interaction between the student and the inves- and type. Most of this research has focused on conceptual
tigator, the depth of the probing, the form of the data ob- understanding or problem-solving performance. The effec-
tained, and the method of analysis of the data. The authors tiveness of laboratory instruction and lecture demonstrations
should indicate awareness of possible weaknesses in the pro- has also been investigated, but to a much more limited ex-
cedures and indicate that they have taken appropriate precau- tent. There also has been some research on other aspects of
tions. student learning, such as the ability to apply mathematics in
The goals and the perspective of the investigators should physics. In addition, several studies have examined student
be explicitly stated. These may influence both the design of attitudes and beliefs.
the experiment and the interpretation of the results by the
authors. The limits of applicability of the results should be
A. Conceptual understanding
made clear. The reader should be able to determine the de-
gree to which the findings have general relevance and are not This subsection is organized into content areas in the way
idiosyncratic. that the traditional introductory course is taught. In each con-
In the selection of references, preference has been given to tent area, the papers have been classified into three overlap-
papers in which the approach and the rules of evidence are ping categories: ~a! identification and analysis of student dif-
close to those traditional in the physics community. How- ficulties, ~b! development and assessment of instructional
ever, experiments in physics education differ in a number of strategies, and ~c! development and validation of broad as-
respects from the idealization of a traditional physics experi- sessment instruments.
ment. Among the differences are: ~1! a limited ability to
identify and control all the variables, ~2! the necessity of 1. Mechanics
using a strongly interacting probe, and ~3! the degree of
quantification that is appropriate. a. Identification and analysis of student difficulties. The
Classrooms, students, and teachers are all complex sys- references below are divided into overlapping subcategories
tems. Experiments with such systems involve many vari- according to their main emphasis: ~1! kinematics, ~2! dynam-
ables, some of which are unknown. It is difficult to deter- ics, and ~3! relativity and frames of reference.
mine the effect of past experience and cultural environment (1) Kinematics. In the following papers, the authors iden-
on students and teachers. The formal education of students tify and analyze specific difficulties that students have with
prior to their enrollment in undergraduate courses may sig- the kinematical concepts and their graphical representations,
nificantly affect how they interpret what is taught. As in and with the relationship of concepts and graphs to the real
traditional physics research, it is sometimes impossible to world.

757 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 757
26. ‘‘Investigation of student understanding of the concept of velocity in one tories of objects,’’ A. Caramazza, M. McCloskey, and B. Green, Cog-
dimension,’’ D. E. Trowbridge and L. C. McDermott, Am. J. Phys. 48, nition 9, 117–123 ~1981!. About 50 undergraduates were asked to trace
1020–1028 ~1980!. the path of a pendulum bob if the string were cut at different positions
27. ‘‘Investigation of student understanding of the concept of acceleration in along its path. Only about one-fourth responded correctly.
one dimension,’’ D. E. Trowbridge and L. C. McDermott, Am. J. Phys. 38. ‘‘Understanding of gravity,’’ R. F. Gunstone and R. White, Sci. Educ.
49, 242–253 ~1981!. 65, 291–299 ~1981!. Simple lecture demonstrations were shown to sev-
The two papers above report on an investigation of student understand- eral hundred first-year university students in Australia. The students
ing of the concepts of position, velocity, and acceleration. Individual exhibited a strong tendency to observe their prediction regardless of
demonstration interviews, conducted with 200 university students, indi- what actually happened.
cated that even after instruction many students confused position with 39. ‘‘Students’ preconceptions in introductory mechanics,’’ J. Clement, Am.
velocity and velocity with acceleration. J. Phys. 50, 66–71 ~1982!. The results of this study indicate that many
28. ‘‘Even honors students have conceptual difficulties with physics,’’ P. C. students believe that motion implies a force, both before and after the
Peters, Am. J. Phys. 50, 501–508 ~1981!. A variety of conceptual diffi- study of introductory mechanics. A detailed comparison is made be-
culties were identified among students in an introductory honors physics tween student quotes and the writings of Galileo.
course. Although mostly about kinematics, the discussion includes dy- 40. ‘‘Rule-governed approaches to physics: Newton’s third law,’’ D. P.
namics, electricity, and magnetism. Maloney, Phys. Educ. 19, 37–42 ~1984!. More than 100 university stu-
29. ‘‘Student preconceptions about vector kinematics,’’ J. M. Aguirre, Phys. dents with different backgrounds in physics were asked to compare the
Teach. 26, 212–216 ~1988!. This paper discusses student difficulties forces that two interacting objects exerted on each other. About two-
with vector kinematics. More detail is given in a related paper: ‘‘Stu- thirds thought that they would be of different magnitude in some cir-
dents’ conceptions about the vector characteristics of three physics con- cumstances.
cepts,’’ J. Aguirre and G. L. Erickson, J. Res. Sci. Teach. 21, 439–457 41. ‘‘Common-sense concepts about motion,’’ I. A. Halloun and D.
~1984!. Hestenes, Am. J. Phys. 53, 1056–1065 ~1985!. The authors found that
30. ‘‘Student difficulties in connecting graphs and physics: Examples from students have many common-sense views about motion both before and
kinematics,’’ L. C. McDermott, M. L. Rosenquist, and E. H. van Zee, after formal instruction. This paper is part of a sequence that led to the
Am. J. Phys. 55, 503–513 ~1987!. A long-term study involving several development of the FCI. ~See Sec. IV A 1 c.!
hundred students helped identify student difficulties in relating kinemati- 42. ‘‘Student understanding in mechanics: A large population survey,’’ R. F.
cal concepts, their graphical representations, and the motions of real Gunstone, Am. J. Phys. 55, 691–696 ~1987!. On a multiple-choice test
objects. Instructional strategies designed to address some of these diffi- given to 5500 high school students, a majority predicted that two equal
culties are described in Ref. 58. masses on an Atwood’s machine would ‘‘seek’’ the same level.
31. ‘‘Student difficulties with graphical representations of negative values of 43. ‘‘Student understanding of the work-energy and impulse-momentum
velocity,’’ F. M. Goldberg and J. H. Anderson, Phys. Teach. 27, 254– theorems,’’ R. A. Lawson and L. C. McDermott, Am. J. Phys. 55, 811–
260 ~1989!. Interviews and written tests conducted at four universities 817 ~1987!. In an investigation conducted after instruction on the work-
probed student understanding of negative velocity. energy and impulse-momentum theorems, most students were unable to
32. ‘‘Displacement, velocity and frames of reference: Phenomenographic relate the algebraic formalism to motions that they observed. ~Further
studies of students’ understanding and some implications for teaching research on this topic is reported in Ref. 70.!
and assessment,’’ J. Bowden, G. Dall’Alba, E. Martin, D. Laurillard, F. 44. ‘‘Students’ concepts of force as applied to related physical systems: A
Marton, G. Masters, P. Ramsden, A. Stephanou, and E. Walsh, Am. J. search for consistency,’’ M. Finegold and P. Gorsky, Int. J. Sci. Educ.
Phys. 60, 262–269 ~1992!. This study involved high school students 13, 97–113 ~1991!. A study involving more than 500 university and
from several countries. It was found that as problems became easier to high school students in Israel examined the extent to which students
solve quantitatively, the level of conceptual understanding became more consistently applied alternative concepts of force in different contexts.
difficult to determine. This paper includes a discussion of a general 45. ‘‘Effect of written text on usage of Newton’s third law,’’ R. K. Boyle
technique used in education research to reliably extract an understanding and D. P. Maloney, J. Res. Sci. Teach. 28, 123–140 ~1991!. The inves-
of what students are thinking from interview transcripts. tigators examined the beliefs about Newton’s third law of 100 university
33. ‘‘Cognition for interpreting scientific concepts: A study of accelera- students before instruction. Half of the students were given a handout
tion,’’ F. Reif and S. Allen, Cogn. Instruction 9 ~1!, 1–44 ~1992!. Dia- describing forces with explicit statements of the third law. No student
grams of trajectories of two-dimensional motions were shown to five without the handout applied the third law correctly and of those with the
students in introductory physics and five physics faculty. Analysis of handout, fewer than half applied it correctly.
how the two groups interpreted the diagrams enabled the investigators to 46. ‘‘Motion implies force: Where to expect vestiges of the misconcep-
identify the underlying knowledge and skills required. tion?’’ I. Galili and V. Bar, Int. J. Sci. Educ. 14, 63–81 ~1992!. This
(2) Dynamics. The references below focus on the identifi- study examined the persistence of misconceptions in a range of popula-
tions from 10th-grade students to pre-service technology teachers.
cation of student difficulties with dynamics, including New- 47. ‘‘Research as a guide for teaching introductory mechanics: An illustra-
ton’s Laws, circular motion, and the concepts of energy and tion in the context of the Atwood’s machine,’’ L. C. McDermott, P. S.
momentum. Shaffer, and M. D. Somers, Am. J. Phys. 62, 46–55 ~1994!. A study of
34. ‘‘Spontaneous reasoning in elementary dynamics,’’ L. Viennot, Eur. J. student understanding of the Atwood’s machine revealed serious diffi-
Sci. Educ. 1, 205–221 ~1979!. This paper presents the results of an culties with the acceleration of the two masses, the internal and external
investigation conducted among European students drawn from the last forces, and the role of the string. The development of a tutorial to ad-
year of secondary school through the third year of university. The stu- dress these difficulties is also described. ~The tutorial can be found in
dents demonstrated a strong tendency to assume a direct linear relation- Ref. 210.!
ship between force and velocity. 48. ‘‘A cross-college age study about physics students’ conceptions of force
35. ‘‘Factors influencing the learning of classical mechanics,’’ A. Cham- in pre-service training for high school teachers,’’ R. Trumper, Phys.
pagne, L. Klopfer, and J. Anderson, Am. J. Phys. 48, 1074–1079 Educ. 31, 227–236 ~1996!. A study conducted in Israel noted difficulties
~1980!. More than 100 students in an introductory university course with the concept of force among pre-service high school physics teach-
were given a short-answer test on force and motion prior to instruction. ers.
Many non-Newtonian ideas were observed, including: a constant force 49. ‘‘A hierarchical model of the development of student understanding of
produces constant velocity and in the absence of forces, objects are momentum,’’ T. Graham and J. Berry, Int. J. Sci. Educ. 18, 75–89
either at rest or slowing down. ~1996!. Observations of more than 500 British 17–18 year old physics
36. ‘‘Curvilinear motion in the absence of external forces: Naive beliefs students provided a basis for classification of the development of the
about the motion of objects,’’ M. McCloskey, A. Caramazza, and B. concept of momentum into stages.
Green, Science 210, 1139–1141 ~1980!. University students, many of 50. ‘‘The effect of context on students’ reasoning about forces,’’ D. Palmer,
whom had studied physics, were asked to predict the motions of objects Int. J. Sci. Educ. 19, 681–696 ~1997!. This study compared how a group
moving in constrained curved paths. Many believed that an object would of high school physics students and a group of pre-service teachers
‘‘remember’’ the curve after it left the constraint. responded to a variety of simple physics questions in which the physics
37. ‘‘Naive beliefs in ‘sophisticated’ subjects: Misconceptions about trajec- was the same but the contexts were different.

758 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 758
51. ‘‘Conceptual dynamics: Following changing student views of force and mechanics how the ~often incorrect! ideas that students bring to a phys-
motion,’’ R. K. Thornton, AIP Conf. Proc. 399, 241–266 ~1997!. ~See ics class can be used as ‘‘anchoring conceptions’’ around which suc-
Ref. 9.! A framework is constructed for identifying the state of student cessful instructional strategies can be built.
understanding of the laws of mechanics and explores the dynamics of 64. ‘‘Overview, Case Study Physics,’’ A. Van Heuvelen, Am. J. Phys. 59,
how student views develop through instruction. 898–907 ~1991!. Results from research guided the design of the Over-
view, Case Study ~OCS! method. This method helps students build a
(3) Relativity and frames of reference hierarchical knowledge structure of mechanics based on a spiral of in-
52. ‘‘ ‘Spontaneous’ ways of reasoning in elementary kinematics,’’ E. creasing sophistication. OCS students performed significantly better on
Saltiel and J. L. Malgrange, Eur. J. Phys. 1, 73–80 ~1980!. A study of the tests described in Refs. 73 and 80 than did a control group that had
700 university students and 80 eleven-year olds identified student diffi- received traditional instruction.
culties with relative motion and reference frames. 65. ‘‘Socratic pedagogy in the introductory physics laboratory,’’ R. R. Hake,
53. ‘‘Alternative conceptions in Galilean relativity: frames of reference,’’ S. Phys. Teach. 33, 1–7 ~1992!. In this laboratory-based approach to teach-
Panse, J. Ramadas, and A. Kumar, Int. J. Sci. Educ. 16, 63–82 ~1994!. ing dynamics, students perform simple experiments that serve as a basis
54. ‘‘Alternative conceptions in Galilean relativity: Distance, time, energy for Socratic dialogues.
and laws,’’ J. Ramadas, S. Barve, and A. Kumar, Int. J. Sci. Educ. 18, 66. ‘‘Using bridging analogies and anchoring intuitions to deal with stu-
463–477 ~1996!. dents’ preconceptions in physics,’’ J. Clement, J. Res. Sci. Teach. 30,
55. ‘‘Alternative conceptions in Galilean relativity: Inertial and non-inertial 1241–1257 ~1993!. The author describes how a succession of analogies
observers,’’ J. Ramadas, S. Barve, and A. Kumar, Int. J. Sci. Educ. 18, can be used to form a bridge for transforming students’ common-sense
615–629 ~1996!. ideas to the Newtonian view.
The three papers above describe a series of studies in which undergradu- 67. ‘‘The impact of video motion analysis on kinematics graph interpreta-
ate students in India were asked questions about transformations be- tion skills,’’ R. J. Beichner, Am. J. Phys. 64, 1272–1277 ~1996!. The
tween different frames. Both kinematical and dynamical issues were author investigated the use of video software in helping students develop
considered and student responses classified. graph-reading skills. Various combinations were tried, ranging from no
56. ‘‘A case study of conceptual change in special relativity: The influence use of video, to video demonstrations, to student-captured videos in
of prior knowledge in learning,’’ Peter W. Hewson, Eur. J. Sci. Educ. 4, laboratory experiments. Greater use and integration with other compo-
61–76 ~1982!. A series of interviews with a graduate tutor in introduc- nents of instruction correlated strongly with improved scores on the
tory physics probed his understanding of special relativity. Implications TUG-K described in Ref. 72.
of this case study are discussed in detail in Ref. 178. 68. ‘‘On the effectiveness of active-engagement microcomputer-based labo-
b. Development and assessment of instructional strategies. ratories,’’ E. F. Redish, J. M. Saul, and R. N. Steinberg, Am. J. Phys. 65,
The primary focus in almost all of the studies cited above 45–54 ~1997!. Gains on multiple-choice and on open-ended questions
was on the nature or prevalence of student difficulties. In were compared for students with tutorials incorporating microcomputer-
based laboratory ~MBL! tools and for students without these experi-
some instances, however, the design of effective instruction
ences. The students with MBL tutorials performed better on both types
was an integral part of the investigation. of questions. A description of the tutorial approach can be found in Ref.
57. ‘‘Diagnosis and remediation of an alternative conception of velocity 47. ~See also Ref. 210.!
using a microcomputer program,’’ P. W. Hewson, Am. J. Phys. 53, 69. ‘‘Using interactive lecture demonstrations to create an active learning
684–690 ~1985!. This paper examines student learning using a computer environment,’’ D. R. Sokoloff and R. K. Thornton, Phys. Teach. 35,
program designed to diagnose and remediate difficulties with kinemati- 340–347 ~1997!. This paper describes a general strategy for increasing
cal concepts. For a more detailed analysis, see ‘‘Effect of instruction student engagement in lectures through the use of microcomputer-based
using microcomputer simulations and conceptual change strategies on lecture demonstrations. Applications in the teaching of kinematics and
science learning,’’ A. I. Zietsman and P. W. Hewson, J. Res. Sci. Teach. dynamics are presented and evaluated.
23 ~1!, 27–39 ~1986!. 70. ‘‘The challenge of matching learning assessments to teaching goals: An
58. ‘‘A conceptual approach to teaching kinematics,’’ M. L. Rosenquist and example from the work-energy and impulse-momentum theorems,’’ T.
L. C. McDermott, Am. J. Phys. 55, 407–415 ~1987!. Results from re- O’Brien Pride, S. Vokos, and L. C. McDermott, Am. J. Phys. 66, 147–
search were used to guide the design of a laboratory-based curriculum 156 ~1998!. Evidence is presented that difficulties with the two theorems
that has been shown to be effective in addressing some of the difficulties extend beyond the introductory level. ~See Ref. 43.! The article de-
in kinematics that were identified in Ref. 30. scribes a research-based tutorial that was developed to address these
59. ‘‘Facilitation of scientific concept learning by interpretation procedures difficulties. ~See Ref. 210.! Issues related to the assessment of student
and diagnosis,’’ P. Labudde, F. Reif, and L. Quinn, Int. J. Sci. Educ. 10, understanding are discussed.
81–98 ~1988!. The authors present a general instructional strategy for 71. ‘‘Do they stay fixed?’’ G. E. Francis, J. P. Adams, and E. J. Noonan,
helping students develop coherent procedures for interpreting scientific Phys. Teach. 36, 488–490 ~1998!. This study probed the extent to which
concepts and for correcting deficiencies in their pre-existing knowledge. student gains on the FCI resulting from interactive-engagement instruc-
60. ‘‘Learning motion concepts using real-time microcomputer-based labo- tion persisted beyond the conclusion of the course. ~The tutorials from
ratory tools,’’ R. K. Thornton and D. R. Sokoloff, Am. J. Phys. 58, Ref. 210 were used.! The study found little decline in FCI scores over
858–867 ~1990!. The authors describe the use of microcomputer-based several years following instruction.
laboratory ~MBL! activities to help students overcome some common
conceptual difficulties in kinematics. Reference 47 also discusses the development of an in-
61. ‘‘Explaining the ‘at rest’ condition of an object,’’ J. Minstrell, Phys. structional strategy to address difficulties with the concept of
Teach. 20, 10–14 ~1982!. The author describes a carefully structured tension in a string.
questioning sequence designed to address the failure of many students to c. Development and validation of broad assessment instru-
recognize that a stationary surface can exert a force on an object with ments. A few comprehensive instruments to assess student
which it is in contact. This study represents a form of ‘‘action research,’’
understanding in mechanics have been published. The papers
through which teachers gain insight into how their students are thinking.
62. ‘‘Modeling instruction in mechanics,’’ I. A. Halloun and D. Hestenes, cited in this subsection relate to four multiple-choice tests
Am. J. Phys. 55, 455–462 ~1987!. An introductory university physics that are easy to administer and grade. Their use with a vari-
course was developed to test an instructional theory that emphasizes ety of student populations has provided compelling evidence
mathematical modeling and study of paradigmatic problems. Nearly 500 that many students who do well on quantitative examination
students were divided into test and control groups. The students in the questions have serious conceptual difficulties. The tests have
test group did substantially better, especially those who performed been used as an indicator of the initial state of different
poorly on the pre-test.
63. ‘‘Not all preconceptions are misconceptions: Finding ‘anchoring con-
populations and in some instances as a standard by which to
ceptions’ for grounding instruction on students’ intuition,’’ J. Clement, judge the effectiveness of instruction.
D. Brown, and A. Zeitsman, Int. J. Sci. Educ. 11 ~spec. issue!, 554–565 In comparing instructors or instructional strategies, any
~1989!. This paper illustrates in the context of a high school class in single instrument must be used with great care since many

759 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 759
variables are involved in any teaching situation. The test may The Mechanics Baseline Test ~MBT! is another multiple-
be incomplete and the questions may be subject to misinter- choice test. It contains a greater range of topics than does the
pretation by the student. As a measure of instructional effec- FCI and is intended for use after instruction.
tiveness, the results from multiple-choice tests alone should 80. ‘‘A Mechanics Baseline Test,’’ D. Hestenes and M. Wells, Phys. Teach.
be viewed with skepticism. @See, for example, the letter ‘‘On 30, 159–166 ~1992!. The test is included in the paper.
not choosing multiple choice,’’ T. R. Sandin, Am. J. Phys. In the study described in Ref. 70, two questions that ap-
53, 299–300 ~1985!.# It is often impossible to tell when in- pear on the MBT were given to students with explanations
correct reasoning leads to a correct answer. Good perfor- required. The results differed considerably when the expla-
mance on broad assessment instruments that do not require nations were and were not taken into account.
explanations should be considered as a necessary, rather than A fourth multiple-choice test is the Force and Motion
sufficient, criterion for meaningful learning. See the com- Conceptual Evaluation ~FMCE!. Multiple questions on each
parison of multiple-choice and open-ended questions in Refs. concept allow this test to be used as a diagnostic for particu-
70 and 79. lar difficulties of individual students.
The Test of Understanding Graphs in Kinematics 81. ‘‘Assessing student learning of Newton’s laws: The Force and Motion
~TUG-K! is a multiple-choice test on the interpretation of Conceptual Evaluation and the evaluation of active learning laboratory
graphical representations of motions. and lecture curricula,’’ R. K. Thornton and D. R. Sokoloff, Am. J. Phys.
66, 338–352 ~1998!. The results from traditional introductory courses
72. ‘‘Testing student interpretation of kinematics graphs,’’ R. J. Beichner, on a subset of the questions on the FMCE are compared with those from
Am. J. Phys. 62, 750–762 ~1994!. The appendix includes the TUG-K. courses in which instruction included activities supported by
Administration of the test to about 900 students in high school and microcomputer-based laboratory ~MBL! tools. The performance of the
college yielded results consistent with those from other types of studies MBL students was significantly better. The FMCE is included in the
on the interpretation of motion graphs. The paper also includes a de- Appendix.
tailed discussion of the development and validation of multiple-choice
tests.
2. Electricity and magnetism
The most widely used and thoroughly tested assessment
instrument is the Force Concept Inventory ~FCI!. Each test Student understanding of concepts in electricity and mag-
item requires that students distinguish between correct New- netism has not been investigated in as great detail as in me-
tonian answers and erroneous ‘‘common-sense’’ beliefs. chanics. Published articles on student difficulties have dealt
Widespread administration of the FCI has raised the aware- primarily with two topics: DC circuits and electric fields.
ness of faculty to the failure of most lectures to promote a. Identification and analysis of student difficulties
conceptual development. @For an anecdote describing the im- (1) dc circuits. Student difficulties with dc circuits have
pact on a university instructor of results from the FCI, see E. been documented in many studies.
Mazur, Peer Instruction: A User’s Manual ~Prentice-Hall, 82. ‘‘Student conceptions of simple electric circuits,’’ N. Fredette and J.
Lochhead, Phys. Teach. 19, 194–198 ~1980!; ‘‘Student misconceptions
Englewood Cliffs, NJ, 1997!, p. 4.#
of an electric circuit: What do they mean?’’ N. Fredette and J. Clement,
73. ‘‘The initial knowledge state of college physics students,’’ I. A. Halloun J. Coll. Sci. Teach. 10, 280–285 ~1981!. These two papers discuss the
and D. Hestenes, Am. J. Phys. 53, 1043–1056 ~1985!. The authors responses of college students to the task: ‘‘Combine a battery, bulb, and
present a multiple-choice instrument, the Mechanics Diagnostic Test, one wire to make the bulb light.’’
that has evolved into the FCI ~next ref.!. Use of the test in an introduc- 83. ‘‘Potential difference and current in simple electric circuits: A study of
tory college physics course is described. The paper also discusses the students’ concepts,’’ R. Cohen, B. Eylon, and U. Ganiel, Am. J. Phys.
construction of effective multiple-choice tests. 51, 407–412 ~1983!. The authors analyzed responses from multiple-
74. ‘‘Force Concept Inventory,’’ D. Hestenes, M. Wells, and G. Swack- choice tests given to 145 high school students and 21 in-service physics
hamer, Phys. Teach. 30, 141–158 ~1992!. This paper contains the Force teachers in Israel. Although the teachers did better than the students,
Concept Inventory ~FCI! and a detailed discussion of the Newtonian many had similar conceptual difficulties.
concepts it is constructed to probe. Results from administration of the 84. ‘‘Conceptions of French pupils concerning electric circuits: Structure
FCI before and after instruction are given for some high school and and evolution,’’ J.-J. Dupin and S. Johsua, J. Res. Sci. Teach. 24, 791–
university classes. 806 ~1987!. A study in France examined the views on electric current
75. ‘‘What does the force concept inventory actually measure?’’ D. Huff- held by students ranging in age from 12 to 22 years. It was found that
man and P. Heller, Phys. Teach. 33, 138–143 ~1995!. some simple misconceptions disappear with instruction, but teaching
76. ‘‘Interpreting the Force Concept Inventory: A response to March 1995 seems to have little effect on others.
85. ‘‘A study of students’ understanding of electricity in five European
Critique by Huffman and Heller,’’ D. Hestenes and I. Halloun, Phys.
countries,’’ D. M. Shipstone, C. v. Rhöneck, W. Jung, C. Kärrqvist, J.
Teach. 33, 502–506 ~1995!.
Dupin, S. Johsua, and P. Licht, Int. J. Sci. Educ. 10, 303–316 ~1988!. A
77. ‘‘Interpreting the Force Concept Inventory: A reply to Hestenes and
study that was conducted among high school students in five countries
Halloun,’’ P. Heller and D. Huffman, Phys. Teach. 33, 503–511 ~1995!.
revealed substantially the same difficulties everywhere.
The three papers above carry on a dialogue on the subject of correlations
86. ‘‘Macro-micro relationships: The missing link between electrostatics
among student errors on the FCI. and electrodynamics in student reasoning,’’ B.-S. Eylon and U. Ganiel,
78. ‘‘Interactive-engagement versus traditional methods: A six-thousand- Int. J. Sci. Educ. 12, 79–94 ~1990!. In a study conducted in Israel, high
student survey of mechanics test data for introductory physics courses,’’ school students who lacked a coherent microscopic model could not
R. R. Hake, Am. J. Phys. 66, 64–74 ~1998!. This paper presents a predict the behavior of transients in simple circuits.
collection of pre- and post-instruction FCI data from instructors at a 87. ‘‘Variable uses of alternative conceptions: A case-study in current elec-
large number of high schools, colleges, and universities. Most ‘‘active- tricity,’’ P. Heller and F. N. Finley, J. Res. Sci. Teach. 29, 259 ~1992!.
engagement’’ classes ~as defined by the instructors! showed much Fourteen in-service elementary and middle school teachers were found
greater improvement than traditional classes. to have a coherent, but incorrect, model of current.
79. ‘‘Performance on multiple-choice diagnostics and complementary exam 88. ‘‘Research as a guide for curriculum development: An example from
problems,’’ R. N. Steinberg and M. S. Sabella, Phys. Teach. 35, 150– introductory electricity. Part I. Investigation of student understanding,’’
155 ~1997!. This paper compares the responses of introductory univer- L. C. McDermott and P. S. Shaffer, Am. J. Phys. 60, 994–1003 ~1992!;
sity physics students on the FCI and on open-ended examination ques- erratum, 61, 81 ~1993!. This paper identifies specific difficulties that
tions that probe the same concepts. Students did not always perform many undergraduate students have with dc circuits. Instructional strate-
similarly on the two types of questions. gies designed to address these difficulties are described in Ref. 100.

760 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 760
89. ‘‘Images of electricity: How do novices and experts model electric cur- S. Shaffer and L. C. McDermott, Am. J. Phys. 60, 1003–1013 ~1992!.
rent?’’ S. M. Stocklmayer and D. F. Treagust, Int. J. Sci. Educ. 18, This paper describes the application of the results from the research
163–178 ~1996!. This study found that experts have images of electric described in Ref. 88 to the development of both a laboratory-based
current that differ significantly both from those of novice students and curriculum for an inquiry-oriented course and a supplementary tutorial
from the models that are usually taught. Experts draw on a field concept curriculum for a lecture-based course. See Refs. 210 and 218.
more than on a particle model. 101. ‘‘Superposition of electric fields and causality: From research to teach-
90. ‘‘Seeking the causal connection in electricity: Shifting among mecha- ing,’’ S. Rainson and L. Viennot, AIP Conf. Proc. 399, 679–687
nistic perspectives,’’ J. Gutwill, J. Frederiksen, and M. Ranney, Int. J. ~1997!. ~See Ref. 9.! Instructional strategies are described for address-
Sci. Educ. 18, 143–162 ~1996!. The authors examined the reasoning ing the difficulties with superposition of fields described in Ref. 96.
used and the evolution of perspectives among 22 high school students as
they solved problems on electric circuits.
91. ‘‘The persistence of students’ unfounded beliefs about electrical circuits: 3. Light and optics
The case of Ohm’s law,’’ A. Métioui, C. Brassard, J. Levasseur, and M. a. Identification and analysis of student difficulties
Lavoie, Int. J. Sci. Educ. 18, 193–212 ~1996!. Interviews and written
tests were used to probe the understanding of Ohm’s law among Elec- (1) Nature of light, color, and vision
trical Engineering Technology students in Quebec. 102. ‘‘Commonsense knowledge in optics: Preliminary results of an inves-
tigation into the properties of light,’’ C. La Rosa, M. Mayer, P. Patrizi,
(2) Electrostatics and magnetostatics and M. Vicentini-Missoni, Eur. J. Sci. Educ. 6, 387–397 ~1984!. Ideas
92. ‘‘Charged poles,’’ D. P. Maloney, Phys. Educ. 20, 310–316 ~1985!. about light, color, and geometrical optics were explored through inter-
Results from a study conducted in an algebra-based physics class views with teachers and open-ended written questions administered to
strongly suggest that, even after instruction, many students are confused high school students. On the basis of their observations, the authors
about the interactions between electric charges and magnetic poles. propose a progression of stages in student thinking about light.
93. ‘‘Students’ understanding of the transfer of charge between conduc- 103. ‘‘Student conceptions of light: A case study,’’ D. M. Watts, Phys.
tors,’’ C. Guraswamy, M. D. Somers, and R. G. Hussey, Phys. Educ. 32, Educ. 20, 183–187 ~1985!. A detailed description is given of the views
91–96 ~1997!. Individual demonstration interviews were used to inves- of a high school student on the nature of light. Many of the common
tigate student understanding of charge and the behavior of charged con- misconceptions are represented in the discussions quoted.
ductors. After instruction, few students were able to identify the forces 104. ‘‘The understanding of the properties of light by students in India,’’ A.
of a charge on a conductor or to describe how charges were shared B. Saxena, Int. J. Sci. Educ. 13, 283–289 ~1991!. This article reports
between touching conductors. the results from a multiple-choice test that was administered to both
secondary school and undergraduate students in India. The results were
(3) Electric and magnetic fields. Since many of the basic similar to those obtained in Refs. 107 and 108.
concepts in electricity and magnetism are not familiar from 105. ‘‘Prospective elementary school teachers’ prior knowledge about
direct experience and are quite abstract, students can be ex- light,’’ S. Bendall, I. Galili, and F. Goldberg, J. Res. Sci. Teach. 30,
pected to have conceptual difficulties. The few published 1169–1187 ~1993!. Preservice elementary school teachers were inter-
studies are quite provocative, but far from complete. viewed about the nature of light.
106. ‘‘Light propagation and visual patterns: Preinstruction learners’ con-
94. ‘‘On the quality of knowledge in the field of electricity and magnetism,’’ ceptions,’’ D. Langley, M. Ronen, and B.-S. Eylon, J. Res. Sci. Teach.
M. G. M. Ferguson-Hessler and T. de Jong, Am. J. Phys. 55, 492–497 34, 399–424 ~1997!. This study explored the ideas about light propa-
~1987!. The authors investigated how first-year university students or- gation and image formation of Israeli 10th graders.
ganized their knowledge of electromagnetism. Successful problem solv-
ers had a more coherent knowledge structure. (2) Geometrical optics
95. ‘‘Novice use of qualitative versus quantitative problem solving in elec- 107. ‘‘Student difficulties in understanding image formation by a plane mir-
trostatics,’’ C. McMillan III and M. Swadener, J. Res. Sci. Teach. 28, ror,’’ F. M. Goldberg and L. C. McDermott, Phys. Teach. 24, 472–480
661–670 ~1991!. Six students in a calculus-based physics class were ~1986!. During interviews, university students were shown an object in
observed as they solved electrostatics problems. The successful students front of a mirror and asked what an observer at various locations would
differed from the others only in mathematical facility, not in qualitative see. Many students could not make correct predictions either before or
understanding. Both groups had difficulty with qualitative questions and after instruction.
had similar misconceptions. 108. ‘‘An investigation of student understanding of the real image formed
96. ‘‘Students’ reasoning about the superposition of electric fields,’’ L. Vi- by a converging lens or concave mirror,’’ F. M. Goldberg and L. C.
ennot and S. Rainson, Int. J. Sci. Educ. 14, 475–487 ~1992!. This paper McDermott, Am. J. Phys. 55, 108–119 ~1987!. Even after instruction,
discusses the difficulties of French and Algerian university students with many students could not apply the formalism of geometrical optics to
Gauss’s Law and with the electric field in an insulator. For a further predict or account for the image formed by a converging lens or con-
analysis that includes Swedish students, see ‘‘Students’ understanding cave mirror.
of superposition of electric fields,’’ S. Rainson, G. Tranströmer, and L. 109. ‘‘The effects of prior knowledge and instruction on understanding im-
Viennot, Am. J. Phys. 62, 1026–1032 ~1994!. Instruction that addresses age formation,’’ I. Galili, S. Bendall, and F. Goldberg, J. Res. Sci.
these issues is described in Ref. 101. Teach. 30, 271–301 ~1993!. Individual demonstration interviews con-
97. ‘‘Confusion by representation: On students’ comprehension of the elec- ducted with students in a college physics course for prospective teach-
tric field concept,’’ S. Törnkvist, K.-A. Pettersson, and G. Tranströmer, ers suggested that, after instruction, students’ prior conceptions of light
Am. J. Phys. 61, 335–338 ~1993!. Analysis of more than 500 written become ‘‘hybridized’’ with the physicist’s model.
responses and nearly 100 interviews revealed difficulties with the con- 110. ‘‘Students’ conceptual change in geometrical optics,’’ I. Galili, Int. J.
cept of electric field lines among second-year students at the Royal Sci. Educ. 18, 847–868 ~1996!. The author discusses how students’
Institute of Technology in Stockholm. conceptual models in geometrical optics change with instruction.
98. ‘‘Mechanics background influences students’ conceptions in electromag-
netism,’’ I. Galili, Int. J. Sci. Educ. 17, 371–387 ~1995!. Difficulties (3) Physical optics
with electromagnetism were identified in a study that included 10th 111. ‘‘An investigation of student understanding of single-slit diffraction
graders and pre-service technology teachers in Israel. and double-slit interference,’’ B. S. Ambrose, P. S. Shaffer, R. N.
99. ‘‘The kinds of mental representations—models, propositions, and Steinberg, and L. C. McDermott, Am. J. Phys. 67, 146–155 ~1999!.
images—used by college physics students regarding the concept of This article identifies specific difficulties that many students have in
field,’’ I. M. Grea and M. A. Moreira, Int. J. Sci. Educ. 19, 711–724 selecting and applying an appropriate model to account for the pattern
~1997!. Brazilian sophomore engineering students participated in the produced on a screen when light is incident on one or two narrow slits.
study. The discussion is within a theoretical educational framework. It also was found that students at introductory and more advanced
levels have seriously mistaken beliefs about photons and the wave
b. Development and assessment of instructional strategies model for matter.
100. ‘‘Research as a guide for curriculum development: An example from
introductory electricity. Part II. Design of an instructional strategy,’’ P. b. Development and assessment of instructional strategies

761 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 761
112. ‘‘Lenses, pinholes, screens and the eye,’’ F. Goldberg, S. Bendall, and situations in which more than two variables change. Some specific
I. Galili, Phys. Teach. 29, 221–224 ~1991!. The authors describe an student difficulties with the ideal gas law can be traced to this compli-
instructional strategy to increase student understanding of real images. cation.
Two demonstrations are used: a real image formed on a screen by a 123. ‘‘Critical review on the research aimed at elucidating the sense that
converging lens and a ‘‘screen reproduction’’ produced by a pinhole. notions of temperature and heat have for the students aged 10 to 16
113. ‘‘Many rays are better than two,’’ D. J. Grayson, Phys. Teach. 33, years,’’ A. Tiberghien, in Research on Physics Education, Proceedings
42–43 ~1995!. Having students draw many rays from each point on an of the First International Workshop, La Londe Les Maures, France,
object appears to help them understand why covering half a lens edited by G. Delacôte, A. Tiberghien, and J. Schwartz ~Éditions du
doesn’t block half the image. ~See Ref. 108.! In a class of 35 South CNRS, Paris, 1983!, pp. 75–90. This article summarizes the published
African university students, improvement on the post-test compared to research on children’s understanding of heat and temperature.
the pretest indicated that this strategy was effective. 124. ‘‘Students’ conceptions of the second law of thermodynamics—an in-
114. ‘‘Making the invisible visible: A teaching/learning environment that terpretive study,’’ S. Kesidou and R. Duit, J. Res. Sci. Teach. 30,
builds on a new view of the physics learner,’’ F. Goldberg and S. 85–106 ~1993!. This paper reports the views of German high school
Bendall, Am. J. Phys. 63, 978–991 ~1995!. The study of light provides students, who have had four years of physics instruction, on thermal
a context in which prospective elementary teachers develop conceptual equilibrium, the concepts of heat and temperature, and the first and
understanding and an awareness of their own learning. second laws of thermodynamics.
115. ‘‘Computer simulations as tools for teaching and learning: Using a 125. ‘‘ ‘Work’ and ‘heat’: On a road towards thermodynamics,’’ P. H. van
simulation environment in optics,’’ B.-S. Eylon, M. Ronen, and U. Roon, H. F. van Sprand, and A. H. Verdonk, Int. J. Sci. Educ. 16,
Ganiel, J. Sci. Educ. Technol. 5 ~2!, 93–110 ~1996!. The authors evalu- 131–144 ~1994!. The difficulties first year Dutch university students
ate the effect of a ray-tracing simulation program on students’ sponta- have with the concepts of thermodynamic system, heat, work, and
neous use of appropriate concepts. They found that the effectiveness of temperature are probed.
the program depends heavily on the learning environment in which the 126. ‘‘Children’s and lay adults’ views about thermal equilibrium,’’ M.
program is used. Arnold and R. Millar, Int. J. Sci. Educ. 16, 405–419 ~1994!. Detailed
116. ‘‘Development and assessment of a research-based tutorial on light and interviews were used to probe views on heating and cooling held by
shadow,’’ K. Wosilait, P. R. L. Heron, P. S. Shaffer, and L. C. Mc- British high school students and university-educated adults not trained
Dermott, Am. J. Phys. 66, 906–913 ~1998!. Evidence is presented that in science. Both groups revealed similar misconceptions.
university students at the introductory physics level and beyond often
cannot account for simple phenomena involving light and shadow. The (2) Pressure, density, and the structure of matter
authors describe the research through which specific difficulties were 127. ‘‘Earth science, density, and the college freshman,’’ J. W. McKinnon,
identified. The article describes the iterative process through which a J. Geol. Educ. 19 ~5!, 218–220 ~1971!. This paper describes how stu-
tutorial to address student difficulties in geometrical optics was devel- dent difficulties with ratio reasoning can lead to difficulties with the
oped and assessed. ~See Ref. 210.! concept of density, even among university students. ~See also Refs. 2
and 8, Instructor’s Guide for Physics by Inquiry, pp. 3–8.!
128. ‘‘Grade 12 students’ misconceptions relating to fundamental character-
istics of atoms and molecules,’’ A. K. Griffiths and K. R. Preston, J.
Res. Sci. Teach. 29, 611–628 ~1992!. The authors report on the views
4. Properties of matter, fluid mechanics, and thermal of 30 randomly selected high school students in Newfoundland about
physics the nature and structure of atoms and molecules.
129. ‘‘Student understanding of the volume, mass, and pressure of air within
Investigations conducted among young children indicate a sealed syringe in different states of compression,’’ K. C. deBerg, J.
Res. Sci. Teach. 32, 871–884 ~1995!. The author studied the responses
that serious misconceptions about heat and temperature are
of high school students in England who had studied physics or chem-
common. Since there is little published research involving istry to qualitative tasks involving pressure, volume, and mass of a gas
university students, many of the references below are to stud- in a syringe. Only about one-third of the students demonstrated a quali-
ies with younger students. tative understanding of these concepts.
130. ‘‘Pupils’ conceptions of matter and its transformations ~ages 12–16!,’’
(1) Heat, temperature, and thermodynamics B. Andersson. See Ref. 6, pp. 12–35. This paper reviews some of the
117. ‘‘The teaching of the concept of heat,’’ J. W. Warren, Phys. Educ. 7, research literature on the ideas of high school students about matter,
41–44 ~1972!. This paper discusses the inability of first-year university including chemical reactions ~such as burning!, phase transitions, con-
students to separate the concepts of heat, internal energy, and tempera- servation of matter, and the nature of atoms and molecules.
ture.
118. ‘‘Misconceptions in school thermodynamics,’’ A. H. Johnstone, J. J.
MacDonald, and G. Webb, Phys. Educ. 12, 248–251 ~1977!. A ‘‘ther-
modynamics approach test’’ was administered to 98 middle and high
5. Waves and sound
school students in Scotland. Eight prevalent ‘‘misconceptions’’ were 131. ‘‘A study of tertiary physics students’ conceptualizations of sound,’’
identified. Several of these pertain to chemical reactions. C. J. Linder and G. L. Erickson, Int. J. Sci. Educ. 11 ~spec. issue!,
119. ‘‘Children’s conceptions of heat and temperature,’’ G. L. Erickson, Sci. 491–501 ~1989!. In this study, many students claimed that sound is not
Educ. 63, 221–230 ~1979!. It was observed in this study that many a wave and created other models to account for sound phenomena.
students aged 11–16 believe that heat and cold are substances and that 132. ‘‘Spontaneous reasoning on the propagation of visible mechanical sig-
temperature is a measure of their amount. Few students were able to nals,’’ L. Maurines, Int. J. Sci. Educ. 14, 279–293 ~1992!. In a study of
distinguish between heat and temperature. student understanding of factors affecting the speed of wave propaga-
120. ‘‘The influence of intellectual environment on conceptions of heat,’’ tion, students were found to emphasize the shape and manner of cre-
M. G. Hewson and D. Hamlyn, Eur. J. Sci. Educ. 6, 254–262 ~1984!. ation of the wave rather than the properties of the medium.
Interviews were conducted with Sotho children and adults from an arid 133. ‘‘University physics students’ conceptualizations of factors affecting
region of South Africa. Sotho subjects were less likely than Western the speed of sound propagation,’’ C. J. Linder, Int. J. Sci. Educ. 15 ~6!,
subjects to use a caloric model. The authors concluded that cultural 655–662 ~1993!. The author investigates student understanding of
metaphors influence the interpretation of physical situations. sound propagation.
121. ‘‘A microcomputer-based diagnostic system for identifying students’ 134. ‘‘Using education research to develop waves courseware,’’ D. J. Gray-
conception of heat and temperature,’’ R. Nachmias, R. Stavy, and R. son, Comput. Phys. 10 ~1!, 30–37 ~1996!. Difficulties with two-
Avrams, Int. J. Sci. Educ. 12, 123–132 ~1990!. The authors describe dimensional kinematics were investigated in the context of mechanical
the structure of their microcomputer-based diagnostic system for inves- waves. A computer program enabled students to investigate differences
tigating students’ conceptions of heat and temperature. between spatial and temporal motion graphs.
122. ‘‘Students’ reasonings in thermodynamics,’’ S. Rozier and L. Viennot, 135. ‘‘Making sense of how students make sense of mechanical waves,’’
Int. J. Sci. Educ. 13, 159–170 ~1991!. A study conducted in Paris Michael C. Wittmann, Richard N. Steinberg, and Edward F. Redish,
analyzed responses of university students and in-service teachers to Phys. Teach. 37, 15–21 ~1999!. This paper reports on an investigation

762 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 762
 of student understanding of pulses propagating along elastic strings. Mem. Cogn. 17, 627–638 ~1989!. The authors observed how 45 nov-
Student responses to multiple questions on closely related topics re- ices and 10 experts categorized and solved problems. They found that
vealed the simultaneous presence of both correct and incorrect inter- the better novices made more use of explanatory statements and phys-
pretations. ics principles in setting up the problems.
136. ‘‘Student understanding of light as an electromagnetic wave: Relating 144. ‘‘Effects of knowledge organization on task performance,’’ B. Eylon
the formalism to physical phenomena,’’ B. S. Ambrose, P. R. L. and F. Reif, Cogn. Instruction 1, 5–44 ~1984!. The results of this study
Heron, S. Vokos, and L. C. McDermott, Am. J. Phys. ~to be published!. suggest that a hierarchical presentation of information improves the
This paper describes an investigation of the difficulties that students ability of students to solve certain types of problems.
have with the interpretation of the diagrammatic and mathematical for-
malism commonly used to represent light as a plane EM wave. Results
from this research were used to guide the development of a tutorial that 2. Development and assessment of instructional strategies
has proved effective in addressing some specific difficulties that were 145. ‘‘Teaching general learning and problem solving skills,’’ F. Reif, J. H.
identified. Larkin, and B. C. Bracket, Am. J. Phys. 44, 212–217 ~1976!. The
authors investigated the abilities needed to understand a relation such
6. Topics in modern physics as a definition or a law. An instructional strategy was developed to
teach a general method for acquiring such an understanding.
To date, there has been little published research on student 146. ‘‘Teaching problem solving—A scientific approach,’’ F. Reif, Phys.
understanding of topics in modern physics. See Sec. Teach. 19, 310–316 ~1981!. The author identifies cognitive issues that
IV A 1 a 3 for a discussion of student difficulties with special need to be addressed in order to develop an effective instructional
strategy for teaching problem solving.
relativity. References on other topics are given below.
147. ‘‘Constraining novices to perform expert-like problem analyses: Ef-
137. ‘‘Modern physics and students’ conceptions,’’ H. Fischler and M. fects on schema acquisition,’’ R. Dufresne, W. J. Gerace, P. T. Hardi-
Lichtfeldt, Int. J. Sci. Educ. 14, 181–190 ~1992!. The authors cite man, and J. P. Mestre, J. Learning Sci. 2, 307–331 ~1992!. The authors
results of a descriptive study of student conceptions in quantum me- describe a computer tool designed to help students become more expert
chanics. problem solvers. The program requires students to consider principles,
138. ‘‘School students’ understanding of processes involving radioactive concepts, and procedures.
substance and ionizing radiation,’’ R. Millar and J. S. Gill, Phys. Educ. 148. ‘‘Teaching problem solving through cooperative grouping. Part 1.
31, 27–33 ~1996!. This paper describes a study that probed the under- Group versus individual problem solving,’’ P. Heller, R. Keith, and S.
standing of British high-school students on the subject of radiation. Anderson, Am. J. Phys. 60, 627–636 ~1992!.
Many could not distinguish between damaging a substance by radiation 149. ‘‘Teaching problem solving through cooperative grouping. Part 2. De-
and making it radioactive by radiation. signing problems and structuring groups,’’ P. Heller and M. Holla-
139. ‘‘Development of a computer-based tutorial on the photoelectric ef- baugh, Am. J. Phys. 60, 637–644 ~1992!.
fect,’’ R. N. Steinberg, G. E. Oberem, and L. C. McDermott, Am. J. The two papers above describe a strategy for teaching problem-solving
Phys. 64, 1370–1379 ~1996!. This article reports on an investigation of skills that is based on collaborative learning. The authors identify sev-
student understanding of the photoelectric effect. The study took place eral important factors, such as the nature of the problems used, the
in a sophomore course in modern physics. The results were used to structure of the group, and the training of teaching assistants.
guide the development of an interactive computer program to address 150. ‘‘Comparing problem solving performance of physics students in
the difficulties that were identified. inquiry-based and traditional introductory physics courses,’’ B.
140. ‘‘Student difficulties in learning quantum mechanics,’’ I. D. Johnston, Thacker, E. Kim, K. Trefz, and S. M. Lea, Am. J. Phys. 62, 627–633
K. Crawford, and P. R. Fletcher, Int. J. Sci. Educ. 20, 427–446 ~1998!. ~1994!. This article presents evidence that performance on quantitative
This paper reports on an investigation of the conceptual structure of problems by students who have had experience in solving qualitative
students who had successfully completed a course in quantum mechan- problems can be as good as ~and sometimes better than! performance
ics at an Australian university. The investigators found that student by students who have spent more time on traditional problem solving.
models were often technically advanced but structurally unsophisti- ~See also Ref. 100.!
cated. 151. ‘‘Using qualitative problem-solving strategies to highlight the role of
conceptual knowledge in solving problems,’’ W. J. Leonard, R. J. Du-
Reference 111 includes a discussion of some student diffi-
fresne, and J. P. Mestre, Am. J. Phys. 64, 1495–1503 ~1996!. An
culties with photons. instructional strategy is described for teaching problem solving. Stu-
dents first write a qualitative description, then identify relevant con-
B. Problem-solving performance cepts and principles, and lastly apply these in finding a solution.
152. ‘‘Problem-based learning in physics: Making connections with the real
The ability of students to solve physics problems has been world,’’ B. J. Duch, AIP Conf. Proc. 399, 557–565 ~1997!. ~See Ref.
the subject of a considerable amount of research, especially 9.! This paper discusses an evaluation of the use of context-rich prob-
in the context of mechanics. Studies have been conducted lems in cooperative group learning. ~See also Refs. 148 and 149.!
not only by physicists but also by other investigators who
have used physics as a context in which to study the thought C. Effectiveness of laboratory instruction and lecture
processes involved in problem solving in a broader sense. demonstrations
1. Investigations of problem-solving behavior Laboratory instruction and demonstrations have tradition-
141. ‘‘Understanding and teaching problem solving in physics,’’ J. H. Lar- ally been considered by physicists to be very important for
kin and F. Reif, Eur. J. Sci. Educ. 1, 191–203 ~1979!. From a case teaching physics. Yet, as the list of references below sug-
study comparing the problem-solving approaches of an expert and a
gests, there have been relatively few systematic efforts to
~good! novice problem solver, the authors identify critical elements
needed for expert problem solving. An instructional strategy is de- assess their effectiveness.
scribed for teaching novices to take a more qualitative, global ap- 153. ‘‘Results of a remedial laboratory program based on a Piaget model for
proach. engineering and science freshmen,’’ R. Gerson and R. A. Primrose,
142. ‘‘Categorization and representation of physics problems by experts and Am. J. Phys. 45, 649–651 ~1977!. This paper demonstrates that a labo-
novices,’’ M. T. H. Chi, P. J. Feltovich, and R. Glaser, Cogn. Sci. 5, ratory designed to improve students’ formal reasoning was more effec-
121–152 ~1981!. This study identified differences in the ways that tive in preparing engineering students deficient in algebra for calculus
experts and novices solve physics problems. It was found that experts than was a traditional college algebra class.
categorized problems according to ‘‘deep structure,’’ while novices 154. ‘‘Teaching physicists’ thinking skills in the laboratory,’’ F. Reif and M.
tended to categorize according to surface features. St. John, Am. J. Phys. 47, 950–957 ~1979!. The authors identify spe-
143. ‘‘The relation between problem categorization and problem solving cific skills that can be taught in the laboratory and demonstrate how a
among experts and novices,’’ P. Hardiman, R. Dufresne, and J. Mestre, carefully structured course can teach those skills effectively.

763 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 763
155. ‘‘The influence of physics laboratories on student performance in a 164. ‘‘Two approaches to learning physics,’’ D. Hammer, Phys. Teach. 27,
lecture course,’’ D. D. Long, G. W. McLaughlin, and A. M. Bloom, 664–670 ~1989!. Case studies of two students in an algebra-based uni-
Am. J. Phys. 54, 122–125 ~1986!. The performance of 2500 students in versity physics course revealed that they differed greatly in their un-
the lecture part of an algebra-based university course was correlated derstanding of what it means to ‘‘understand’’ physics.
with whether or not the students took the laboratory component. The 165. ‘‘Cognition in scientific and everyday domains: Comparison and learn-
laboratory seemed to have little effect for students at the top and bot- ing implications,’’ F. Reif and J. H. Larkin, J. Res. Sci. Teach. 28,
tom of the class but a significant positive effect for the middle 60%. 733–760 ~1991!. The spontaneous cognitive activities that occur in
156. ‘‘Learning statistical analysis of measurement errors,’’ M.-G. Séré, R. everyday life are compared with those required for learning science.
Journeaux, and C. Larcher, Int. J. Sci. Educ. 15, 427–438 ~1993!. A The authors pinpoint differences and show how application of every-
study was conducted in France to determine what 20 students in a day cognitive expectations in a science class causes difficulties.
first-year physics laboratory course had learned about the statistical 166. ‘‘Students’ beliefs about conceptual knowledge in introductory phys-
concepts taught. Students had specific difficulties in understanding the ics,’’ D. Hammer, Int. J. Sci. Educ. 16, 385–403 ~1994!.
role and value of statistical tools in assessing confidence in a measure- 167. ‘‘Epistemological beliefs in introductory physics,’’ D. Hammer, Cogn.
ment. Instruct. 12, 151–183 ~1994!.
157. ‘‘Why may students fail to learn from demonstrations? Social practice The two papers above report on studies in which the author explored
perspective on learning in physics,’’ W.-M. Roth, C. J. McRobbie, K. students’ views about the nature of physics knowledge and their ap-
B. Lucas, and S. Boutonné, J. Res. Sci. Teach. 34, 509–533 ~1997!. proaches to the cognitive content of physics. The author characterized
The authors observed a class of Australian high-school seniors and their attitudes and beliefs along several dimensions.
conducted interviews and post-tests to probe their response to demon- 168. ‘‘How novice physics students deal with explanations,’’ J. S. Touger,
strations. They classify general difficulties that could cause students to R. J. Dufresne, W. J. Gerace, P. T. Hardiman, and J. P. Mestre, Int. J.
miss the point of a demonstration and make suggestions for how to Sci. Educ. 17, 255–269 ~1995!. Introductory physics students were
improve its effectiveness. asked to explain open-ended problem situations and to select which of
158. ‘‘First-year physics students’ perceptions of the quality of experimental a variety of types of explanations they preferred. Their recognition of
measurements,’’ S. Allie, A. Buffler, L. Kaunda, B. Campbell, and F. appropriate concepts was highly situation dependent. They were fre-
Lubben, Int. J. Sci. Educ. 20, 447–459 ~1998!. The paper reports an quently unable to interpret explanations given in everyday terms.
investigation of student understanding about the reliability of experi- 169. ‘‘Models in physics: Perceptions held by prospective physical science
mental data. The research was conducted with first year science stu- teachers studying at South African universities,’’ J. J. A. Smit and M.
dents at a university in South Africa. The investigators analyzed the Finegold, Int. J. Sci. Educ. 17, 621–634 ~1995!. A study was con-
types of reasoning used by the students and found a strong dependence ducted to determine how 200 pre-service physical science teachers in
on context. South Africa and Namibia interpreted the word ‘‘model’’ in a physics
context. Many interpreted the term as a physical construct rather than
as an abstract idea. This confusion exacerbated difficulties with the
D. Ability to apply mathematics in physics interpretation of physical models for light.
A minimum level of mathematical proficiency, as deter- 170. ‘‘Guest comment: Why undergraduates leave the sciences,’’ E. Sey-
mour, Am. J. Phys. 63, 199–202 ~1995!. The author reports on the
mined by prescribed prerequisite courses, is usually assumed results of an extensive three-year study on the reasons why under-
for an introductory physics course. Instructors frequently as- graduates leave science-based disciplines. More than half of the stu-
sume that students will be able to apply the mathematics dents who intended to major in physical science did not complete a
taught in these courses to physics problems. However, both major in science. Those who left did not differ in measured ability
research and teaching experience indicate that many students from those who remained.
lack this ability. The papers below address this issue. 171. ‘‘Differences in students’ perceptions of learning physics,’’ M. Prosser,
P. Walker, and R. Millar, Phys. Educ. 31, 43–48 ~1996!. The authors
159. ‘‘Translation difficulties in learning mathematics,’’ J. Clement, J. conducted open-ended pre- and post-surveys of first-year physics stu-
Lochhead, and G. S. Monk, Amer. Math. Monthly 88, 286 ~1981!. This dents at an Australian university. Most students had a superficial and
paper reports on the pitfalls freshman engineering majors encounter inappropriate view of physics learning.
when they are asked to construct equations to match situations de- 172. ‘‘Views about science and physics achievement: The VASS story,’’ H.
scribed in words. Halloun, AIP Conf. Proc. 399, 605–613 ~1997!. ~See Ref. 9.! The
160. ‘‘The mathematical knowledge of physics graduates: Primary data and author describes the development of the Views About Science Survey
conclusions,’’ E. Breitenberger, Am. J. Phys. 60, 318–323 ~1992!. The ~VASS! to probe student attitudes about the nature of science. He
author discusses a survey of the mathematical sophistication of enter- classifies student attitudes in four broad profiles of increasing sophis-
ing physics graduate students at a major university. tication and correlates the profiles with performance.
161. ‘‘Teaching algebraic coding: Stakes, difficulties and suggestions,’’ G. 173. ‘‘Student expectations in introductory physics,’’ E. F. Redish, J. M.
Rebmann and L. Viennot, Am. J. Phys. 62, 723–727 ~1994!. The au- Saul, and R. N. Steinberg, Am. J. Phys. 66, 212–224 ~1998!. The
thors discuss the difficulty of many university physics students in ap- authors developed a survey to probe student cognitive attitudes and
plying and interpreting algebraic sign conventions consistently. Ex- beliefs about physics. The Maryland Physics Expectations ~MPEX!
amples from dc circuits, thermodynamics, and optics are given. Survey is included in the appendix. Results from 1500 students at 6
162. ‘‘The vector knowledge of beginning physics students,’’ R. D. Knight, colleges and universities indicate that student attitudes about physics
Phys. Teach. 33, 74–78 ~1995!. A study involving about 300 university tend to deteriorate, rather than improve, as instruction progresses.
engineering students probed their understanding of vectors. After
mathematics and physics courses in high school and a semester of
college calculus, only one-third indicated familiarity with finding mag-
nitudes or recognizing vector components.
F. Reflections on research into student reasoning
E. Attitudes and beliefs of students There are some papers that take a broad view on the in-
terpretation or implications of experimental studies that do
The attitudes and beliefs that students bring with them to not easily fit into a content-oriented categorization.
class may influence what they learn in a physics course. The
174. ‘‘Analyzing students’ reasoning: Tendencies in interpretation,’’ L.
papers below report on studies conducted with university
Viennot, Am. J. Phys. 53, 432–436 ~1985!. This paper discusses the
students in introductory courses. danger of interpreting student responses through the filter of a physi-
163. ‘‘Learning physics vs. passing courses,’’ H. Lin, Phys. Teach. 20, cist’s perspective. Two examples from dynamics are cited.
151–157 ~1982!. The author interviewed 25 students who were doing 175. ‘‘Research and computer-based instruction: Opportunity for interac-
poorly in a university calculus-based physics course. He determined tion,’’ L. C. McDermott, Am. J. Phys. 58, 452–462 ~1990!.
that many of their difficulties were related to inappropriate attitudes 176. ‘‘Use of the computer for research on student thinking,’’ D. J. Grayson
about learning and the nature of what is learned in a physics course. and L. C. McDermott, Am. J. Phys. 64, 557–565 ~1996!.

764 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 764
 The two papers above describe the use of the computer as an instruc- problem solving. In some instances, physics is used as a
tional aid and as a research tool to examine student reasoning. context to develop a model for problem-solving in a more
177. ‘‘More than misconceptions: Multiple perspectives on student knowl-
edge and reasoning, and an appropriate role for education research,’’
general sense. The models for problem-solving performance
D. Hammer, Am. J. Phys. 64, 1316–1325 ~1996!. The author reflects discussed in the references below focus on physics and re-
upon what physics education research can bring to the discussion of flect a range of expertise that varies from novice to expert.
instructional goals and strategies. 182. ‘‘Expert and novice performance in solving physics problems,’’ J. H.
Larkin, J. McDermott, D. Simon, and H. A. Simon, Science 208,
V. THEORETICAL PERSPECTIVES 1135–1142 ~1980!. The authors examine the role of physical intuition
in problem solving and conclude that experts use highly structured
As is appropriate in the early stages of any scientific field, patterns of information to index and apply their knowledge.
183. ‘‘Knowledge structures and problem solving in physicists,’’ F. Reif and
most of the research in physics education has been empirical J. I. Heller, Educ. Psychol. 17, 102–127 ~1982!. The authors give a
rather than theoretical. At present, there are no models of detailed description of a theoretical approach to problem solving in
mental processes or theories of instruction nearly as well mechanics.
developed as the models and theories of physics. In order to 184. ‘‘Acquiring an effective understanding of scientific concepts,’’ F. Reif,
build a theory of student learning in physics, it is necessary in Cognitive Structure and Conceptual Change, edited by L. H. T.
~in addition to a strong command of the subject! to have an West and L. Pines ~Academic, Orlando, FL, 1985!, pp. 133–151. Prob-
lem solving is described in terms of three main stages: description and
understanding of human thought processes in a more general
analysis of the problem, construction of a solution, and testing of the
sense. solution. The ability to solve problems depends not only on the learn-
The relevant concepts for describing mental processes are ing of procedures but also on the ability to draw on appropriate ancil-
not easily identified, operationally defined, or readily quan- lary knowledge.
tifiable. Theories of instruction do not have the same predic- 185. ‘‘Non-formal reasoning in experts and science students: The use of
tive capability nor are they falsifiable in the same sense as analogies, extreme cases and physical intuition,’’ J. Clement, in Infor-
theories that pertain to the physical world. Despite these dif- mal Reasoning and Education, edited by J. F. Voss, D. N. Perkins, and
J. W. Segal ~Lawrence Erlbaum, Hillsdale, NJ, 1991!, pp. 341–381.
ferences, a theoretical perspective can be useful for interpret- The author studied the uses of analogy by expert problem solvers and
ing, organizing, and generalizing observations. Models for developed an instructional strategy in which analogies are used to help
how students develop conceptual understanding and the abil- students build a ‘‘bridge’’ from their spontaneous conceptions to a
ity to solve physics problems can help guide the develop- more scientific understanding.
ment of instructional strategies. As in all sciences, compre-
hensive theories may reveal previously unrecognized
relationships, identify questions for further investigation, and VI. PAPERS FROM RELATED FIELDS
set new directions for research. Knowledge of relevant aspects of cognitive science, cog-
A. Concept development nitive psychology, and neuroscience are likely to play an
essential role in the eventual development of accurate and
In the references cited in this subsection, a major goal of useful theories. The extensive literature in these fields con-
the research has been the development of mental models that tains information relevant to physics education research.
can be used to describe the process of conceptual change in None of the references cited here requires an extensive back-
students. ground in either education or psychology.
178. ‘‘Accommodation of a scientific conception: Toward a theory of con-
ceptual change,’’ G. J. Posner, K. A. Strike, P. W. Hewson, and W. A.
Gertzog, Sci. Educ. 66, 211–227 ~1982!. A model that identifies ele-
A. Cognitive studies and physics education research
ments needed for conceptual change is illustrated with an example in
A number of physicists have considered how findings
which students begin to make sense of special relativity.
179. ‘‘The role of conceptual conflict in conceptual change and the design of from cognitive psychology can help us understand how
science instruction,’’ P. W. Hewson and M. G. A’Beckett-Hewson, people learn in general and how they learn physics in par-
Instr. Sci. 13, 1–13 ~1984!. The authors present a model for learning ticular. The papers below draw on relevant research in cog-
that describes conceptual change in terms of conflict between existing nitive psychology.
conceptions and new conceptions. The learner may adopt a new con-
186. ‘‘Can physics develop reasoning?’’ R. G. Fuller, R. Karplus, and A. E.
ception if it is ‘‘intelligible, plausible, and fruitful.’’
Lawson, Phys. Today 30 ~2!, 23–28 ~1977!.
180. ‘‘Studying conceptual change: Constructing new understandings,’’
187. ‘‘Wherefore a science of teaching?’’ D. Hestenes, Phys. Teach. 17,
D. I. Dykstra. See Ref. 7, Research in Physics Learning: Theoretical
235–242 ~1979!.
Issues and Empirical Studies, pp. 40–58. Conceptual change is char-
188. ‘‘Solving physics problems—how do we do it?’’ R. G. Fuller, Phys.
acterized by stages of ‘‘differentiation, class extension, and reconcep-
Today 35 ~9!, 43–47 ~1982!.
tualization.’’ @For a more detailed discussion, see D. I. Dykstra,
189. ‘‘Implications of cognitive studies for teaching physics,’’ E. F. Redish,
‘‘Studying conceptual change in learning physics,’’ Sci. Educ. 76,
Am. J. Phys. 62, 796–803 ~1994!.
615–652 ~1992!. This paper, which is published in a widely distributed
journal, is more oriented toward science educators than the paper in the A number of books provide useful overviews for those
Bremen conference proceedings.# interested in learning more detail about cognitive science.
181. ‘‘Facets of students’ knowledge and relevant instruction,’’ J. Minstrell.
See Ref. 7, ibid., pp. 110–128. Student knowledge is described in 190. Readings in Cognitive Science, A. Collins and E. E. Smith ~Morgan
terms of ‘‘facets’’ that relate to content, strategies or reasoning. In- Kauffmann, San Mateo, CA, 1988!. This is a collection of articles in
struction is viewed as an effort to help students modify existing facets, cognitive science.
add new facets, and incorporate existing and new facets into a correct 191. The Mind’s New Science: A History of the Cognitive Revolution, H.
conceptual framework. Gardner ~Basic, New York, 1987!. This is a brisk and entertaining
review of the history of cognitive science up to 1985. Contributions
ranging from anthropology to linguistics are covered.
B. Problem-solving performance 192. The Growth of Logical Thinking from Childhood to Adolescence, B.
Inhelder and J. Piaget ~Basic, New York, 1958!. This classic work by
Some theoretical research has been directed toward eluci- one of the founders of the cognitive approach contains many examples
dating the process through which students develop skill in of how young children interpret the physical world.

765 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 765
 A few references from educational specialists also give a ‘‘Concepts first—A small group approach to physics learning,’’ R.
useful overview of the relevant psychology. Gautreau and L. Novemsky, Am. J. Phys. 65, 418–428 ~1997! dis-
cusses an implementation and evaluation of the OCS materials.
193. Educational Psychology, D. Ausubel ~Holt, Rinehart and Winston, 205. Understanding Basic Mechanics, Text and Workbook, Frederick Reif
New York, 1968!. A general introduction to the application of psycho- ~Wiley, New York, 1995!. Problem solving is taught through an in-
logical ideas in education, this comprehensive book discusses concept structional strategy that consists of three steps. An initial analysis in-
development and discovery learning. cludes a description of the problem situation, a summary of the goals,
194. Styles of Integrated Learning and Teaching: An Integrated Outline of and a redescription of the situation in technical terms. The problem is
Educational Psychology for Students, Teachers and Lecturers, N. then decomposed into subproblems. The third step consists of check-
Entwistle ~Wiley, New York, 1981!. This is one of the more accessible ing the solution. The steps are repeated if necessary. ~See Refs. 141,
studies of the variability of styles and ways of approaching learning 144, 145, and 146.!
preferred by college students. 206. Tools for Scientific Thinking, David Sokoloff and Ronald Thornton
195. ‘‘Reassessment of developmental constraints on children’s science in- ~Vernier Software, Portland, OR, 1995!.
struction,’’ K. E. Metz, Rev. Educ. Res. 65, 93–127 ~Summer, 1995!. 207. RealTime Physics, David R. Sokoloff, Ronald K. Thornton, and
This article is a good review of the current state of understanding of the Priscilla W. Laws ~Wiley, New York, 1999!.
process of cognitive development. In the two curricula above, microcomputer-based laboratory activities
engage students in graphing motions, including their own, in real time.
Instant feedback helps students relate motions to graphical representa-
B. Applications of cognitive studies to education tions. ~See Refs. 62, 69, 81, and 209.!
208. Physics by Inquiry, Vols. I and II, L. C. McDermott and the Physics
A number of references from education are particularly Education Group at the University of Washington ~Wiley, New York,
relevant to physicists interested in specializing in physics 1996!. Physics by Inquiry is a set of laboratory-based modules in
education research. Following are a few books and collec- which the emphasis is on the development of concepts and scientific
tions that can give the reader an entry into this extensive reasoning skills. Students work collaboratively in small groups, con-
literature. duct investigations with simple equipment, and use their observations
as a basis for constructing scientific models. These instructional mate-
196. Mental Models, edited by D. Gentner and A. L. Stevens ~Lawrence rials are especially appropriate for preparing prospective and practic-
Erlbaum Associates, Hillsdale, NJ, 1983!. ing teachers to teach physics and physical science at the pre-college
197. Cognitive Science and Mathematics Education, edited by A. H. level. ~See Refs. 26, 27, 30, 58, 88, 100, 107, 108, and 116.!
Schoenfeld ~Lawrence Erlbaum Associates, Hillsdale, NJ, 1987!. 209. Workshop Physics Activity Guide, P. Laws ~Wiley, NY, 1997!. Instruc-
198. Toward a Scientific Practice of Science Education, edited by M. Gard- tion is based on a four-part learning sequence. Students make predic-
ner, J. G. Greeno, F. Reif, A. H. Schoenfeld, A. diSessa, and E. Stage tions about a phenomenon, reflect on their observations and try to
~Lawrence Erlbaum Associates, Hillsdale, NJ, 1990!. reconcile any differences; they develop definitions and equations from
199. Handbook of Research on Science Teaching and Learning, edited by theoretical considerations; they perform experiments to verify predic-
D. L. Gabel ~MacMillan, New York, 1994!. tions based on theory; they apply their understanding in solving prob-
200. Cognitive Process Instruction, edited by J. Lochhead and J. Clement lems.
~Franklin Institute Press, Philadelphia, PA, 1979!. ‘‘Millikan lecture 1996: Promoting active learning based on physics
201. Cognitive Structure and Conceptual Change, L. H. T. West and A. L. education research in introductory physics courses,’’ P. Laws, Am. J.
Pines ~Academic, New York, 1984!. Phys. 65, 14–21 ~1997!.
202. Problem Solving and Comprehension, A. Whimbey and J. Lochhead ‘‘Calculus-based physics without lectures,’’ P. Laws, Phys. Today 44
~Lawrence Erlbaum Associates, Hillsdale, NJ, 1991!. ~12!, 24–31 ~1991!.
The two papers above describe the Workshop Physics curriculum and
its effectiveness in some detail.
VII. RESEARCH-BASED INSTRUCTIONAL 210. Tutorials in Introductory Physics, preliminary edition, L. C. McDer-
MATERIALS mott, P. S. Shaffer, and the Physics Education Group at the University
of Washington ~Prentice–Hall, Upper Saddle River, NJ, 1998!. This
The results of research in physics education are gradually supplementary curriculum can be used in conjunction with any stan-
beginning to be incorporated in the development of new cur- dard introductory physics textbook. The tutorials are designed to be
used in small group sessions in which three or four students work
ricula for students and handbooks for instructors. This sec-
together collaboratively. Worksheets guide students through the rea-
tion contains a short list of materials that have been recently soning required to develop and apply important concepts and prin-
published in the United States. In some instances, these have ciples. ~See Refs. 26, 27, 30, 43, 47, 58, 70, 88, 100, 107, 108, 111,
been developed by individuals and groups in conjunction and 116.!
with research. In other cases, the materials draw on research 211. Minds on Physics, Activities and Reader ~6 volumes!, W. J. Leonard,
by others. R. J. Dufresne, W. J. Gerace, and J. P. Mestre ~Kendall/Hunt,
Dubuque, IA, 1999–2000!. These volumes contain many activities to
help students explore their existing concepts and learn to reason sci-
entifically.
A. Instructional materials for students
Some of the instructional materials listed above formed
For each of the student materials listed below, evidence of the basis of sample classes given at the 1996 ICUPE. These
the research base is in published papers. We have not in- ~and others! are described in greater detail in the proceedings
cluded materials ~1! which are not yet published, ~2! in of that conference. ~See Ref. 9.!
which the basis in physics education research is undocu-
mented in the literature, and ~3! in which reference to edu-
cation research is not specific to physics. B. Guidance for instructors
203. ALPS: Mechanics (Vol. 1), Electricity and Magnetism (Vol. 2), A. Van Below are a few references on teaching physics that in-
Heuvelen ~Hayden-McNeil, Plymouth, MI, 1994!. structors may find useful. Although some of the instructor’s
204. Overview Case Study (OCS) Study Guide, A. Van Heuvelen ~Hayden- guides have been developed for implementing the instruc-
McNeil, Plymouth, MI, 1995!.
The above two items contain materials for a course in which students,
tional materials above, their applicability extends beyond a
guided by worksheets in interactive lectures, analyze physical situa- particular curriculum.
tions. The first encounter with a topic is qualitative. Quantitative 212. A Guide to Introductory Physics Teaching, A. B. Arons ~Wiley, New
analysis follows. ~See Ref. 64.! York, 1990!.

766 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 766
213. Homework and Test Questions for Introductory Physics Teaching, A. 221. ‘‘Resource Letter: ColR-1: Collateral Reading for Physics Courses,’’
B. Arons ~Wiley, New York, 1994!. A. M. Bork and A. B. Arons, Am. J. Phys. 35, 71–78 ~1967!.
214. Teaching Introductory Physics, A. B. Arons ~Wiley, New York, 1997!. 222. ‘‘Resource Letter: EP-1: Educational Psychology,’’ J. W. George
The two volumes above and a new section on energy and momentum Ivany, Am. J. Phys. 37, 1091–1099 ~1967!.
have been combined into a single volume. 223. ‘‘Resource Letter: PCP-1: Pre-College Physics Curriculum Materials,’’
Drawing on his extensive classroom experience, in the three items L. G. Paldy and C. E. Swartz, Am. J. Phys. 41, 166–178 ~1973!.
above, the author provides guidance for physics teachers on the nature 224. ‘‘Resource Letter: PhD-1: Physics Demonstrations,’’ J. A. Davis and
of student difficulties and on instructional methods that he has found B. G. Eaton, Am. J. Phys. 47, 835–840 ~1979!.
effective.
215. Preconceptions in Mechanics: Lessons Dealing with Conceptual Diffi- IX. CONCLUSION
culties, C. J. Camp, J. Clement, D. Brown, K. Gonzalez, K. Kudukey,
J. Minstrell, J. Schultz, K. Steinberg, M. Veneman, and A. Zietsman Traditionally, physics instruction has been based on the
~Kendall/Hunt, Dubuque, IA, 1994!. This volume discusses student instructor’s view of the subject and perception of the student.
preconceptions in mechanics and contains a series of lesson plans that
As many of the references included in this Resource Letter
are designed to build a bridge from common preconceptions to a more
scientific view. demonstrate, the same instruction may appear very different
216. Instructor’s Manual for Understanding Basic Mechanics, Frederick to the instructor and to the student. Improving the match
Reif ~Wiley, New York, 1995!. This guide to the author’s mechanics between teaching and learning requires knowledge about
text and workbook ~Ref. 205! discusses problems and pitfalls involved how students think. Results from research have proved to be
in teaching mechanics. It also gives an overview of general cognitive extremely useful as a guide to the development of effective
and pedagogical issues, as well as many references. instruction.
217. Peer Instruction, A User’s Manual, Eric Mazur ~Prentice–Hall, Upper
Saddle River, NJ, 1997!. The author describes a general strategy for
In the past two decades, research in physics education has
promoting intellectual engagement by students in large courses. At emerged as a field of scholarly inquiry in which physicists
several points during the lecture, the instructor presents a qualitative are actively engaged. They are conducting systematic inves-
question and multiple-choice responses that together are designed to tigations that are contributing to a steadily growing research
reveal common conceptual difficulties. Many examples are provided. base. For this resource to be useful to the physics teaching
218. Instructor’s Guide for Physics by Inquiry, L. C. McDermott and the community, however, studies must be documented in the lit-
Physics Education Group at the University of Washington ~Wiley, New
York, 1998!. The Instructor’s Guide outlines the goals of particular
erature and subjected to the scrutiny and challenges of peers
exercises and experiments in Ref. 208. as in traditional areas of physics research. Only in this way is
219. Instructor’s Guide for Tutorials in Introductory Physics, L. C. McDer- cumulative progress possible.
mott, P. S. Shaffer, and the Physics Education Group at the University
of Washington ~Prentice–Hall, Upper Saddle River, NJ, 1998!. The ACKNOWLEDGMENTS
Instructor’s Guide provides pretests, sample examination questions,
and additional information on individual tutorials in Ref. 210. The authors would like to thank Paula R. L. Heron and
Peter S. Shaffer for their invaluable assistance in preparing
VIII. OTHER RESOURCE LETTERS RELEVANT TO this Resource Letter. Thanks are also due for contributions
PHYSICS EDUCATION from the other members of the Physics Education Group at
the University of Washington and from the members of the
Of the approximately 120 Resource Letters that have been Physics Education Research Group at the University of
published in the past 30 years, only a few have physics edu- Maryland. In particular, Bradley Ambrose, Christian Kautz,
cation as their primary focus. Although the ones cited below Michael Loverude, Rachel Scherr, John Thompson, Michael
are not on research, they address important related issues. Wittmann, and Karen Wosilait served as reviewers. The au-
220. ‘‘Resource Letter: AT-1: Achievement Testing,’’ H. Kruglak, Am. J. thors also gratefully acknowledge the ongoing support of the
Phys. 33, 255–263 ~1965!. National Science Foundation.

B2FH—BURBIDGE, BURBIDGE, FOWLER, AND HOYLE

Protons and neutrons are collectively called nucleons. Their production, nucleogenesis, oc-
curred at a still hotter and denser phase of which we have relatively little knowledge. Nucleosyn-
thesis is the array of processes by which they are assembled into nuclei.
The catch phrase ‘‘God made hydrogen and helium; Burbidge, Burbidge, Fowler, and Hoyle
made all the rest’’ is a summary of those processes.

Virginia Trimble, Visit to a Small Universe ~The American Institute of Physics, New York, 1992!, p. 120.

767 Am. J. Phys., Vol. 67, No. 9, September 1999 L. C. McDermott and E. F. Redish 767