Inquiry experiences as a lecture supplement for preservice elementary teachers andgeneral education studentsJill A. Marshall and James T. Dorward Citation: American Journal of Physics 68, S27 (2000); doi: 10.1119/1.19516 View online: http://dx.doi.org/10.1119/1.19516 View Table of Contents: http://scitation.aip.org/content/aapt/journal/ajp/68/S1?ver=pdfcov Published by the American Association of Physics Teachers Articles you may be interested in The Effect of an InquiryBased Early Field Experience on PreService Teachers’ Content Knowledge and AttitudesToward Teaching AIP Conf. Proc. 1179, 253 (2009); 10.1063/1.3266729 Modeling Aspects Of Nature Of Science To Preservice Elementary Teachers AIP Conf. Proc. 883, 69 (2007); 10.1063/1.2508693 Students’ Cognitive Conflict and Conceptual Change in a Physics by Inquiry Class AIP Conf. Proc. 818, 117 (2006); 10.1063/1.2177037 Oersted Medal Lecture 2001: “Physics Education Research—The Key to Student Learning” Am. J. Phys. 69, 1127 (2001); 10.1119/1.1389280 The challenge of science education: Teaching physics to elementary educators AIP Conf. Proc. 399, 99 (1997); 10.1063/1.53201

This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25

Inquiry experiences as a lecture supplement for preservice elementaryteachers and general education students

Jill A. Marshalla) and James T. DorwardUtah State University, Logan, Utah 84322-4415

~Received 2 December 1997; accepted 20 March 2000!

The study reported here was designed to substantiate the findings of previous research on the use ofinquiry-based laboratory activities in introductory college physics courses. The authors sought todetermine whether limited use of inquiry activities as a supplement to a traditional lecture anddemonstration curriculum would improve student achievement in introductory classes for preserviceteachers and general education students. Achievement was measured by responses to problemsdesigned to test conceptual understanding as well as overall course grades. We analyzed the effecton selected student outcome measures in a preliminary study in which some students engaged ininquiry activities and others did not, and interviewed students about their perceptions of the inquiryactivities. In the preliminary study, preservice elementary teachers and female students showedsignificantly higher achievement after engaging such activities, but only on exam questions relatingdirectly to the material covered in the exercises. In a second study we used a common exam problemto compare the performance of students who had engaged in a revised version of the inquiryactivities with the performance of students in algebra and calculus-based classes. The students whohad engaged in inquiry investigations significantly outperformed the other students. ©2000

American Association of Physics Teachers.

I. INTRODUCTION

In recent years a substantial and growing body of researchhas demonstrated that interactive engagement~IE! allowsstudents to construct and implement appropriate mental mod-els of physical phenomena better than the traditional passivelecture~or lecture with prescriptive laboratory! approach tophysics education. McDermott and Redish1 have compiledan exhaustive overview. Basic precepts of cognitive sciencesuggest the importance of IE for all physics students,2 but theneed is particularly acute in the case of preservice elemen-tary teachers, especially given the expectation that these stu-dents will go on to teach science in the same way that theyhave been taught.

Logically, one might expect a hands-on approach to bebetter for science education in the primary grades. Elemen-tary students are not likely to be engaged by a lecture ordemonstration in which they do not participate. Researchsupports this assertion. Students who regularly engage inhands-on activities have been shown to outperform studentswho do not.3 Further, students who engaged in inquiry ac-tivities ~hands-on activities oriented toward discovery learn-ing! outperformed students in programs that used laboratoryactivities only as verification exercises.4 Perhaps equally im-portant, fourth and fifth graders’ enjoyment of science hasbeen shown to increase after inquiry exercises.5 This wasparticularly true for female students, supporting a wide-spread contention6 that hands-on experiences are key to re-taining girls’ interest in science. Lack of teacher preparation,however, has been a major stumbling block in the implemen-tation of inquiry-based curricula.

Studies have shown that a lack of content knowledge willprevent teachers from using the inquiry approach with theirprimary school students, but even solid content knowledgehas been shown to be insufficient to guarantee that teacherswill adopt this approach.7–9 McDermott has made a convinc-ing case that physics classes for preservice teachers shouldbe taught by physics department faculty using an inquiry

approach.10 She describes a 20-year development effort forsuch a course, beginning with the work of Arnold Arons inThe Various Language11 and culminating in the publishedversion ofPhysics by Inquiry.12

Physics by Inquiryhas been shown to be a highly effectiveapproach to science learning for both preservice and in-service teachers.10 Thackeret al. report that elementary edu-cation majors at Ohio State who were taught usingPhysicsby Inquiry significantly outperformed other students, includ-ing those in a calculus-based course and an honors course, oncommon quantitative and conceptual exam problems.13 Leareports that elementary education majors in that samePhys-ics by Inquiryclass at Ohio State developed more positiveattitudes toward teaching physics and intended to use inquiryactivities when they went on to teach.14

The Physics by Inquiryapproach enables students to de-velop a more robust conceptual framework, but it requires acommensurately higher commitment of resources on the partof the teaching institution and of students. ThePhysics byInquiry course for preservice teachers as taught at the Uni-versity of Washington and Ohio State consists of six hours aweek10 and three hours twice a week,14 respectively, in alaboratory setting, over the course of two~presumably ten-week! quarter terms. The method requires both a lowstudent-to-instructor ratio and a laboratory setting, resultingin limited class sizes. Physics departments that typicallyteach large numbers of elementary education majors~greaterthan 200, for example! each year would be hard pressed tocommit these necessary resources.

Students are also required to commit two three-hourblocks per week for two~ten-week! terms to complete thePhysics by Inquirycurriculum, whereas most traditional lec-ture courses for elementary education majors~or generaleducation students! provide a survey of physics in only onequarter or one semester. Upper division female students, inparticular, have expressed concern over the time commit-ment for some inquiry-based programs. These students ex-

S27 S27Phys. Educ. Res., Am. J. Phys. Suppl.68 ~7!, July 2000 © 2000 American Association of Physics Teachers This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25

pressed frustration with a method that was at variance withtheir expectations of learning as straightforward fact gather-ing or memorization.15 Some researchers have argued that‘‘constructivist’’ curricula such as these may in fact fail tomeet student needs because they do not take into accountstudent expectations and goals.16

A strict inquiry approach will also result in coverage offewer curriculum topics in the same amount of time. Stu-dents discussed in Ref. 13 covered only electrical circuitsand light and optics in their one-quarter course. Many pro-ponents of science education reform are calling for just sucha trade-off of ‘‘mile-wide, inch deep’’ coverage for a morenarrowly focused, in depth curriculum, particularly in lightof the recent TIMSS results.17 Yet, the fact remains that theelementary science curriculum in many states requires teach-ers to teach many topics within science. McDermott reportsthat for preservice teachers in a small practice teaching pro-gram at the University of Washington about five~presum-ably ten-week! quarters of work in~physics! courses are re-quired before the students can prepare and teach material thathas not been previously studied, and that the situation issimilar for inservice teachers, although the time requireddoes not appear to be quite so long.9

Given these constraints~laboratory setting, low student-to-instructor ratio, reduced coverage of topics! on a coursetaught exclusively by the inquiry method, some physics edu-cators have instituted a compromise approach, supplement-ing a traditional lecture with limited exposure to IE. There isevidence that such a combined approach yields improve-ments over traditional instruction alone. In Hake’s compre-hensive survey of IE and traditional introductory physicscourses,18,19 some university courses that employed peer in-struction and concept tests20 during lectures achieved Hakefactors nearly double that of any class with traditional in-struction alone. The Hake factor is the normalized gain~ratioof actual gain to possible gain! between pre- and postcoursescores on the Force Concept Inventory~FCI!,21 and is awidely used figure of merit for the effectiveness of instruc-tion in introductory mechanics courses.

Traditional university physics lecture courses supple-mented with inquiry activities outside of lecture have alsobeen shown to yield higher Hake factors than courses withno inquiry activities. Two examples areTutorials in Intro-ductory Physics22,23 and Group Problem Solving.24 Thesecurricula were both developed in an iterative cycle of re-search, curriculum development, and instruction. TheTuto-rials approach augments a traditional lecture with one hourper week in which small groups of students work onresearch-based worksheets, replacing the traditional recita-tion or ‘‘problem session’’ in which teaching assistantsmodel problem solving skills and students usually do notactively participate.Group Problem Solvingreplaces a tradi-tional recitation session with one hour per week of IEthrough problem solving in small groups.

Both these methods have been shown to be effective. Re-dish and Steinberg recently reported a systematic study~withmatched pairs! of more than 2000 university physics studentsat eight institutions.25 These students were enrolled either intraditional lecture courses~no inquiry!, lecture coursessupplemented withTutorials, or Workshop Physics. Work-shop Physicsreplaces lecture, recitation, and laboratory withtwo three-hour sessions per week of research-based,hands-on activities and discussion, and is considered a fullexposure to the inquiry method. A report by Saul26 extended

the comparison to courses usingGroup Problem Solving.27

These studies again reported normalized percentage gainsfrom pre- to postcourse administration of the FCI, i.e., Hakefactors.Workshop Physics, which uses the inquiry approachexclusively for six hours each week, yielded the highestHake factor, 0.4160.02. The two limited approaches, how-ever, also achieved significantly higher Hake factors~0.3560.03 and 0.3460.01! with only one hour of inquiry perweek, as compared with traditional, non-inquiry courses(0.1660.03).

At the time of the study reported here, we knew of nomatched-pair study comparing a limited inquiry approach toa traditional approach for elementary education majors orstudents in non-algebra-, non-calculus-based courses de-signed to fulfill general education requirements. In this con-text, limited inquiry is an average of one hour per week orless of inquiry-based, hands-on activities as a supplement toa regular lecture curriculum.

To determine the effectiveness of limited exposure to in-quiry activities for elementary education majors and otherstudents in introductory~non-algebra, non-calculus! courses,we implemented a preliminary study of the effectiveness of~1! two-hour inquiry sessions six times during ten weeks forelementary education majors and~2! one-hour inquiry ses-sions six times during ten weeks for general education stu-dents. Following the preliminary study and revision of theinquiry activities based on formative assessment, limited in-quiry activities were implemented for both groups of stu-dents during a third ten-week term.

We give details of the implementation and a description ofthe inquiry exercises in Sec. II. In Sec. III we describe thevarious formative and summative assessments used. In Sec.IV, we present detailed results, and, in Sec. V, we presentour discussion and conclusions.

II. EXPERIMENTAL DESIGN

In order to determine the extent to which limited exposureto inquiry activities affects student mastery of concepts forelementary education majors and others in introductory~non-algebra, non-calculus! courses, we incorporated selected in-quiry activities into the curriculum of a large~140 students!lecture class at a large land grant institution. During twoconsecutive ten-week terms~winter and spring quarters,1996! we performed a preliminary study of the inquiry ac-tivities. The activities were then revised and institutionalizedinto our curriculum. During a third ten-week term~winterquarter, 1997!, we performed a comparison study, involvinga common exam problem, with algebra- and calculus-basedclasses at the same institution.

During each term, the class comprised two groups of stu-dents:~1! those who were taking the course to satisfy generaleducation science requirement and~2! those who were takingthe course to satisfy a laboratory science requirement~pri-marily elementary education majors!. There were no prereq-uisites for the introductory course in our study. Many of thegeneral education students had never had an algebra courseand had only rudimentary mathematics skills. Elementaryeducation majors were required to maintain a minimumgrade point average prior to their acceptance into that pro-gram and to take a college algebra course~although not nec-essarily prior to taking the physics survey course!.

All students attended the same 50-minute lectures~nomi-nally five times a week!. Students in the second category also

S28 S28Phys. Educ. Res., Am. J. Phys. Suppl., Vol. 68, No. 7, July 2000 J. A. Marshall and J. T. Dorward This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25

attended six two-hour laboratory sessions with mandatoryattendance. All students, both with and without a laboratorysession, took the same exams during the lecture period.

A. Preliminary study

The existence of two student populations allowed for asimple division of the class into students who would be as-signed to participate in inquiry activities and those whowould not ~the ‘‘inquiry’’ and ‘‘non-inquiry’’ groups!. Dur-ing the first term, students who did not register for the labo-ratory performed inquiry exercises and those registered forlabs did not. In the second term this assignment was re-versed, so that students with labs performed inquiry exer-cises and the students without labs did not.

The inquiry group clearly constitutes a conveniencesample under this procedure, but resources did not permitsubdivision of the two populations by inquiry and non-inquiry within each term to create a truly random sampling.Such convenience samples have been widely used in physicseducation research. For example, Hake18 and Redish andSteinberg25 both compared IE versus traditional methods us-ing data entirely obtained by convenience samples, that is,the teaching method varied on a class-by-class basis and stu-dents were not randomly assigned to one method or theother.

We sought to mitigate the shortcoming of a conveniencesample by carefully comparing students in the two terms onall measures available to us~grade point average prior totaking the class, gender, and major!. We found no significantterm-to-term variation in either subgroup. Further, studentsdid not know whether they would be assigned to inquiryactivities when they registered for the course, eliminating thepossibility of self-selection on the part of the students.

During the first term, students who were registered for alab (N547) attended lectures five days a week and com-pleted six traditional physics labs outside of lecture hours.These laboratory exercises were prescriptive in nature, listinga series of experimental steps to be performed and calling fora well-described data analysis procedure~for example, mea-sure this, graph this, etc.!. The labs were intended for prac-tice in measurement skills rather than concept developmentand were not designed with a constructivist approach inmind. Students did not have to make any predictions, drawany qualitative conclusions, or explain their thinking. A por-tion of each student’s final grade was based on laboratoryperformance as recorded in a laboratory notebook.

During the first term students who were not registered fora lab performed selected inquiry exercises~described in de-tail below! during six one-hour lecture periods. Studentsworked in self-selected groups of four to six. The laboratorystudents were excused from attendance during these six classperiods. Students performed the activities either in their seatsor on the floor in the lecture auditorium or in nearby halls oroutdoors. The instructor and an undergraduate assistant whohad taken the class before circulated among the groups pro-viding guidance in the form of suggestive questions and ap-proving students’ work at designated check points in theworksheets. A portion of the final grades for students whodid not attend laboratory~corresponding to the laboratorygrade for the students with labs! was based on worksheetscompleted during these ‘‘hands-on’’ periods.

During the second ten-week term, the situation was re-versed. Students who registered for labs performed the in-

quiry exercises during their assigned two-hour laboratory pe-riods in place of the prescriptive, measurement-oriented labs.A portion of their final grade was based on their performanceof these exercises. During this term, students who did notregister for a lab did not participate in inquiry exercises.Instead, they were required to complete extra homeworkproblems each week. Scores on homework comprised 35%of their final grade as compared to 25% for the students withlabs. The extra homework problems were both conceptualand quantitative, and were representative of problems on ex-ams.

In the second term, there were no lecture periods set asidefor inquiry activities, resulting in five extra lecture periods.~Spring term is one day shorter than winter term.! This timewas used to cover selected topics in subatomic physics.These topics were not covered in our inquiry activities dur-ing the preliminary study and were addressed only in thefinal exam. Coverage of all other topics was approximatelyequal, in terms of lectures, during the two quarters. Table Isummarizes the treatment of the two groups of students dur-ing the two terms of the preliminary study.

B. Comparison study

Following the preliminary study, the inquiry activitieswere revised based on students’ comments on class evalua-tions, problems reported by teaching assistants, and evidenceof persisting misconceptions in student work. During a thirdten-week term~winter 1997! the revised exercises wereimplemented into the curriculum for all students in our in-troductory class. Students with a lab assignment~primarilyelementary education majors! completed the inquiry exer-cises during six two-hour laboratory sessions. Students with-out a lab assignment completed a shorter version of the ex-ercises during six one-hour lecture periods; students withlabs were excused from class during these periods. One prob-lem on the final exam was based on Fig. 3 from Ref. 28,shown here as Fig. 1.

A version of this problem was also administered to acalculus-based physics class as part of a final exam, and to analgebra-based physics class as an ungraded quiz during thelast week of the term. Both were classes at the same univer-

Fig. 1. Students were asked to rank the five bulbs in the circuits shown hereby brightness, assuming that all bulbs are identical and all batteries areidentical and ideal~after Fig. 3 in Ref. 28!.

Table I. Summary of the treatment of students during the preliminary study.

Studentpopulation

Term 1 ~winter 1996!preliminary study

Term 2 ~spring 1996!preliminary study

Registered forlab ~primarilyelementary education!

Prescriptive labs„non-inquiry …

Six two-hourinquiryexercises during labperiod

Not registeredfor lab~general education!

Six one-hourinquiryexercises duringlecture period

Extra homeworkproblems„non-inquiry …

S29 S29Phys. Educ. Res., Am. J. Phys. Suppl., Vol. 68, No. 7, July 2000 J. A. Marshall and J. T. Dorward This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25

sity, and both were part of a year-long series of three ten-week terms. The common question was administered at theend of the second term, during which electric circuits werecovered in lecture and laboratory sessions. At the request ofthe instructor of the calculus-based class, we changed theproblem to read ‘‘Rank the five resistors in terms of powerdissipation’’ instead of ‘‘Rank the five bulbs in order ofbrightness,’’ and redrew the diagram to show resistor sym-bols rather than light bulbs. The algebra- and calculus-basedstudents had studied the behavior of generic resistors~ratherthan actual light bulbs! in their laboratories and the instructorwas concerned that students might not be able to expresstheir knowledge of currents in terms of light bulb brightness.The issue of power dissipation had been discussed in bothclasses.

A shortcoming of our design is that we did not include acommon quantitative problem for comparison, as per Ref.13. As our final exams are comprehensive, covering manyadditional topics, we felt that two problems addressing resis-tive circuits would be excessive. We had given a quantitativeversion of this problem to students in our preliminary studyand found that both those who had experienced the inquiryexercises, and those who had not, had higher mean scores onthe quantitative version than on the qualitative versionshown in Fig. 1.

Table II summarizes the various laboratory experiences ofall students in the comparison study.

C. The inquiry exercises

Each exercise was designed to address certain misconcep-tions in a particular subject area. These misconceptions hadbeen identified, partly from experiences~informal discus-sions and test questions responses! with previous groups ofstudents, but also from the literature.28,29 The ‘‘elicit, con-front, resolve’’ educational paradigm30,31 was used. Thisstrategy first requires that students make predictions or pro-vide explanations about a physical system to be studied. Stu-dents then investigate the system using a simple physicalmodel. They follow a set of guideline questions and activi-ties designed to expose misconceptions and develop an ap-propriate conceptual model that the students can then use topredict the results of changes to the system and the behaviorof similar systems. Finally, students are asked to describe orexplain the behavior of the system in their words, that is, toexplicitly express their mental model.

In this study, the elicit phase of the program did not in-clude a formal pretest. Rather, students were asked ques-tions, either in the introductory part of the exercises or aspart of the lecture portion of the class. For example, studentswere asked to predict the motion of a ball leaving a circularchannel prior to the activities on circular motion~question 6

from the Force Concept Inventory21!. Students then collectedthe necessary equipment and worked through a short work-sheet in self-selected groups of two to six people. The exer-cises covered eight topics: constant velocity and acceleratedmotion in one dimension, circular motion, conservation ofenergy, heat transfer, density and the buoyant force, light~reflection!, standing waves, and resistive circuits.

For some topics, namely resistive circuits and one-dimensional motion, the activities were shortened versions ofthe Physics by Inquiryactivities developed for elementaryeducation majors by McDermottet al.12 The circuit activitiesfollowed the outline of McDermott, but time constraints didnot allow for the entire McDermott ‘‘Batteries and Bulbs’’unit to be implemented. Students performed the well-knownexercise of lighting a small bulb with a battery and one pieceof wire and then progressed to comparing the brightness ofbulbs and the effect of removing or adding bulbs to seriesand parallel resistive circuits. The concepts of conductorsand insulators and current flow had been previously intro-duced in lecture and were not developed in inquiry exercises.Students did not investigate more complicated circuits suchas resistors in series with parallel elements or the use ofvoltmeters and ammeters. Similarly, the activities on one-dimensional motion were abbreviated versions of those inPhysics by Inquiry.

Other activities~conservation of energy and circular mo-tion! had been developed as part of a workshop on Amuse-ment Park Physics.32 The conservation of energy activitiesused a low-friction model roller coaster, made of BBs andplastic tubing. Students investigated the concepts of a changein gravitational potential energy versus an absolute value ofgravitational potential energy. McDermott and Shaffer hadidentified failure to distinguish between potential and poten-tial difference as a difficulty commonly experienced by stu-dents in introductory electricity,28 and we had observed thatour students experienced similar difficulties with gravita-tional potential. Students also investigated the lack of depen-dence on intermediate path of energy conservation from finalto initial state. The circular motion activities investigated theidea of a centrifugal versus centripetal force and the relationbetween velocity and centripetal force using simple modelsof the channel described in question 6 from the FCI21 andanother rotating system.

Finally, some activities~reflection, heat transfer, andstanding waves! were developed especially for these classes,but were based in part on suggestions by Arons.33

The students in this study were not given formal posttests,but were instructed to consult with instructors at specificpoints in the exercises. Instructors reviewed the students

Table II. Summary of the laboratory experiences of the four groups of students during the comparison study.

Our studentsregistered forlab ~primarilyelementary education!(N570)

Our students notregistered for lab~General Education!(N546)

Students in traditionalalgebra-based course (N559)

Students in calculusbased course (N5150)

Six two-hourinquiryexercises duringlab period

Six one-hourinquiryexercises duringlecture period

Six nominally three-hourprescriptive labs

Six nominally three-hourcomputer-based labs

S30 S30Phys. Educ. Res., Am. J. Phys. Suppl., Vol. 68, No. 7, July 2000 J. A. Marshall and J. T. Dorward This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25

work and posed additional questions. Conceptual and quan-titative questions on the material covered in the exerciseswere included on exams.

In every case, an attempt was made to cast the activities interms of an interesting theme. For example, the activities onone-dimensional motion wereMarble Races, conservation ofenergy wasBB Roller Coasters, and heat transfer wasIceCream Sundaes. This was done partly to remove the stigmaof a formal physics laboratory experiment as being perhapsboring, difficult to understand, and unlikely to produce theexpected result.

We also hoped that the preservice teachers among our stu-dents would view these activities in the light of preparationfor activities with which they might engage their own stu-dents in the future. While these activities are not appropriatefor young children as they stand, many actually have theirroots in activities specifically developed for primary stu-dents. For example, our units on density and the buoyantforce and resistive circuits correspond toClay BoatsandBat-teries and Bulbs, respectively, from Elementary ScienceStudy~ESS!.34 ESS was developed under the sponsorship ofthe National Science Foundation as a model inquiry curricu-lum for elementary school. Our conservation of energy ac-tivity ~BB Coasters! has been used in a modified form at themiddle school level. Inquiry activities, although of a lessguided nature, have been generally shown to be appropriatefor elementary audiences.5

D. Constraints and limitations

Constraints to our research design limit the generalizabil-ity of our results. Our use of a convenience sample~i.e.,without random sampling! raises the distinct possibility thatinquiry- and non-inquiry-based groups were not representa-tive of their respective populations. We mitigated this limi-tation to some degree by randomly assigning each class tofollow an inquiry-based curriculum or a traditional curricu-lum. With random assignment, a test of statistical signifi-cance can address whether groups under analysis can be re-garded as samples from the same population.35 When weinvestigated the effect of the inquiry exercises on subgroups,such as female students or elementary education majors, oursample sizes became fairly small. For this reason, we reporteffect size metrics to assess practical significance.

III. ASSESSMENT

We used multiple methods to assess the effectiveness ofthe two limited inquiry approaches in our experiment. Ref-erence 36 suggests that interpretations from triangulation ofinformation collected using multiple methods on differentsamples at different times can be more credible than conclu-sions based upon one-dimensional data collection tech-niques. In the preliminary study, a number of student out-comes were compared for students who experienced inquiryactivities and those who did not.

The outcome measures we selected for analysis includedfinal exam grades, course grades, and scores on midtermexam questions related to topics covered in the inquiry exer-cises. We chose these outcomes measures, as opposed tostandardized tests of conceptual understanding such as theFCI, first because widely accepted standardized questions arenot available for all the topics we covered in the course.Second, we believe that these measures are of primary value

to our students. Unfortunately, course grades may be moreimportant to our students in their careers than a more robustunderstanding of physics.

That being said, with proper assessment instruments, examand course grades should reflect student understanding. Wechose to look at both comprehensive measures, such as finalexam and course grades, as well as scores on problems di-rectly related to the topics we were able to cover in theinquiry activities. This allowed us to investigate the possibil-ity of a secondary effect, in which student performance ontopics not covered in the inquiry activities might somehowbe affected, possibly through heightened interest in or com-mitment to the course.

All exams consisted of both qualitative and quantitativeproblems in a variety of formats. Students were required toset up and solve numerical problems, write out explanationsand predictions, and select from multiple choice responses.Some of the problems used on exams were taken directlyfrom the physics education research literature~for example,problems cited in Ref. 13 and problems from the FCI! aswell as problems of one of the author’s~JAM! invention.While the tests were not identical from term to term, prob-lems were comparable from one term to the next. For ex-ample, one problem presented a graph of an object’s velocityversus time and asked students to identify when the objectwas stopped, when and how much it was accelerating, howmuch distance it covered in the time period, etc. A secondpart of the problem asked students to choose a description ofa motion that might correspond to the one shown in thegraph. The problem was the same in both terms except thatthe details of the graph were changed.

The final exam was cumulative. Course grades were basedon homework and exam scores, and either inquiry work-sheets, lab reports from prescriptive labs, or extra homeworkdepending on the class as discussed earlier. Roughly one-third of the material covered on exams had been the subjectof an inquiry exercise. The rest had been covered in lectureonly. A separate tally was kept of scores on the midtermexam questions that related directly to the inquiry exercises.Because of time constraints in returning exams, we did notevaluate final exam questions on topics covered in the in-quiry exercises separately.

At the beginning and end of courses in the preliminarystudy we conducted one-hour focus group interviews of vol-unteer students, both those who were registered for labs andthose who were not. These interviews were designed to elicitstudent attitudes toward the inquiry activities and toward sci-ence and scientists in general.

Table III. Multivariate analysis of variance summary for majors by instruc-tion method~inquiry versus non-inquiry!.

Multivariate ANOVA Univariate ANOVAQuestions ontopics covered

in inquiryexercisesSource F Final exam Course grade

Student major 25.68a F 0.05 0.27 0.62Instructionmethod

47.11a F 0.28 1.78 7.75a

Student majorby instructionmethod

9.57a F 0.08 1.18 6.16a

ap,0.05.

S31 S31Phys. Educ. Res., Am. J. Phys. Suppl., Vol. 68, No. 7, July 2000 J. A. Marshall and J. T. Dorward This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25

In the comparison study, the performance of students whoparticipated in limited inquiry activities~of both one- andtwo-hour duration! was compared with that of students intraditional algebra- and calculus-based lecture courses withprescriptive labs.

IV. RESULTS

A. Preliminary study

To determine the effect of the inquiry exercises in thepreliminary study, we performed a multivariate analysis ofvariance, MANOVA, on outcome measures from all fourclasses combined~171 students!.37 This technique allows forcontrolling for grade point average and gender. Table IIIpresents the results of the MANOVA for three outcome mea-sures~final exam score, course grade, and scores on examquestions related to topics covered in the inquiry exercises!as a function of student major, instruction method~inquiryversus non-inquiry!, and a combination of the two.

The MANOVA results in anF statistic~somewhat analo-gous to the more familiar chi-square statistic which is usedwhen data are in the form of frequency counts!. A significantF statistic indicates that the means of the three~or more!samples in the MANOVA are significantly different~see p.355 of Ref. 37 for further details!. The F statistics shownhere suggest that there are differences in outcomes betweenstudents who are elementary education majors and those whoare general education majors, and also between those whoexperienced inquiry activities and those who did not. In bothcases, differences were significant only on exam questionsrelated to material directly investigated in the inquiry exer-cises.

In order to determine which factors were responsible forthese differences, we performed a series ofT-tests~the sta-tistical test of choice when small samples are studied37!.Table IV shows the results of aT-test comparing the scores~mean percent correct! on exam questions related to topicscovered in the inquiry exercises for elementary educationmajors who participated in inquiry investigations and thosewho experienced a prescriptive lab. The inquiry students out-scored those doing prescriptive labs by seven percentagepoints.

The relatively high effect size, a metric measuring themagnitude of results that is independent of sample size andscale of measurement,35 suggests this result has practical, aswell as statistical, significance.

In contrast, there was no observed difference in the per-formance of general education majors who experienced onehour each week of inquiry-based laboratory exercises andgeneral education majors who experienced extra homeworkproblems on this same outcome measure~see Table V!.

Student’s performance on exam questions dealing withtopics investigated in our inquiry activities did not seem tobe determined by their majors. AT-test comparing this out-come measure by major bore out this result~see Table VI!.There was a difference in the amount of time that elementaryeducation majors and other students in our study spent en-gaging in inquiry activities, two hours every other week forelementary education majors compared with one hour for theothers. Therefore, these results also indicate that the amountof time spent in the inquiry activities was not the predomi-nant factor in whether these activities effected a change.

This led us to suspect that gender, which is related tomajor in that elementary education majors in this study weremore than 90% female, might have played a major role. Asmentioned earlier, prior research has suggested that youngerfemale students may benefit from inquiry-based laboratorystrategies.5 Our study provided support for the conjecturethat women at the college level have higher achievement onsome measures when they participate in inquiry exercises.Table VII reports the results of a MANOVA for female stu-dents by major~elementary education and others!, instruc-tion method~inquiry and non-inquiry!, and a combination ofthe two.

Analysis of data from female students revealed that thosewho experienced the inquiry-based laboratory exercises alsohad higher achievement on exam questions on inquiry topicswhen compared with women experiencing the non-inquirylaboratory exercises~see Table VIII!. In comparison, differ-ences between means on all dependent measures for the cor-responding two groups of male students were not significant.

Table IV. T-test comparisons of percent correct on exam questions on in-quiry exercise topics for elementary education majors experiencing inquiryand non-inquiry activities.t523.51,p50.001,ES50.68.

Variable No. of cases Mean percent s.d. SE of mean

Inquiry 33 91.24 7.54 1.31Non-inquiry 47 84.23 10.30 1.50

Table V. T-test comparisons of percent correct on exam questions on in-quiry exercise topics for general education majors experiencing inquiry andnon-inquiry activities.t520.25,p50.804,ES50.055.

Variable No. of cases Mean percent s.d. SE of mean

Inquiry 48 88.96 9.74 1.41Non-inquiry 42 88.46 9.12 1.41

Table VI. T-test comparisons of percent correct on exam questions on in-quiry exercise topics for education and general education majors experienc-ing inquiry-based activities.t521.19,p50.238,ES50.23.

Variable No. of cases Mean percent s.d. SE of mean

General Education 48 88.96 9.74 1.41Education Majors 33 91.24 7.54 1.31

Table VII. Multivariate analysis of variance summary for females by stu-dent major and instruction method (n5104).

Multivariate ANOVA Univariate ANOVA Questions ontopics covered

in inquiryexercisesSource F

Finalexam

Cumulativegrade

Student major 31.36a F 0.01 0.73 0.99Instructionmethod

53.10a F 0.40 1.61 1.36

Student majorby instructionmethod

13.31a F 0.04 0.54 6.92a

ap,0.05.

S32 S32Phys. Educ. Res., Am. J. Phys. Suppl., Vol. 68, No. 7, July 2000 J. A. Marshall and J. T. Dorward This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25

When a T-test of the same outcome measure was per-formed for female students by major~elementary educationor general education!, there was not a statistically significantdifference~see Table IX!. This again confirms that gendermay play a more important role than a student’s chosen ma-jor in whether that student will benefit from inquiry exer-cises.

B. Comparison study

Following Ref. 13, we administered a common problem toall students in our class, as well as those in the algebra- andcalculus-based classes at Utah State. The problem, shownhere as Fig. 1 and taken from Fig. 3 in Ref. 28, asks studentsto rank the order of brightness of bulbs in three differentcircuits. All bulbs are identical and all batteries are ideal.One circuit has a single bulb, one has two bulbs in series,and one has two bulbs in parallel. As a correct answer, weexpected students to state that 15455.253.

We expected a complete explanation to indicate~1! thatthe current through~and therefore the brightness of! any bulbis independent of the existence of parallel branches in thecircuit so that bulb 1 is in an identical situation to bulbs 4and 5 and therefore would be equally bright.~2! Studentsneeded to mention that the same current that flows throughbulb 2 must flow on through 3; therefore these bulbs must beequally bright.~3! Students were required to state that thecurrent through 2 and 3 would see higher resistance than thecurrent through 1, 4, or 5, and therefore the current through 2and 3 would be less, and bulbs 2 and 3 would be dimmerthan the others.

Comparisons of responses from our students, who experi-enced inquiry-based activities, and from the students in thealgebra- and calculus-based classes, who did not, are shownin Table X. Only 9% of the non-inquiry students~3% of thealgebra students and 11% of the calculus students! gave acorrect answer with an adequate explanation. Reference 28reports that typically 15% of students in a standard calculus-based course are able to produce a completely correct re-sponse to this question. Our algebra-based students may haveproduced a particularly low number of correct responses asthe question was given to them as a nongraded quiz. It ispossible that some students who might have been able toprovide an explanation simply did not take the time to do so.

Of our ‘‘inquiry’’ students, 26% were able to give a com-pletely correct response to the question with an adequate

explanation. Reference 23 reports that students who had ex-periencedPhysics by Inquirytutorials were able to give com-pletely correct responses 45% of the time to a similar prob-lem. This difference may be due to a difference in time spenton the subject of resistive circuits. Our students typicallyspent only one hour on this specific topic. Students who per-formed the exercises during a two-hour lab period also in-vestigated batteries and Ohm’s law during the same session.In the Physics by Inquiryapproach resistive circuits are partof a much more comprehensive and carefully orchestratedseries of steps toward building a mental model of electriccircuit, and therefore it is reasonable to expect that approachto yield a better result. Our result could in fact be a continu-ation of a trend toward better results from more extensive IEhinted at in Refs. 25 and 26.

We saw many of the same misconceptions in the explana-tions given for wrong answers that were reported in Ref. 28.In particular many students indicated that the current in bulb3 would be less than the current in bulb 2 because bulb 2would have ‘‘used up some of the current.’’ Likewise, manystudents indicated that the current through bulbs 4 and 5~while equal to each other! would be less than the currentthrough bulb 1, indicating a belief that the battery is a fixedcurrent source. These responses were much more commonamong the non-inquiry students~algebra- and calculus-basedclasses! than among our students who had experienced in-quiry exercises with batteries and bulbs.

Among the students who had experienced the inquiry ex-ercises there was no evidence of confusion between total~equivalent! resistance of the entire circuit and the resistanceof individual bulbs. Reference 28 had reported that somestudents expect bulbs 4 and 5 to be brighter than bulb 1because ‘‘a parallel circuit has lower resistance.’’ Our stu-dents were taught to find the total current in a circuit byadding the currents through individual branches and then tofind an ‘‘equivalent resistance’’ using Ohm’s law. They werenot taught a formula for the equivalent resistance of resistiveelements in parallel. Some students in the algebra- andcalculus-based classes used such a formula as justificationthat bulbs 4 and 5 would be brighter.

A small but disturbing number of our students did statethat bulbs 2 and 3 differed in brightness or that bulbs 4 and5 were dimmer than bulb 1 because they ‘‘saw it that way inthe lab.’’ It is possible that they did see it that way during theinquiry exercises. Irregularities in bulbs sometimes result indiffering brightness for bulbs in series. An inadequatelycharged nonideal battery can exceed its maximum currentlimit when attached to two bulbs in parallel, and thereforethe two bulbs in parallel may also be dimmer than the onebulb by itself in an actual demonstration.

Shaffer and McDermott argue that these variations fromthe ideal situation ‘‘can be exploited to help deepen concep-tual understanding.’’23 This is true in theory. Our teachingassistants should have been able to catch the problem at thepoint where students were required to have their work

Table VIII. T-test comparisons of percent correct on exam questions oninquiry exercise topics for female students in inquiry-based and non-inquiry-based classes.t522.61,p50.01,ES50.45.

Variable No. of cases Mean percent s.d. SE of mean

Non-inquiry 56 84.90 11.13 1.49Inquiry 58 89.93 8.93 1.17

Table IX. Comparisons of percent correct on exam questions on inquiryexercise topics for female education and general education majors experi-encing inquiry-based activities.t521.86,p50.069,ES50.42.

Variable No. of cases Mean percent s.d. SE of mean

General education 29 87.7 10.02 1.86Elementary education 29 91.96 7.24 1.35

Table X. Chi-square test comparison between groups on resistive circuitquestion. Chi-square559.36,p,0.05.

Variable No. of casesPercent with completely

correct response

Non-inquiry 209 9.1Inquiry based 116 25.9

S33 S33Phys. Educ. Res., Am. J. Phys. Suppl., Vol. 68, No. 7, July 2000 J. A. Marshall and J. T. Dorward This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25

checked and suggest ways in which students might investi-gate the true cause of the variations in brightness. The per-sistence of these misconceptions could have one of two ex-planations. Either our assistants were less than completelydiligent in checking students’ work~which may well havebeen true in the general education class where the student toinstructor ratio was particularly high!, or these misconcep-tions were so strongly rooted that students misrememberedwhat they had seen during the exercises. To avoid this prob-lem, we now require our TAs to check the bulbs systemati-cally for irregularities and the batteries for proper voltageimmediately before each implementation of the batteries andbulbs activities.

C. Interviews

Quantitative assessment data suggested that limited in-quiry approach did contribute to an increase in student un-derstanding of the topics covered in the inquiry activities, atleast for some students. Improved performance on severaloutcome measures was particularly evident among femalestudents and elementary education majors. Anecdotal com-ments from the focus-group interviews provided additionalinsights on these findings.

Students who volunteered to participate in the focus-groupinterviews agreed to attend a pre- and postcourse session inplace of one homework assignment. All 22 interview partici-pants experienced inquiry-based activities. Questions askedby an external evaluator during the four pre- and four post-treatment interviews were designed to assist interpretationand validation of quantitative data.38,39 Responses from theinterviews were analyzed and codified.

An emergent theme from discussions about the hands-oncomponent of the introductory physics class was value asso-ciated with concept confirmation. Concrete activities, regard-less of whether they were laboratory experiences or class-room demonstrations, were perceived as beneficial.

‘‘I love to actually be able to work with the material andactually see how it comes out. Also I think it’s not just work-ing with it, it’s also the examples she gives, like seeing isbelieving.’’ ~female general education student!

Some students indicated that the inquiry activities weremore beneficial when the concepts under investigation weredirectly linked to previous lecture topics. A previous studyalso found that open-ended inquiry sessions were most ben-eficial when they followed combined lecture and demonstra-tion sessions.40

‘‘I felt sorry for those people that had hands-on or theirlaboratory at the beginning of the week when we didn’tcover the material in class until later...I have to see it onpaper and see it in the laboratory. It’s a reinforcing experi-ence for me.’’~male general education major!

There was some indication from the interviews that stu-dents in classes experiencing the longer inquiry-based exer-cises in the laboratory valued those experiences more thanstudents who participated in abbreviated activities during thelecture hour. However, findings from the interviews sug-gested that this value was attributed more to ‘‘increasedtime’’ and ‘‘being in a lab setting’’ than a fundamental dif-ference in instructional strategies.

‘‘I think if I would have been in the lab environment~rather than the classroom or hallway!, it would have been alot easier. Because when you’re in a lab setting you’re al-ways constantly doing experiments.’’~female general educa-tion major, inquiry-based exercises in the lecture room!

A second emergent theme from interview and observationinformation was the value associated with inquiry-basedstrategies of concept acquisition. Students approached theintroductory physics classes with common expectations ofwhat goes on in a science classroom. These perceptions wereillustrated when students were asked to describe differencesbetween scientists and science teachers:

‘‘The scientist is actually involved. They have a lot moreknowledge of the deeper stuff, the more scientific things thatyou wouldn’t explain to a student. A science teacher has tocover a lot of material in a short period of time, so I wouldthink that they would have more of a basic knowledge. Oneknows a lot about a little and the other knows a little about alot.’’ ~female general education student, precourse!

‘‘Science teachers want students to measure somethingthat is going to be a certain weight...a scientist working in alab can try and discover new things.’’~male general educa-tion major, precourse!

While students at the beginning of each quarter were veryclear that science teaching consisted of transmission ofknown facts and prescriptive laboratory exercises, commentsat end of the quarters were less definitive. This narrowing ofa perceived gap between doing science and learning aboutscience was attributed to several variables.

‘‘In an idealistic sense, I think of myself as more of ascientist in her class. I think her being a woman teacheraffected me a lot; and because it~the lab! was more realis-tic.’’ ~female education major!

It was evident that some students participating in theinquiry-based exercises had begun to challenge their earlierperceptions about the nature of science teaching and learn-ing. The more realistic nature of inquiry-based approachesand use of a variety of instructional mediums to complementthe inquiry-based exercises may have contributed to thesechanges.

The value students placed on concept discovery appears tohave influenced their acquisition of the concepts covered inthe inquiry exercises. Some students who experienced theinquiry-based exercises were reluctant to challenge their per-ceptions of a distinction between learning science and doingscience. One noticeable characteristic of inquiry-based exer-cises that reinforced this distinction was the absence of timedevoted to detailed outline of procedure. For several studentswith strong prescriptive expectations of science labs, a per-ceived lack of direction resulted in frustration and with-drawal, rather than challenge and active involvement inproblem solving.

On the other hand, students who valued the concept dis-covery aspect appeared to appreciate the exercises more.Comments within this theme were characterized by suchwords as ‘‘dynamic,’’ ‘‘exciting,’’ and ‘‘alive,’’ and labgroups were typically actively working and communicating.

‘‘You could see the field and how it’s progressing. It getsyou excited. It makes the field come alive. Whereas mychemistry class is just the same old, same old. It’s like adrill.’’ ~female, education major!

Student comments about the inquiry exercises on courseevaluations were very nearly universally positive, providinganecdotal evidence that these activities improved student at-titudes toward the course. There were infrequent commentsindicating that students preferred to have lecture coverage ofa topic prior to their inquiry investigations. Course evalua-

S34 S34Phys. Educ. Res., Am. J. Phys. Suppl., Vol. 68, No. 7, July 2000 J. A. Marshall and J. T. Dorward This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25

tion scores and instructor evaluation scores were both higherduring quarters when inquiry exercises were included in thecurriculum.

One interpretation of comments related to the value ofconcept discovery is that these students view the content asmore dynamic and themselves as more active participants inthe learning process. In this scenario, concept discovery isclosely linked to responsibility for learning. For students in-volved in this project, the aspect of discovery in the inquiry-based exercises was one motivating factor that contributed toacceptance of responsibility in the learning process.

V. CONCLUSIONS AND IMPLICATIONS

Results from this study indicate that implementation oflimited ~one or two hour every other week for a ten-weekterm! inquiry-based laboratory exercises increases under-standing of concepts treated in the exercises for some stu-dents. In particular, female students and female preserviceteachers in an introductory class for elementary educationmajors and general education students showed increased un-derstanding compared with their peers who had received noinquiry training. Possible reasons for these observed differ-ences include the validation or confirmation value ofhands-on activities, and value associated with alternativeways of acquiring knowledge in science, particularly discov-ery.

It should be noted, however, that differences in the inquiryand non-inquiry groups were significant only on assessmentmeasures that dealt directly with concepts investigated in theinquiry exercises. We saw no ‘‘cascading effect’’ throughwhich student performance on topics not directly covered inthe inquiry exercises was enhanced. This result suggests that,for optimum preparation, preservice teachers should be ex-posed to inquiry activities on as many topics as possible,especially on topics which they will be required to teach aspart of a state elementary science curriculum. This needmust, of course, be balanced by the need for in-depth andpossibly repeated exposure.

Our study was not able to distinguish between studentsexperiencing one hour of inquiry exercises as a replacementfor a lecture and extra homework once every two weeks andstudents experiencing two hours of inquiry activities as areplacement for traditional prescriptive laboratory activitiesonce every two weeks. Student comments did provide someanecdotal evidence that students perceived the longer expo-sure to be more beneficial. Further study of the effectivenessof inquiry activities versus the length of exposure time isneeded.

Gender differences also appeared to play a role in ourstudy. In general, the physics education research literaturehas not addressed gender as a variable. There are some no-table exceptions. As mentioned earlier, Laws has reportedthat some female students may be particularly resentful ofthe time commitment required for an inquiry approach.Brown et al. found gender differences in student response toa task with batteries and bulbs.41 We found the effect of ourinquiry activities to be statistically significant on female stu-dents but not on male students. In contrast, one study of highschool students in the Netherlands found that girls did notperform as well as boys under the active inquiry approachbut did under vicarious inquiry.42 Clearly, this issue meritsadditional study.

Recognizing risks inherent in interpretation of findings

from education research, we suggest that physics educatorswho teach introductory classes for preservice elementaryteachers consider the importance of including inquiry-basedexercises into their courses, even if it is only possible to doso on a limited basis. Activities such as those described ear-lier may be of particular value to the largely female popula-tion of prospective elementary teachers. Efforts to increasefuture teachers’ conceptual understanding and attitudes to-ward science are of particular importance in that they mayresult in improved elementary science instruction, thus af-fecting large numbers of future science learners.

ACKNOWLEDGMENTS

The research was conducted as part of a comprehensiveevaluation of a curriculum development effort supported bythe National Science Foundation~Grant no. DUE-9555017!.The authors recognize the contribution of Trevor Willey intallying scores on the common exam problem and thankDavid Peak and Balraj Menon for evaluating the selection oftest problems relevant to topics covered in the hands-on ex-ercises.

a!Electronic mail: marshall@cc.usu.edu1Lillian C. McDermott and Edward F. Redish, ‘‘Resource letter PER-1:Physics education research,’’ Am. J. Phys.67 ~9!, 755–767~1999!.

2Edward R. Redish, ‘‘Implications of cognitive studies for teaching phys-ics,’’ Am. J. Phys.62 ~9!, 796–803~1994!.

3Patricia M. Stohr-Hunt, ‘‘An analysis of frequency of hands-on experienceand science achievement,’’ J. Res. Sci. Teach.33 ~1!, 101–109~1996!.

4James A. Shymansky, Larry V. Hedges, and George Woodworth, ‘‘A re-assessment of the effects of the inquiry-based science curricula of the 60’son student performance,’’ J. Res. Sci. Teach.27 ~2!, 127–144~1990!. Ona mean composite and individual measures of performance~includingachievement, perceptions, and process skills!, students in inquiry-basedcurricula outperformed other students at all grade levels, with the differ-ence being significant at all grade levels except 4–6.

5J. B. Kahle and A. Damnjanovic, ‘‘The effect of inquiry activities onelementary students’ enjoyment, ease, and confidence in doing science: ananalysis by sex and race,’’ J. Women Minorities Sci. Eng.1, 17–28~1994!.

6Ameican Association of University Women,How Schools ShortchangeGirls ~American Association of University Women, Washington DC,1992!, Executive Summary, p. 4.

7Daniel C. Dobey and Larry E. Schafer, ‘‘The effects of knowledge onelementary science inquiry teaching,’’ Sci. Educ.68 ~1!, 39–51~1984!.

8J. E. Pedersen and D. W. McCurdy, ‘‘The effects of hands-on, minds-onteaching experiences on attitudes of preservice elementary teachers,’’ Sci.Educ.76 ~2!, 141–146~1992!.

9Lillian C. McDermott, ‘‘Teacher education and the implementation of el-ementary science curricula,’’ Am. J. Phys.44 ~5!, 434–441~1976!.

10L. C. McDermott, ‘‘A perspective on teacher preparation in physics andother sciences: The need for special science courses for teachers,’’ Am. J.Phys.58 ~8!, 734–742~1990!.

11A. B. Arons,The Various Language~Oxford U.P., New York, 1977!.12Lillian C. McDermott,Physics by Inquiry~Wiley, New York, 1996!, Vol.

1, pp. 3–42; Vol. 2, pp. 383–418 and 639–669.13B. Thacker, E. Kim, K. Trefz, and S. M. Lea, ‘‘Comparing problem solv-

ing performance of physics students in inquiry-based and traditional intro-ductory physics courses,’’ Am. J. Phys.62 ~7!, 627–633~1994!.

14S. M. Lea, ‘‘Adapting a research-based curriculum to disparate teachingenvironments: how to succeed in transporting inquiry-based physics in-struction from university to university by really trying,’’ J. Coll. Sci.Teaching Feb 1993.

15P. W. Laws, P. J. Rosborough, and F. J. Poodry, ‘‘Women’s response to anactivity-based introductory physics program,’’ Am. J. Phys.67 ~7!, S32–S37 ~1999!.

16H. Wildy and J. Wallace, ‘‘Understanding teaching or teaching for under-standing: alternative frameworks for science classrooms,’’ J. Res. Sci.Teach.32 ~2!, 143–156~1995!.

17For example, William H. Schmidt and Curtis C. McKnight, ‘‘What can we

S35 S35Phys. Educ. Res., Am. J. Phys. Suppl., Vol. 68, No. 7, July 2000 J. A. Marshall and J. T. Dorward This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25

really learn from TIMSS?’’ Science282, 1830–1831~1998!. The ThirdInternational Math and Science Study~TIMSS! reviewed science educa-tion at three different grade levels in over 40 countries. The completereports are available at www.csteep.bc.edu/TIMSS.

18Richard R. Hake, ‘‘Interactive-engagement versus traditional methods: Asix-thousand student survey of mechanics test data for introductory phys-ics courses,’’ Am. J. Phys.66 ~1!, 64–74~1998!.

19R. R. Hake, ‘‘Interactive engagement versus traditional methods: A six-thousand student survey of mechanics test data for introductory physicscourses,’’ Am. J. Phys.66 ~1!, 64–74~1998!.

20E. Mazur, Peer Instruction: A User’s Manual~Prentice–Hall, UpperSaddle River, NJ, 1997!.

21D. Hestenes, M. Wells, and G. Swackhammer, ‘‘Force concept inven-tory,’’ Phys. Teach.30 ~3!, 141–158~1992!; revised by Ibrahim Halloun,Richard Hake, and Eugene Mosca, http://modeling.la.asu.edu/modeling/R&E/Research.html~1995!.

22Lillian C. McDermott and Peter S. Shaffer,Tutorials in Introductory Phys-ics, preliminary ed.~Prentice–Hall, Upper Saddle River, NJ, 1998!, pp.93–107.

23Peter S. Shaffer and Lillian C. McDermott, ‘‘Research as a guide forcurriculum development: An example from introductory electricity. PartII: Design of instructional strategies,’’ Am. J. Phys.60 ~11!, 1003–1013~1992!.

24P. Heller, T. Foster, and K. Heller, AIP Conf. Proc.399, 913–934~1997!.25Edward F. Redish and Richard N. Steinberg, ‘‘Teaching Physics: Figuring

Out What Works,’’ Phys. Today52 ~1!, 24–30~1999!.26Jeffery M. Saul and Edward F. Redish, ‘‘A comparison of pre- and post-

FCI results for innovative and traditional introductory calculus-basedphysics classes,’’ AAPT Announcer28, 80 ~July 1998!.

27P. Laws,Workshop Physics Activity Guide~Wiley, New York, 1997!.28Lillian C. McDermott and Peter S. Shaffer, ‘‘Research as a guide for

curriculum development: An example from introductory electricity. Part I:

Investigation of student understanding,’’ Am. J. Phys.60 ~11!, 1003–1013~1992!.

29Ibrahim Abou Halloun and David Hestenes, ‘‘The initial knowledge stateof college physics students,’’ Am. J. Phys.53 ~11!, 1043–1065~1985!.

30Lillian C. McDermott, ‘‘Guest comment: How we teach and how studentslearn—A mismatch?’’ Am. J. Phys.61 ~4!, 295–298~1993!.

31Lillian Christie McDermott, ‘‘Millikan Lecture 1990: What we teach andwhat is learned—Closing the gap,’’ Am. J. Phys.59 ~4!, 301–315~1991!.

32Paula Scott and Marion Dennison, ‘‘Marble Coasters,’’ inAmusementPark Physics Curriculum Manual, 2nd ed., edited by J. R. Dennison~UtahState University Physics Department, Logan, UT, 1996!, pp. F17–F20.

33Arnold B. Arons,A Guide to Introductory Physics Teaching~Wiley, NewYork, 1990!, Chap. 5, pp. 118–121; Chap. 9, pp. 202–208.

34Elementary Science Study~Education Development Center, Newton, MA,1968!.

35J. P. Shaver, ‘‘What statistical significance testing is, and what it is not,’’J. Exp. Educ.61 ~4!, 293–316~1993!.

36M. D. Shipman,The Limitations of Social Research, 3rd ed.~Longman,London, 1988!.

37W. Borg, M. Gall, and J. Gall,Educational Research: An Introduction, 6thed. ~Longman, London, 1996!.

38J. P. Goetz and M. D. LeCompte,Ethnography and Qualitative Design inEducational Research~Academic, Orlando, FL, 1984!.

39M. Q. Patton,Qualitative Evaluation Methods~Sage, Beverly Hills, CA,1980!.

40Marcia Linn, ‘‘Free-choice experiences: how do they help childrenlearn?’’ Sci. Educ.64, 237–248~1980!.

41Thomas R. Brown, T. F. Slater and J. P. Adams, ‘‘Gender differences withbatteries and bulbs,’’ Phys. Teach.17 ~3!, 311–323~1995!.

42G. D. Thijs and G. M. Bosch, ‘‘Cognitive effects of science experimentsfocusing on student’s preconceptions of force: a comparison of demonstra-tions and small-group practicals,’’ Int. J. Sci. Educ.36, 526–527~1998!.

As our mental eye penetrates into smaller and smaller distances and shorter and shorter times,we find nature behaving so entirely differently from what we observe in visible and palpablebodies of our surroundings that no model shaped after our large-scale experiences can ever be‘true’. A completely satisfactory model of this type is not only practically inaccessible, but noteven thinkable. Or, to be precise, we can, of course, think of it, but however we think it, it iswrong; not perhaps quite as meaningless as a ‘‘triangular circle’’, but more so than a ‘‘wingedlion’’.

Erwin Schroedinger quoted inThe Universe in a Teacupby K. C. Cole

S36 S36Phys. Educ. Res., Am. J. Phys. Suppl., Vol. 68, No. 7, July 2000 J. A. Marshall and J. T. Dorward This article is copyrighted as indicated in the article. Reuse of AAPT content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

158.121.247.60 On: Tue, 11 Nov 2014 00:40:25