DOCUMENT RESUME ED 383 206 FL 023 006 AUTHOR Goertzen, Philip; Howard, Ron Creating Multimedia Courseware for ESP. TITLE PUB DATE 95 NOTE 11p.; For complete volume, see FL 023 004. Descriptive (141) Reports Journal Articles (080) PUB. TYPE JOURNAL CIT Edinburgh Working Papers in Applied Linguistics; n6 p18-26 1995 EDRS PRICE MFO1 /PCO1 Plus Postage. Authoring Aids (Programming); *Clinical Diagnosis; DESCRIPTORS *Computer Software Development; *Courseware; *English for Special Purposes; Foreign Countries; Higher Education; Language Research; Linguistic Theory; *Medical Education; *Multimedia Instruction; Physical Examinations; Second Language Instruction ABSTRACT This report details the development of computer software to teach English-as-a-Second-Language skills for medical diagnosis. Design of a prototype unit of multimedia materials using a common authoring tool required about 300 hours of work over a period of 9 months. The report describes the overall design of the courseware, hardware used, stages of development (preliminary planning, choice of authoring software, structuring and creation of the activity, graphics and sound, programming), planned improvements, problems and issues encountered, piloting with five learners, and time requirements for software development. General issues that needed to be addressed during the process included: minimization of user anxiety; effective screen layout and color use; the format to be use to simulate doctor-patient interaction; the degree of freedom to give the learner in manipulating the materials; and efficient learner navigation through the materials. It is concluded that while creation of this program was time-consuming, much was learned and the development of additional courseware will be more efficient. (Contains 26 references.) (MSE) *********************************************************************** * Reproductions supplied by EDRS are the best that can be made * from the original document. *********************************************************************** U.S. DEPARTMENT OF EDUCATION "PERMISSION TO REPRODUCE THIS Oftice of Educahorial Research and Improvement MATERIAL HAS BEEN GRANTED BY EDUCATIONAL RESOURCES INFORMATION CENTER (ERICI c'Z Nr ,a,,---, hts document has been reproduced as dived from the person or organization originating it f .RC...3"`-1V. 0 Minor changes nave Won made to improve rTh5--)s, Ir._ ZLI reproduction isublity Points of view or OrmOnS stated in this docu. TO THE EDUCATIONAL RESOURCES merit do not necessarily represent official OE RI point*n or policy INFORMATION CENTER (ERIC)." Creating Multimedia Courseware for ESP Philip Goertzen and Ron Howard (IALS) 2 BEST COPY AVAILABLE I CREATING MULTIMEDIA COURSEWARE FOR ESP Philip Goertzen and Ron Howard (IALS) Abstract Multimedia courseware can be produced with relatively little technical expertise. We produced a prototype unit of multimedia materials in Microsoft Access 1.10, taking about 300 hours over a period of nine months. A second unit has token under 100 hours. Our procedure is described and some of the problems and issues which arose during the design stage are discussed. Introduction 1. Many ,language teachers have welcomed the arrival of multimedia. Fox et al. (1990: 37) point to 'considerable enrichment of the learning environment ... as the spoken word and the "real world" can be brought in'. For Evans (1993: 214): 'The integration of text, sound and visual data clearly is of great benefit to the learner as this will reinforce comprehension, pronunciation and contextual use in a way that traditional materials are unable to do.' Wolff (1993: 27) states of multimedia packages that: They can be used to create a rich learning environment which will motivate learners to use the foreign language in an authentic way. They can lot .ised to provide learning materials adapted to the individual learner's learning strategies and learning styles. They can be used to help learners discover and practice processing and learning strategies. They can be used to help learners develop learner autonomy. (Wolff 1993: 27) The promise such writers see in multimedia has so far had little empirical substantiation, as Wolff admits (op. cit.: 25). Until multimedia courseware is readily available it is impossible to evaluate it. Although some materials are now appearing on the market, they are inevitably oriented to the biggest market areas, notably English for Business. Materials for other specialist areas are unlikely to be produced commercially for some time to come, if ever. In January 1994, we decided to see if it was possible to develop appropriate and effective multimedia materials with resources (hardware, software and staff expertise) already to hand. The first part of this objective has been achieved: a prototype unit (William Evans: a medical case problem) has been developed using Microsoft Access as an authoring tool. We now plan to pilot and evaluate the material and to see how far it can be used as a template for other units in the same area (English for Medicine), and in others (English for Law). This article briefly describes the unit and outlines the stages of its development and some of the problems and issues which arose during the process. The material consists of a problem-solving activity for doctors and medical students wanting to improve their English. Many of the decisions we took will apply to other types of problem-solving activity in multimedia format. EDINBURGH WORKING PAPERS IN APPLIED LINGUISTICS, number 6, (1995) ISSN 0959-2253 3 The material 2. making underlying the diagnostic process. The activity simulates the information gathering and decision variety of forms: letters, written reports. Learners are presented with information about a patient in a collecting all the appropriate information, telephone reports, dialogues, X-rays. If they are successful in diagnosis. The material therefore provides they should be able to solve the case problem and arrive at a integrated skills practice in a semi-authentic way. It is intended mainly for self-access use, but it could also working in pairs (if enough computers be used by two learners working together, or indeed a whole class are available). and digitised sound. Learners have The package consists of text, graphics (including colour photographs), contains about 2,000 words of written and taken from 60-90 minutes to work through the material, which occupies about 20 megabytes of space spoken text, plus three gap filling exercises. The file containing it on the hard disk (including sound files). Hardware 3. 14" SVGA monitors, 200 megabyte hard disks (8 megabytes We used Dell 486SX 50Mhz machines with connected by way of Windows of RAM), Sound Blaster sound cards, and headphones. The machines are facilitate fast response times, for Workgroups 3.11. The main Access file runs off a central server; to of accompanying sound files (WAV) are located on local machines where they occupy about 10 megabytes hard disk space. including a colour scanner for some of the In the production stage, some additional equipment was used graphics, a microphone, and the recording software provided with Sound Blaster. Staging 4. suggested by Keith and We approached the task in three stages corresponding quite closely to those Lafford (1989:126). Preliminary planning 4.1 instance. Two members of staff (the Four weeks (140 hours) were allocated for the project in the first advise. A Medical English authors) were mainly responsible, with two others assigned to monitor and could benefit from problem-solving lesson was chosen for the project since we thought this type of activity plan the general content of the multimedia adaptation. It also had the advantage that no time was needed to unit. Choosing software 4.2 available. HyperCard has A number of software packages designed for authoring multimedia programs are educational software developers (e.g. Evans 1993; Gibbons 1992), so far been the most popular with However, at IALS we do not largely perhaps because it came free with the Apple Macintosh computer. option. Some packages were excluded because of cost use Apple Macs, so HyperCard was not an worked with Guide, which is less (Authorware Professional, Icon Author). Both authors had previously visually pleasing finish to our expensive, but we had found it difficult to use. We also wanted a more Too lbook. Colleagues in other departments in the screens than Guide could provide. We considered (Sec Darby (ed.) 1992a, Fox University had achieved impressive results with it, but it is said to be slow. (1994) and Marshall (1995) for reviews of software.) et al. 1992: 37-52, Donnelly (ed.) 1994, Windeatt 19 4 We decided, finally, not to use any of these specific authoring packages. King and Murray (1992) have shown how Microsoft Windows 3.1 can be used to create multimedia materials without any additional software. Spreadsheets have been used by other authors (e.g. Young, Heath and Cuttle 1994). We chose a database, Microsoft Access. As far as we know, we are alone in using Access in this way. In many respects, however, Access is an easier authoring tool to use than purpose-built programs such as Toolbook or Guide. Individual screens are created from Access 'forms', macros take care of most of the manipulation of these forms (without the need for programming), and student file input/output is largely taken care of by Access tables. On-screen menus are fully customisable. For more complicated features like gap-filling exercises and navigational maps, Access Basic provides an industry standard programming interface that does not require learning a proprietary language. 4.3 Structuring the activity As we have said, it was not necessary to spend time in the pre-design phase on planning the overall content of the program, since it was to be based on a well-tried classroom activity. Some care was taken, however, in choosing the particular medical case problem. (We used one from Taylor, Armstrong and Carroll 1987. See Ferguson 1991 for a discussion of ways such case problems can be used.) We looked for a case that was interesting and challenging but not too difficult, and that gave scope for the use of graphics (X-rays, for example). We also gave considerable thought to getting the right balance between tasks that were purely communicative and those that were more language focused (Johns and Davies 1987). We finally opted for weighting the former. We have included three gap-filling exercises, one at the end of each section (history-taking, examination, laboratory tests), but even in these it is content words which are deleted, and their main function is to provide the learner with feedback on the success of information gathering. 4.4 Creating the activity One of us (RH) next created the text of the program on paper, form by form, with an indication of the links between them. Figure 1 shows an example of the preliminary design for one form. Form 3: History You have decided to obtain more information from the patient. C loose one or more of the following: Personal details I. form 4 2. Social history form 5 3. Main complaint form 6 4. Family history form 7 5. Past history form 8 6. ,Systems enquiry form 9 7. Quit Figure 1: Sample of the preliminary design of a form 20 Graphics and sound 4.5 proformas, line graphics and sound for each form. Graphics consist of The next step was to collect the proformas, are integral to the activity while others are drawings and photographs. Some of these, e.g. look more attractive to the learner. simply intended to make the program of the patient's abdomen. The learner is required to 'palpate' the One graphic consists of a line drawing When the painful area is touched, the patient is heard abdomen by clicking the mouse on different areas. This particular task does not make any linguistic demands on the saying: 'Ouch! That hurts!' ( Figure 2). but we believe it serves the purpose of helping to make learner (beyond understanding the instructions), the activity enjoyable. t"..Q.r..74;';,,;:irraTVAZIffro:, en you dick on each of the zlour quadrants you WII hear the roc-i patient's reply to the question: "Is hat sore?" 2: Screen from William Evans Figure from local hospitals, were created directly in Access. Profonnas, based on authentic examples taken taken by one of the authors (PG). Others, e.g. the Colour photographs, e.g. of hospital departments, were clip-art files. In the form illustrated by Figure 2, a speech photograph of the 'patient', were taken from Paintbrush. balloon was added to a clip-art photograph using the learner's questions, and spoken reports of laboratory Sound consists of responses from the patient to through a microphone directly onto the hard disk. tests. These were scripted and recorded Programming 4.6 forms by PG. Decisions on graphics and layout for each The 'storyboard' was next converted into Access later. storyboard stage, sometimes during programming, sometimes screen were mace sometimes at the 6 21 BEST COPY AVAILABLE- Although we met periodically to evaluate the work-in-progress, the division of labour meant that for most of the time we could work independently and were not constrained by the need to find time when we were both free to work on the project. On the other hand, the fact that PG has taught medical English and that lore efficient. RH is moderately familiar with computers made the collaboration Further developments planned 5. The unit is now functional, although we hope to continue making improvements - some minor, some more substantive. For example, we will eventually add a local glossary with contextually appropriate meanings of key words. EmblemaAndiuuks 6. In the course of design and implementation a number of problems arose, some general, and some specific to the material. In this article, we will restrict ourselves to discussing the former. 6.1 Design The effects of user anxiety when interacting with computers are fairly well documented (see Todrnan and Lawrenson 1992). To try and minimise the stress of using a computer, several design issues were agreed on at the outset: The user would interact with the computer primarily by means of the mouse and would therefore need no knowledge of the operating system. Options would appear as a series of 'buttons'. The user chooses by clicking on a button. Where users were required to perform tasks (such as filling in proformae, answering questions, etc.) they would have the option of doing so on screen or on paper. 6.2 Screen layout We accepted as given that to make the screen as clear and attractive as possible, text should be kept to a minimum. A basic design of four frames was adopted (title, instructions, graphic, option buttons) with a is never necessary. We experimented toolbar at the bottom, but variations are possible. Scrolling informally with the placing of text in relation to graphics. When more than one frame contains text, we have found that an enlarged capital (a 'dropped' capital) at the beginning of the first word draws the eye to the text which should be read first. We used different fonts and different colours sparingly, following the two-colour two-font rule. High contrast colours were chosen. The guidelines on design in the Microsoft Access manual were useful. Learner input 6.3 In the history-taking stage of a medical consultation, the doctor obtains information by interrogating the patient. In our simulation, the learner plays the part of the investigating doctor and all other roles including that of the patient are assigned to the computer. It is a simple matter to reproduce the patient's responses as digitised speech. But how to have the learner-doctor address questions to the computer-patient posed more of a problem. The potential solutions are: The learner speaks the questions. This is not possible with the technology available at present. 1. The learner types the questions. This is off-putting for some learners (see Section 7) and, although 2. technically possible, getting the computer to accept all the possible valid questions would have involved far more programming than we were able to undertake. The learner selects questions from a menu. This was the method used. 3. 22 7 Even in the case of the option we chose, variations are possible. We considered the possibility of offering learners a choice of questions, which could be (1) acceptable, (2) grammatically incorrect, or (3) pragmatically inappropriate. However, this was eventually rejected, partly to grammatically acceptable see also below on the number of questions), and partly because we thought limit the size of the progro confusing to the learner.2 having to switch from a co.nmunicative task to a language-based one may be situation. To include them all would have taken A very large number of questions could be asked in this We therefore felt we had to limit the number of questions in up an inordinate amount of time and space. particular part of the history-taking section was to present a batch some way. One solution we chose for a questionnaire, an idea taken from a medical source (Short of information in the form of a patient response 1986). questions, each intended to elicit In William Evans, the learner is therefore presented with a limited set of also Gibbons 1992.) The question forms will normally have been a different piece of information. (See additional exposure to these forms to taught in classroom lessons; our simulation is intended to provide been included to make choice more reinforce classroom levrning. A few redundant questions have questions (medical meaningful, since observation shows that otherwise learners simply select all reproaches the learner. If thoroughness?). When one of the redundant questions is selected the computer it could have earned the learner a negative mark. we had ad,oted a scoring system (see below), Interactivity 6.4 of the main issues we faced. The view that The question of how much freedom to give the learner is one be a gaining ground in the literature (e.g. Sciarone and too much freedom is counterproductive seems to of its chief advantages. Meijer 1993), although initially the autonomy possible with CALL was seen as one it allows them to make clearly wrong Freedom of choice could be undesirable if it bewilders learners or if choices. We hope we have avoided both dangers. keyboard to enter responses (see 6.1), As well as being able to choose whether to use pen and paper or the learners have a number of other choices. diagnostic problem. If we were Choice of path: They can choose which path to take in order to solve the doctor or medical student would teaching medical diagnostic skills, this would not be a real choice since a then examination and finally be expected to follow the well-established pith of history-taking first, incorrect. Medical laboratory investigations. There would be one correct path, and all others would be optimal steps in the optimal sequence, and penalise programs of this type often award points for taking the include a scoring system since this would any deviation. We have so far resisted the temptation to designed to teach medicine rather than language. We feel it encourage the impression that this program is tried to make is important that the learner is aware of the distinction, although at the same time we have the program medically exact. Learners can choose to Transcript: An onscreen transcript is available for the longer listening passages. listen first without the transcript, read this before, while or after they listen. While they are encouraged to superiority of this approach for all learners and have therefore we do not know of any evidence for the different decision (Moses 1991). The allowed them to choose. Other multimedia producers have taken a the transcript, but it has to be few learners who have so far tried the material all listened first without have been an influence. After admitted that the presence of one of the authors (see Section 7) may decision. evaluating the material with a larger number of learners we will of course review our formal feedback, if they wish. Feedback: Learners are free to complete the activity without obtaining any in the form of gap-filling Optional listening and reading comprehension checks are, however, provided these exercises on screen or exercises. As mentioned in Section 6.1, learners are given the option of doing of a completed version of the exercise which can be printed on paper. Feedback in the latter case consists 23 8 BEST cppy,AvAILAKE.:,.,... ,te word, and the computer out. In the former case, after typing in an answer the learner clicks on responds with: 'Your answer is correct' or 'Your answer doesn't match the computer's answer. The computer's answer is xxx.' This is a very simple way round the complex problem of how to deal with incorrect and near correct answers. (See, for example, Race and Brown 1994). Ideally, one would deal with errors in a more constructive way but it was necessary to compromise in view of the limited time available. We chose a version of gap-filling in which the first or even the first two letters are given. This has the advantage in a computer program of limiting the possible number of correct answers to one in most cases, although the program handles two possible answers in one or two cases. It would be possible to design the program in such a way that learners are forced to follow a fixed path through the activity, and complete each section including its gap-filling exercise before they are permitted to go on to the next. We were reluctant to impose this type of constraint since we felt it would upset the delicate balance between user control and computer control too much in favour of the computer. Navigation 6.5 Because learners will normally want to follow the standard path (see Choice of path, above), navigation is observed that learners quite not as much of a problem in this simulation as it often is. However, we have often wish to retrace their steps, for example to check on information they did not quite take on board in gap-filling exercises after solving the an earlier section. In addition, they may decide not to do the problem. A map showing the main sections of the activity can be obtained by clicking on Tools at any time. This provides an overview of the simulation, shows learners where they currently are (large arrow), and allows them to move to any other section by clicking on the appropriate box. Evaluation 7. To date the material has been tested on five learners. They worked individually, with the ubserver (RH) sitting beside them making notes on such things as problems with navigation, time spent on particular tasks, choices made, etc., and intervening only when requested. This procedure is time-consuming but worthwhile with a small number of learners. The feedback obtained has already led to improvements in the program. For example, following a comment from one doctor, the map will be shown to the lenner at the start to provide an overview. All five learners preferred to fill the gaps in a printed version rather than in place for the moment but it may on screen. We have decided to leave the option of on-screen filling Unnecessary choices increase turn out that it is not worth retaining this option if it continues to be ignored. the risk of confusion (see 6.4) and should be eliminated as far as possible. Time 8. hours of development time for a single estimates Johnsol (cited in Keith and Lafford 600-1000 1989:126) simulation unit. Gibbons (1992:26) claims that he and his team took about 1800 hours to create their legal the Murder One: 'Four people executed Murder One in the interstices of otherwise busy schedules over would not claim to have brought ours up to course of ten months.' Their product is being marketed; we that standard. In a study cited in Reinhardt (1995:70), the mean number of hours requi.ed to create a experience. single hour of courseware was 228 hours, and this corresponds more closely to our would be 250 Officially, we had 140 hours for our project, but in fact a better estimate of the time taken another 50 hours. This excludes time taken for PG to become more familiar with the softwareprobably first as a template hoursand some secretarial time. A second unit has almost been completed, using the and changing only the data3; time spent so far is about 70 hours. 24 9 Cone lufinn 9. materials tailored to the needs of a special We have demonstrated that it is possible to create multimedia relatively simple and inexpensive tools. 300 hours may seem a lot of time group of language learners with of material. But we learned a great deal in doing so, and the materials to spend creating one hour's worth units will certainly be easier and faster to could not have been obtained any other way. Subsequent produce. Acknowledgements their advice and help with the project, and to We are grateful to Eric Glendinning and John Mc Ewan for the lines of dialogue. We would also like to Eileen Dwyer, David Cole and Ncni Goertzen for speaking available to us. thank Edinburgh University Computing Services for making the scanner Notes Each screenful of information corresponds to an Access 'form'. 1. reluctant to abandon the information This impression was confirmed during trials. Learners seemed 2. gathering process to do a doze exercise. format and procedure. Not entirely true: it was impossible to resist making a few changes in 3. References Development. Oxford: CTISS Publications. Darby J. (ed.) 1992a. Authoring Systems for Courseware Darby J. (ed.) 1992b. Multimedia. Oxford: CTISS Publications. Publications. Darby J. (ed.) 1994. Distance Education. Oxford: CTISS Teaching Packages. University of Donnelly P. (ed.) 1994. Constructing and Implementing Mutimedia Glasgow: TILT. learners'. Sy= Evans M. 1993. 'Nicolas: Using HyperCard with intermediate-level French 21/ 2: 213-229. English'. EMP Newsletter Ferguson G. 1991. 'Case conference materials in the teaching of medical 8: 7-14. Technology in Modern Fox J., A. Matthews, C. Matthews and A. Rope. 1990. Educational Trust for the Training Agency. Language Learning. University of East Anglia and Bell Educational Schostak. 1992. New Perspectives in Modern Fox J., B. Labbett , C. Matthews , C. Romano-Hvid and J. Sheffield: Department of Employment. Language_Learning.. Learnin2 Methods Project Report. for teaching law' in Gibbons H. 1992. 'Murder Onede.reloping interactive simulations Darby (ed.) 1992b:24-28. information: the classmom use of written texts Johns T. and F. Davies. 1983. 'Text as a vehicle for Foreign Language 1/1: 1-19. in teaching reading in a foreign language'. Reading in a 25 10