Interactive features for HTML-based tutorials in distance learning programs

Introduction

Interaction & Distance Learning

Students in distance education are learners who have chosen this mode of study usually because of the flexibility it offers; students are not bound by time and location. This is evident considering that most distance learners are not isolated geographically from an institute of higher learning (Geddes, 1993) and could have chosen a program in traditional education instead. But interaction is often sacrificed for this gain in flexibility as a consequence of the barrier of distance between student, peers, and instructor. This interaction has individual and social dimensions (Bates, 1990a & 1990b) where individual interaction is interaction between the individual and the learning material and social interaction is interaction between two or more people concerning the learning material. Both forms of interaction are necessary for learning and should be ensured through the technology of the media and the design of the distance learning presentation.

In 1993, the School of Computer Information and Science at Nova Southeastern University organized a multimedia research group to examine system and network issues and courseware development issues associated with the new technology on the Web. Within the Courseware Development Team, a committee of the research group, group members are looking at multimedia uses in traditional education for applicability to the computer media communications (CMC) distance learning environment (DLE). An initial review of how on-line tutorials were implemented at NSU revealed that most uses followed a hierarchical menu level approach with some interaction between student and tutorial and very little interaction between student and teacher (other than through outside communications) despite the potentiality of the hypertext environment through the Hypertext Transfer Protocol (HTTP) and associated HTML. Consequently, the team is investigating on-line CAI designs to improve individual and social interaction. Traditional CAI learner control strategies are being looked at for improving individual interaction and Common Gateway Interface (CGI) program support for social interaction.

Interaction Through Learner Control

Learner control is a characteristic of CAI where students are allowed to make decisions on the content and context of the CAI presentation, as opposed to program control where the learner has no opportunity for this type of interaction. This is not to say that program control CAI does not allow individual interaction. It is well understood that an individual can mentally interact with the knowledge domain of such programs, just like one interacts with a good novel, but cognitive learning theories suggest an interaction with the environment that shapes schemata and learner control is used to help facilitate this type of interaction. The interaction implementation strategies in traditional CAI have primarily included content strategies where interaction is achieved through learner control techniques such as content control, sequence control, pace control, display (strategy) control, and feedback control (Chung, 1992; Friend & Cole, 1990). Context strategies are also used to give the learner control over the background setting and text density (Higginbotham- Wheat, 1990).

Content control allows the learner to control the selection of the content he or she wishes to learn. The CAI has a set of topics from which the student can select. Sequence control, like content control, allows the selection of subjects, but in this case, all the subjects or topics have learning objectives related to the overall course goal and the learners' control is over the sequence of the topics to be learned.

Pace control is typically considered one of the strongest advantages of CAI. It is a form of individualized instruction which allows the learner to take as much time as necessary to learn the lesson concepts. Usually, this type of control involves some kind of student tracking system so users can take up where they left off from the last CAI session. It also includes features that allow the student to control movement from one screen to another (Hazen, 1985).

Display control allows the learner to choose one of several displays which cover the same subject matter or concept. A rule, an example, and simulation, all covering the same material, can be provided to allow the learner to select which options they personally need in order to grasp the lesson concepts. Feedback control allows the learner to either defer to computer controlled feedback or select options on the feedback mechanism.

Context control includes learner control over the background of the material and text density. Principles and concepts can be presented in different environments which serve as the backdrop to the lessons and also provide a form of situated learning ( Brown et al, 1989; Brown et al, 1993). Mathematical concepts, for example, can be presented as business, or scientific problems and the user is free to choose the situation (Higginbotham-Wheat, 1990). Text density refers to the density of the material presented--bullet statements as opposed to explanatory paragraphs, etc.

Implementation Strategies of Learner Control

Hicken, Sullivan & Lein (1992) approached learner control classifications by examining the strategy of the controlling implementation as opposed to the control mechanisms. They found that whatever type of control mechanism used, the implementation strategy generally fell into two categories. The first is characterized by an approach where the entire lesson is presented to the student and the student has positive control to eliminate unwanted material. The second is characterized by an approach where only basic and fundamental principles are presented to the student and the student has positive control to add to the material. They called these two approaches as "FullMinus" and "LeanPlus" respectively.

In a study using these two approaches and variations with incentive, the FullMinus group performed only marginally better than the LeanPlus group; however, learner behavior showed that the FullMinus group was able to cover more material in the same amount of time and this group had a more favorable attitudinal response than the LeanPlus group so there would seem to be a case for a FullMinus approach in the learner control strategy (Hicken et al, 1992).

Can Learner Control be Implemented in HTML-Based Tutorials?

It appears that both content and context methods of learner control can be applied through a Web structure although some content methods may be more difficult than others. The HTTP protocol allows branching of documents and can be used for both content control and sequence control. Pace control is already inherent in distance education and providing a student tracking system within the CAI would require additional programming or CGI support. Feedback control is limited by the basic HTML standard but can be improved greatly with the addition of CGIs.

Display control and context methods of control (background and text density controls) appear well within the educator's ability to implement using HTML. Learner control strategies like FullMinus and LeanPlus also seem possible.

Another challenge using HTML in CAI design actually results from the strength of the Web. The Web was originally designed to facilitate the growth of non-linear knowledge bases throughout the Internet. Users are free to enter works (an organized collection of documents) at different locations (documents) and they can establish these entries within their own home pages. A CAI, even one with extensive branching, tends to be a single work that expects students to begin at the same location and end at some defined criteria level. The paths between the two may be different for each student but the beginning is usually the same. Students' power to enter the works at any location may not meet the CAI designer's intent so the design of each document within the work must be able to stand alone so that confusion is not caused by references to previous topics not visibly seen. This capability to enter the work at any document location, on the other hand, might be a method of learner control unique to the distance learning environment, yet to be explored.

When one examines past studies on learner control in CAI in traditional settings, a general impression is formed that interaction, discussed in these studies between a student and CAI, is a representation of interaction (social) between the student and teacher/designer who authored the CAI program. It probably goes without disagreement, this implied social interaction may actually occur by other means; communications with peers or SME.

The central problem in the distance learning environment is that the effectiveness of the programmed interaction through learner control strategies may not be sufficient, in itself, to improve social interaction in the absence of other communications. There has been unanimous agreement among distance education theorists that two-way communications (between teacher and student) is an essential element in distance education (Mood, 1995). By looking at the learning situation, it helps to visualize where social interaction strategies are needed to improve student-teacher interaction in HTML-based CAIs.

Improving Social Interaction

First consider the need for student-to-SME knowledge base interaction. In this scenario, the student is taking the CAI and is confronted with an issue of special interest the student wishes to explore further by browsing related literature immediately. If the instructor has been maintaining a home page with an index into the knowledge domain, HTML links could be strategically placed into the tutorial to allow the student to deviate from the planned course of instruction to investigate related material.

This is a unique form of learner control that can be implemented on the Web which goes beyond traditional CAI uses of branching. With Web tools, totally unplanned and unprogrammed excursions are made available to the student. The student is literally interacting with the instructor's knowledge and the instructor's Internet links to the knowledge base of the subject matter.

Now consider the need for student-to-SME direct communications. In this scenario, the student is taking the CAI and is confronted with material that is confusing, or the student wants to add information, or take issue with the presentation. The student wishes to communicate with the instructor on the subject. In traditional CAI, the student can normally have face-to-face human interaction with the CAI subject expert at the next scheduled class session but in distance education, this is not always possible. The distance learner must make a mental note and then contacts the instructor by telephone, e-mail, or written correspondence and this delay may cause the spirit of the message to be diluted. With HTML/CGI forms technology, it would be possible to add hypernotes write capability to the on-line CAI that would allow email communications with the instructor while taking the CAI.

Finally, consider the need for instructor-to-student communications. Just as any teacher wishes to monitor student progression in the class room in order to help redirect when necessary, in this scenario, the instructor wishes to monitor the progress of students as they interact with the CAI. HTML/CGI might provide the capability to input student performance measures and CAI file access times into a data file which can be reviewed periodically by the instructor, and used to gauge student progression through the CAI, thus identifying student problems early on. The instructor would be able to generate follow-up messages to students.

HTML CAI design characteristics & considerations

In light of the above, the following considerations are offered:

1. It is possible to have no learner control by establishing the work (presentation) as a single document (file). The entire work would look much like an electronic version of a printed paper and would have no links to any other documents. Very basic learner control could be added to the document with a table of contents that links to internal document anchors. The advantage of this traditional approach to on-line Web documentation are the file can be easily printed or downloaded by students and less links may offer less access time; an economical concern. Improving interaction will lessen these advantages and should be cautiously considered in light of different learning styles but used where appropriate.

2. Once the work is divided into linked documents, learner control over content and sequence greatly improves, but will be unrestricted in that the user can establish their own home pages and links to augmented sources. The author should exploit this capability in the learner control strategy instead of considering it a liability, and should realize that the work may evolve into a complex web of information (improving social interaction in the cyber community) by the learner's actions as additional links are established to other sources. Such a learning environment may likely be reserved for advanced learning situations (Jacobson & Spiro, 1991).

3. Since a FullMinus learner control strategy is apparently more effective than a LeanPlus strategy in traditional uses of tutorials, an on-line tutorial that introduces new concepts should contain as much required information as possible and students can use the learner control capabilities of item 2 above to filter unwanted information. On the other hand, a LeanPlus strategy might prove more useful than the FullMinus strategy in a tutorial that is designed to reinforce concepts previously learned. The tutorial could list other Web source documents or search engines that the student could linked to if desired.

4. Text density is another area of great promise for on-line tutorials. Tutorial authors can develop two or more versions of each document within the work that would allow the learner the choice between abbreviated or expository information.

5. Despite the unrestricted content and sequence control inherent in an HTML work, authors should establish navigation features within each document that help users to move within the design originally envisioned by the author. The user, of course, is free not to use the navigation features, but for beginners, these features might be an indispensable help aid.

6. An orientation strategy should be used to add visibility to the structure of the entire work. Graphical presentations have been shown to be more effective at this than text based (Barba, 1993). A map of the presentation should be included and navigation methods to the map should be available within each document. The graphical map, however, must use HTML alternative methods since some students may only have text based browsers or have elected to turn off the graphical capabilities of advanced browsers in order to improve performance in data transfer.

7. The protocol for HTML forms allows data input by the user into a database file or electronic mail. Link objects could be added to the tutorial that students could activate when they need to contact the instructor. The instructor could review comments daily and respond to students as necessary.

8. Using HTML supported by CGI, implement tracking methods to gauge student performance through CAIs. Design them so that student problems can be easily identified and instructors can give feedback to students as early as possible.

Conclusion

The use of learner control strategies in traditional CAI has been study extensively (Williams, 1993) and is still on-going as technologies change the learning environment. We can use this research to help design learner control strategies for on-line tutorials. Additional research in learner control for on-line applications, however, will be needed to understand individual interaction in the on-line environment, with all its own technological challenges and constraints. Also needed are studies in the design of on-line CAI to see if there is any impact on social dimension of interaction.

References

Barba, R.H. (1993). The effects of embedding an instructional map in hypermedia courseware. Journal of research on computing in education. 25(4), 405-412.

Bates, A.W. (1990, March). Application of new technologies (including computers) in distance education: Implications for the training of distance educators. Paper presented at the Round Table on Training, Commonwealth of Learning, Vancouver, British Columbia. (ERIC Document Reproduction Service No. ED 332 683).

Bates, A.W. (1990, September). Interactivity as a criterion for media selection in distance education. Paper resented at the Annual Conference of the Asian Association of Open Universities. (ERIC Document Reproduction Service No. ED 329 245).

Brown, J.S., Collins, A, & Duguid, P. (1989). Situated cognition and the culture of learning. Educational researcher., January-February. 32-42.

Brown, J.S. & Duguid, P. (1993). Stolen knowledge. Educational technology. March., 10-15.

Chung, J.S. (1992, February). An advance toward instructional management: Prescriptive knowledge base of learner control. Proceedings of selected research and development presentations at the convention of the Association for Educational Communications and Technology and sponsored by the Research and Theory Division. (ERIC Document Reproduction Service No. ED 347 981).

Friend, C.L. & Cole, C.L. (1990). Learner control in computer-based instruction: A current literature review. Educational technology. November, 47-49.

Geddes, T. (1993, July). The higher education distance learner and technology. Paper presented at the Australian and South Pacific External Studies Association Biennial Forum on Distance Education Future, Adelaide, Australia. (ERIC Document Reproduction Service No. ED 365 309).

Hazen, M. (1985). Instructional software design principles. Educational technology. November. 18-23.

Hicken, Sullivan & Lein (1992). Learner control modes and incentive variations in computer-delivered instruction. Educational technology research & development. 40(4), 15-26.

Higginbotham-Wheat (1990, Feb). Learner control: When does it work?. In proceedings of selected paper presentations at the convention of the Association for Educational Communications and Technology.

Jacobson, M.J. & Spiro, R.J. (1991). A framework for the contextual analysis of computer-based learning environments. (Report No. 527). Washington D.C.: Office of Education Research and Improvement. (ERIC Document Reproduction Service No. ED 329 938).

Kenny, R.F. (1992). The generative effects of instructional organizers with computer-based interactive video. Proceedings of selected research and development presentations at the convention of the Association for Educational Communications and Technology and Sponsored by the Research and Theory Division. 430-451.

Kinzie, M.B., Sullivan, H.J., Beyard, K.C., Berdel, R.L. & Haas, N.S. (April 1987). Learner versus program control in computer assisted instruction. Paper presented at the annual meeting of the American Educational Research Association, Washington D.C.

Mood, T.A. (1995). Distance education: an annotated bibliography. Englewood, CO.

Williams, M.D. (1993). A comprehensive review of learner-control: The role of learner characteristics. In Simonson, MR. & Dristen, A.'s (Ed.) Proceedings of Selected Research and Development Presentations at the Convention of the Association for Educational Communications and Technology, New Orleans, LA.

Hypertext References

HREF1

http://scis.acast.nova.edu/~eatonm - Mark Eaton's Home Page

Copyright