EMST 291B Class Project


Table of Contents


Project Participants

Eunice Yi

Chris Schneider


Development Proposal

Abstract

We plan to build Sputnik, an html-based tutorial that will guide high school physics students toward the solution of a real-world engineering problem: placing a satellite in geostationary orbit. Sputnik will draw on the concepts the students have hopefully learned in class (e.g. circular motion, centripetal acceleration, universal gravitation, etc.), but will provide hints and scaffolding when necessary, based on the selections students make in response to various questions it poses to them. Thus, it will not be a vehicle for introducing subject material; rather it will provide an application that will hopefully help students integrate new subject material, foster improved performance on subsequent traditional assessments, and promote better long-term retention of force & motion concepts.

Cognitive Goal

Textbook problems generally don't incorporate more than one concept, and therefore don't reflect the kind of real-world problem solving that gets done by engineers. Instead, these problems typically involve the mapping of a single formula onto an idealized (and often abstract) setting. This means that students are less motivated than they would be if they could see a concrete and useful purpose for the exercise. Without an overriding goal it is also difficult for students to place each concept they've learned into a practical framework of knowledge.

We feel that via technology-base scaffolding, students might be able to harness multiple concepts covered via traditional means (e.g. textbook, lecture, lab, etc.) to solve these more interesting, motivating problems. Since real-world engineering problems significantly increase the cognitive load and the concepts to be employed would still be relatively new to the students, an expert (instructor) would normally be required to lead them through the process, either one-on-one (impractical) or through a class-wide problem-solving exercise (which wouldn't necessarily involve all students directly).

What kinds of technology would provide the best scaffolding for individual or small group activity (i.e. 2 or 3 students)? We hope that a hypertext-based tutorial could provide the average student with enough hints, direction, and supporting information to solve engineering problems involving multiple concepts. This kind of activity will also make the importance of each concept clearer by situating it within a larger, real-world context.

Implementation Overview

After covering several units on Newtonian force & motion, two dimensional motion, etc. students would be given 2-3 lab days to work individually or in small groups at a computer. They would use Netscape to surf their way through Sputnik, an interactive html-based tutorial directed toward solving a large realistic engineering problem: placing a satellite in geostationary orbit. The session would begin with a loosely defined goal (We'd like to keep an eye on the South American rain forest around the clock from space, so how would we go about solving this problem?) The first step for the students would be to develop this into a more concrete question to work on and eventually solve (What would the altitude and velocity of a satellite need to be for it to circle the Earth exactly once every 24 hours?) After the question was well-formulated, Sputnik would guide the student through identifying the relevant variables and choosing/applying appropriate formulas to arrive at numerical solutions.

Sputnik would accomplish each of these tasks by conducting a running dialog with the students. It would first pose a question and then ask the students to come up with as complete an answer as possible. If the question contained terms that might not be immediately familiar to all students, the terms would have hypertext links to supporting pages of information. After collecting the students' answer to a question, Sputnik would provide several responses of its own, each hyperlinked to a unique page. Incorrect selections would provide the students with scaffolding pages containing hints, diagrams/animations, explanations, etc. that would help put them back on the right track. The correct response would require the student(s) to type in a justification for their selection which would ensure that they didn't randomly select the response (or exhaust all possible responses).

In addition to making multiple choice selections, we also envision students placing check marks or other symbols beside elements in order to classify them. For example, students might be asked to identify which variables (e.g. the mass of the Earth, the mass of the satellite, etc.) are relevant to the problem. In this case, Sputnik might also collect text from the students characterizing the dependency (e.g. "The greater the mass of the planet you're orbiting, the faster you have to go to stay in orbit.") This kind of activity would also be iterative, with opportunities for revision throughout the session, initiated either as a specific student request or in response to prompting from Sputnik.

Periodically, Sputnik will also ask students to conduct self-assessment/reflection on their progress through the task, hopefully prompting them to evaluate and employ successful heuristic strategies. For example, it could ask whether the students have organized their intermediate results clearly, checked to ensure that the units make sense, analyzed dependencies between variables in a formula, etc.

All text typed by the students during the session (along with text identifying the multiple-choice selections they've made) would be collected into a single text file which would then be provided (automatically e-mailed?) to the instructor for assessment. Incentives could be given to the students for completing each part of the problem with a minimum of extra help. For example, extra credit could be given to students whose initial responses are largely correct, who follow a directed path through the activity, and who don't need to refer to the supporting information on the scaffolding pages. We also hope that by collecting this data from students, Sputnik will automatically generate the information required for us to revise and refine it.

Challenges and Opportunities

One particular danger with all computer-based group work is that one student may monopolize the keyboard, and thus the entire activity. We have several ideas for how Sputnik can prevent this from happening. When a group of students begins a Sputnik session, each will be required to enter his/her name. Throughout the session, Sputnik will attempt to balance keyboard time between students by asking each a question or two and then prompting them to switch places. For key questions, Sputnik may require all members of the team to respond, each in a separate box. Since the students will be asked to type a substantial amount of text as a first answer to a question and then more as a justification of multiple-choice or other quick selection, it should be relatively easy for the instructor to ensure that all students have contributed to the activity (i.e. by analyzing differences in writing styles). In addition, the breadth of concepts covered over the course of the session will hopefully motivate/require all students to contribute their own ideas.

We hope that by giving students the opportunity to work collaboratively on an engineering project, we can counteract the negative feelings many female students have toward engineering. However, because we hope to use student responses for assessment purposes, we're unable to allow the kind of anonymous contributions which seem to be so effective for female students. Hopefully, though, the fact that these responses will be seen only by the instructor will mitigate their reluctance.

As with most of classical mechanics, formulas are central to the solution of problems. Collecting these types of formal representations from students will be a serious challenge for us. For example, we need some way for a student to "type in" a response to a question like "How does the centripetal acceleration of the satellite depend on the radius of its circular orbit and the mass of the Earth?" We're currently investigating whether there's any simplistic equation editing Java code available. Our experience with the equation editor in Ron Avitzur's Graphing Calculator suggests that even if we do obtain good software, becoming adept at editing equations via a keyboard might prove to be a significant challenge for our students. We also hope to consult with Dr. Gifford's group on this issue as well.

The dependence of Sputnik on the browser and hypertext links means that proficiency at surfing through web pages will be useful for completing the task. An important design consideration will be how much extra scaffolding will be required to avoid distracting them from the subject matter. For example, can we rely on a student to press the browser's Back button, or will we need a "Back to [xxx]" link on each sub-page? It would be great if students could practice their browsing while working on the activity. Perhaps the best solution would be to provide a "How to I get back to [xxx]" link on each sub-page that linked to a page explaining how to use the browser's Back button.

Finally, we hope to scaffold all potential valid solution paths, so that the students can act as autonomously as possible throughout the activity. Some questions will therefore have more than one "correct" answer, leading the students down alternate solution paths. Potentially, we might also want to follow "invalid" solution paths farther as well, rather than immediately explaining their error and posing the question again. This should make the problem solving activity a little more authentic, since experts often must follow an approach for some time before they realize that it will not be fruitful.

Scaffolding Outline

Here's a brief outline of the scaffolding Sputnik would provide for the students as they worked their way through the problem. Note that the text below was chosen to be both terse and specific, and will be completely rewritten (expanded into paragraphs with associated diagrams, etc.) to make the dialog appropriate for high school physics students:

Simulating Problem Solutions

After coming up with a solution to the problem, it would be nice if students could use simulation software like Interactive Physics to "check their math." We've successfully built a simulation of the solution to the problem using this software, but we hope to extend it with appropriate graphics, analysis, etc. If you've got a copy of Interactive Physics, and you'd like to check it out, you can do the following (Note: the demo version 2.5, on which these instructions were based was recently removed from the Knowledge Revolution web site - bummer!): Note that the items in red above would be the values calculated by the students as part of the activity. We wonder how much of the simulation should be canned and how much should be designed by the students. We could take another step in that direction by having them enter values for the mass of the Earth and the satellite. We could also make the activity a little more fun by letting them draw a picture of the satellite and then use it in the simulation.

Presenting Results

Once the students have solved the problem and tested their solution, it would be productive to have them collect textual information saved in their Sputnik session and relevant data from the simulation environment, and then make a presentation to the rest of the class. This would obviously be much more useful if there were several different Sputnik projects to choose from, so that everyone wasn't making the same presentation. In this way, the whole class could learn from each group's work and have an opportunity to critique it.

Summary

We hope that by scaffolding the solution of a relatively large engineering problem, Sputnik might contribute to the skill development necessary for more autonomous engineering work, both within the domain and hopefully beyond it. In addition, we think an activity that makes use of several new physics concepts at once (as opposed to drilling concepts in isolation) could help students identify relationships between the concepts and fit them into a framework of knowledge useful for solving real-world problems.


Evaluation Plan

The main cognitive goal for our topic is to provide students with a larger, real-world context where they can incorporate the knowledge they've gained via a traditional physics curriculum. Often students wonder what inclined planes have to do with the real-world. We feel that by allowing the students to solve a problem more like those faced by engineers in the real-world, they can begin to see how the physics concepts they've learned in class can be applied to the real-world. We feel that hypertext is a good format for this because it allows the students come up with their own solutions through guided direction rather than just having the instructor put it on the board.

High school physics teachers and developers of future hypertext tutorials are the intended audience for our evaluation. Hopefully, our results will convince teachers that Sputnik is something that they want to incorporate at the end their unit dealing with two-dimensional motion in order to integrate all of the individual concepts that the students learned through traditional learning methods. Our evaluation should also allow other developers of curriculum to add to and/or change the current format to improve on the capacity for Sputnik to teach.

We will try our best to predict all possible outcomes of Sputnik, but inevitably there will be unanticipated consequences. We think a first step toward resolving these before they arise in the classroom is to have peers and high school physics teachers test it out. The second step is to get typical high school physics students to try it out before incorporating it into a real classroom. One area we're particularly worried about is scaffolding as many different solution paths as possible. We want to allow students to go about the task in the way that makes sense to them, thereby encouraging and developing autonomy.

One way to determine whether the complete curriculum is effective is to do a comparison test with another class that reviews the material without the use of Sputnik, preferably led by the same teacher. Since Sputnik is not a vehicle for introducing subject material, but rather a vehicle for integrating the knowledge that they learned in the classroom, it should be easy to see whether or not having the activity has had any effect under the conditions stated previously. Perhaps when the unit is completed, both classes could be given an exercise to solve which, like the Sputnik activity, requires them to employ multiple concepts. Since this activity would not be scaffolded, it would have to be much simpler than the Sputnik activity. However, if designed carefully it could still measure whether Sputnik has accomplished its primary goal: helping students apply the knowledge gained in class to solve real-world problems incorporating multiple concepts, thus proving that they've integrated the knowledge in a meaningful way.

Since Sputnik will constantly be collecting text and other responses from the students, there will be a rich set of data available for further analysis. This data will be most useful for incremental tuning of the activity itself, but we also intend to use it to evaluate the success of the intervention as a whole. Throughout the session, Sputnik will constantly be referring back to concepts covered earlier, thus forcing students to revisit questions which required the most scaffolding for them to answer. We hope that by comparing their initial responses to those they come up with later in the activity, we will be able to see incremental progress toward knowledge integration.

Sputnik could also periodically request feedback from the students on the activity itself, asking them whether or not they're finding it helpful and what in particular they consider most useful. In addition, there's a "Comment on This Page" button on every Sputnik screen for the student to click whenever they find the content particularly confusing. We hope that this will make it easy for our reviewers to suggest improvements during the design and initial evaluation phases.

Finally, in order for an intervention to be practical in a high school physics class, we assert that it will have to be successful as measured by the currently employed assessment methods in that class. It seems naive to expect the science department to accept both a radically modified curriculum and a new set of performance metrics. For this and other reasons, we have decided to position Sputnik as a review activity just before the standard unit exam on two dimensional force and motion. We hope that the performance of the Sputnik students will be at least as good as the performance of students who reviewed the material via more traditional means, and that this will lend weight to our argument that the intervention fits well within the physics teacher's curriculum.

How will physics teachers respond to our evaluation? Some teachers are going to be reluctant to use technology regardless of its merits. However, since Sputnik doesn't replace their entire curriculum and since it doesn't require much intervention on their part, we hope to minimize these concerns. In addition, by tying up most students in a self-directed activity, the teachers should be free to focus their attention on students who need individual attention. Furthermore, if the Sputnik students do as well as (or better than) the students in the control group using the teacher's own assessment methods, then we expect teachers to be more willing to seriously consider our analysis of student responses, which we hope will show incremental knowledge integration, and therefore more practical knowledge. This should make it easier to convince teachers to use Sputnik in their classrooms.


Geostationary Orbit Activity

Click here to see what we've got running so far.


Comments on Our Project

People are talking about our project! Follow the link below if you'd like to read what they've been saying:

Comments on Our Project


Web Sites Relevant to This Topic

Glossary of Satellite Terms - A pretty reasonable glossary. Click OK in response to the 2 JavaScript errors, and you're on your way.

Orbital Motion Theory - An overview of celestial mechanics.

Physics Hypertext Site List - A bunch of links to college Physics courses conducted on or supplemented by the web.

Rotational Motion Tutorial - A complete tutorial on rotational motion. There are a few self-test questions, but you just get a "correct!" or "no, you should have done it this way..." answer, rather than having different information and/or questions provided in response to the specific answer given.

Edward (Joe) Redish - University of Maryland Physics professor who's done a fair amount of web-based tutorial and assessment development. Unfortunately, I didn't find much content online.

The Interactive Physics Textbook - Doesn't seem very interactive to me, but I didn't try the Java version.

ACT Group - This is John Anderson's web site, which contains lots of information about his intelligent tutoring systems.

Knowledge Revolution - The company that wrote Interactive Physics, simulation software for classical mechanics.


Back to Schmed's EMST 291B Work

Back to Schmed

Back to EMST 291B

Please e-mail comments regarding this web site to:

Schmed@TransPac.com and Eunice@UGAstro.Berkeley.edu.

© 1997-2000, Eunice Yi and Chris Schneider. All Rights Reserved World Wide.

Last updated 20 March 2000