This year's programme for the Allied Social Science Associations meeting in San Francisco included one session on Computer Aided Instruction (as they still like to call it) and a "What's on the Internet for Economists?" update which was available in three different sessions at the conference.
The first paper was by Arnie Katz of the University of Pittsburgh. As a revised version of the paper is printed in full elsewhere in this issue of CHEER I will keep my comments short. Katz stressed the point that the technology we now have can be used not just to deliver learning material to students, but also to keep track of how they learn. He illustrated his point by describing work that he has been doing with students who have used the Smithtown collection of software. The tracking routines are able to provide information on the different types of errors that students make when answering questions, how many attempts they have to get the correct answers, the time spent on each question etc.. It was particularly noticeable that students were making errors because of an inability properly to distinguish between endogenous and exogenous variables (over 40% of the errors were attributable to this source). It was also clear that some students become obsessional in their attempts to get all answers correct, even though their results do not count towards their grade. Another interesting feature revealed by the study was the pattern of errors over the learning phase. In the first part of the course there were only a small number of errors as students drew on what they knew before the course. As they got into more unfamiliar territory the errors increased, before reducing towards the end as the work they put in helped to put things into place. It was also possible to study responses in a sequential way, looking at consecutive right or wrong answers. From the information extracted a lot could be discovered about the diversity of learning styles and predictions could be made about final achievement on the basis of in-course performance.
Katz stressed that the data collection was unobtrusive through the recording of mouse clicks, although students were aware that their performance was being monitored. (On reflection could this explain their obsessional behaviour in trying to get everything right in that they were trying to impress upon their instructors how keen they were?).
The second contribution was from Ted Osborne of the University of Western Ontario. He talked about the use of the program Canvas 3.5 as a tool for the construction and manipulation of graphs for economics applications and demonstrated its use for the construction of marginal and average cost curves. Through the use of tangent control points and Bezier curves the package draws smooth and mathematically exact curves. Further applications involving the construction of demand curves from an indifference map were given to illustrate the point that here is a simple and effective way of producing accurate graphs for on-screen tutorials.
The third paper was from Pauline Crichton and me on the evaluation of WinEcon. We stressed the importance of evaluation for the project and distinguished between the quality assurance phase of the project prior to general release and the on-going post release phase which would attempt to monitor the way that the package was being used in teaching and learning and look at how effective it was for different types of learners and in different learning environments. We made a further distinction between subjective evaluation responses (the appeal of the product to users) and objective measures (the effectiveness of the product in improving understanding and efficiency in course delivery). Pauline explained how we have made use of a computerised version of the Kolb learning style inventory to classify students before they use WinEcon so that learning style scores (together with other variables representing student characteristics such as age, gender etc.) could be used in regressions to account for variations in student performance on the course and measures of the attractiveness of the software to them.
Preliminary results appear to confirm the view that students who score highly on the Active Experimentation dimension of the inventory have a natural predisposition towards interactive software like WinEcon. It also seems that WinEcon provides greater assistance to such students in developing their understanding of a topic than those students who score highly in other areas of the Kolb test such as Abstract Conceptualisation. The presentation ended with a brief outline of future work that is to be undertaken on the project. The questionnaire is to be revised and automated so that student responses can be collected automatically. An Internet discussion list is to be introduced for WinEcon users (staff and students) to provide an efficient means of collecting comments and problems raised by users and in feeding back suggestions to other users.
The last presentation in the CAI session was by Richard MacMinn of the University of Texas who gave an illustrated talk about his use of the Internet in teaching business finance. MacMinn has three years of experience in using the Internet in teaching, delivering course material, connecting students to data sources, testing them on-line and establishing an electronic communications link between students and teachers. Some course assignments have been set and submitted via e-mail, with Word and Excel attachments. Information on grades was returned to students though the same medium. MacMinn also provides material via the World-Wide Web. Readers might like to point their browsers at http://kiwiclub.bus.utexas.edu/finance/kiwiserver/fin357m/fin357m.html [No longer available- Web Editor] to see BizFinWeb where MacMinn has his course material marked up in html form, together with selected problems, grade distributions etc.. You might also want to take a look at http://medweb.bham.ac.uk/http/caa/newdb/macminn.html to see a couple of quizzes on Capital Budgeting that MacMinn has set up using the MedWeb Computer Assisted Assessment framework created by David Davies at the University of Birmingham Medical School.
MacMinn talked briefly about some of the problems of using the Internet in this way. First, there is the investment which is needed, not only in hardware but in the development of software skills. It is always worth looking to see if there are facilities elsewhere on which you can piggy-back - the MedWeb software at Birmingham provides a good example. There is, of course, the problem of servers going down and poor response times during busy periods. The former problem can be mitigated by sharing resources between universities and establishing mirror sites to provide alternative sources for some information. Another hazard is that students who can send you e-mail messages expect immediate responses and this can be a problem with big groups. MacMinn noted that a fully computerised course which is delivered via the Internet requires a lot of organisation if you want to be organised. However course material can be quickly and seemlessly updated. Returning to Arnie Katz's theme, when you run your course via the Internet you can keep track of users' progress and generate a lot of useful data automatically.
MacMinn said that his brief presentation had only been able to scratch the surface of what can be done. He is planning to move to Lotus Notes, which is used by many large corporations, and also to make use of the Adobe Acrobat portable document format which can have embedded links for a more efficient delivery of course material.
The panel of discussants was made up of Betty Blecha, Tod Porter, Michael Lovell, and Paul Grimes. Betty Blecha said that she liked the way the software was developing, giving us the ability to collect data on how students learn. She too was interested in investigating different student learning styles and referred back to her paper at a previous meeting where she had used the Myers- Briggs test to analyse the learning styles of students using Mathcad. She was interested in the different ways that students work on the Web - some read the whole page while some click immediately to jump to another site. We should pay some attention to how students use time, particularly if we move towards a more student-centred approach.
Tod Porter, commenting on Ted Osborne's paper said that the main problem with it is that it takes away our excuses for not drawing AC/MC curves accurately! Porter uses AmiPro (which also makes use of Bezier curves) in a similar way. Paul Grimes had seen the WinEcon software (which was on the Blackwell's stand in the publishers' exhibition area) and he said he was very impressed and encouraged everyone to go and take a look. He was rather more critical of the WinEcon evaluation paper but appeared not to have recognised fully the preliminary nature of the work. Hopefully he will be be more satisfied with our analysis when the results of a wider survey of WinEcon users are available.
Mike Lovell admitted that he was a "klutz" at mastering new technology, but said that the Netscape Navigator software had opened up a whole new world to him, and he was most impressed with what MacMinn had managed to do on the Web. However what we desperately need now, he said, was a coach to help us in this new ballpark. He found that he couldn't rely too much on the support personnel at his university. This was a cue for the Goffe, Greenwade and Parks sessions on the Internet and provides further justification for the steps taken in the UK to set up the SOSIG service at Bristol.
When the discussion was opened up to members of the audience there was some discussion about student use of e-mail. Blecha noted that e-mail could empower some students who for cultural or other reasons didn't speak up in class, but who would contribute on-line. A member of the audience was concerned, however, that if e-mail contributions were to be made compulsory this could disadvantage other students who had difficulty expressing themselves in the medium. Concern was also expressed that if all students were to start sending regular e-mail messages to staff then we would quickly become overwhelmed with e-mail. I suggested that if automatic facilities for separating mail from different users into different folders were not available then lecturers should maintain two e-mail addresses - one for student mail and one for other messages.
I also attended the second of three identical sessions given by Bill Goffe, George Greenwade and Bob Parks on "The Internet for Economists: Update" (see CHEER Volume 9 Issue 1 pp 19-21 for a printed version of last year's original presentation). Goffe, who is well known for his "Resources for Economists on the Internet" guide (available at http://www.rfe.org), conducted us through most of the presentation (with Bob Parks on keyboard and mouse). The presentation was intended to provide a point of entry for new users and an update on new material for people who came last year (about half the talk was devoted to new stuff). A four page handout was provided which listed the services covered, together with a separate two page handout on the Economics Working Paper archive at Washington. The talk itself was a mix of Power Point images, recorded sessions and live on-line experiences. The latter were restricted because the partial government shutdown which was then in force meant that some services were temporarily unavailable.
Goffe began with a general outline, giving a brief history of the Internet and distinguishing some of the different tools and services that you come across on it. The number of networks linked to the Internet had roughly doubled since his presentation last year, he said, and he put on screen a map showing connectivity around the world. He showed graphs displaying the growth in users, Web sites and services by country and by type and quoted a number of key indicators to demonstrate the increasing importance of the Net (over 24m adult users in the US, 5% growth per month, nearly 13 thousand known mailing lists, increasing use of URLs in advertisments etc.).
He spend some time explaining the structure of Host names and IP addresses and talked briefly about the Domain Name System for translating human readable versions into the numerical formats. He talked about the latest trend in interesting and personalized host names, mentioning several of the more striking ones (badbreath.com for Proctor and Gamble, geeksrus.com and most.weird.com).
Goffe talked about the culture of the Net and the importance of the work of volunteers who obtained non-pecuniary returns from their competitive hobby, mentioning too some of the services which had begun as unpaid activities but which had then turned commercial. He touched on matters of "netiquette", security and encryption, and the way in which mailing lists operate.
At this point Bob Parks started up Netscape and went to a few newsgroups and Usenet groups, starting with the clari.biz.earnings which is part of the ClariNet service. He explained that, unlike the mailing lists, this is a paid subscription service so that some sites would not have access. The next section of the talk contrasted telnet, ftp, gopher and the Web. Using a visit to the Iowa Electronic Markets, Greenwade showed how you can now telnet from within Netscape. To illustrate ftp, and to show how some journals have established archives of data sets used in published papers, we went to the Journal of Business and Economics Statistics archive at Duke University (ftp://ftp.duke.edu/jbes/). He guided us through to the data from an October 1993 paper by Engle and Kozicki and showed how it could be saved to disk and loaded into Excel. This was all accomplished very quickly and showed the ease with which it can now be done.
Goffe briefly illustrated gophers by taking us to the NBER table of Business Cycle Reference dates (again from within Netscape - gopher://nber.harvard.edu/70/11/) [This seems to be no longer available- Web Editor] illustrating the use of Bookmarks to speed things up, before the talk concentrated on the World-Wide Web. Bob Parks took us to his favourite movie database before turning less frivolously to the Applied Economics Letters site on the Chapman-Hall publishers computer (http://hermes.chaphall.co.uk/al.html) [now at http://www.tandf.co.uk/journals/routledge/13504851.html -Web Editor]. Here we saw how to configure Netscape to view the file and we looked at a paper by Geraint Johnes. This led on to a discussion about various file types and associated software: .exe (self-extracting), .ps (PostScript), ,pdf (Adobe Acrobat) and .zip (WinZip). There was also a discussion about helper software and how to obtain and install it.
The next part of the presentation looked at where to find information. As well as referring people to his own guide to Internet resources (http://wuecon.wustl.edu/EconFAQ/EconFAQ.html), Goffe talked about the NetEc project (http://netec.wustl.edu/~adnetec/local/NetEc.html) and Bob Parks' Working Paper Archive (http://econwpa.wustl.edu/wpawelcome.html). He also discussed a number of information gateways and search engines, including the University of Washington's lengthy list of links to indices and searching tools which can be found at http://wuecon.wustl.edu/other_www/searchwww.html [Link out of date - Web Editor].
We then visited a number of database sites including FRED (the Federal Reserve Economic Database at the St. Louis Federal Reserve Bank - http://www.stls.frb.org/fred/), the NBER searchable Macro-Historical Database (http://nber.harvard.edu/macrohist.html [This seems to be no longer available -Web Editor]) where we searched for (and found!) information on kitchen sinks, the Panel Study of Income Dynamics at the University of Michigan - a longditudinal study of a representative sample of US individuals and families going back to 1968 (http://www.umich.edu/~psid), and the Penn World Tables (http://cabsim.epas.utoronto.ca:5680/pwt/pwt.html). [From the Web Editor: For UK users see also the BizEd site for a highly accessible and user friendly front end to this excellent source of data!]
He talked about the benefits of having data on-line and the advantages of having simple and easy to use access protocols via the Web.
The presenters (I'm not sure which - the pace was relentless and by now I was having difficulty in keeping track!) then returned to journals and the Internet. Particular reference was made to JSTOR, the Journal Storage Project [now mirrored in the UK at http://www.jstor.ac.uk/ -Web Editor, May 1998] for the development of a digital library in support of the arts and sciences funded by the Mellon Foundation. Initially ten journals in the areas of economics and history are being scanned in, including the AER, Econometrica, the JPE and QJE. Access is currently restricted to the five sites which are participating in the experiment, but there are plans to broaden access to the database in the near future.
Lastly we took a quick trip around a number of other interesting sites including the JAVA programming site (http://java.sun.com/ - look in particular at the Financial Portfolio page and the StockTicker applet), CNN (http://www.cnn.com) and Pasi Kuoppamaki's economics joke page (http://www.etla.fi/pkm/joke.html - there's even one there from CHEER!). Of course when one is in San Francisco one must visit all the real sites - the Golden Gate Bridge, Fisherman's Wharf, Ghirardelli Square etc. (not to mention taking a trip out to the Napa Valley!). But economists in the city should also make a visit to the Federal Reserve Bank of San Francisco to see their lobby display "The World of Economics". Here with a variety of exhibits the bank attempts to explain to the general public how the economy works and what it means to "think like an economist". There is a good mixture of practical hands-on interactive displays to convey an understanding of economic concepts and charts and newspaper cuttings to illustrate historical events. With a touch screen robot to guide you around - its a bit like a walk-through WinEcon!
Final note : a bookmark file containing the links covered in the Goffe, Greenwade and Parks presentation is available from http://wuecon.wustl.edu/demo/aeabkmk.htm [No longer seems to be available at that address- Web Editor].
Guy Judge
University of Portsmouth