Why use computers to teach Physics Labs?

Microcomputer-Based Laboratories (MBL)

MBL effects upon student learning and conceptual development in undergraduate physics have been studied by Thornton, Laws and their associates at Tufts University and Dickinson College (Laws, 1989; Thornton, 1989; Thornton and Sokoloff, 1990). These researchers and instructors have developed quantitative instruments designed to measure changes in the physics-related graph interpretation skills and kinematics conceptual understanding of undergraduate students. Their research attempts to contrast such skills and knowledge acquisition of students using their own locally-developed MBL materials working in small groups with typical undergraduate students in traditional physics laboratory curricula. Their large scale testing at various sites indicate that their own laboratory curricula incorporating MBL and the instructional strategies proposed by Arons and McDermott are considerably more effective in teaching basic kinematics (mechanics) concepts than standard lectures (Thornton, 1993).

Another group of science education researchers at the University of California (Berkeley) have also examined the role played by MBL-based activities in science education, but at the middle-school level. Linn, Nachmias, Songer and associates (Leiberman & Linn, 1991; Linn, 1988; Linn & Songer, 1989; Nachmias and Linn, 1987; Stein, 1987) have examined the roles of changing curricular expectations and MBL based activities on middle school student conceptual development and graphical skills acquisition. Their research has indicated that curricular activities and expectations play a pre-eminent role in student science laboratories where MBL technology is used. Linn, et al., claim that curricular evolution taking advantage of several characteristics of MBL technology can achieve profound conceptual changes amongst students.

Other researchers (Amend & Furstenau, 1992; Amend, 1991, 1989; Nahkleh & Krajcik, 1991; Lehman & Campbell, 1991; Heck, 1990) have all examined various aspects of MBL implementation in educational laboratory settings. Prevalent findings include significantly higher levels of both student and instructor motivation, and increased level of curricular control by both students and instructors. Yet others (Beichner, 1990; Stuessy & Rowland, 1989) have tried to examine the effects of delayed MBL information presentation practice in greater depth.

These studies examine small groups of students who perform laboratory activities in carefully controlled environments with considerable access to MBL apparatus. Such research has not concentrated upon large-enrollment university mechanics laboratories using locally-developed materials, which is the case with PHYS 152L. Several of these studies have suggested that the maximum quality of student experience and greatest conceptual change can be achieved through careful incremental modification of curricular expectations, exploitation of technology during instructional activities, and improved pedagogical materials (Linn & Songer, 1989). However, this study [of PHYS 152L] is one of very few to explicitly document such practice as carried out by a group of participants (including students) actively involved in curricular reformation incorporating MBL.

Cognitive Psychology and MBL

This section addresses issues raised by an examination of MicroComputer-Based Laboratory (MBL) characteristics and human learning theory. Different investigators have identified a long list of cognitively desirable instructional characteristics of MBL. These include (Tinker, 1984a):

Environmental Simplicity

One unexpected characteristic of MBL activities is a simplified experimental environment, due to fewer repetitive chores such as data collection and graphing. More of the student"s concentration can be spent upon the phenomena and relationships, reducing cognitive overhead. Cognitive overhead refers to the constraints of human short term memory (Gagne & Glaser, 1987) which make learning difficult when experiencing an extended series of discontinuous events. Human learning performs best with a sharply limited series of contiguous events. This means that timely interpretation of experimental data is required for learning, and that next-day discussion of graphically analyzed laboratory results (common in school experiments) is far from optimal. MBL technology may also provide a situational simplification in reducing all data collection, storage and transformation to a single device. Different experiments may make use of a common, modular design, repeating similar procedures with similar equipment for calibration, data collection and analysis over a wide variety of experiments (Tinker, 1984a).

Fast Feedback

The immediacy of feedback during MBL activity is an empowering characteristic of the technology. Data that is immediately available in a comprehensible form may make more time available for the Piagetian processes of accommodation and self-regulation. These terms describe the development of cognitive structures that resolve apparently aberrant phenomena. With MBL, more opportunity for these regulatory processes may be experienced with more iterations in laboratory experience. The importance of high quality situational feedback is recognized by almost all learning theories.

There is an oft-perceived disadvantage to fast feedback in the science laboratory -- that students do not perform as many detailed numerical data manipulations and calculations when using computers. There are appropriate laboratory situations for extensive student calculation for pedagogical purposes, but after the required mathematical manipulations are mastered by the student, repetition can become drudgery. An analogy can be made to the use of calculators in mathematics courses; while several hundred repeated calculations might be appropriate in an elementary mathematics class, high school students are widely encouraged to use a calculator for arithmetic operations (Tinker, 1984a).

Direct Experience

Gagne and Glaser (1987) describe the ability of the mind to accept data from several sensory channels of different limitation. They refer to the visual channel as having the greatest bandwidth, that is the greatest transmissive ability of information in a single chunk. Such chunks are then theorized to be processed in a sharply limited short term memory and go on to be learned. MBL provides an enormous throughput of information in this readily learnable visual form, allowing the complexities of laboratory situations to be analyzed by students in the laboratory. MBL increases access to phenomena, allowing new experiments that are otherwise too technically or conceptually difficult (Linn & Songer 1989).

Student Control

MBLs represent a novel situation making use of familiar media (television and computers) to encourage student control, ownership and involvement. MBL technology is attractive, dynamic and interesting. Again, the majority of learning theories indicate that student interest is of considerable import (Tinker, 1989a).

Student control is a situational characteristic inherent in MBL practice and in the instructional philosophy known as constructivism. Thornton describes student control in a constructivist MBL curriculum:

Immediate feedback supports collaborative learning and collaborative work provides immediate feedback. Learning is also enhanced by encouraging students to express their predictions and to discuss unexpected results with their peers.
(Thornton, 1990, p. 866)

In the MBL laboratory, the locus of experimental control shifts to the student and away from the teacher and the curriculum. MBL gives the student a tool that encourages even poorly prepared students to become active participants in a scientific process which invites them to ask and answer their own questions. Students are encouraged through an increased ability to control the experimental environment and develop an understanding of a specific phenomenon before attempting to progress to more abstract concepts.

Data Transformation and Graphs

This is possibly the single most studied cognitive characteristic of MBL technology because student graphical interpretation abilities are quantifiable. There is a wealth of information readily available for analysis even in simple graphs, while in contrast, tabular numeric data are of much more limited use to experimenters and students. The effects of MBL graphical data presentation upon student learning have been closely examined and found to be very encouraging (Brazil, 1987; Nachmias & Linn, 1987; Stein, 1987). Gagne and Glaser (1987) describe the process of appropriately encoding data so as to emphasize important features as essential to learning.

Graphical interpretation and evaluation have been extensively studied by many MBL investigators. These studies conclude that the real-time graphing features of MBL are effective in improving student graphic interpretation performance (Brasell, 1987; Nachmias & Linn, 1987; Stein, 1987). Graphs permit the ready display of key information in relationships. A graph presents a starting point, endpoints, slopes, cusps, asymptotes and so forth, all containing interpretable information. Information must be extracted from the graphs through active student interpretation.

The majority of graphs examined by students during the instructional process are presented in textbooks and therefore graphs are rarely disputed -- they are largely accepted upon faith. Student-produced graphs, however, are typically evaluated in terms of inaccurate data, poor labeling and messiness rather than by comparison with experimental phenomena. Graphs produced with MBL are typically evaluated by comparison with subject knowledge or by instrumental effects (Nachmias & Linn, 1987). The interpretation of MBL graphs by students can be said to increase the level of critical appraisal in two ways -- both in terms of the subject matter itself and in terms of the data collection and presentation methods.

The formal abstraction of meaning from data via visual transformations has also been addressed by Novak and Gowin (1984). Pictorial representations of information represent simple dissociations of student thought from the experimental context, which can be carried on into the mapping of concepts and relationships themselves.

Curricular Reformation and Technology

The "crisis" of ineffectiveness in U.S. science education is recognized as a major concern and technological innovation is being heralded as at least a partial solution. Prescriptive plans to integrate technology into the science curriculum have mentioned possible improvements as due to the following facts:

Linn goes on to suggest that the implementation of technology for instructional purposes moves through three major stages of acceptance (Linn, 1988):

This would suggest that MBL adoption will catalyze significant changes in science curricula by making apparent present procedural shortcomings in instructional delivery, then by changing the curriculum content to surmount these limitations and finally by supporting reforms in the curricular paradigms of science pedagogy. Such reforms are already apparent in the constructivist movement in science pedagogy, which embraces many of the characteristics of free investigation and student empowerment ascribed to technological innovation.

Technological Innovation in Laboratory Science Education

In this day and age very few real experiments are conducted without employing the latest technology -- sophisticated measurement instruments supplying large amounts of accurate data to a computer for storage, analysis and display.
(MacKenzie, 1988, p. 13)

Typical high-school science laboratories attempt to approximate research methodologies using processes similar to that in Figure 8. In the research lab, the researcher chooses an experimental problem and designs the experiment; in the school lab these activities are usually prescribed by the curriculum or text due to time constraints. Students usually do not participate in experimental design. Students follow the given directions in the lab procedure, acquire data, perform calculations to treat the raw data appropriately and then complete some form of analysis, usually including graphical procedures. Then a generalization of some form is extracted (usually including an explanatory theory in active research) and results are documented for a report.

MBL procedures most notably affect those steps in the laboratory experiment sequence involved in data acquisition and analysis. MBL procedures are an adaptation from research use of the same technology for similar tasks:

MBL laboratories typically involve the use of sensors or probes to directly collect data in an electrical form and to display it in both numerical and graphical form as it is collected. This real-time display greatly abbreviates analysis and allows for immediate observation and control of experimental variables (Amend et al., 1989).

Students set up their apparatus and sensors, set scaling and display options on the microcomputer and then calibrate their sensors using known standards. Data are then collected using a series of real-time "runs", with continuous observation of the computer screen and the physical process. After a run is complete, data are saved to disk and/or printed, results are discussed and compared with others and decisions regarding experimental repetition or variable control are made. Usually, some variable is modified and the experiment repeated, with results juxtaposed and examined. When complete, the experiment is written up into a report (Amend et al., 1989).

Laboratory Advantages to MBL Procedures

The advantages for science students inherent in the use of MBL technology in the laboratory are twofold: (Amend et al., 1989)

The computer becomes a tool which allows repeatability -- the reliable and untiringly accurate collection of data -- in volumes not otherwise possible due to time and attention constraints. Events become more easily quantified, and those events which happen too quickly to examine otherwise may be analyzed. Additionally, experiments involving a number of simultaneous measurements may be easily performed.

The data collected can be displayed instantaneously, and in any numerically processed form desired. This rapid processing and analysis allow the testing of user suggestions and conjectures not otherwise possible due to time constraints. The amount of data throughput is greatly increased. The data can be meaningfully examined while being collected, encouraging investigation by discovery. More time than before can be spent examining relationships, postulating relationships, controlling experimental variables and redesigning the experiment. MBL technology has the ability to free the user from the drudgery of quantification and graphical analysis and allow active investigation (Amend et al., 1989).

MBL technology also introduces students to scientific measurement. This includes errors of measurement, graphical interpretation, instrumental effects (calibration, accuracy, repeatability, error of quantification, resolution, scaling) and control of extraneous variables. These topics are not typically treated in the school laboratory because of the nature of "precooked" experiments, the lack of available precision and time. They are nonetheless valuable laboratory science skills (Linn, 1989). Instrumental effects refer to the inherent distortions in data due to the collection process. When using MBL, data can be made unreliable by five major instrumental causes: inappropriate graph scaling (in software), inappropriate setup, poor probe calibration (and resulting inaccuracy), inadequate probe resolution (where the equipment cannot discriminate fine enough gradations in the phenomena) and experimental variation (due to random error or invalid procedures). Students can be trained to recognize and correct these problems (Nachmias & Linn, 1987). Such training should be an integral part of MBL laboratory instruction.

Philosophical Changes in Science Pedagogy

Recently science education has been turning from the content-based curriculum established by the revolutions of the 1950s and 1960s (Duschl, 1985) with voluminous transmission of information and attendant laboratory exercises stressing the replication of proven concepts to a more process-oriented curriculum stressing skills of analysis, questioning, synthesis and problem solution via laboratory experience.

As an example, the National Science Teachers Association (NSTA, 1983) has identified the following concerns regarding science education:

  1. The textbook IS the curriculum.
  2. The goals of individual classes are not related to previous or subsequent classes.
  3. The lecture is the main form of instruction with laboratories used for verification.
  4. Science is evaluated in the traditional method.
  5. Science is removed from the world outside of the classroom.

Several investigators suggest that the adoption of MBL techniques provides a response to several of these concerns:

Teachers and students will be active participants in the science process. Teachers will utilize methods of moving away from the text towards laboratory experiences which may be more directly related to the world of the student outside of the classroom. As a result, teachers will lecture less, and students will be involved in the active seeking of information. This will necessarily cause a change in the classroom evaluation procedures utilized.
(Woerner, 1987, p. 35)

Tinker has also decried the science laboratory in its present form,

[Science labs invoke] unpleasant memories of boredom or fear in most high school survivors. The lab is a place where normal common sense is suspended in favor of that ineffable "scientific method" which seems to consist of lots of numbers, lab books and funny equipment. For many, the whole process resembles having to cook a meal for someone with terrible taste, cleaning up afterwards, and then doing penance with pages and pages of arithmetic.
(Tinker, 1984a, p. 24)

while suggesting solutions provided by MBL:

...use the computer as a laboratory instrument. Make it into a tool that allows students to quantify the world around them. Give students a fantastically powerful tool that no science teacher could have dreamed of having in class only a few years ago -- an instrument that can measure ... ...give students these tools and you will see a (pardon the expression) revolution in science education -- a true embodiment of Piagets' notion that children learn best by discovering and creating the world for themselves.
(Tinker, 1984a, p. 26)

The pedagogical basis upon which science is taught is currently changing from a teacher-oriented presentational style to a participatory style involving the negotiation of meaning (constructivism) wherein teachers must surrender a large degree of situational control. MBL technology and methods can provide a route to this style of interaction by encouraging student control centered upon the experimental relationships under study rather than instructor and textbook direction (Linn, 1988).

Additionally, MBL technology has been seen to enhance the qualities of on-task inter-student communications:

Students' propensity to monitor and compare their results to others, which was made possible by the fact that results were displayed graphically on the computer screen. Students compared results with one another constantly and thus were alerted to disparities. The sharing of data also encouraged cooperative remediation of problems, with students forming into consulting groups of increasing size according to the difficulty of the problem at hand.
(Stein, 1987, p. 233)

So there are at least a few good reasons to try using computers to teach physics labs. References

Dan MacIsaac, 1996 (http://www.physics.nau.edu/~danmac)