As the freezing rain coated the outside of the restaurant window, I could understand my waitress emphatically asking me why I had come to New York in the winter. Economists all over the east coast of the United States must have been asking themselves the same question, as they struggled with disrupted air schedules and rental cars to get to the 1999 ASSA meetings. Tod Porter and I breathed a sigh of relief as the weather improved, and all of the presenters for the Computer Assisted Instruction (CAI) sessions let us know that they had arrived.
Betty’s session, titled "Evaluation of Learning," included four presenters.
Carol Johnson presented the results of a study conducted in her department to assess whether group-based international case studies and computer learning modules affected student attitudes and performance. The cases were available through the internet and incorporated a set of tasks for students to perform, in addition to providing case materials. Student responses to the cases and modules were assessed through a student attitude survey, individual student interviews, observations of the group case studies, a mid-semester review, exit surveys, exam performance, module performance, and a quality of teaching questionnaire. A regression analysis showed that the computer modules had a positive effect on student exam performance for students repeating the introductory macro course but had no significant effect for students taking the course for the first time. The group case studies did not affect student performance for either group of students. The teaching evaluations of the instructors using both innovations were slightly higher, and student attitudes were slightly more positive toward the course material taught with the cases and modules. Finally, Carol noted that peer pressure effects were apparent in student responses to the group cases. The department’s work with technology and samples of the cases and modules can be seen at http://www.economics.unimelb.edu.au/rdixon/101/101home.html.
David Brown discussed which computer innovations had yielded the highest benefits for students on the Wake Forest campus in general and for his first year seminar course in particular. Wake Forest is one of the few campuses in the United States with "ubiquitous computing." Students receive laptop computers when they enroll as freshman, and the laptops are replaced every two years. All classrooms and dorms are wired for the student laptops. The general findings from surveys of Wake Forest students over the last several years have been remarkably consistent. When asked to assess the impact of the Wake Forest approach, 20 percent find that computers improve classroom presentations, 35 percent find that the laptops provide more opportunities to practice and analyze problems, and 43 percent feel that the laptops provide more access to source material via the internet. Most significantly, 87 percent report that the wired campus provides more communication among students and faculty. Adding data from other heavily wired universities in the United States, David argued that the major impact of computers has been to facilitate collaboration and build communities. To date, email has proven to be the most effective instructional tool, while PowerPoint slides and treaded discussions have proven to be the least effective. Many instructors who have tried PowerPoint slides and treaded discussions have dropped both from their current courses.
David also discussed the organization of his first year seminar course and the degree of interactivity allowed by the Wake Forest approach. Email is used to update the syllabus, elaborate homework, and send out announcements. Students take a one-minute quiz at the end of a lecture and hand in the quiz by email. Groups of students edit rough drafts of each other’s papers by email. When students were asked to assess the seminar relative to other first year courses, 2/3rds reported that they learned more and 1/3rd reported that they learned about the same, while 2/3rds reported that they spent the same amount of time and 1/3rd reported that they spent less time. Virtually all the students responded that they enjoyed the wired course more. A link to David’s course materials can be found at http://www.wfu.edu/~brown.
Jim Clark and Jan Wolcutt presented the internet course they are developing to assist K-12 social studies teachers to prepare their students for the Kansas Social Studies Assessment. Organized around the Voluntary National Content Standards, the course requires no campus visits and is structured to maximize interactivity in a web-based instructional environment. Software used to create the course includes Front Page, Power Point, Real Audio, and Macromedia’s Flash and Director. Jim and Jan showed several examples of how they use graphics, humor, and animation to maintain student interest while completing course assignments. Their course materials can be viewed at http://www.mrc.twsu.edu/economics.
Jim and Jan also shared their "top ten" list of things that they had learned about doing classes over the internet: (10) No matter how much time you allow to get it up and running, it isn’t enough. (9) You can’t do all the newest, flashiest stuff because your target group probably won’t have the hardware or software. (8) No matter how much RAM your computer has, it isn’t enough. (7) No matter how big your hard drive, it isn’t big enough. (6) No matter how much you pad your cost estimates to the administration, it will cost more. (5) The funding never comes when they say it will. (4) If you are facing an important deadline, the technical support people disappear. (3) No matter how easy a technology is, it won’t work easily for you. (2) The Pareto Rule applies. The last 20 percent of the work takes 80 percent of the time. (1) Internet courses are not the way to generate mass credit hours with little time input.
Geoffrey Gerdes gave an overview of the Instructional Enhancement Initiative currently underway at the University of California at Los Angeles. The initiative was launched in 1997 at the directive of the UCLA Chancellor and has come to represent an ambitious plan to strengthen undergraduate instruction through integration of computing technology with the curriculum. The Economics Department is pursuing the increased use of information technology as part of a long-run plan to revamp the entire undergraduate curriculum in response to the initiative.
Geoff summarized the results of the first year surveys of students and faculty for courses making extensive use of new web materials and presented examples of his collaborative efforts with Trudy Cameron. Students have responded favorable to the web materials, but not surprisingly object to the student fee that was imposed to pay for the initiative. In terms of the actual instructional materials, Geoff argued that effective web pages need to include truly interactive features. Using Java applets, or something similar, web pages should incorporate at least display interactivity that allows students to vary parameters and observe results. Interactive materials should be supplemented with testing materials to challenge students. Interactive quizzes can lead students through the pedagogical design of the applet, asking questions that require manipulation of the economic model and ultimately promoting conceptual understanding. Geoff and Trudy’s paper can be viewed at http://jevons.sscnet.ucla.edu/gerdes/cai/ [Now moved to http://www.econtools.com/jevons/cai/ - Web Editor]. The paper includes several Java applets and a full discussion of the work being done in various economics courses.
Scott Simkins, North Carolina A&T State University, was the discussant for the Johnston and the Clark-Wolcutt presentations. Scott noted that the results for the Melbourne experiments show the challenges of trying to use computer technology to improve student learning. As in most studies measuring cognitive gains related to computer assisted instruction, the gains remain elusive. Particularly disappointing was the failure of the technology to reduce the high non-pass rate for the course. Scott had some difficulties with the conduct of the study. The study design did not include a pretest and posttest and relieved on only one or two questions to measure student learning outcomes. Although student grades in previous economics courses were included in the regression to measure the effectiveness of the interventions on performance, there was no attempt to control for the overall level of preknowledge (such as through the use of something similar to SAT scores in the United States).
In commenting on the Clark-Wolcutt presentation, Scott noted that there is a massive movement to migrate toward the web in academia prompted by prospective cost savings, the desire to promote distance learning, and the desire to provide increased access to college courses. The Clark-Wolcutt presentation is the story of two teacher’s personal trek to the web with a result that is becoming all to familiar; it takes a lot of time, effort, and money to provide Wed-based instruction, with uncertain benefits. Web-based instruction is most likely to work when there is a clearly defined need (in this case state testing requirements) and the spatial separation of participants with significant time constraints. This is why we have seen web-based instruction occur first in MBA programs. However, in general, web-based courses present a sever challenge that few have as yet mastered.
Kim Sosin, University of Nebraska at Omaha, was the discussant for the Brown presentation and the Gerdes presentation. In discussing the Brown presentation, Kim gave several examples from her experience that confirmed the general conclusions about the efficacy of computer technology. In her experience, threaded discussions had proved to be particularly ineffective. The only exception to this, noted with tough in cheek, was when she suggested that members of her department could avoid a department meeting if they could resolve an issue by using a threaded discussion. She also commented on the very special nature of Wake Forest. Few public institutions could come close to providing the ubiquitous computing environment associated with the first year seminar, which clearly gives students a superior educational experience. Finally, Kim indicated that she thought Geoff’s paper went right to the most significant issues all economic departments are currently confronting in deciding what role the web should have in their instructional program. The question of how much software development a department should do on its own is especially troubling and is much more easily answered by a doctoral granting department like UCLA with graduate assistants like Geoff.
Tod’s session, titled "Course Applications," also had four presenters.
Craig Marcott demonstrated several Mathematica notebooks that featured animated economic models. He argued that such notebooks encourage active learning by allowing students to change parameters and set simple VCR-like controls to change the direction and speed of the animations. His presentation was given using a CD, and he distributed copies of the CD to the audience. Several of the Mathematica notebooks are also available at http://milkweed.econ.stthomas.edu/~csmarcot/my.html [This address is no longer valid. -Web Editor]. The CD distributed at the session can be obtained by contracting Craig. Instructors, students, and undergraduate teaching assistants at the University of St. Thomas are currently using the notebooks. Since all classrooms, offices, computer labs, and dormitory rooms are equipped with high speed T-1 internet connections, users generally load the notebooks directly from the Economics Department server. Craig has observed that the notebooks are especially effective in helping students during office hours.
Two particularly interesting notebooks demonstrated by Craig illustrated the level curves of a CES utility function and the Lorenz Curves for several countries over several years. Mathematica’s ContourPlot command allows level curves to be animated, and the CES notebook showed how the indifference map changed as rho changed, something that is very difficult to show with traditional teaching methods. The map changed quickly as rho varied between –1 and 1, and varied much less for changes in larger values of rho. The animation nicely illustrated the limiting Leontief shape as rho increased to infinity. Using data from Deininger and Squire (1996), the Lorenz Curve notebook used a Gini coefficient "speedometer" to show the changes in the Gini coefficient and cumulative quintile shares over the years 1947-1991.
Winfried Reiss demonstrated a set of Java applets for interactive teaching on the web called the JEWEL system (Java Enabled Worldwide Economic Learning). The program featured a set of "books" for different topics. Winfried had some difficulties with his computer setup, but he was able to show most of the general features of the books. Particularly interesting were the context sensitive help files, which appear as HTML files. The workbooks allow for a high degree of student interactivity, creating true computer texts for different economic topics. The best way to see the structure of the JEWEL system is to walk through the presentation itself, which can be viewed at http://econscience.uni-paderborn.de/Java/1-1/Books/Index-0.htm (JEWEL has now moved to http://pbfb5www.upb.de/www/vwl/vwl8/vwl8.nsf/JEWEL?OpenView -Web Editor).
Miles Cahill and George Kosicki argued that spreadsheet programs have improved to the point that spreadsheets can be used to make a wide range of advanced topics more accessible to students. Their paper gave several Excel examples from the intermediate theory courses and showed how models can be explored more thoroughly in the classroom because of technological advances in spreadsheet software. The examples showcase the range of spreadsheet features, from three-dimensional graphics to constrained optimization. The examples included utility functions and production functions, substitution and income effects, consumers’ surplus, cost minimization, oligopoly models, price indices, the consumption function, the IS-LM model, the AS-AD model, and the Solow growth model. Miles and George only had time to demonstrate two examples during their presentation; how to use Goal Seek to assess substitution effect bias in the consumer price index, and how to use Interation and Solver to evaluate the Solow growth model. The paper is currently under journal review and a revised copy can be obtained directly from either Miles and George.
Lydia Ortega demonstrated a multimedia product called ProEcon and discussed the software features needed for effective student learning. Specifically, ProEcon is designed to mimic a carefully presented classroom lecture. Student attention remains focused on key elements as the "lecture" develops on the computer screen. The demonstration used the basic demand function as an example. Each point was carefully detailed, with highlighting used to emphasize the most significant points. The software design kept the focus on the development of the key demand concepts and reinforced the presentation with voice and graphical material. The end product of the software is to try to get students to "think like economists." Lydia also discussed a self-test program for economics students called Patterns in Problem Solving. Anyone wishing more information about these programs, their current state of development, and projected availability should get in touch with Lydia.
Joseph Daniels, University of Delaware, was the discussant for the Marcott presentation and the Reiss presentation. In commenting on Marcott’s work, Joe discussed the general strengths and weakness of Mathematica, which he argued were mirrored in Craig’s notebooks. Mathematica provides a reasonable learning curve for producing high quality graphs and animations. It is also possible to include equations and parameter values so that students can experiment with changes in the graphs. But, while admitting he was a big fan of Mathematica, Joe went on to describe some significant difficulties. Mathematica has a large computing overhead. You simply must have lots of memory, processor speed, and disk space to use the program efficiently. Mathematica is also significantly more expensive than most similar software products and has a steep leaning curve for students. It does not do real-time animations. It calculates all the graphs first, then displays them sequentially, making it difficult to experiment with changes in an animation. The program also has limited multimedia features. These features of the program are responsible for some of the limitations associated with Craig’s notebooks. Craig must hide the equations and parameters, so the animations remain "black-boxes." Consequently, students cannot manipulate parameters and observe the results. Student interaction with the models is thus quite limited.
In commenting on Winfried’s work, Joe stressed that he believes Winfried’s basic approach uses the correct developmental model, i.e., the production of interchangeable software components that can be customized by passing parameters through statements that call the program. It is highly desirable that such software be distributed on the web due to the associated simplicity, flexibility and ease of distribution. Joe had a few specific comments about the design of Winfried’s software. He felt that it would be better if the components were designed for display directly on the web page rather than in a separate frame. This would improve the flexibility for displaying the software components by allowing HTML to handle the layout and integration with the text. It would also improve the portability of the components by forcing them to be designed as stand-alone applets.
Michael Lovell, Wesleyan University, was the discussant for the Cahill-Kosicki presentation and the Oregeta presentation. Michael talked about the way spreadsheet design had improved over the years and noted some of the strong pedagogy associated with both presentations. He also discussed the need to have a much stronger organizational structure for the dissemination of information about computer assisted instruction in the economics profession. The recent decision of the Journal of Economic Education to start a software review section is a big step in the right direction.
Once again this year, Bill Goffe and Bob Parks gave a session on "What’s on the Internet for Economists? A Demonstration and Update." They added significantly more information this year on the future of internet technology and its implications for educational use. Their demonstration can be viewed at http://wuecon.wustl.edu/~goffe/ASSA.99.html. New to the site this year is a automatic "slide-show" presentation that allows you to sit back and watch the ASSA presentation as if you had been in the room when it was first projected.
Anyone interested in attending next year’s CAI sessions can see the preliminary program at the CAI website (http://online.sfsu.edu/~bjblecha/cai/cai.htm). Please contact Betty Blecha (bjblecha@sfsu.edu) or Tod Porter (tsporter@cc.ysu.edu) for any additional information.