University of West Indies (UWI) was established in 1948 to cater for the English-speaking population in the Caribbean. It was formed as a college in collaboration with University of London at Mona, Jamaica (Brandon, 1999). The university consists of three campuses located in Mona (Jamaica), St. Augustine (Trinidad and Tobago), and Cave Hill (Barbados). In addition, twelve non-campus countries contribute to the university for whom the university runs distance courses. The distance education program at UWI began in 1983-84 under the name University of West Indies Distance Teaching Experiment (UWIDITE). Since then, UWI began experimenting with the use of telecommunications to deliver courses to off-campus students through UWIDITE (Perraton, 2000). Satellite links were established between campuses and off-campus centres. However, technical weaknesses as well as managerial difficulties led to UWIDITE exploring new potentials. UWI lags behind USP in integrating its distance education programme with other activities.

The regional models used by both USP and UWI are only applicable to small island nations where neighbouring nations are also small island nations. Even with these collaborations the host countries dominate and the countries with fewer resources suffer as a consequence. This disparity will be higher when small island nations are situated alongside huge countries where total control will be taken by the larger power players making this sort of collaboration impossible. Maldives is a prime example of a small island nation with less than 300 thousand people surrounded with countries having millions and billions of people. Hence the nations like Maldives have to seek alternative models.

Maldives is a prime example of small island nations facing a number of barriers due their size and economic status. As it is not economically feasible to build a secondary and high school in each of the islands one of the ways to achieve economies of scale is to have regional centres where the students travel on a daily basis to get their education. However, the absence of regular ferries between the islands of Maldives makes it impossible for students to travel to the regional centre on a daily basis if the centre is not situated in their own island. For this alternative to work Maldives needs to invest heavily on establishing regular ferry services between the islands in addition to the human resources development and other infrastructure development costs. Alternatively boarding houses can be established in the islands where the regional schools are located and students can reside in these houses during the school term which again requires huge financial investments. This leads to the alternative of reaching the students wherever they are through distance mode. Hence a study into the best technology and delivery systems in Maldivian context was carried out.

THE DISTANCE EDUCATION MODEL
The distance education model was designed and developed focussing on accessibility, cost minimisation, and interactivity. Access to education was the main criteria in developing the model as there were no venues of access to secondary education in most parts of Maldives within the existing infrastructure. Hence, the distance education model designed and developed was aimed at providing access to secondary education for the students in the outer islands of Maldives. While providing access was a major criterion, cost minimisation in terms of both initial investments as well as recurrent costs was a huge factor due to the economic conditions of Maldives. Consequently, the model was designed based on the existing infrastructure of Maldives without requiring additional investments on infrastructure. The third criterion was the level of interactivity provided to the students within the model. Distance education programs are highly criticised was the lack of interactivity and isolation of students. Hence, it was prudent that an adequate level of interactivity was incorporated into the model. The model provides adequate interactivity for the students to interact with both content as well as the teacher.

A descriptive case study approach led to a deep understanding of the barriers and constraints for delivering distance education in Maldives. A process of elimination of the different technologies used for distance education was used to identify the most suitable choice for Maldives. This process will not explained in this paper as it beyond the scope of this paper. The model developed uses the existing computer networks to deliver the instruction, with Internet for teacher-student interaction. Three-tier architecture of communication is used to provide effective interaction between the teachers and students. In essence, the current infrastructure allows us to use computers as a medium of delivery without any access to Internet for individual students. However, at a regional level, Internet access is available and can be used for a limited amount of time without having to incur huge costs.

The distance education model developed in this study is based on the assumption that the students will not have personal Internet connections and therefore cannot be required to go online at any time on their own computers during their study. However, they need to communicate with the teacher who would be based in a different location (most probably in the main island); hence a link needs to be made between the students and the teacher. The model is based on three-tier architecture, which portrays these three entities: the student, the teacher, and the link between the two. The three tiers comprise the student module, regional module, and the headquarters module. The student module is the component that the student uses to learn and interact with the content. The headquarters module is the component with which the teachers interact to monitor students’ progress and provide feedback. The regional module is the link between these two modules and provides a bridge between offline and online modes. All the communication between the students and the teachers are done through the regional module. The student module is sent to each student, the regional module is set up in each of the regional centres, and the headquarters module resides at the central location where teachers are available. Figure 1 shows the communications structure of the model. All the communications between the student module and the regional module are offline; hence the students are not required to have Internet connectivity. All the communications between the regional module and the headquarters module are online, but they take place in batches, hence not requiring continuous online connection.

Figure 1: Structure of the model

The model in this study uses a CD-ROM/Web hybrid model to deliver distance education. A hybrid model is mainly used to get the benefits of both CDROM as well as Internet, while compensating the drawbacks of each other. Use of multimedia on the Internet poses major problems in terms of bandwidth which can be overcome using a CDROM with all the multimedia content. Similarly, using only a CDROM will provide static content which can be made dynamic and adaptive through the use of Internet. According to Metcalf (1996) a hybrid model will provide timely information, real-time interaction, and centralised user activity.

A hybrid model was used in this research mainly to provide an adequate level of adaptivity in content. Although Internet is not directly accessible to the students, Internet is used to provide the adaptivity. CDROM is used to deliver the content and Internet is used for regular updates to the content. Due to the lack of human resources in Maldives, especially teachers, the possibility of having teachers for each subject in each island is impossible. Hence, this model proposes human teachers to be located in the capital providing instruction to a number of islands at the same time.

EVALUATION
A learning system based on the theoretical model was designed and developed to implement the model. This learning system was then implemented to find out student accessibility within the remote islands of Maldives. The main purpose of the evaluation was to find out whether the proposed model is suitable in the Maldivian context. In order to find this out, three questions needed to be answered:

• Will the proposed model provide physical access to learning to the students in the islands?
• Will the students be able to learn through the proposed model?
• Will the students accept the proposed model as the basis for a form of teaching?

The evaluation was carried out at three different centres in the Maldives. This included two of the smallest islands in Maldives in terms of population as well as school enrolments. Furthermore, these two islands were considered very remote due to limited transport services to them. The third centre was at the capital where most of the facilities are available.

The completion of the evaluation in all the three centres proved that physical access to students in remote islands was possible using the model. Furthermore, the student responses from all the three centres strengthen the claim that this distance education model was suitable in delivering education to the remote islands of Maldives.

Physical accessibility is only one aspect of the students’ accessibility provided by the proposed model. The other two aspects were how much the students learn using the learning system, and whether the students would accept the proposed model as the basis for a form of teaching. Statistical analysis was used to measure these aspects of the learning system. T-tests and Correlation Analysis were used to answers these questions.

One aspects of the evaluation of this research is to identify whether the students were able to learn using the learning system. The most logical way to determine this was to identify their knowledge before and after using the learning system and compare them to find out whether they performed better in later case. Hence pre-test and post-test were used during the evaluation and the scores from these two were compared. Since there were no control groups, a control measure was used to check the reliabilities of the test scores. That is, the average scores from the students’ last grade at school were compared with their final scores in the evaluation to see if there is any relationship. The statistical tests determined that the students were able to learn using the learning system. Furthermore, results showed that the students increased their knowledge by 135% after using the learning system. When compared with their average school scores the results determined that the students scored 8% higher when using the learning system.

While it is important that any model of education is able to demonstrate that it can create effective learning, it is also important that students themselves are willing to accept the new method. Hence the evaluation also focused on determining whether the students accepted the learning system. Technology acceptance model was used as the basis to identify how well students accepted this learning system. Davis (1993) proposed the technology acceptance model where he argued that the perceived ease of use and perceived usefulness of a technology would formulate users’ attitudes towards using a particular technology and consequently to actual use of the technology.

Once the students completed their course using the learning system they were given a questionnaire which consisted of questions regarding their perceptions of the software. The questionnaire was divided into several scales, each scale measuring a different aspect of students’ perception of the learning system.

The research also looked at the students’ perceptions of the learning system. One of the aspects the study focused on was the students’ overall perception of the learning system. Most students liked using the learning system and were ready to use the learning system again for other subjects. Sixty four percent of the students stated that they liked using the software, while only 20% expressed some dislike. When asked whether they would use the learning system for any other subject 52% indicated that they would use the system for other subjects. These results were then compared with other factors including relating them to the students’ attitudes to and experience with computers.

Further analyses were done on the student acceptance of the model. Student acceptance was measured using following attributes of the technology acceptance model:

• Ease of use
• Perceived Usefulness
• System Design Features and Tools

In measuring the ease of use, three different aspects were measured: student perception of prior knowledge requirement to use the learning system, ease of use of the learning system, and ease of use of the layout of the learning system. The statistical tests showed that there was weak to no relationship between students’ attitudes towards and experience with computers, and their perception of learning using the system. Hence, it can be concluded that computer usage or experience has little effect on the students’ belief that they learnt using the learning system. Consequently, it can be concluded that the students are likely to accept the proposed model as a reliable form of teaching.

CONCLUSION
The main aim of this model is to provide access to secondary education throughout the country with a minimal expenditure on the infrastructure development. The model capitalises on the existing infrastructure and technology in the country reducing the costs required for implementation. The model also provides students interactivity and adaptivity without asking for direct Internet connectivity. One of the major advantages between this model and other commercially available systems is that this system consists of a two-layered CD/Web Hybrid approach where the students with and without online connections are catered for. The results of the evaluation showed that this model is both physically accessible and acceptable to the students in the remote islands of Maldives.

Although this research focussed on Maldives, the distance education model can be used in different small island nations with minor adaptations. Many small island nations in the South Pacific, like Fiji and Solomon Islands, share similar geographical constraints to Maldives. Hence, this distance education model can be easily adapted to cater for students in these island nations. These nations have limited accessibility to education in many parts of the country and a hybrid model can reach some of the remotest areas with extreme conditions. Further research need to be carried out in other small island nations to find out how the model can be adapted in these countries.

The implications of this research are enormous for the education in Maldives. In addition, to providing secondary education in the outer islands it can also be used to provide skill based training to the existing labour force. Currently, most of the students in the tertiary education sector are employed and they generally attend evening classes after work (personal observation). The proposed model will allow these students a more flexible learning environment.

The next stage of development will be to create comprehensive content on the secondary school curriculum using the model. Furthermore, more personnel will need to be trained, at least one per centre, to implement the model. One of barriers to providing secondary and higher education in the smallest islands is the lack of skilled human resources. This can be overcome using our model. In addition, the model will have most impact where there is a lack of teachers in those islands affected by the Tsunami. It will benefit the local population in the islands, where there is a lack of education due to inadequate human resources. Further research will also be carried out to find out how this model can be used in other small island nations, especially in the South Pacific.

REFERENCES
Brandon, E. (1999) “The University of West Indies” in K. Harry (eds) Higher Education through Open and Distance Learning, New York: Routledge.

Davis, F. D. (1993) User Acceptance of Information Technology: System Characteristics, User Perceptions and Behavioural Impacts, International Journal of Man-Machine Studies, 38, 475 – 487

Matthewson C. and Va’a, R. (1999) “The South Pacific: kakai mei tahi” in K. Harry (eds) Higher Education through Open and Distance Learning, New York: Routledge.

Metcalf, D. (1996) WebCD: A Model for CD-ROM Authoring Development with Internet Integrated Data, AusWeb96, July 1996, Gold Coast: Australia.

Perraton, H. (2000) Open and Distance Learning in the Developing World, London: Routledge.

Schlosser, L. A. and Simonson, M. (2002) Distance Education: Definition and Glossary of Terms (Draft), Sponsored by the Definitions and Terminology Committee, Association for Education Communications and Technology, Bloomington, IN