The Viability of Distance Education Science Laboratories
##AUTHORSPLIT##<--->Galileo Galilei established experimentation as a foundation of modern science through the simple act of dropping two iron balls from the Tower of Pisa. Though debatable whether he actually performed that experiment, discussion in his
Dialogues Concerning the Two New Sciences shows clearly the power and importance of experimental observations in convincing others of the correctness of a particular scientific theory or hypothesis. The history of science from Galilei on has primarily been the reconciliation of theory with imperfect experimental data. Providing a similarly compelling laboratory experience for a student, especially to one not physically present, is problematic.
Experimentation is obviously central to scientific investigation, but what compelling reasons are there for students to perform experiments, particularly the textbook laboratory exercise in which the results are already well known? There are two main elements to an experiment: its design and analysis of the data. But no experiment can be performed without error, so one must determine with what degree of certainty the data supports a particular hypothesis. Coming to terms with the inaccuracy and imprecision of results requires knowledge of the interplay between experimental design and data analysis. Some laboratory skills, such as the statistical analysis of data, can be learned in the abstract outside of the laboratory. Experimental design, however, can only be learned from using real equipment in real experiments, often through a certain amount of trial and error. It is no great surprise that student practice of experimentation is needed to understand science; educational abstractions alone are not enough. In addition to other laboratory goals, it is crucial that this skill, the art of experimental design, be communicated to the student through the laboratory experience.
Current Delivery Technologies
Distance education attempts to achieve traditional educational goals with the added challenge of connecting the instructor and student by some delivery technology. No matter how unobtrusive, any delivery technology can conceal as well as illuminate an educational point. It can also impose a high cost on instructors in learning the requisite technical skills. Complex delivery technologies will attract few instructors, thereby reducing the use of distance education. In contrast, simple delivery technologies will be more easily mastered by a greater number of instructors and likely produce a clearer experience for the student. Whether distance education is successful in maintaining experimentation as an important part of science education depends to a large degree upon the transparency, for both instructor and student, of the delivery technology. The three delivery technologies that to date have most often been employed for performing experiments in distance education are computer simulations, videos of real laboratories and laboratory kits sent to a student.
There is clearly a place for computer simulations in science education with many excellent programs available, which are useful in exploring real and hypothetical scenarios. They can also provide exposure where reality is too dangerous, expensive, complex, fast or slow, such as exploring human anatomy, controlling a nuclear reactor or examining the supersonic crack of a whip. Although in watching a video the student is often only a passive observer, videos can offer a way to actively participate in an experiment.
Both delivery technologies can enhance the science educational experience and readily lend themselves to the distance education environment. However, they are ill-suited for delivering a realistic environment to conduct experiments, measure results and determine error. Simulations tend to restrict students to a narrow investigative path, are physically unconvincing and never provide ambiguous results as occurs with real instruments. Similarly, by providing the student with limited examples in unambiguous conditions, a pre-packaged video often affords few opportunities to practice, explore, fail and subsequently learn. For these reasons, the laboratory experience cannot be entirely replaced by simulations or videos of real experiments. Sending equipment kits to students is aimed at duplicating traditional lab experiments at home, but this raises cost and safety concerns. Typical equipment for a college introductory physics course might include lasers, radioactive elements, high-voltage power supplies and costly optical apparatus such as spectral tubes. A college chemistry course might also require a Bunsen burner and at least small amounts of somewhat dangerous chemicals.
Pros and Cons of Distance Labs
In many ways, distance labs can provide a superior educational experience to a purely in-residence laboratory. Distance labs are not restricted to synchronized attendance by instructors and students; they have the potential to provide constant access whenever needed by students. Safety issues that would otherwise limit the kinds of laboratory experiences available to undergraduate science students are diminished in distance labs. They can also transcend the restrictions of time and space, allowing experiments that monitor geographically distant phenomena such as weather and seismographic data. And because students have greater access to experimental equipment, fewer lab stations are needed, thereby mitigating the costs associated with purchasing and maintaining lab equipment.
Conducting experiments aided by a computer has a long and venerable history in many scientific areas, particularly physics and astronomy. Most high school and college physics labs already control experiments, and collect and analyze data using laboratory computers. Given this history and the ability of the Internet to connect the experimenter to the experiment, it seems logical to ask whether viable laboratory experience is possible by conducting real experiments through distance labs.
From past experience, the authors have identified three key requirements for a distance laboratory to compare favorably with an in-residence laboratory:
- Students must have enough control of lab equipment to start and stop an experiment and make appropriate adjustments.
- The experiment should be no more difficult to conduct than with the equipment physically present.
- Students need appropriate feedback.
By satisfying these requirements, the key elements of experimental design and error analysis remain viable student learning goals.
Key impediments to the growth of distance labs are the absence of an educational model for distance science laboratories, the lack of delivery technology standards for instrument hardware and software, and the considerable technical difficulty and expense of development. Due to the lack of standard development practices and technologies, a pessimistic view that distance laboratories are not worth the effort and expense may be understandable. However, we believe that with a general development approach based on technically simple and familiar tools, and a distance education model based on sound pedagogical principles, selected distance labs can be developed that compare favorably with traditional laboratories in development effort, expense and student learning.
A Simple Approach
The most promising and technologically simple approach for distance lab development to date uses a standard spreadsheet application with access to distant instruments over a network. Spreadsheets have long been used in science laboratories to analyze data, but require some minor modifications to access instruments due to the lack of instrumentation standards.
For the instructor, such a familiar and relatively easy-to-use approach lowers learning costs and effort in creating a lab. To create a lab, the instructor might do a few of the preliminary lab steps, such as how to get data from a distant instrument into the spreadsheet, leaving the remaining steps of performing the experiment and data analysis to the student. Using such a general tool, a student can then open the same spreadsheet from a Web page at home or in the traditional lab setting, and essentially pick up where the instructor stopped. Students have the freedom to choose their own investigative approach, much as in a traditional lab. As in a traditional lab, students would need to determine appropriate parameters for the measurement instrument and adjust for errors. Instructions, graphics or audio can also be provided with the spreadsheet on the Web page.
Unforeseen Benefits
In the following examples, the instrumentation involved was relatively inexpensive – less than $300 each for Vernier software, well within the reach of most high school and introductory college laboratories. Note that all examples can be done with a standard spreadsheet, although some were initially implemented as a computer program before recognizing the spreadsheet potential for distant laboratories.
The first example involves a physics experiment in which students made noise to study the frequency of sound. A microphone captured sound waves that were digitized and inputted into a computer. Students began the experiment with an instructor-prepared spreadsheet that collected and graphed the raw digital sound data as the noise was created. The spreadsheet included the Fourier analysis and graphics tools necessary to analyze the frequency spectrum of the sound data. Students directly controlled the parameters by changing appropriate cells of the spreadsheet. The full lab instructions and spreadsheet were part of a Web page, and could be used from any computer with the spreadsheet application and an Internet connection.
This lab was conducted in a traditional setting, during a scheduled time with the instructor, computer, apparatus, musical sound sources and students at the same location. However, in addition to sounds made by students locally, a sound source located in another room was included for students to analyze. By changing a cell in the spreadsheet to the distant device address, students collected data from the unknown sound source. Determining the frequency of that source required students to experiment with the measurement instrument sample rate and the number of data points collected. As part of the laboratory report, students explained the accuracy and error inherent in the measurements, requiring an understanding of instrument limitations, experimental design and the spreadsheet tools used to analyze the data. Unanticipated benefits from using the spreadsheet approach were the very low development effort for the instructor and the opportunity for students to follow an investigative approach not anticipated by the instructor.
In another experiment, computer science students in a networking class used a Web browser that ran a Java-applet as an oscilloscope analyzing a digital signal. This started as a traditional laboratory with one computer running an oscilloscope program, another generating the digital signal and an analog-to-digital converter instrument measuring the signal. Though there was no scheduled laboratory time, stu-dents often had to queue for access to the single-equipment setup. The replacement distance laboratory was developed mainly for presenting a networked application example, not directly to benefit waiting students. However, because the oscilloscope was part of the laboratory Web page, all instructions, tools and equipment were readily accessible to students over the Internet, making the single setup of equipment sufficient. As another unintended benefit, because any computer could be the oscilloscope and students preferred to do the experiment from home, the setup back in the lab needed only one computer with the analog-to-digital converter instrument.
Our third example presents a potentially enormous benefit of distance laboratories: the possibilities for new science education models to ignore time and place. The setup consisted of three antennas designed to receive extremely low-frequency electromagnetic signals. They were buried at the edge of our campus, making direct student and instructor access physically difficult. In this case, students could not change instrument parameters, but could only collect and analyze the signal data so that an observation by one student using the same instrument did not interact with another’s observation. This model would be generally useful where many students monitored equipment that is otherwise inaccessible, such as remote weather stations.
Distance laboratories also offer a unique opportunity for educational models that take advantage of geographical features. In our last example, one student used multiple- instrument setups simultaneously over an uninterrupted one-week period. Several volunteers at sites in Nebraska, New York, Indiana and Pennsylvania attached a light-sensing instrument to their Internet-connected computer and ran a small communications program to give access to the instrument. Over the week, the student’s computer program controlled instrument parameters and recorded the light measured at each distant site every second from our campus.
With inexpensive instruments and technically simple means, other data could easily be collected from multiple distant sites. One unanticipated discovery was the logistics of locating and coordinating distant volunteer sites that were far more challenging than the technical elements of the project. The current limits of using the Internet to perform large-scale distant laboratory exercises seems more dependent on limitations of human resources than with technical challenges.
Conclusion
Distance education offers the possibility for very different kinds of learning opportunities, such as large-scale collaboration. With student collaboration at distant schools, experiments based on sharing local weather data of barometric pressure, temperature and wind speed are quite possible. Other collaborations, such as seismographic data, ambient particulate contamination and air quality in different parts of a building or in an entire organization become very feasible. Imagine if students had been able to monitor local radiation levels in various parts of Europe after the Chernobyl incident, or the nuclear tests in Pakistan and India. Such projects appear to require large amounts of equipment and organization, but with the Internet as a collaborative tool and cooperation at the distant sites, it is quite practical to perform some of these collaborations cheaply and effectively.
Among the unanswered questions in distance education is the role of the science laboratory. As an alternative to computer simulations, videos and laboratory kits sent out to the student, we have investigated some facets of conducting real laboratory experiments at a distance. While we do not believe distance laboratories can or should replace other approaches entirely, distance laboratories can offer considerable educational advantages, even when compared to traditional laboratories. Greater student access to equipment can be provided that would not otherwise be available due to time restrictions by the student or the instructor. Physical location and expense of the equipment can also be made less important. Web-based delivery offers a means to conveniently package a complete laboratory, including written guides, data collection and data analysis tools, whatever the location of the student or experiment. Finally, by keeping control of an experiment in the hands of the student, distance laboratories can achieve educational goals important in the traditional laboratory.
For appendices associated with this article click here.
Kyle Forinash, Ph.D., is a professor of physics at Indiana University Southeast in New Albany, IN. His research interests include computer modeling of the nonlinear dynamics of solids such as bio-polymers. He has also investigated the application of computers for data collection in the student laboratory, including remote control of experiments using the Internet.E-mail:
[email protected]
Raymond Wisman is an associate professor in the Computer Science Department at Indiana University Southeast.E-mail:
[email protected]Galileo Galilei established experimentation as a foundation of modern science through the simple act of dropping two iron balls from the Tower of Pisa. Though debatable whether he actually performed that experiment, discussion in his Dialogues Concerning the Two New Sciences shows clearly the power and importance of experimental observations in convincing others of the correctness of a particular scientific theory or hypothesis. The history of science from Galilei on has primarily been the reconciliation of theory with imperfect experimental data. Providing a similarly compelling laboratory experience for a student, especially to one not physically present, is problematic.
Experimentation is obviously central to scientific investigation, but what compelling reasons are there for students to perform experiments, particularly the textbook laboratory exercise in which the results are already well known? There are two main elements to an experiment: its design and analysis of the data. But no experiment can be performed without error, so one must determine with what degree of certainty the data supports a particular hypothesis. Coming to terms with the inaccuracy and imprecision of results requires knowledge of the interplay between experimental design and data analysis. Some laboratory skills, such as the statistical analysis of data, can be learned in the abstract outside of the laboratory. Experimental design, however, can only be learned from using real equipment in real experiments, often through a certain amount of trial and error. It is no great surprise that student practice of experimentation is needed to understand science; educational abstractions alone are not enough. In addition to other laboratory goals, it is crucial that this skill, the art of experimental design, be communicated to the student through the laboratory experience.
X@XOpenTag001X@XOpenTag000Current Delivery TechnologiesX@XCloseTag000X@XCloseTag001
Distance education attempts to achieve traditional educational goals with the added challenge of connecting the instructor and student by some delivery technology. No matter how unobtrusive, any delivery technology can conceal as well as illuminate an educational point. It can also impose a high cost on instructors in learning the requisite technical skills. Complex delivery technologies will attract few instructors, thereby reducing the use of distance education. In contrast, simple delivery technologies will be more easily mastered by a greater number of instructors and likely produce a clearer experience for the student. Whether distance education is successful in maintaining experimentation as an important part of science education depends to a large degree upon the transparency, for both instructor and student, of the delivery technology. The three delivery technologies that to date have most often been employed for performing experiments in distance education are computer simulations, videos of real laboratories and laboratory kits sent to a student.
There is clearly a place for computer simulations in science education with many excellent programs available, which are useful in exploring real and hypothetical scenarios. They can also provide exposure where reality is too dangerous, expensive, complex, fast or slow, such as exploring human anatomy, controlling a nuclear reactor or examining the supersonic crack of a whip. Although in watching a video the student is often only a passive observer, videos can offer a way to actively participate in an experiment.
Both delivery technologies can enhance the science educational experience and readily lend themselves to the distance education environment. However, they are ill-suited for delivering a realistic environment to conduct experiments, measure results and determine error. Simulations tend to restrict students to a narrow investigative path, are physically unconvincing and never provide ambiguous results as occurs with real instruments. Similarly, by providing the student with limited examples in unambiguous conditions, a pre-packaged video often affords few opportunities to practice, explore, fail and subsequently learn. For these reasons, the laboratory experience cannot be entirely replaced by simulations or videos of real experiments. Sending equipment kits to students is aimed at duplicating traditional lab experiments at home, but this raises cost and safety concerns. Typical equipment for a college introductory physics course might include lasers, radioactive elements, high-voltage power supplies and costly optical apparatus such as spectral tubes. A college chemistry course might also require a Bunsen burner and at least small amounts of somewhat dangerous chemicals.
X@XOpenTag003X@XOpenTag002Pros and Cons of Distance LabsX@XCloseTag002X@XCloseTag003
In many ways, distance labs can provide a superior educational experience to a purely in-residence laboratory. Distance labs are not restricted to synchronized attendance by instructors and students; they have the potential to provide constant access whenever needed by students. Safety issues that would otherwise limit the kinds of laboratory experiences available to undergraduate science students are diminished in distance labs. They can also transcend the restrictions of time and space, allowing experiments that monitor geographically distant phenomena such as weather and seismographic data. And because students have greater access to experimental equipment, fewer lab stations are needed, thereby mitigating the costs associated with purchasing and maintaining lab equipment.
Conducting experiments aided by a computer has a long and venerable history in many scientific areas, particularly physics and astronomy. Most high school and college physics labs already control experiments, and collect and analyze data using laboratory computers. Given this history and the ability of the Internet to connect the experimenter to the experiment, it seems logical to ask whether viable laboratory experience is possible by conducting real experiments through distance labs.
From past experience, the authors have identified three key requirements for a distance laboratory to compare favorably with an in-residence laboratory:
- Students must have enough control of lab equipment to start and stop an experiment and make appropriate adjustments.
- The experiment should be no more difficult to conduct than with the equipment physically present.
- Students need appropriate feedback.
By satisfying these requirements, the key elements of experimental design and error analysis remain viable student learning goals.
Key impediments to the growth of distance labs are the absence of an educational model for distance science laboratories, the lack of delivery technology standards for instrument hardware and software, and the considerable technical difficulty and expense of development. Due to the lack of standard development practices and technologies, a pessimistic view that distance laboratories are not worth the effort and expense may be understandable. However, we believe that with a general development approach based on technically simple and familiar tools, and a distance education model based on sound pedagogical principles, selected distance labs can be developed that compare favorably with traditional laboratories in development effort, expense and student learning.
X@XOpenTag005X@XOpenTag004A Simple ApproachX@XCloseTag004X@XCloseTag005
The most promising and technologically simple approach for distance lab development to date uses a standard spreadsheet application with access to distant instruments over a network. Spreadsheets have long been used in science laboratories to analyze data, but require some minor modifications to access instruments due to the lack of instrumentation standards.
For the instructor, such a familiar and relatively easy-to-use approach lowers learning costs and effort in creating a lab. To create a lab, the instructor might do a few of the preliminary lab steps, such as how to get data from a distant instrument into the spreadsheet, leaving the remaining steps of performing the experiment and data analysis to the student. Using such a general tool, a student can then open the same spreadsheet from a Web page at home or in the traditional lab setting, and essentially pick up where the instructor stopped. Students have the freedom to choose their own investigative approach, much as in a traditional lab. As in a traditional lab, students would need to determine appropriate parameters for the measurement instrument and adjust for errors. Instructions, graphics or audio can also be provided with the spreadsheet on the Web page.
X@XOpenTag007X@XOpenTag006Unforeseen BenefitsX@XCloseTag006X@XCloseTag007
In the following examples, the instrumentation involved was relatively inexpensive – less than $300 each for Vernier software, well within the reach of most high school and introductory college laboratories. Note that all examples can be done with a standard spreadsheet, although some were initially implemented as a computer program before recognizing the spreadsheet potential for distant laboratories.
The first example involves a physics experiment in which students made noise to study the frequency of sound. A microphone captured sound waves that were digitized and inputted into a computer. Students began the experiment with an instructor-prepared spreadsheet that collected and graphed the raw digital sound data as the noise was created. The spreadsheet included the Fourier analysis and graphics tools necessary to analyze the frequency spectrum of the sound data. Students directly controlled the parameters by changing appropriate cells of the spreadsheet. The full lab instructions and spreadsheet were part of a Web page, and could be used from any computer with the spreadsheet application and an Internet connection.
This lab was conducted in a traditional setting, during a scheduled time with the instructor, computer, apparatus, musical sound sources and students at the same location. However, in addition to sounds made by students locally, a sound source located in another room was included for students to analyze. By changing a cell in the spreadsheet to the distant device address, students collected data from the unknown sound source. Determining the frequency of that source required students to experiment with the measurement instrument sample rate and the number of data points collected. As part of the laboratory report, students explained the accuracy and error inherent in the measurements, requiring an understanding of instrument limitations, experimental design and the spreadsheet tools used to analyze the data. Unanticipated benefits from using the spreadsheet approach were the very low development effort for the instructor and the opportunity for students to follow an investigative approach not anticipated by the instructor.
In another experiment, computer science students in a networking class used a Web browser that ran a Java-applet as an oscilloscope analyzing a digital signal. This started as a traditional laboratory with one computer running an oscilloscope program, another generating the digital signal and an analog-to-digital converter instrument measuring the signal. Though there was no scheduled laboratory time, stu-dents often had to queue for access to the single-equipment setup. The replacement distance laboratory was developed mainly for presenting a networked application example, not directly to benefit waiting students. However, because the oscilloscope was part of the laboratory Web page, all instructions, tools and equipment were readily accessible to students over the Internet, making the single setup of equipment sufficient. As another unintended benefit, because any computer could be the oscilloscope and students preferred to do the experiment from home, the setup back in the lab needed only one computer with the analog-to-digital converter instrument.
Our third example presents a potentially enormous benefit of distance laboratories: the possibilities for new science education models to ignore time and place. The setup consisted of three antennas designed to receive extremely low-frequency electromagnetic signals. They were buried at the edge of our campus, making direct student and instructor access physically difficult. In this case, students could not change instrument parameters, but could only collect and analyze the signal data so that an observation by one student using the same instrument did not interact with another’s observation. This model would be generally useful where many students monitored equipment that is otherwise inaccessible, such as remote weather stations.
Distance laboratories also offer a unique opportunity for educational models that take advantage of geographical features. In our last example, one student used multiple- instrument setups simultaneously over an uninterrupted one-week period. Several volunteers at sites in Nebraska, New York, Indiana and Pennsylvania attached a light-sensing instrument to their Internet-connected computer and ran a small communications program to give access to the instrument. Over the week, the student’s computer program controlled instrument parameters and recorded the light measured at each distant site every second from our campus.
With inexpensive instruments and technically simple means, other data could easily be collected from multiple distant sites. One unanticipated discovery was the logistics of locating and coordinating distant volunteer sites that were far more challenging than the technical elements of the project. The current limits of using the Internet to perform large-scale distant laboratory exercises seems more dependent on limitations of human resources than with technical challenges.
X@XOpenTag009X@XOpenTag008ConclusionX@XCloseTag008X@XCloseTag009
Distance education offers the possibility for very different kinds of learning opportunities, such as large-scale collaboration. With student collaboration at distant schools, experiments based on sharing local weather data of barometric pressure, temperature and wind speed are quite possible. Other collaborations, such as seismographic data, ambient particulate contamination and air quality in different parts of a building or in an entire organization become very feasible. Imagine if students had been able to monitor local radiation levels in various parts of Europe after the Chernobyl incident, or the nuclear tests in Pakistan and India. Such projects appear to require large amounts of equipment and organization, but with the Internet as a collaborative tool and cooperation at the distant sites, it is quite practical to perform some of these collaborations cheaply and effectively.
Among the unanswered questions in distance education is the role of the science laboratory. As an alternative to computer simulations, videos and laboratory kits sent out to the student, we have investigated some facets of conducting real laboratory experiments at a distance. While we do not believe distance laboratories can or should replace other approaches entirely, distance laboratories can offer considerable educational advantages, even when compared to traditional laboratories. Greater student access to equipment can be provided that would not otherwise be available due to time restrictions by the student or the instructor. Physical location and expense of the equipment can also be made less important. Web-based delivery offers a means to conveniently package a complete laboratory, including written guides, data collection and data analysis tools, whatever the location of the student or experiment. Finally, by keeping control of an experiment in the hands of the student, distance laboratories can achieve educational goals important in the traditional laboratory.
X@XOpenTag010For appendices associated with this article click here.X@XCloseTag010
Kyle Forinash, Ph.D., is a professor of physics at Indiana University Southeast in New Albany, IN. His research interests include computer modeling of the nonlinear dynamics of solids such as bio-polymers. He has also investigated the application of computers for data collection in the student laboratory, including remote control of experiments using the Internet.E-mail:
[email protected]
Raymond Wisman is an associate professor in the Computer Science Department at Indiana University Southeast.E-mail:
[email protected]
This article originally appeared in the 09/01/2001 issue of THE Journal.