CALCULUS-BASED PHYSICSWITHOUT LECTURES

Computer tools ond kinestheticapparatus play key roles in a novel

approach to introductory physics thattakes into account both time-honored

ideas about learning and findingsfrom recent educational research.

Priscilla W. Laws

Priscilla Laws, a professor of physics at Dickinson College inCarlisle, Pennsylvania, has directed the Workshop Physics

project since 1986.

Every fall several hundred thousand students enroll incalculus-based "engineering" physics courses throughoutthe United States. Informal statistics tell us that over halfof them will fail to complete the sequence of introductorycourses. These students complain that physics is hard andboring. The most compelling student critique of tradition-al introductory physics and chemistry courses comes fromcollege graduates in the humanities who were engaged bySheila Tobias to take introductory science for credit.1These students paint a devastating portrait of introduc-tory courses as uninteresting, time consuming, narrowlyfixated on the procedures of textbook problem solving,devoid of peer cooperation, lacking in student involvementduring lectures, crammed with too much material, andbiased away from conceptual understanding.

Why aren't students who take introductory sciencedoing better? Why are they turning away? It is temptingfor frustrated introductory physics instructors to seeksimple answers such as "High schools are no longer doingtheir job" or "If students were only smarter and willing towork harder, we could teach them successfully." Thereare probably many reasons for the apparent decline inperformance of introductory physics students: A largerpercentage of 18-year-olds are enrolling in colleges; manystate universities have open-admissions policies; there is ashortage of properly trained high school teachers; college-bound high school students spend less than one hour a daystudying; they come to physics with little experienceworking with their hands; there are more extracurricularactivities and campus jobs to distract college students fromacademics; and so on. Whatever the reasons, mostinstructors agree that at present many introductoryphysics students seem unprepared and unmotivated.

Workshop Physics philosophyAt Dickinson College we have attempted to analyze theproblems associated with the teaching of calculus-basedcourses, to set new goals and to achieve these goals bychanging the way we teach. After receiving a three-yeargrant from the Department of Education's Fund for theImprovement of Postsecondary Education, John Luet-zelschwab, Robert Boyle, Neil Wolf and I began planningthe Workshop Physics program at Dickinson College inthe fall of 1986, in collaboration with Ronald Thorntonfrom the Tufts University Center for the Teaching ofScience and Mathematics and David Sokoloff from the

2 4 PHYSICS TODAY DECEMBER 1991 © 1991 Americon Institute of Physics

Analog to projectile motion. A WorkshopPhysics student hits a bowling ball repeatedlyin one direction with a baton to approximate

a continuous force. Beanbags dropped atregular intervals record locations of the ball.

The ball moves with a constant velocity in onedirection and constant acceleration in the

other, and students obtain a "musclememory" of having to chase the ball faster

and faster.

University of Oregon.The implicit goals in most traditional introductory

physics courses center around teaching students to solvetextbook problems. In developing Workshop Physics weassumed that acquiring transferable skills of scientificinquiry is a more important goal than either problemsolving or the comprehensive transmission of descriptiveknowledge about the enterprise of physics. Arnold Aronsrefers to this aim as the "development of enoughknowledge in an area of science to allow intelligent studyand observation to lead to subsequent learning withoutformal instruction."2

There were two major reasons for emphasizingtransferable inquiry skills based on real experience. First,the majority of students enrolled in introductory physicsat both the high school and college level do not havesufficient concrete experience with everyday phenomenato comprehend the mathematical representations of themtraditionally presented in these courses. The processes ofobserving phenomena, analyzing data and developingverbal and mathematical models to explain observationsafford students an opportunity to relate concrete experi-ence to scientific explanation. The second reason forfocusing on the development of transferable skills is thatwhen one is confronted with the task of acquiring anoverwhelming body of knowledge, the only viable strategyis to learn some things thoroughly while acquiringmethods for independently investigating other things asneeded. This approach follows the adage "Less is more."

We incorporated three new elements into our plan-ning: We took findings from science education researchinto account; we designed and adapted integrated comput-er tools for the introductory classroom; and we developeddevices that allow students to experience motions andforces with their own bodies (kinesthesic apparatus).

We used several criteria in choosing topics to becovered in the Workshop Physics courses. To helpstudents prepare for further study in physics and engi-neering, we decided to select topics normally covered inthe introductory course sequence. Most of these topicsinvolve phenomena that are amenable to direct observa-tion, and the mathematical and reasoning skills needed toanalyze observations and experiments in these topics areapplicable to many other areas of inquiry. We did not addtopics, such as relativity and quantum mechanics, thatrequire levels of abstract reasoning we believe to be

beyond the abilities of the majority of introductorystudents.

Since we wanted to eliminate several topics, we choseto omit those that are covered in our second-year program,such as waves, ac circuits, and geometric and physicaloptics. We did develop two new units on contemporarytopics for the introductory program, one on chaos and theother on radon monitoring. Thus we found ourselveseliminating about 25% of the material we used to cover inour traditional introductory calculus-based physics se-quence. There is nothing sacrosanct about our choices.Indeed, several other institutions that have adopted ourprogram have chosen to delete different material, optingfor a sequence of topics tailored to the particular needs oftheir students and the special interests of their faculties.

To allow students the time to make observations, doexperiments and discuss their findings, we decided toeliminate formal lectures and teach the courses in aclassroom-laboratory environment outfitted with comput-ers and scientific apparatus. Although lectures anddemonstrations are useful alternatives to reading for

PHYSICS TODAY DECEMBER 1991 2 5

••DD•DDD

••D•D•D

•D•DDDDD

D•DDDD•

i234567

A

(m.)0.826.32

10.7113.5517.01

Bi

(sec.)0.441.151.491.701.91

Ct*2 ;

(see'2)0.19:1.32!

2.22:2.893.65

Freefall Data for S vs. t

o.o 2.0

Svs.tA2y = 0.0739 + 4.6777X R = 1.00

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0tA2 (secA2)

Linearizing data using spreadsheet andgraphing software, a: Students drop an objectfrom different heights 5. b: Data are enteredinto a spreadsheet, c: Data are transferred toa graph and plotted, d: Vertical distance 5 isplotted as a function of t2, and the curve islinearized.

transmitting information and teaching specific skills,their value as vehicles for helping students learn how tothink, conduct scientific inquiry or acquire real experiencewith natural phenomena is unproven.3 In fact, someeducators believe that peers are often more helpful thaninstructors in stimulating original thinking and problemsolving on the part of students.4 Thus we believe that thetime students now spend passively listening to lectures isbetter spent in direct inquiry and discussion with peers.The role of the instructor in our program is to help createthe learning environment, lead discussions and encouragestudents to engage in reflective discourse with oneanother.

Workshop Physics in practiceWorkshop Physics was first taught at Dickinson Collegeduring the 1987-88 academic year, to students in both thecalculus- and non-calculus-based courses. It is taught inthree two-hour sessions each week, with no formallectures. Each section has one instructor, two undergrad-uate teaching assistants and up to 24 students. Inaddition, the workshop labs are staffed during evening andweekend hours by undergraduate teaching assistants.

Pairs of students share the use of a computer and anextensive collection of scientific apparatus and othergadgets. Among other things, students pitch baseballs,whack bowling balls with twirling batons, break pineboards with their fists, pull objects up inclined planes,build electronic circuits, explore electrical unknowns,ignite paper with compressed gas and devise engine cyclesusing rubber bands.

Hans Pfister, who joined the Workshop Physicsteaching staff this fall, has designed a series of carts thatstudents can ride and in which they can experience withtheir own bodies one- and two-dimensional motions andcollisions. The range of kinesthetic experiences availableto students is expanding as new apparatus is designedand tested.

The topics have been broken up into units lastingabout one week, and students use a specially preparedWorkshop Physics Activity Guide, which includes exposi-tion, questions and instructions as well as blank spaces forstudent data, calculations and reflections (see the box onpage 29). In general a four-part learning sequence is used:Students begin a week with an examination of their ownpreconceptions and then make qualitative observations.After some reflection and discussion by the students, theinstructor helps with the development of definitions andmathematical theories. The week usually ends withquantitative experimentation centered around verifica-tion of mathematical theories. Readings and problems areassigned in a standard textbook, but only after studentshave discussed phenomena and made predictions andobservations in class. In adapting computers for use inWorkshop Physics, we have attempted to mimic some ofthe ways that physicists use computers to understandphenomena. Thus the computer is used in almost everycapacity except that of computer-assisted instruction.

Although the MUPPET project at the University of

26 PHYSICS TODAY DECEMBER 1991

Maryland has reported great success in teaching introduc-tory students to program in PASCAL (see Gerhard Sa-linger's article in PHYSICS TODAY, September, page 39), ourexperience in developing computer-based laboratories atDickinson has led us to use the spreadsheet as the majortool for calculation. Previous attempts to incorporate aprogramming language into the introductory lab left uswith the feeling that we were using physics to teachcomputing rather than the other way around.

The computer application most frequently used inWorkshop Physics involves the use of spreadsheets fordata analysis and numerical problem solving. Data arereadily transferred to graphics software with curve fittingroutines. (Curve fitting is considered to be one of theessential transferable skills associated with WorkshopPhysics.) Using the microcomputer for curve fitting,linearization and least-squares analysis, students discoversimple functional relationships empirically or verifymathematical theories (see the figure on page 26).

In one unusual application of linearization, a parallelarray of nails on a wooden base represents "flux lines"associated with a uniform electric field. The number ofnails passing through a wire hoop (used to represent asurface area) as a function of the angle between the hoop'snormal vector and the direction of the nails can becounted. Plotting the number of nails subtended versusthe cosine of the angle yields a straight line. Thus the stu-dents "discover" that flux through an area can berepresented as a dot product of the field and the normalvector.

Spreadsheet calculations are also used as a tool forperforming numerical integrations. In some cases spread-sheet calculations are used for mathematical modeling.For example, spreadsheet relaxation calculations workbeautifully for modeling the pattern of electrical poten-tials surrounding the "electrodes" on electric field map-ping paper. Mathematical functions representing travel-ing waves can be plotted in position space at threedifferent times, and the velocity of the wave can bemeasured on the graph. This helps students explore thereal meaning of the expression Y = f(x ± vt).

We are beginning to explore the potential of symbolicand numerical equation solvers in the Workshop Physicsprogram. Students are taught to enter simple commandsinto Maple, a computer algebra system capable of symbolicmanipulation, to determine integrals, solve simultaneousKirchhoff's law equations and plot functions. Because weconsider the spreadsheet operations to be more obvious tostudents and less demanding with regard to syntax, wehave no plans to expand the use of programs like Mapleand Mathematica at the introductory level.

Use of MDL toolsAs part of the Tools for Scientific Thinking project basedat Tufts University, which—like Workshop Physics—issupported by the Education Departments's Fund for theImprovement of Postsecondary Education, Thornton andhis colleagues have collaborated with the WorkshopPhysics staff in the design and testing of hardware and

software to allow students to collect and display graphs ofdata in real time. These microcomputer-based laboratorytools are used extensively in the Workshop Physicsprogram and in high school and university physicscourses throughout the United States. An MBL stationconsists of a sensor or probe plugged into a microcom-puter via a serial interface. With appropriate softwarethe computer can perform instantaneous calculations orproduce graphs.

The MBL software is used in two ways. First, incases where the user can observe or control changes in asystem directly, the microcomputer is particularly power-ful when it is set up to display a real-time graph of thesystem changes. Thornton and Sokoloff, and HeatherBrassell, working independently at the University ofFlorida, have demonstrated that the use of MBL tools tocreate real-time graphs yields impressive results inhelping students develop an intuitive feeling for themeaning of graphs and for qualitative characteristics ofphenomena they are observing.5 For example, a timetrace of the position of one's own body as monitored by anultrasonic motion detector is unparalleled for learninghow the abstraction known as a graph can represent thehistory of change in a parameter. MBL software has beendeveloped at Tufts for logging motion, force, temperature,sound and voltage data.

In addition, MBL software has been developed atDickinson for radiation detection and photogate timing. Areal-time frequency distribution produced using a Geigertube with a radioactive source, affords students the sameopportunities to explore and develop intuitive notionsabout both the meaning of frequency distributions and thenature of counting statistics. The MBL photogate softwareis pedagogically oriented and uses a raw plotter to allowstudents to see the times when real events switch one ormore photogates on or off. A real-time raw plot, which isone of Robert Tinker's many innovative ideas, lends itselfto students' discovering how to use operational definitionsin the measurement of velocity and acceleration.

Until the past year or so it was difficult for us tostreamline the acquisition and analysis of two-dimension-al motion data. The availability of computer-based videotechnology has solved that problem. This fall we beganintroducing students to computer analysis of motionsrecorded on videodisc and on student-generated video-tapes. We have been collaborating with Jack Wilson atRennselaer Polytechnic Institute and Joe Redish at theUniversity of Maryland on adapting the video tools underdevelopment as part the CUPLE project (see Salinger'sarticle). For example, video analysis is invaluable forstudying vertical free-fall and projectile motion. It alsomakes it possible to track the center of mass of a system ofpucks on an airtable or of a high jumper passing over a bar.

In select cases where acquiring real data is notfeasible or is too time consuming, we have resorted to theuse of simulations. One such simulation is a programdeveloped by David Trowbridge of Microsoft, Graphs andTracks,6 which simulates position, velocity and accelera-tion graphs for a ball rolling down a set of inclined ramps.

PHYSICS TODAY DECEMBER 1991 2 7

Coulomb, a program developed by Bias Cabrera ofStanford University, is capable of displaying electric fieldlines associated with a collection of charges.7 Studentsenjoyed creating strange and unique charge configura-tions on the computer screen and watching the patternsgenerated by the field lines. This simulation allowedstudents to discover that in two-dimensional "Cabrera-land," the flux enclosed by a loop is always proportional tothe net charge enclosed by the loop. In another simula-tion, students can analyze the motion of a moleculebouncing around in a two-dimensional box as part of thekinetic theory derivation relating the pressure andvolume in a box to the kinetic energy of the molecule.

Using a visual simulation program such as Knowl-edge Revolution's Interactive Physics package8 studentscan create an idealized impulse curve that might resultwhen a pair of rigid objects connected by a spring collideswith a wall. They can then compare this idealized curvewith the actual impulse curve obtained when a rolling cartcollides with a force probe. This exercise providesstudents with an illuminating glimpse at the process ofmodeling physical phenomena as idealized systems.

Last but not least, our students use the computer forword processing and creating apparatus drawings forformal laboratory reports. Since we hold written commu-nication skills to be quite important, students are requiredto hand in a formal lab report each semester. Instructorsreview these reports carefully and then return them to thestudents for extensive revisions. Students can create anentire laser-printed lab report, with computer-logged dataas well as computer-generated tables, graphs, diagramsand prose, without ever picking up a pencil or pen.

Student learning and attitudesAn extensive program is under way to assess the impact ofWorkshop Physics activities and teaching strategies onstudent learning and attitudes at Dickinson College andthe University of Oregon. We have administered severalconceptual tests and tracked the performance of studentson course examinations before and after the WorkshopPhysics program was instituted at Dickinson. We alsohave conducted a survey of student attitudes toward thestudy of introductory physics among about 1600 studentsat 16 colleges and universities. Here are some of ourpreliminary findings.t> Students at Dickinson College express a preference forthe workshop method of teaching. Based on the writtenresponses to the college-wide course evaluation forms atDickinson, about two-thirds of all students who have takenWorkshop Physics in our calculus-based courses express astrong preference for the workshop approach over whatthey imagine the lecture approach to be like. About halfthe students in the algebra-based courses serving premedi-cal students state a preference for the method.> In Workshop Physics, a greater percentage of studentsmaster concepts that are considered difficult to teachbecause they involve classic misconceptions. This im-proved mastery is demonstrated by improvements in thescores on selected concept-oriented questions developed atArizona State University, Tufts University, University ofWashington and the University of Oregon. These im-provements in basic conceptual understanding are the

result of students acquiring direct experience with phe-nomena. For example, a question on the mechanicsconcepts examination developed at Arizona State Univer-sity10 asks students to identify the path of a rocket after itsengines have fired at a constant rate (see the figure onpage 30). Prior to the introduction of the WorkshopPhysics program 73% of Dickinson students who hadcompleted the mechanics portion of physics got the wronganswer. Only 31% of students who had completed themechanics portion of Workshop Physics in the fall of 1983failed to answer the question correctly. The kinestheticactivity in which students apply "constant" forces bygiving moving bowling balls small taps with twirlers'batons gives them a base of experience and allows them tovisualize two-dimensional motion in which one dimensionhas no acceleration and the other dimension undergoes auniform acceleration (see the figure on page 25).

A senior woman who came to Dickinson as aninternational studies major and switched to physicsdescribed this experience in an oral interview: "The firstexam was going to be problems, and I said, 'How can I pos-sibly take an exam which is problems when all we've beendoing is playing with toys?'. . . Well, I got the exam, andthe first problem was a rocket problem, and it talked abouta rocket going up, and it has a constant wind hitting it, andyou had to guess the path of the rocket. I have learnednothing about rockets and I'm not a rocket scientist, and sohow am I ever going to do this problem? .. . All of a suddenI remembered sitting in the Kline Center with a baton anda bowling ball and hitting this bowling ball, and so if wethought this baton was the wind and the bowling ball wasa rocket—wow! I did this problem and I got it right... Itwasn't in a book or anything, but I saw it in my head."t> Performance of Workshop Physics Students in upper-level physics courses and in solving traditional textbookproblems is as good as that of students who took ourtraditional lecture courses. For assessment purposes, wedevote the same amount of time to textbook reading,homework assignments and textbook-style problems onexaminations as we did when teaching traditional courses.Performance is judged based on grades on textbookproblems, scores on the problem portions of our introduc-tory examinations, and the impressions of instructors inupper-level courses who are teaching our former students.We see no signs that students' problem-solving skills havediminished.D> We know by observation that students who completeWorkshop Physics are considerably more confortable work-ing in a laboratory setting and working with computers.This competency with the tools of exploration and analysisis often noted by visitors from other institutions who visitour classrooms during the second semester of our two-semester sequence. In the spring of 1988 a freshmancommented: "The intellectual challenge and quality ofthis course were excellent. Some days after doing anexperiment that worked out really well, I would feel as if Iaccomplished so much. Even after struggling over anexperiment for the whole period, finally getting it was agreat feeling. I received a lot more from the course thanan understanding of physics.... Just the experience withthe computers and equipment has helped me a lot. I hadstayed away from computers and been afraid to play

2 8 PHYSICS TODAY DECEMBER 1991

Sample Exercises from Workshop Physics Unit on Collisions in One DimensionWhat's Your Intuition?

Your are sleeping in your brother's room while he is awayat college. Your house is on fire, and smoke is pouringinto the partially open bedroom door. The room is somessy that you cannot get to the door. The only way toclose the door is to throw either a blob of clay or asuperball at the door—there's not time to throw both.

What Packs the Biggest Wallop—Clay or the Superball?

Assuming the clay blob and the superball have the samemass, what would you throw to close the door? The clayblob, which will stick to the door, or the superball, whichwill bounce back at almost the same velocity as it hadbefore it collided with the door? Give reasons for yourchoice. Remember, your life depends on it.

Observing the Wallop!

Let's check out your intuition by dropping a bouncy ballon a scale and then dropping a dead ball of approximate-ly the same mass on the scale from the same height. Wecan associate the maximum force on the scale with themaximum force a thrown ball can exert on a door. Wewould like to investigate how the maximum force isrelated to the change in momentum of the ball in eachcase. To do these observations you'll need the followingequipment:t> a small live ball (of mass m)\> a small dead ball or blob of clay (also of mass m)> a platform scale.As a warm-up to the observations let's consider themathematics of momentum changes for both inelasticand elastic collisions. Recall that momentum is definedas a vector quantity that has both magnitude and

direction. Mathematically, momentum change is givenby the equation

Ap = p, - p,

where p: is the momentum of the object just before acollision and pr is the momentum of the object just after acollision.

Calculating ID Momentum Changes

(a) Suppose a dead ball is dropped on a table and "sticks"to the table so that it doesn't bounce. Suppose that justbefore it bounces it has an initial momentum p, = — p\along the negative y axis where j is a unit vector pointingalong the positive y axis. What is the final momentum ofthe ball in the same vector notation?

(b) What is the change in momentum of the ball as a resultof the collision of the ball with the table? Use the sametype of i,j,k vector notation to express your answer.

Ap =

(c) Suppose a live ball is dropped on a table and"bounces" on the table in an elastic collision so that itdoesn't lose any kinetic energy. Suppose that just beforeit bounces it has an initial momentum p, = — p\ alongthe negative y axis where j is a unit vector pointing alongthe positive y axis. What is the final momentum of theball in the same vector notation? Hint: Does themomentum vector point along the + or — y axis?

(d) What is the change in the momentum of the ball as aresult of the collision of the ball with the table? Use thesame type of i,j,k vector notation to express your answer.Hint: The answer is not zero. Why?

Ap =

around with equipment before, but now I'm not and I canjust dig in."

Our attitudes survey indicates that students feel morepositive about the mastery of computer applications thanany other aspect of the Workshop Physics courses.Students view computer skills as useful in many contextsoutside of physics.\> Students in Workshop Physics rate a whole range oflearning experiences more highly than their cohorts takingtraditional courses. For example, when students areasked to assess the value of 15 learning opportunities,including attending lectures, using computers, watchingdemonstrations, solving textbook problems and doingexperiments, Workshop Physics students rate all exceptworking out text problems, reading the textbook andattending lectures more highly than do students takingintroductory physics courses at other liberal arts colleges.They express significantly more positive feelings about thevalue of observations and laboratory experiments thanstudents taking traditional courses do. This differencereflects the fact that more observational and experimentalactivities are available to Workshop Physics students andthat performance of these activities counts for a largerproportion of their grade.

Negative feedbackIn addition to positive outcomes from Workshop Physics,we have encountered several problems.

\> Some students complain that Workshop Physics coursesare too complex and demand too much time. The studentsreported that in addition to the six hours in class eachweek, they spent an average of seven hours outside of classto complete activities and assignments. On pollingstudents at 16 other colleges, we discovered that six-and-a-half hours of outside activity was the median for theircourses.

We remain undisturbed about the time demands theWorkshop Physics course makes on students; however, wedo not want the courses to be overwhelming for our lessable students. We recognize that even in WorkshopPhysics courses in which the number of topics covered hasbeen reduced, a wider range of learning abilities isrequired than in traditional courses. These includereading textbooks, solving problems, mastering computerapplications, observing, experimenting, discussing materi-al with peers, composing essays, doing mathematicalderivations, analyzing data, and writing and revisingformal laboratory reports.

We continue to struggle to eliminate less essentialmaterial and to simplify the demands on our students andourselves without losing the educational advantages wefeel we have achieved.t> A small percentage of students thoroughly dislike theactive approach. Some students state emphatically thatthey would prefer a return to the lecture approach.Although the vast majority of freshmen prefer the

PHYSICS TODAY DECEMBER, 1991 29

workshop approach, roughly half of the upper-classchemistry majors express a desire to have us return to thelecture method.

One junior pre-health student wrote on a courseevaluation: "It was discouraging to know that if I didn'tlike the format of teaching of this course .. . there was notanother being taught in a different format that I couldswitch into. . . . There needs to be more lecturing.... Wedon't need to do so much experimenting to deriveequations.... I need textbook questions with textbookequations to solve anything that's not intuitive.... I spentso much time doing out of class work that my other classessuffered and for that I am resentful."

Many of the students who think they would preferlectures resent having to "teach themselves everything."Fortunately students who have always depended onpassive learning and memorization to succeed in coursesconstitute only a small percentage of our students.Although the percentage of such students is less than thepercentage of students who used to be hostile toward ourtraditional lecture-based courses, we are attempting toachieve a better understanding of why some WorkshopPhysics students feel so negatively.O The conceptual gains of students are sometimes disap-pointing. Although we have reported with pride onselected conceptual gains, in other areas we apparentlyneed to give much more attention to appropriate curricu-lar changes. For example, we were disappointed to find

that students at the University of Oregon who completedWorkshop Physics laboratories on circuits did not dosignificantly better on a number of questions thanstudents who only enrolled in the lecture part of thecourse. We have noted among students at both Dickinsonand the University of Oregon that they have several of thesame preconceptions Lillian McDermott's physics educa-tion group at the University of Washington has discoveredand successfully overcome. One of the most interesting isthe tendency of students to visualize a battery as aconstant-current source even after they have learned howto use Ohm's law to analyze simple dc circuits mathemat-ically (see the figure on page 31). We are looking forwardto consulting with McDermott's group on restructuringour activities to take these preconceptions about circuitsinto account.

This process of experimenting with student learning,developing theories and designing new instructionalstrategies is not unlike physics research. It can berewarding and exciting as well as frustrating.t> It is difficult to learn to teach in a workshop format.The transformation of instructors from authorities whodeliver didactic lectures to designers of creative learningenvironments is extremely challenging. Instructors haveto assimilate new understandings of how different stu-dents learn and have to break themselves of the habit ofwinding into long explanations at every turn. We mustmaster the art of nurturing reflective discourse among

A A-» *» >•

7f\The accompanying figure shows a rocket coasting in space in the direction of the line. Between

A and B no outside forces act on the rocket. When it reaches point B, the rocket fires itsengines as shown and at a constant rate until it reaches a point C in space.

Rocket problem f rom a mechanicsconcepts examination developed at

Arizona State University. Students whohave taken Workshop Physics answer

this question correctly much more oftenthan students from traditional courses.

Which of the paths below will the rocket follow from B to C?(d) C (e) c

3 0 PHYSICS TODAY DECEMBER 1991

A bulb and a battery are connected as shown below.

Which is true about the current at various points in this circuit?A. The current is largest at A.B. The current is largest at B.C. The current is largest at C.D. The current is largest at D.E. The current is the same at A, B, C and D.F. The current is the same at A and B; the current is the same at C and D

and smaller than at A and B.

Sample circuit problem adapted byDavid Sokoloff at the University ofOregon from an examination used atthe University of Washington.Performance of students on suchproblems helps Workshop Physics staffimprove curricular materials.

students about physics. Since many of us model ourteaching instinctively on what our own teachers haveprovided, it is hard to break out of the traditional mold.We still have a tendency to drone on at times. A juniorwho took my Workshop Physics course last springreminded me of this: "Lectures were rarely needed....Eliminate the talks before class."

Adaptations and outlookThe hardware, software and curricular materials devel-oped for the Workshop Physics and Tools for ScientificThinking programs are available commercially,9 and over400 colleges, universities and high schools have purchasedsome or all of the materials.

Thornton, Sokoloff and Laws have given a sequence ofworkshops at both the winter and summer meetings of theAmerican Association of Physics Teachers for the pastfour years, in cooperation with Pat Cooney of MillersvilleUniversity (in Millersville, Pennsylvania). One-, two- andthree-week-long seminars have been offered to highschool, college and university instructors during the pastfour summers. Over 700 instructors have taken theworkshops.

A number of small universities, liberal arts colleges,community colleges and high schools have adoptedWorkshop Physics programs and dropped formal lecturesessions.

Large universities with high enrollments do not havethe personnel and financial resources to adopt the full-blown Workshop Physics program. They can, however,adopt some elements. Edward Adelson, David Andereckand Bruce Patton at Ohio State University, for example,have been attempting to use fewer lectures and more MBLactivities in their laboratories. And George Horton, BrianHolton and Chris Borkowski at Rutgers University haveused Tools for Scientific Thinking and Workshop Physicsactivities by having students drop into the school'sinnovative Math and Science Learning Center.

Sokoloff has adapted the calculus-based WorkshopPhysics Activity Guide units for use in algebra-basedcourses at the University of Oregon. He has beencoordinating MBL and Workshop Physics laboratoryactivities with interactive lecture demonstrations.

Our experience with implementing Workshop Physicsat Dickinson College and a number of other institutionshas been exhilarating, for it represents a blending of time-honored ideas about learning with new laboratory toolsand educational technology. The Workshop Physicsenvironment has given students unprecedented power toexamine and revise their "common sense" understandingsof science in the light of experience and to connect those

understandings in a more formal, mathematical frame-work. At the same time, while we think we have some ten-tative answers about how to improve the teaching ofintroductory physics, the Workshop Physics program isfar from perfect.

What lies in the future for Workshop Physics and forother programs that might be designed for use at theintroductory level? Although the "science of teachingphysics" had its origins over 2000 years ago, it hasundergone tremendous growth only recently, in tandemwith the emergence of new computer tools and newunderstandings of the learning process and the nature ofphysics itself. The application of the young science ofcurricular design to the physics classroom is in its infancy.We have not yet proposed laws of learning, and theoutcomes of new teaching strategies are rarely tested.

The nature of our quest as we continue to develop amore scientific approach to teaching was aptly describedby philosopher of science Karl Popper when he wroteabout science in general, "Its advance is . . . toward aninfinite yet attainable aim of ever discovering new, deeper,and more general problems, and of subjecting its evertentative answers to ever renewed and ever more rigoroustests."11

References1. S. Tobias, They're Not Dumb, They're Different: Stalking the

Second Tier, Research Corp., Tucson, Ariz. (1990).2. A. Arons, A Guide to Introductory Physics Teaching, Wiley,

New York (1990), ch. 12.3. O. Milan, Alternatives to the Traditional, Jossey-Bass, San

Francisco, Calif. (1972). D. A. Bligh, What's the Use of Lec-tures?, Penguin, Baltimore (1971).

4. P. U. Treisman, in Mathematicians and Education Reform,Conference Board of the Mathematical Sciences Issues inMathematics Education 1, N. Fisher, H. Keynes, P. Wageich,eds., Am. Math. Soc, Providence, R. I. (1990), p. 33.

5. R. Thornton, D. Sokoloff, Am. J. Phys. 58, 858 (1990). H.Brassell, J. Res. Sci. Teaching 24, 385 (1987).

6. Graphs and Tracks is distributed by Falcon Software; call(603) 764-5788.

7. For information about Coulomb VI.2, call Intellimation at(800) 346-8355.

8. Interactive Physics is distributed by Knowledge Revolution;call (800) 766-6615.

9. Vernier Software, 2920 SW 89th Street, Portland OR 97225;(503) 297-5317.

10. LA. Halloun, D. Hestenes, Am. J. Phys. 53, 1043 (1985),question 26.

11. K. R. Popper, The Logic of Scientific Discovery, Basic Books,New York (1959). ' •

PHYSICS TODAY DECEMBER 1991 31