CO 68.3 [Spring 1991], pp. 100-102: Tutorial software discussed; Crown dissertation; Flex, FasText, EZ-Comp, Latina Nouns/Verbs; Paidagogos, reVERBerations; Macintosh LC with Apple IIe card; Text Encoding Initiative

In the last two columns I have attempted to define and discuss the types of instructional software commonly referred to as simulations and drill-and-practice programs. According to the classification scheme which I have found to be most predominant, the third and last category is known as tutorial software.

This category is easily the least available of the three, largely because it involves the most complicated programming. By the definition which I set out earlier, tutorial software is designed to automatically reorder its sequence (of questions or challenges) according the user's varying degree of success. It may also attempt a more complex form of evaluation than simple percentage or point scoring.

To expand on this definition: tutorial software is programmed to mimic some aspects of the human tutor. This means that the program will be given some very specific objectives for student performance and the ability to change its own course if the student does not respond successfully enough. Practically speaking, this means that the program must keep more detailed records of student performance so that it can verify that all prerequisite topics have been passed satisfactorily before a new topic is presented.

As you might imagine, the general objective of tutorial software--to mimic a human tutor--is one of the more controversial areas of educational technology. Although I have heard very few instructional designers claim that any current software can actually replace a human tutor (or teacher) on an equal footing, I know there are still high-level researchers in the field of artificial intelligence who hold that to be their ultimate challenge (and, therefore, their goal). One of the more gratifying things I have learned from my own survey of the research literature is that the achievement of this goal is still subject to severe doubt by a significant contingent of researchers--enough so, I would say, to make it unexpectable in the foreseeable future.

Of course, this does not prevent software designers, including myself, from taking on the challenge of providing as much tutorial assistance in a program as possible. In fact, my committee colleague, Rickie Crown, and I have spent a good bit of our own research efforts in the last several years determining what types of tutorial software are actually useful and effective in the Latin classroom, given the current practical constraints of hardware and software and the psychological constraints of everyday teaching environments.

Rickie recently completed her doctoral dissertation in Educational Psychology at the University of Michigan, with much assistance from Glenn Knudsvig in the Department of Classics. The dissertation is entitled "A Comparison of the Effects of Traditional Instruction, Tutoring, and Software Tutorials in the Latin Classroom," and it is probably the most thoroughly documented piece of research directly applicable to instructional microcomputing in our field. (Contact: Rickie Crown, Baker Demonstration School, 2840 Sheridan Rd., Evanston, IL 60201-1796)

Because there was no real tutorial software available when she began her research, Rickie wrote her own program to teach the topic of direct and indirect discourse. She tested the program in two similar high schools in the Chicago area and compared its effectiveness against that of both full classroom instruction and one-on-one human tutoring.

While Rickie was doing her work under a rather strict statistical format, I was conducting somewhat less formal research at both the high school and college levels in Madison, Wisconsin. Not too surprisingly, some of our general conclusions turned out to be rather similar.

What we both found was that: 1) Students quickly get frustrated and annoyed with tutorial software which maintains strict control of progress through the material. 2) Student response to tutorial Latin software depends very much on each student's attitude toward both computers and Latin in general, based on previous experience. Thus, requiring the use of such software as part of the curriculum can be deleterious to some students' progress in the course, while just making it optionally available can spur others on to unexpected advances.

Rickie's recommendations to anyone considering the use of tutorial software in the Latin classroom: be sure that the methodology of the program is in tune with your own teaching approach, and check to see that students are proficitent in any prerequisite skills or knowledge before setting them loose with a new program.

Let it be known that there is a new 1991 update of the Survey of Latin Instructional Software for the Microcomputer available from the ACL's Teaching Materials and Resource Center (Item #B219). It now lists 40 programs and includes a critical review of each program, along with a general introduction to evaluating instructional software.

I would particularly like to draw your attention to several new programs in the listing which were written and published by Latin teachers themselves. They may not have all of the professional glitz and polish of some of the "big name" publications, but their prices are not as "big time" either.

Those of you who attended the ACL Institute computer workshops in 1988 and 1989 may recall the set of IBM-compatible programs that Allan Wooley demonstrated there. He has developed them for his own classes at Phillips Exeter Academy and is constantly making improvements to them. Flex is a verb form drill which randomly generates items based on customizable parameters. FasText is designed to assist new Latin readers by providing assistance on syntax and vocabulary for a passage with a color-coding and pop-up window scheme. EZ-Comp tests a user on Latin composition, using full sentences from Fabulae Faciles. Roots is an etymological matching quiz program which demonstrates the connections between Latin and Greek words and their English derivatives.

Those who are looking for IBM materials (sparse as they are!) may want to give them a try, especially since they are very reasonably priced. Each program is $15, and a "combo" demo disk, containing demo versions of Flex, FasText, and Roots, is $10 (+$1.50 postage/handling). Contact: Allan Wooley, 10 Whitley Rd., Exeter, NH 03833.

Another New England colleague has written a very handy pair of Apple II-compatible programs. Marcia Jones has been trial-testing her programs, Latina Nouns and Latina Verbs, at Berkshire Country Day School for more than four years now. The two drill-and-practice programs complement each other and follow similar formats. The user may specify one or more declensions (or conjugations) for drilling. In the verb program the user also specifies active or passive voice and present tense only, present system, or all tenses. Each English form is randomly created from a vocabulary list of 245 nouns or 150 verbs, and the user must create the Latin counterpart. There is also the option of separate quizzing on stem and ending for each form, and the verb drill will check to see if the user knows the first principal part before requesting the necessary stem. After each form-item the user is given the opportunity to quit.

The programs are loaded into the computer in their entirety at the start, so it is possible to use one disk to load the program into many computers one after the other--effectively providing most of the convenience of a site license. One backup copy may be made. The cost of the programs is $39.95 each. Contact: MVJ Software, 64 Bartlett Ave., Pittsfield, MA 01201.

It took awhile to sneak a microcomputer programmer into the bastion of classical Greek instruction, but it's finally happened! And it happened on an IBM-compatible, no less! Michael Swanson of Vanderbilt University has written Paidagogos to provide computerized exercises for teachers of introductory Greek. It is not correlated with a particular text, but seems fairly accomodating as an elementary workbook.

The program covers a very comprehensive variety of 26 topics, from recognizing the alphabet to conjugating in the middle voice. There are 15-32 multiple choice questions included in each topic quiz. A student receives one chance to select the answer; a stinging beep notifies the world of a wrong choice. At the conclusion of each drill, a percentage score is provided, along with the opportunity to print out a short record of the session.

The Greek is quite readable, though the character sizing is somewhat uneven. Breathing marks are used but not accents, due to the low resolution of IBM-CGA graphics. Basic graphic capability (CGA) is required. The publisher offers several purchase options: individual pack, $25; lab pack (5 copies, 1 manual), $50; site license, level 2 (4-year college) $375; level 1 (other institutions), $250. Contact: National Collegiate Software, Duke University Press, 6697 College Station, Durham, NC 27708, (919) 684-6837.

Allan Wooley (mentioned above) has also written a Greek verb drill program for IBM-compatibles, called reVERBerations. This program has several different modes: passive review, form creation, morphological labeling, and error detection; it is probably best used as a review for more advanced students. See contact/cost info above.

Of course, the Macintosh has generally become known as the easiest and best machine to use for Greek because of its superior graphic capabilities (see "Random Access," CO 67.1). However, I am aware of only one instructional program for Greek on the Mac, and that is a Koine Greek program from the University of Minnesota. If anyone is aware of other publicly available Greek instructional programs for micros, please let me know about them so that we can expand the Survey to include Greek, too. (Please note that, by the definitions of instructional and educational software I have used [see "Random Access," CO 67.3], the Perseus Project would be considered an educational, rather than instructional, software tool and resource, albeit an extremely powerful one.)

Apple has finally produced their long-awaited "crossover" computer, and it is called the Macintosh LC. The "LC" supposedly stands for "Low Cost," but that appellation may be debatable for educators. It does, in fact, pack a lot of "bang for the buck," but it doesn't quite fit into the old Apple II-Mac Plus under-$1,000 category.

The Macintosh LC is comparable in many ways (such as power and speed) to the original Macintosh II (released in 1987 at a much higher price tag) all wrapped up in a much smaller box. Like the whole Mac II line it provides the capacity for color on a large separate monitor (unlike the original "compact" Macintosh, still available as the newly upgraded Macintosh "Classic"). The Apple IIe compatibility is provided through an add-on card.

If you have any particular software that you want to use on it, it would be best to find an opportunity to try it out on one of the new machines before making any commitment to buy. There was no demo machine available to me at this writing; the arrival date is set for early 1991.

The list price of the Mac LC is $2,499 (while the Classic is only $1,499, including the monitor); that price includes a 40MB hard disk and a keyboard. Schools may be able to get a double-floppy drive setup at a lower cost. The color monitor is an additional $999 (monochrome: $399), and the Apple IIe card is another $199. Keep in mind that educational purchases have generally received a 30-50% discount from Apple. For more info, contact: Apple Computer, 20525 Mariani Ave., Cupertino, CA 95014.

The Text Encoding Initiative (TEI) has just announced the availability of the first draft of its Guidelines for the Encoding and Interchange of Machine Readable Texts. Sponsored by the Association for Computers and the Humanities (ACH), the Association for Literary and Linguisitc Computing (ALLC), and the Association for Computational Linguistics (ACL), the TEI has as its goal the development of a set of standards for the encoding of texts. Such standards would make it easier for researchers and software designers to use machine-readable texts for a variety of purposes. Developed over the last two years of study, the Guidelines are now scheduled to be tested and revised over the next two years. A single copy of the draft version may be requested at no cost. Contact: C.M. Sperberg-McQueen, Computer Center M/C 135, University of Illinois, Box 6998, Chicago, IL 60680 (Bitnet: U35395@UICVM).