Online English Language Testing

For Aberdeen College

by

Duncan I. MacLeod

 

 

Thesis presented as part of the requirement for the Master of Science degree in Information Systems at The Robert Gordon University, Aberdeen.

August 1999

 

DECLARATION

I confirm that the material presented in this thesis is my own work. Where this is not the case the source of the material has been acknowledged.

Signed: .............


 

CONTENTS


Acknowledgements

Supervisor: Graeme Dunbar, Supervisor; School of Electronic and Electrical Engineering, The Robert Gordon University.

Client: Anne Bain, Languages Team Leader; Communication, Languages & Media Studies, Aberdeen College.

Initial idea: David Morley, Head of Information Services; Aberdeen College.

Audio clips read by: Joan Cooper and Maria Moni of Aberdeen College

Audio clips recorded by: Alan Glendinning, Technician; Communication, Languages & Media Studies, Aberdeen College.

Testing by: Wai Han Cheung, Young Taek Yoon, Noorhaidawati T. Yarnall, Franzon Waldi, Jie Mei Fong, Nilsa Ferreira, Azhara Mukhamejanov - students of Aberdeen College

Advice and assistance: Denise Boyne, Paul Martin and Alistair Blackadder, IT Centre, Aberdeen College.

I would also like to thank my family, friends, and colleagues for their support and encouragement, but mostly for their patience.


Abstract

This thesis begins by introducing the project itself and then goes on to discuss the background to the project: the use of the Internet to deliver tuition.

Following this is an induction into the software packages and computer languages used during this project.

Key elements of the program itself are then presented, and a complete listing of the coding is available in the Appendix. Conclusions are then drawn from this.

Finally, recommendations for future work are discussed.


Chapter 1 - Introduction

1.1. Introduction to Problem

1.2. Background Information

1.3. Project Aims and Objectives

 

1.1. Introduction

The aim of this project is the computerisation of Aberdeen College’s English as a Foreign Language test. At present, the test is supervised by lecturers, who read out the questions for the listening test, and invigilate the rest of the tests. The College seeks to automate as much of the test as possible, so that a lecturer does not need to be present, and the tests can be invigilated by an instructor, thereby freeing the lecturer for other duties.

 

1.2. Background

Aberdeen College is Scotland’s largest further education college, providing training for approximately 20,000 people in 1997-8. [1-1] International students undergo the test of English as a Foreign Language (EFL), which is used by the Languages team to place students in a band applicable to their level of knowledge of the English language.. The EFL test is composed of 4 parts, each part contributing to the final 'placement' score:

Aberdeen College is interested in open learning over the Internet and are keen on types of assessment which integrate well into a web environment. The College is also working towards the development of a learning intranet of which the present project could be part.

 

1.3. Aims & Objectives

Aim: To design, implement and test a suite that uses Internet technology to test a student's skill in English as a Foreign Language.

The computerisation of Aberdeen College's English as a Foreign Language (EFL) test will involve the use of HTML, JavaScript, and some Java. The investigation of similar projects will bring to light some of the problems that will need to be overcome, and also the bonuses envisaged by using the Internet for the delivery of courses and assessments.

Many of the problems in this project could be overcome by using conventional languages (eg. C++), however, these could not be delivered over the Internet, and are of no use in the open learning environment that Aberdeen College hopes to develop.


Chapter 2 - Discussion

2.1. Introduction

2.2. Teaching

2.3. Employee Training

2.4. Online Learning

2.5. Conclusion

 

2.1. Introduction

This project came about because of the changing attitudes in the education system. Both higher and further education establishments are making better use of existing resources and are seeking to develop new markets. Online education comes into both of these categories. By utilising modern technology, courses can be offered over the Internet to a wider audience, and without direct lecturer supervision.

Bill Gates of Microsoft said in 1995, that the Internet "will alter the focus of education from the institution to the individual. The ultimate goal will be changed from getting a diploma to enjoying lifelong learning." [2-1] Prime Minister Tony Blair wrote in 1997, that "technology has revolutionised the way we work and is now set to transform education." [2-2] Like any modern business, educational institutions are having to improve their use of resources and expand their markets. For example, Aberdeen College is expanding its customer base by building local information technology centres and developing on-line learning resources. The use of new technology and materials developed for it will be integrated back into the classroom teaching environment, following the constructivist movement in education. [2-3]

The focus of teaching is changing.

 

2.2. Teaching

"The purpose of teaching is to bring about learning. Learning is a process which enables us to do something that we could not do before we learned it." Mary Howard [2-4]

Modern teaching practices are moving away from the Dickensian attitude of the learner as an impassive vessel, ready to be filled, with knowledge, towards a student-centred model with the instructor acting as 'facilitator' helping students to learn the subject. [2-5] Part of the momentum for this move is being provided by the increased numbers of adults returning to education. With their real-world knowledge and attitudes, these mature students are less likely to put the tutor on a pedestal, as to seek a more interactive lesson. [2-1]

Mary Howard's book, "How to Teach Adults", points out the following key factors for learner success: [2-4]

This change in focus is part of the constructivist movement. The constructivist movement in education sees the aim of teaching as not so much to transmit information, but rather to encourage knowledge formation and development of the necessary mental processes for judging, organising, and acquiring new information. Students will then become educated, not just trained. [2-6]

 

2.3. Employee Training

This new focus and use of new technology will assist in the tapping of a major market sector: the training of employees. At present, employees generally have to travel to the place of study, with only the larger firms being able to afford on-site training. Online education is changing this, making on-site training a reality for small and medium enterprises.

Management guru, Peter Drucker advises businesses to "train the least-skilled people and make them functional." [2-7] A study by the American Society of Training and Development predicted that 74% of workers will require retraining within 7 years. [2-8] Businesses require fast, consistent, comprehensive and comprehensible corporate training; but cost is an important factor. By sending an employee on a training course, the firm will spend approximately 66% of training costs are spend on travel and accommodation; and this does not include the time lost to the firm when the employee is travelling to and from the course. By using the Internet, costs are minimised, training is consistent, and can be delivered to the employee's nearest workstation. A recent survey indicates that 93% of major corporations are considering online learning. [2-9]

In the UK, the government established the University for Industry (UFI) in 1997. Its aim is to make it easy for people to learn, whether at home or in the workplace. David Brown, a UFI director and chairman of Motorola UK, notes that "modern industry understands that its greatest asset is knowledge, and learning is the key which unlocks knowledge". [2-10]

Computer giant, Microsoft, takes a different approach, believing that community colleges are in a better position to deliver this training. Colleges also have more experience at delivering the basic skills training that many employees require. [2-8]

Nothing frustrates managers more than having employees who need training but not having the time or the money to let them take that training. By using online training (over the Web or company Intranet) employees can learn about the task from a modularised tutorial, learning what they need for their work. [2-11]

 

2.4 Online Learning

"The new online educational environment has unlimited, untapped potential for academic quality, rigor, interactivity, and cost-effectiveness." Dee Stallings [2-12]

2.4.1. Background

a. Distance Learning

The term 'distance learning' [DL] covers the process of off-campus learning, and uses a variety of communication systems, including the Internet, television, telephone and the postal service. [2-13] DL tends to be aimed more at adults who find it inconvenient to attend a college or university; for example:

At present, there are relatively few courses that are run completely over the Internet, but many DL courses use e-mail for communication, chat rooms for discussions, and have their own Web pages. [2-14]

Boettcher and Conrad propose that distance learning should be based on following interactions: faculty-to-student, student-to-student, and student-to-resources. [2-15] They have used the 'ACCEL' model to improve learning outcomes:

A 1997 survey estimated that there were 1 million people taking online courses in America, compared to 13 million traditional college students. [2-16] Nearly every study examined by Thomas Russell of North Carolina State University found that distance-learning students who apply themselves learn as much as traditional students. [2-16] Peter Drucker wrote: "Already we are beginning to deliver more lectures and classes off-campus ... at a fraction of the cost. The college won't survive as a residential institution. Today's buildings are hopelessly unsuited and totally unneeded." [2-7]

b. The Internet

The Internet is a global network of computers that uses standard communication protocols. The system grew out of the late 1960's US Department of Defense's ARPAnet network, designed to withstand 'partial outage', i.e. having some communication centres attacked by nuclear weapons. [2-17] Sending a message on the network meant that data was put into a packet, and transmitted to the next computer in the network where the address was checked and sent on. [2-18]

From those few early connected computers, the network expanded to 180 million users in 1999, and is expected to reach 327 million by the end of 2000. [2-19] At a conference, Noel Chidwick of SCET pointed out that 7% of the Scottish population are in Further Education, which means that the potential audience for online courses are 7% of 180 million: 12,600,000 people on the Internet are interested in continuing their education. [2-20]

 

c. The World Wide Web

The Internet started to leave the domain of the engineers and academics when, in 1990, Tim Berners-Lee wrote the World Wide Web prototype, allowing the retrieval of linked documents. [2-21] This facility was envisaged by Ted Nelson in the 1960's - a time when people could get any type of information, images, sound, text, films all by selecting the appropriate icon within an on-screen document. [2-22] Making use of hypertext encoded into documents written in HTML and browsers to interpret the code and display the document to the user, the Web is the richest source of information on the planet. Search engines make it possible to find and retrieve information from this global library, skills which should be taught to the students in a similar way that students are shown how to use the campus library.

 

2.4.2 Online Learning

Until the development of the World Wide Web, information technology in education was likely to have maintained the status quo of the "talking head" lecture. [2-1] However, many academics have been noting that technology can humanise education, making students active participants in their own education, and following a student-centred constructivist approach. For example, in their online learning environment package, Teknical includes learning-to-learn tutorials. [2-23]

 

2.4.3. Problems for Student

a. Motivation

An important job of the tutor is to motivate the student. By their very nature, distance learning and online education do not have tutors present, so the student must motivate themselves, or obtain motivation from other sources. One solution is to get into the habit of working on the course every day, until it becomes part of the routine. [2-14]

b. Lack of communication

Although courses have their own online chat facilities, newsgroups, mailing lists, a common problem is that students were not very communicative. Various approaches to get the students 'talking' have been tried, some have found that an online discussion between two expert tutors worked the best. However, a student-to-student discussion is one of the best ways for students to learn [2-24], and this is encouraged within Aberdeen College's Best Practice Arrangements [2-25], and should be encouraged online. Oakes also noticed the lack of communication, recommending that mailing lists and newsgroups be incorporated early as a means to improve class interaction. [2-26] Dr Maddox of Penn State University, noted that mailing lists and newsgroups "opened the door to sharing knowledge with others as a way to self-education". [2-27] Some students are more likely to make contributions electronically than in a classroom environment.

c. Diploma or degree mills

These are establishments where the student can obtain a qualification for little or no work. This brings to light the value of a qualification. If the qualification is important to you, look for accredited courses. These have been validated by an external body (e.g. a professional institute or the state's education department), and will be nationally recognised. [2-28]

d. Lack of social interaction

"A university education is more than taking courses; it's the social experience of encountering new ideas and new people": Jim Gregory, University of Washington. [2-29] Until technology takes a vast leap forward into the realms of virtual reality, electronic communication will never replace the social interaction of the classroom. [2-30] Cheaper WebCams may provide part of the solution, with students being able to see and hear each other. Another solution is for students to club together geographically, and have regular face-to-face meetings.

 

2.4.4. Problems for Providers

a. Under-planned & Over-committed

There can be a problem of over-committing to new technologies, especially the 'lone ranger' approach, where one person's vision and lack or research leads the training provider into trouble. Dees Stallings [2-12] noted that some trainers are dispensing online training like a "vending machine", with little or no study of the cost or effectiveness; and advises on the use of pilot schemes and prototypes. Barbara Horgan feels that universities are being pressurised to control costs, improve quality and respond to customer needs - seeing IT as a way to solve these problems. [2-31] An example cited is 1998's launch of Western Governors University, sponsored by networking giant Novell and others, initially only 10 people enrolled on courses. [2-16]

b. Identity

This is an important matter in distance learning: can the tutor be sure that the person receiving the qualification is the same person who is handing in the work? With non-accredited courses, this is not really a problem, because the student is enrolled to learn the subject rather than obtain a qualification. Many DL courses have regular 'physical' meetings, so, again, identity is not a problem. It is only when the course is completely online and accredited that verifying that enrolled student is capable of carrying out the work becomes critical. If the course organisers cannot do that, then the course will have little value to the community, and may even lose it's accreditation. At present, the only solution has been for the student to travel to a validated centre and take a final exam in person. Steve Levicoff notes: "if even one person can question the legitimacy of your degrees, it's not worth it. When it comes to your credentials, credibility is everything." [2-28] At the University of North Carolina-Chapel Hill (UNC-CH), the students sign an agreement promising that they will not break the University's "Honor Code". However, UNC-CH also intends to use timers for its online quizzes, as paper-based quizzes have time constraints. Some faculty at the University of British Columbia use a timer that starts as soon as the page is accessed - in a similar manner to students being told that they can now turn over the paper! [2-32]

c. Evaluation of Written Work

Although evaluating written submissions is time-consuming for the tutor, at present, computers are not very good at carrying out this task. [2-12] Due to the complexity of language, anything beyond the fill-in-the-blanks exercise would not be cost-effective. This limits the level of qualification that can be obtained online, before the human tutor has to appear. Programs currently exist that will compare written work against a master document, but that has limited use, for example, when marking word processing exercises.

d. Staff Resistance

Not all staff will welcome the move to online courses. The new technology or the pace of chance will intimidate some, others will see it as a threat to staffing levels and a final move away from traditional teaching methods. [2-23] Some see it as a fad, taking funding away from traditional education. [2-16] In 1997, the faculty of York University in Toronto went on strike, one of the reasons for this was that staff were irate about plans to deliver sponsored courses delivered over the Internet. [2-16] Only training and familiarity with online education will alleviate these worries, as the number of staff required will be determined by the success or failure of the course. In this respect, there is little difference between campus-based and DL methods.

e. Cost

The initial investment in time and resources will cost more than the continued provision of traditional teaching methods. This is true of most investments for the future. The University of Phoenix has been providing distance learning since 1974, and is at the fore-front of online education provision, however, Terri Hedegaard-Bishop, vice-president for distance learning says that "technology education frequently costs more". [2-16]

 

2.4.5. Benefits for students

a. Self-paced, on-demand training

The student is able to work at their own pace, making mistakes in the privacy of their own virtual working environment. The learner no longer has to worry about missing a lecture, or having to collect handouts from the tutor's office, as materials will be available to them online.

 

b. Cost savings

Savings on travel and accommodation expenses, relocation expenses for the equivalent full-time courses, possibly lower course fees, possible exemption from government student fees: all these costs are may be an incentive for the student when considering whether to take an online course.

c. Personalised content

Electronic documents are easier to edit and add notes to than paper-based ones. This leaves more time for the learner to actually read the documents rather than re-writing them.

d. Reduced training time

As the learner is working at their own pace, they may finish the course early. This option is not generally available when the student is working at the pace of the class, or when assessments are on set days.

e. Consistent delivery

The student may find that they are able to subscribe to a number of different modules without having to accommodate different teaching styles. They can then get on with the business of learning, having become familiar with the method of delivery.

 

2.4.6. Benefits for providers

a. Standardisation

Templates can be used for the creation of Web-based materials, where the writer provides the content, graphics, audio clips, and movies of relevant materials, and they are integrated into a standard format. [2-12] This makes it easier for the student to follow, as all materials will have a standard format, and the on-screen buttons will be in the same place, and have the same effect when activated.

b. Streamlining

As courses are being re-written for online delivery, it is a good opportunity for a review of materials and assessments. Assessments may be streamlined so that they can be rapidly and flexibly assessed, perhaps using checklists, multiple choice questions, or other techniques that are easy.

c. Cost

Although not proven yet, any form of automation will reduce long-term costs if productivity and quality can be increased. Profits do not necessarily mean that the research and implementation costs have been recouped.

 

2.4.7. Online methods on-campus

Interactive distance learning methods can also provide flexibility for on-campus students. Some teachers already publish their lecture notes on the Web, so that students can download their own copies, and catch up of they are off (e.g. Dr A D Marshall at Cardiff University). [2-33] Notes can also be edited by the teacher after publishing, as new information becomes available. [2-34] Paper-based notes have a tendency to remain unchanged from year to year until a major re-write is required.

 

a. Computerised assessments and delivery

The benefits of online delivery have been discussed previously, and many are relevant to on-campus delivery. A report on the development of online materials and delivery can be found in Appendix 6.

The University of Luton has been using computerised assessments since on-campus 1993. Over this time they have developed the following procedures: [2-35]

 

b. Putting students work online

Kate Egerton of the University of North Carolina's Chapel Hill (UNC-CH) found that the quality of her students' work improved when they were told that it would be published on the Internet. [2-34] When publishing on the Internet, the audience is no longer just a few class-mates and staff, but potential employers, relatives, and anybody else who may find the page, and judge the author only on the quality of their work. Dr Leuthold of the University of Illinois agrees that global projection improves the quality of projects: "it's fun for the students to see their work 'in print'. It provides an inventive for them to do good work." [2-36]

Linking to students' homepages is an easy way to promote community. If they have not set pages up yet, Sheila Englebardt at UNC-CH developed an online form that created student profiles, this gave a more personal touch, giving more information that a personal Web page might. [2-27]

 

2.5. Conclusion

Evidence for the success of online education has not been conclusive [2-23] [2-37], and may not benefit every type of learner or every type of course, however it is a worthwhile addition to the arsenal of teaching methods available, especially for adult learners and businesses. Teaching institutions will look to hiring teaching staff that have passed qualifications using distance learning, as the personal experience they bring will assist the students. [2-38]


Chapter 3 - Tools

3.1. Introduction

3.2. HTML

3.3. JavaScript

      1. Java
      2. Graphics
      3. Multimedia
      4. Web Design

 

 

3.1. Introduction

There are many proprietary tools that can be used on the Internet - this project has attempted to use only standard tools that are recognised by as many browsers as possible. This is a problem that the World Wide Web Consortium (W3C) was set up to overcome this problem by ratifying and developing common protocols that are then available free of charge. [3-1] W3C recognises that this is the only way for the Web to reach its full potential, by ensuring its interoperability.

 

3.2. HTML

W3C defines the HyperTextMarkup Language (HTML) as: "a simple mark-up language used to create hypertext documents that are portable from one platform to another." [3-2] More simply put "HTML - the Hypertext Mark-up Language - is the lingua franca for publishing on the World Wide Web." [3-3]

In practice, this means that by writing a Web page using a standard form of HTML, the page will be able to be viewed by any Web browser, and displayed on the user's screen in a similar layout. The browser can be the latest from Microsoft or a two-year old version of Opera, running on the latest Silicon Graphics workstation, or a classic Apple Macintosh - so long as the HTML complies with the W3C standards, and the browser is not older than the HTML standard - the page will be available to the user.

HTML uses tags in a similar way to those used by DOS-based word processors like WordStar. For example, to make a word bold, it has to be encased between a bold-on tag and a bold-off tag, like so:

HTML code

This is in <B>bold</B>

Browser display

This is in bold

This project has used HTML version 3.2 as many users have not yet changed their browsers to accept HTML version 4. There is also a credo in Web design that says that the author should always write for the last version of HTML, making their pages available to as many people as possible. For this reason, many of the leading edge technologies are not used on the Web, as it would alienate many users.

For this project, the author used a variety of authoring tools, forgoing much of the repetitive HTML coding. Initially, Microsoft FrontPage Express was used, as it provided a standard Microsoft interface, and its results were consistent across a variety of browsers. Later, a demonstration version of Macromedia Dreamweaver was used – this provided the ability to clean up the HTML coding, and make sure that the coding would work in version 3 browsers. Also, it would not attempt to change any JavaScript that it did not recognise.

 

3.3. JavaScript

"JavaScript is a compact, object-based scripting language for developing client and server Internet applications." Invented by Netscape Communications in 1990, it was originally released as Livescript 1.0, but with the popularity of the Java language, the name was changed. [3-4]

As with HTML, there are many versions and flavours, so this project uses the most common one, Netscape JavaScript version 1.0. Again, there are features in later versions that cannot be used, however, in the final version of the program, only the date functions had to be re-written to work in this earlier version.

The language is similar to C, and is included in the Web page within the <SCRIPT> </SCRIPT> tags. Functions are called from within the body of the HTML, e.g. the following uses a button to run the ScoreTest() function when clicked:

<input type="button" name="B1" value="MARK THE TEST" onclick="ScoreTest()">

JavaScript therefore provides additional functionality to HTML, some of which is made redundant by later releases of HTML (e.g. changing text colours when the mouse passes over it).

 

3.4. Java

Java developed by Sun Microsystems in 1995 as a cross-platform programming language. When used on the Internet, small applications ('applets') are embedded into the HTML and downloaded by a Java-enabled browser. [3-5] Where Java differs from the languages it is derived from (e.g. C) is that the code is run by the Java Virtual Machine [JVM] - an interpreter that is on the user's computer. This allows code to be sent quickly over the Internet instead of the user downloading a compiled executable program.

The language is simpler to learn than C++, and free copies of the Java Developer's Kit [3-6] and JVM means that anyone can create their own Java applets for distribution over the Web.

 

3.5. Graphics

The main problem with graphics on the Web are:

1. Size: the bigger the file size is, the longer it takes to download.

2. Colours: there are only 216 colours that work on both Windows and Macintosh monitors set to the earlier 256 colour depth.

3. File formats: the only two common standards for graphics on the Web are GIFs (for simple images) and JPEGs (for photographs, etc).

Therefore, graphics have to be used carefully on the Web. When graphics were created specifically for this project, they were created using '216-friendly'. As the program was to be used by a variety of users, from inexperienced to experienced, the graphical interface was kept to a minimum, using text instead. Where required, only simple graphics were used, like the arrow at the bottom of the page, or those in the listening test.

 

3.6. Multimedia

Current standards for animation and audio are very rudimentary. Animated GIFs have a larger file size, as one image overwrites another. 'Onion-skinning' is not used, since the format cannot handle transparencies in this manner. The Web's oversight body, W3C, is considering a proposal from Macromedia to make the Flash format an official standard. This format uses vector graphics and advanced animation techniques to allow the programmer to create animation with very small files sizes and quicker download times than the equivalent animated GIF.

For more complex animation, Macromedia's Shockwave is available, and can be viewed via the same browser plugin as Flash animation. Shockwave files are produced by Macromedia Director, a powerful multimedia tool where the designer can "combine graphics, sound, animation, text, and video into compelling content." [3-7]. Macromedia Director was going to be used in this project to create practice exercises for users with no mouse or keyboard experience, but in the end, Director was only used to produce high quality audio files for the listening test and the practice exercise.

The only standard for audio at this time is the Java .au format. This outputs sound in an 8 bit, 8000 Hz mono format, and is inappropriate for voice or other sounds - it is really only useful for low quality sounds. All other sound formats are non-standard, proprietary or system dependant.

 

3.7. Web Design

Designing for the Internet is basically no different than designing any other form of software. The designer should follow a basic software life-cycle layout, such as that in Appendix 4, which follows Sommerville's Waterfall Model, illustrated below.

Web design crosses the boundaries of software engineering and graphic design, with a few other skills thrown in for good measure. Much of what can be achieved in software engineering projects under languages like C++ would be too slow to download over the Internet. According to Vincent Flanders [3-8] , the maximum size of a page, including graphics, should be 35k. The reason for this is that this is a 15 second download time when a 28.8kbps modem is used. Surveys have shown that 15 seconds is the maximum time that a user will wait for a page to download, unless they have a reason to wait for the data.

There is plenty of advice to be found on Web page design on the Internet, and many Internet Service Providers have their own newsgroups where inexperienced and experienced Web page builders can discuss their problems. Much of the design aspect is down to fashion - whether sidebars are 'in' or the appropriate choice of colour scheme, the perennial use of black backgrounds by artistic people, and if the use of a white background is boring.


Chapter 4 - Solution

4.1. Introduction

4.2. Requirements Analysis and Definition

4.3. Problems

4.4. Initial Design

4.5. Implementation and unit testing

4.6. System Testing

4.7. Operation and Maintenance & Feedback

 

 

4.1. Introduction

Initial meetings were held with the candidate's manager and supervisor to discuss the project - minutes can be found in Appendix 2. The program is to replace as much of the current English as a Foreign Language tests as possible. It is envisaged that a CD-ROM containing the program could be sent out to authorised test centres, where a tutor would oversee the students taking the test, returning the completed results to the College. At present, a paper-based test is used, a copy of which can be found in Appendix 3.

 

4.2. Requirements Analysis and Definition

The requirements analysis below follows the European Space Agency's criteria: [4-1]

Functional requirements: the users will proceed through the English as a Foreign Language [EFL] tests without a tutor being present.

Performance requirements: if there is too much of a requirement for a tutor to be present, the program will not be cost-effective.

Interface requirements: the software will use a version 3 browser as an interface, allowing the program to be incorporated into the client's virtual campus at a later date. Additionally, the system must not rely on Internet access, as not all computers will have Internet access.

Operation requirements: the program will introduce itself, and then direct the user through a practice area, so that they can try out similar questions before going on to the real test pages. At the end of the test, the user will have their scores and a page about themselves that will be sent to the College for analysis.

Resource requirements: the client's language laboratories have Evesham Vale Prestige IBM compatible PCs, using Microsoft Windows 95, with Microsoft Internet Explorer 3 as their browser, which includes the Java Virtual Machine. The computers are not connected to the Internet. The program should not be restricted to one particular browser, and should be compatible with version 3 browsers, using JavaScript 1.0 if required.

Verification requirements: the client is to be kept informed of any developments, and will test the program at various stages of its development.

Acceptance testing requirements: the client will also test the program before its final release.

Documentation requirements: a users manual will be written to accompany the program.

Quality requirements: the program will be easy to use, but does not have to comply with any official standards.

Safety requirements: none.

Reliability requirements: the program must return the correct scores to the questions answered by the user.

Maintainability requirements: all coding is commented, and the program uses agreed World Wide Web standards. This will allow for future maintenance by another programmer.

 

4.3. Problems

4.3.1. Online Assistance

The University of Glasgow's Teaching with Independent Learning Technologies [TILT] project noted that "When students go through a new piece of courseware, they may spend time mainly in getting oriented: finding out how the controls work, and what to expect. Only if they then went through similarly organised material could they even begin to work actively and effectively." [4-2] It was therefore important for the user to be given practice questions before proceeding to the real tests.

The user’s computer ability should form part of the test, so introductory tutorials covering keyboard and mouse skills was to have formed part of the suite, however, the candidate did not have time to develop and integrate these modules. The client did not feel that this was too much of a problem, as the user could always use one of the College’s CD-ROM based tutorials, e.g. "Computers Don’t Bite" or "IT For All". If necessary, the user could always sit the paper-based version of the test. As mentioned earlier, the level of IT literacy is always going to be a problem for computer-based learning.

Every part of the test has instructions, and the user can obtain copies of these instructions in a number of different languages by clicking ‘help’ on the sidebar. The number of languages made available was determined by the Client’s ability to call upon translators.

 

4.3.2. Automatic Marking

In the paper-based version of the test, there were two tests that could not be automated.

a. Free writing

Due to the complexity of the English language, it is impossible to program a computer mark most written passages beyond a simple ‘fill-in the blank’ or choose the correct statement. Since the latter format was already being used in the Grammar test, the Client decided on a more complex written test, where the user has to find and correct mistakes in a piece of correspondence.

b. Spoken English

The final test is the interview, when the assessor judges the student’s conversational spoken English. This was impossible to automate, and the Client was informed that for this to be computerised, something akin to video-conferencing would be required. A simpler and cheaper alternative would be for the interview to take place over the telephone.

4.3.3. Sound Quality

Clarity of voice was essential for the listening tests. To obtain a clear recording, Alan Glendinning of Aberdeen College’s Media department was approached, and he operated the recording equipment whilst Anne Bain and her group of readers clearly read out the relevant passages. The completed digital recording was then copied on to a Zip disk, as it was 59MB in size.

The final format was important. To be used by Java, the format would be dropped to 8-bit, m law, mono at 8khz sampling frequency. This resulted in a very poor recording and would not have suited the playback for the purposes of testing a user’s English language ability. The table below records the different file sizes

Table 4.1

Sample "The total bill comes to £17.60"

File details/format

File Size

22050Hz 16bit stereo (.wav)

351k

11025Hz 16bit stereo (.wav)

176k

11025Hz 16bit mono (.wav)

88k

11025Hz 8bit mono (.wav)

44k

8012Hz 16bit mono (.au)

32k

If the file was kept in a .WAV format, the playback would rely on proprietary media players, for example, Microsoft Media Player. The user could then replay the question until they were sure of the answer. This was not what the client wanted.

The only remaining option was to embed the sound file in a second Web page, and control its opening with JavaScript. This would permit the user to hear the clip only once, but the quality of the sample was chosen by the programmer. To embed a sound file, a 1 pixel by 1 pixel Shockwave movie whose only task was to play the sample was embedded in a Web page that would be opened by a JavaScript function, and the opening recorded so that later attempts to open the file would be met only be a friendly message stating that the question could be heard only one.

There were two reasons for choosing this method. At present, only the Microsoft Internet Explorer browser can embed sounds, this would have limited the cross-platform ability of the project. Secondly, the user was familiar with Macromedia Director, having learnt the basics for the failed attempt to incorporate the keyboard and mouse tests; and having access to Aberdeen College’s copies of Macromedia Director.

The only problem with using Shockwave is that the user’s browser will have to have the Shockwave plug-in. However, this is free to download from CD-ROM or over the Internet, and installs easily. A copy has been included on the CD-ROM for the Client.

 

4.3.4. Concealing the Answers

The lack of CGI access meant that the answers would have to be stored within the Web pages. Any form of concealment would not stop the most determined cheat from discovering the answers, so some form of encryption was also required. It was decided that the best place for the answers to be was in a frame as part of the sidebar.

A 1 pixel high frame was ruled out as the user would notice the thin bar downloading, so the College logo was placed in the frame – following the adage "if you can’t hide it, make a feature of it". The answers to the multiple choice questions are simply encrypted by transferring the letter of the alphabet into numerical format (A=1, B=2, etc), and multiplying the result by 12. When the questions are being marked, the program calls these numbers into an array, and divides them by 12 to give the correct answer for comparison with the users’ response.

If Internet access could have been guaranteed, the answers could have been stored on the College’s server, and accessed using CGI scripting or JavaScript commands.

 

4.4. Initial Design

The project was proceeded by a period of background reading and learning the fundamentals of Web page design. Most of the introductory work was carried out on the author’s own Web site, learning about the use of frames and tables, monitoring newsgroups and speaking to colleagues at work. The main influence comes from seeing what works and what does not work on the Internet itself.

The initial design, illustrated in Figure 4-1 below, shows how the practice area was initially going to be integrated into the suite. Even at this early stage, it was clear that JavaScript could not carry out complex commands without being browser dependant, as having a button that would print the window would only work under one particular browser. Even catering for the two most popular browsers (MS Internet Explorer and Netscape Navigator/Communicator) would rule out a number of minority browsers.

4.4.1. Overall design concepts

a. Screen Resolution.

The Web pages are not designed to be viewed at any particular screen resolution, but some of the text boxes work better at 800 x 600 resolution.

b. Frames are Used for Navigation

Although this means that early browsers cannot view the program, it is not envisaged that this is a real problem as there are freely available browsers that can view frames.

c. Simple site navigation

The initial site plan was streamlined to that shown in Figure 2-2 below. Each page of information is viewed beside the sidebar (which comprises two frames, the top one and the one containing the answers to the questions).

d. Use of the Client’s Colour Scheme

The shade of blue used in the sidebar was changed to a background colour rather use the three-dimensional graphic from the Client’s Web site. The colour chosen was one of the 216 browser-friendly colours (#000099; RGB 0-0-153). Similar colours were chosen for the links and arrows on the main pages.

 

4.5. Implementation and unit testing

4.5.1. Frame Architecture

Frames are used to split up the screen into individual areas; the layout of the frames are defined in a page called a frameset document. The contents of an individual frame are set by the Web page that the frame refers to. In the example below, the user sees one Web page, but it is actually made up four: the frameset document, and the three individual frames. To make it easier to refer to frames, they are named. This is useful when the program has to open a new page in a frame.

Figure 4-3 Screen Shot Showing Frames

 

JavaScript can refer to frames by using a simple numbering system. In the JavaScript example below, a loop is used to read in the answers into an array, decrypting them as well.

for( z = 0; z <= 24; z++ )

{

real_answer[ z ] = ( ( top.frames[ 1 ].listening[ z ] ) / x );

}

 

4.5.2. Input and Output

The user can input their choice by selecting a radio button or by typing in their answer into a text box.

a. Radio Buttons

These selection buttons are grouped together, so that only one choice can be accepted. In this way, the user can only select one answer for a given question.

b. Text boxes

Also used for the user to input answers in the Practice and Writing test, and to type in data about themselves in the 'Goodbye' section.

Text boxes are also used when outputting data for the user's information, e.g. scores. This is carried out as part of the JavaScript program that is invoked when the user presses the "Mark The Test" button.

<input type="button" name="B1" value="MARK THE TEST" onclick="ScoreTest()">

The ScoreTest function is similar for each of the tests.

1. The variables are initialised and the text box is cleared to make way for the final score.

function ScoreTest()

{

var score = 0; // running total

var count = 0; // loop counter

var z; // loop counter

var x = 12; // codebase

var real_answer = new Array; // correct answers

var todays_date = new Date(); // todays date

var today; // string to hold dates

// clear text boxes

document.test.textbox_output.value = " ";

document.test.textbox_output1.value = " ";

document.footer.textbox_name.value = " ";

2. The correct answers are read into an array.

for( z = 0; z <= 24; z++ )

{

real_answer[ z ] = ( ( top.frames[ 1 ].listening[ z ] ) / x );

}

3. The given answers (already stored in an array read in from the radio buttons) are checked against the correct answers and a score is calculated.

for( count = 0; count <= 24; count++ )

{

if( real_answer[ count ] == given_answer[ count ] ) // if answered correctly

{

score++;

}

}

4. The final score and an adjusted score (required by the Client) are output to text boxes.

// display score and adjusted score

document.test.textbox_output.value = score; // final score

document.test.textbox_output1.value = score; // adjusted score for listening test is score * 1

5. Finally the date is output to a text box so that when the page is printed, the date of the test is printed out too.

// insert today's date in to text box

// NB: getFullYear is a function call for JS1.0 replication of a JS1.2 command

today = ( todays_date.getDate() + "/" + ( todays_date.getMonth() + 1 ) + "/" + get_Full_Year(todays_date) );

document.footer.textbox_date.value = today;

}

4.6. System testing

Version 4 of the program was tested by English language students from Aberdeen College. The students were selected by the Client to give a range of English ability, and did not refer to IT ability at all. This provided a range of language and computer literacy.

Analysis of the students’ feedback is given in Appendix 7, however, the main points were:

a. Listening Test

Sound took too long to download, catching students unawares when it did start to play. To combat this, chimes were added to warn the user that the question is about to start.

b. Reading Test

Some students had difficulty understanding the context of the question. After discussion with the Client to decide if the users’ English language comprehension was the problem, the solution was to add thumbnail pictures to ‘set the scene’.

A timer was also added, which allows the user 50 minutes before bringing up a warning message in a small window.

c. Writing Test

This caused problems for the students. The main problem was that the user had to make corrections in a text box. This required more computer literacy that the author expected. As a result, some of the work was marked wrong because the user typed in too many spaces, for example. The author noted that the Client would, therefore, have to read over the computer marked answers. After discussion with the Client, the test was changed to a ‘free-writing’ test. The topics for this are chosen at random, and a toggle prevents the user from changing the selection.

The changes were incorporated into Version 5 of the program, submitted as part of this project in Appendices 8 and 9.

4.7. Operation and maintenance & Feedback

This has not yet been carried out, as the Client is on holiday for the Summer. As the coding is open-source, and the author works for the same College, updates and further testing can be carried out at regular intervals.


CHAPTER 5 - Conclusions

The purpose of this thesis was the computerisation of Aberdeen College’s English as a Foreign Language test. This has been partly achieved, and the program will have additional working trials when the students return to the College for the next teaching block.

The testing carried out on Version 4 of the program showed that the students preferred the computer-based test to the paper-based test. Joan Cooper, one of the Language lecturers who spoke to the students during part of the test, noted how engrossed they were. Although the students were aware that they were participating in a test, many were wanting to finish the questions; despite the fact that they had a limited time due to their timetabling.

This attitude ratifies the comments made by educationalists about the different attitude that people have towards computer-based training. Students can proceed at their own pace, and can fail in their own way. One of the students noted that they were not informed of their final score, a limitation due to the fact that the written test is too complex to be marked by the program; and that the student still has to undergo a final interview.

In conclusion, it is the nature of the current diversity and lack of standards that have limited this project. Even text boxes have to be narrow to allow for low resolution printers – if not, part of the student’s work does not print out! Hopefully, in the near future, Internet standards will be followed by the main browser manufacturers, and the majority of the limitations imposed on this project will be removed.


Chapter 6 - Future Work

The background to this project is constantly changing. Much work is being implemented by the UK Government into lifelong learning. The Department for Education and Employment (DfEE) has a dedicated lifelong learning site which has a number of initial, intermediate and final reports into the field; as well as an online magazine. [6-1] Educational establishments are being pulled into what the Government terms the "learning age", when everyone who wants to learn will have the opportunity to learn.

The technology involved is changing. The 'Millennium Bug' will force establishments to upgrade their computer systems, resulting in more modern computers for users, with up-to-date operating systems and browsers which are better able to handle online learning. As new versions of HTML are agreed, more will be able to be written that will work on popular browsers, rather than to either write your Web pages for one company's browser, or have to use coding written for 1995's version 3 browsers. These browsers will be replaced by more modern ones when the computers they are running on are replaced because they are not Year 2000 compatible.


References

[1-1] "Development Plan 1999-2002" by Rae Angus

Aberdeen College, June 1999

[2-1] "Technological Enhancement of Instruction: A Humanistic Approach" by Stephen R Buchanan, September 1998

http://horizon.unc.edu/TS/cases/1998-09.asp

[2-2] Prime Minister Tony Blair's foreword to "Connecting the Learning Society: Consultation Paper on the National Grid for Learning", 7 October 1997

http://www.dfee.gov.uk/grid/consult/foreword.htm

[2-3] "Development Plan 1999-2002" by Rae Angus

Aberdeen College, June 1999

[2-4] "How to Teach Adults" by Mary Howard, How To Books Ltd, Plymouth, 1996. ISBN 1-85703-193-8

[2-5] "Evaluation and Selection of Web Course Management Tools" by Sunil Hazari, 21 June 1998

http://sunil.emd.edu/webct

[2-6] "Cognitive Versus Behavioural Psychology" by Fred T Hofstetter, 1997

http://www.udel.edu/fth/pbs/webmodel.htm

[2-7] "Seeing Things as They Really Are" by Robert Lezner & Stephen S Johnson, 10 March 1997

http://www.forbes.com/forbes/97/0310/5905122a.htm

[2-8] "Extending Community Colleges through the Internet" by Microsoft Corporation

http://www.microsoft.com/education/HED/online/extendcc.htm

[2-9] "Tree of Knowledge" Macromedia Magazine, 1998

[2-10] "Look, no Campus" by Martin Whittaker, Times Educational Supplement: Business Links, 4 June 1999, p30-1

[2-11] "New Online Training a Boon, But Won't Replace the Classroom" by Sandra Gittlen, CNN, 9 June 1998

http://cnn.com/TECH/computing/9806/09/training.idg/index.html

[2-12] "Applying Taylor's Efficiencies in Cyberspace" by Dee Stallings (December 1997)

http://horizon.unc.edu/TS/vision/1997-12.asp

[2-13] alt.education.distance Newsgroup Frequently Asked Questions Number 1, 23 April 1999

http://www.faqs.org/faqs/education/distance-ed-faq/part1

[2-14] alt.education.distance Newsgroup Frequently Asked Questions Number 3, 23 April 1999

http://www.faqs.org/faqs/education/distance-ed-faq/part 3

[2-15] "Distance Learning: A Faculty FAQ" by Judith V Boettcher and Rita-Marie Conrad, August 1997

http://horizon.unc.edu/TS/other/ol1997-08.asp

[2-16] "Virtual Campus" by Hiawatha Bray, The Boston Globe Magazine, 11 April 1999

http://www.boston.com/globe/magazine/4-11/feauturestory3.shtml

[2-17] "FYI on 'What is the Internet?'" by E. Kroll and E. Hoffman, May 1993.

http://burks.bton.ac.uk/burks/internet/general/fyi20.htm

[2-18] "The Whole Internet User's Guide & Catalog" by Ed Krol, O'Reilly & Associates, Inc, California, 1994. ISBN 1-56592-063-5

[2-19] "Global Internet Statistics" Global Reach, 10 April 1999

http://www.euromktg.com/globstats/

[2-20] SCET's Noel Chidwick speaking at the SCET seminar: "Online Learning Environments in Focus" 14 May 1999

[2-21] "Information Management: A Proposal" by Tim Berners-Lee, 1989

http://www.w3.org/History/1989/proposal.html

[2-22] "All You Need to Know About The World Wide Web" by Davey Winder. Future Publishing Limited, Bath. 1994. ISBN 1-898275-40-8

[2-23] Teknical's John Waite speaking at the SCET seminar: "Online Learning Environments in Focus" 14 May 1999

[2-24] "Maximizing Collaboration" Microsoft Corporation, 1 June 1999

http://www.microsoft.com/education/hed/online/collab.htm

[2-25] "Best Practice Arrangements for Delivery of Individual Module/Unit/Subject/Short Course" by Aberdeen College, 3 December 1998

[2-26] "Incorporating Electronic Technology into a Distance Learning Course" by Phyllis B Oakes, October 1997

http://horizon.unc.edu/TS/cases/1997-10.asp

[2-27] "Promoting Community" New Chalk, edited by Roger Akers, et al, December 1996

http://www.unc.edu/courses/newchalk/archive/ncv1n3.html

[2-28] "Distance Education: The Unofficial FAQ" by Steve Levicoff

http://members.tripod.com/~levicoff/de-unfaq.htm

[2-29] "Academics Rebel Against an Online Future" by Todd Woody, CNN, 12 June 1998

http://cnn.com/TECH/computing/9806/15/academics.idg/index.html

[2-30] "Does Using Technology in Instruction Enhance Learning?" by Ed Neal, June 1998

http://horizon.unc.edu/TS/commentary/1998-06.asp

[2-31] "Transforming Higher Education Using Information Technology: First Steps" by Barbara H Horgan, January 1998

http://horizon.unc.edu/TS/vision/1998-01.asp

[2-32] "Online Testing" New Chalk, edited by Roger Akers, et al, November 1996

http://www.unc.edu/courses/newchalk/archive/ncv1n2.html

[2-33] "Internet Computing" by Dr A D Marshall

http://www.cm.cf.ac.uk/Dave/Internet/

[2-34] "Encouraging Student Use of Internet Resources" New Chalk, edited by Roger Akers, et al, November 1996

http://www.unc.edu/courses/newchalk/archive/ncv1n1.html

[2-35] "The Luton Experience" by Stan Zakrzewski, 26 September 1997

http://www.icbl.hw.ac.uk/ltdi/assessit/luton.htm

[2-36] "Student Projects" New Chalk, edited by Roger Akers, et al, March 1998

http://www.unc.edu/courses/newchalk/archive/ncv1n8.html

[2-37] "Letters to the Editor" by Ed Neal and Jerald Schutte, September 1998

http://horizon.unc.edu/TS/letters/1998-09.asp

[2-38] "Distance Learning: Implications for Higher Education in the 21st Century" by Nancy M Levenburg and Howard T Major, November 1998

http://horizon.unc.edu/TS/commentary/1998-11.asp

[3-1] World Wide Web Consortium

http://www.w3.org

[3-2] World Wide Web Consortium’s HTML 3.2 specification

http://www.w3.org/TR/REC-html32.html

[3-3] World Wide Web Consortium’s HyperText Markup Language Activity Statement

http://www.w3.org/MarkUp/Activity.html

[3-4] Netscape’s Introduction to JavaScript

http://home.netscape.com/eng/mozilla/3.0/handbook/javascript/getstart.htm

[3-5] Sun Microsystem’s History of Java

http://java.sun.com/features/1998/05/birthday.html

[3-6] Sun Microsystem’s Java Developer’s Kit

http://java.sun.com/products/jdk/1.2/index.html

[3-7] Macromedia Director 7 Shockwave Internet Studio

http://www.macromedia.com/software/director/

[3-8] "Web Pages That Suck" by Vincent Flanders & Michael Willis

http://www.webpagesthatsuck.com/

[4-1] "Methods for Requirements Elicitation" by Stephen Armitage 1996

http://www.cs.ucl.ac.uk/staff/S.Armitage/project/project.html

[4-2] The University of Glasgow's Teaching with Independent Learning Technologies

http://www.elec.gla.ac.uk/TILT/TILT.html

[6-1] The Individual Learning Division of the Department for Education and Employment (DfEE)

http://www.lifelonglearning.co.uk


Bibliography

"JavaScript: The Definitive Guide, 3rd Edition" by David Flanagan

http://www.oreilly.com/catalog/jscript3/

Java tutorial

http://java.sun.com/docs/books/tutorial/

"Java in a Nutshell, 2nd Edition" by David Flanagan

http://www.oreilly.com/catalog/javanut2/

Graphics tips

http://www.zeldman.com/faq2.html

"Photoshop for the Web" by Mikkel Aaland

http://www.ora.com/catalog/webphoto/

Macromedia to Make Flash™ File Format Available as Open Internet Standard

http://www.macromedia.com/macromedia/proom/pr/1998/flashstandard.html

Introduction to Multimedia

http://www.zeldman.com/faq3.html

Java Sound Files

http://java.sun.com/docs/books/tutorial/applet/appletsonly/sound.html

Web Design Articles

http://www.efuse.com/

"Web by Design" by Molly E. Holzschlag

http://www.designstudio.net/

"Web Navigation" by Jennifer Fleming

http://www.oreilly.com/catalog/navigation/

Frames tutorial

http://www.newbie.net/sharky/frames/intro.htm

Referring to frames in JavaScript

http://sharkysoft.com/tutorials/jsa/

Form tags

http://www.w3.org/TR/REC-html32.html#form


APPENDICES

Appendix 1 - Statement of Research Problem

Appendix 2 – Paper-based EFL Tests

Appendix 3 – Minutes of Meetings

Appendix 4 – Software Life-Cycle

Appendix 5 – Screen Shots

Appendix 6 – Online Materials

Appendix 7 - Testing Of The EFL Program – Version 4

Appendix 8 - Printout of Program Coding

Appendix 9 - Program


APPENDIX 1 - Statement of Research Problem

School of Electronic and Electrical Engineering

The Robert Gordon University

MSc in Information Systems

Project Specification - Session 1998-99

Project Title: English as a Foreign Language Testing Suite

Student: Duncan I MacLeod

Supervisor: G Dunbar

 

 

BACKGROUND:

The English as a Foreign Language (EFL) test is used by the Languages team at Aberdeen College to place students in a band applicable to their level of knowledge of the English language. The EFL test is composed of 4 parts, each part contributing to the final 'placement' score:

Aberdeen College is interested in open learning over the Internet and are keen on types of assessment which integrate well into a web environment. The College is also working towards the development of a learning intranet of which the present project could be part.

 

AIM:

The computerisation of Aberdeen College's English as a Foreign Language (EFL) test.

 

OBJECTIVES:

1. To undertake the necessary background reading and to become familiar with:

2. To investigate other similar projects.

3. To design, implement and test a suite that uses Internet technology to test a student's skill in English as a Foreign Language.

 

MSc CHALLENGE:

This project brings together computer-based learning techniques and Internet technologies. The student will have to consider the issues of browser compatibility; feedback to examiner and student; teaching the students how to use the computer; the integrity of the test (i.e. does not allow cheating); ease of use; visual attractiveness; minimising associated lecturer workload; and introducing the user to the test styles, to put them at their ease.

Many of the problems in this project could be overcome by using conventional languages (eg. C++), however, these could not be delivered over the Internet, and are of no use in the open learning environment that Aberdeen College hopes to develop.

 

SPECIALIST EQUIPMENT REQUIRED:

None.

 

INDUSTRIAL COLLABORATOR:

Aberdeen College.


APPENDIX 2 - Paper-based EFL Tests


APPENDIX 3 - Minutes of Meetings

 

Date

1 December 1998

 

Attending

David Morley [Head of Information Services, Aberdeen College]

Duncan MacLeod [Candidate]

 

Aim

Discuss the MSc project, how we were to take it forward.

 

Points

Project would involve the following:

1.Details of how the software would:

gather input

analyse input

mark input

give feedback to student

2.A managerial element would be involved - and not just producing the software.

3.Background research was required, ie:

by working with teachers

carrying out a literature review

4.The software would need to be tested and a pilot run before implementation.

 

Further Action

1.Duncan MacLeod to arrange a meeting with Anne Bain (David Morley will attend if it is felt necessary).

2.Duncan MacLeod will then produce a Project Specification for approval by RGU.


Date

8 December 1998

 

Attending

Graeme Dunbar [Supervisor, RGU]

Duncan MacLeod [Candidate]

 

Aim

Progress meeting.

 

Points

Discussed project so far, and some future proposals.

1. Must make sure that program includes a user manual and a manual for the overseeing tutor. Both can be online, but candidate would prefer that the tutor's manual would also be in printed form.

2. RGU has a CD writer, so that the finished program could be published on CD. Perhaps include a copy of Netscape Communicator (Win95/NT), as the program will require a version 4 browser for layering animation. The tutor's manual could, therefore, be in PDF format, and include the free reader program on the CD too.

Further Action

1. Duncan MacLeod to post working notes to the Web so that supervisor could monitor progress more efficiently.


Date

15 December 1998

 

Attending

Anne Bain [Languages Team Leader, Aberdeen College]

Duncan MacLeod [Candidate]

 

Aim

Initial fact-finding meeting.

 

Points

Project: The computerisation of Aberdeen College's English as a Foreign Language (EFL) test.

The EFL test is composed of 4 parts, each part contributing to a final 'placement' score. The conversion factor is used to pro-rata the test's score to a final percentile placement score.

Table 1: EFL Placement Test Scoring System

Part

Type

Marks

Conversion factor

1

Listening

25

x 1

2

Grammar/reading

50

/2

3

Writing

5

x 5

4

Speaking/interview

5

x 5

 

The placement score is used to place students in a band applicable to their level of knowledge of the English language.

Table 2: Level of Knowledge

Score (%)

ECLES Levels

Description

0-20

ESOL 1

Complete Beginner

21-39

KET

False Beginner

40-60

PET

Elementary

61-80

FCE

Intermediate

81-90

CAE

Advanced

91-100

CPE

Very Advanced

 

Suite is to include:

 

Further Action

1. Duncan MacLeod to draft out a demonstration suite.

2. Anne Bain will produce a list of closed passage exercises for the writing test.

3. Duncan MacLeod to check the specification of the 10 PCs that Languages uses, in case browser technology will be a problem.

4. Next meeting to be arranged after the New Year.


Date

2 February 1999

 

Attending

Anne Bain [Languages Team Leader, Aberdeen College]

Duncan MacLeod [Candidate]

 

Aim

Progress meeting.

 

Points

Project: The computerisation of Aberdeen College's English as a Foreign Language (EFL) test.

A sans serif font, like Arial, is easier for users to read than a serif font, like Times New Roman.

Users may have difficulty with terms like "site map". Use icons where possible. Standard icons are 'pen' for writing, 'ear' for listening and an 'open book' for reading.

Use graphics to make the screens more interesting.

10 PCs in language labs: Evesham Vale Prestige, Microsoft Windows 95, Microsoft Internet Explorer 3 (with Java Virtual Machine). 32MB RAM. Matshita CD-ROM CR-585. 1.96GB Hard Drive (1.54 GB free space).

Foreign language fonts must be included with the installation disk.