F. Layne Wallace (lwallace@unf.edu)
Department of Computer and Information Sciences
University of North Florida
Doreen Radjenovic (dradjeno@unf.edu)
Department of Nursing
University of North Florida
Keywords: Remote training, HTML, diabetes, in-service education, computer versus paper training
When school age children are diagnosed with diabetes mellitus, the school teachers' are faced with additional responsibilities. Diabetic children spend a large part of each week day with the teacher as the primary supervising adult. This means that the teacher must be aware of warning signs and implications of diabetic reactions in children. Sadly, studies have shown that public school teachers' knowledge of diabetes is lacking (Bradbury and Smith, 1983; Gesteland, et al., 1989; Hodges and Parker, 1987; and Lindsay, et al., 1987). A partial reason for this lack of basic knowledge is the current nature of the in-service training on diabetes.
At present, teachers learn about diabetes in a lecture setting from a diabetes trainer using a paper booklet. This technique works well for the short term but has a number of constraints that make the application of the information difficult. Specialized training with a diabetes education professional is geospecific in that the training must take place in a classroom or auditorium setting. Teachers must travel to this location to receive the information. For teachers in small or rural schools, this travel time can be substantial, taking teachers out of the classroom for extended periods of time.
Current diabetes training is also time constrained. The teachers must attend the training sessions at specific times. These training periods are not necessarily scheduled for the teachers' convenience. If an emergency arises and the teacher cannot attend the training session, then the teacher must wait until the next session. Another aspect of the time constraint is that of availability. Current lecture training techniques are given infrequently and the teachers, understandably, forget the diabetes information.
The last constraint of interest to this research is that of material constraint. While the material presented during most in-service diabetes training has adequate content, the presentation is linear and under the control of the diabetes education trainer instead of the teacher. Thus if a teacher doesn't understand a particular point, the entire training session must stop and go over the material again. This tends to inhibit any teachers who would like additional information on a particular topic. It is also rare when in-service diabetes training sessions can bring in diabetes experts, children with diabetes, or parents of children with diabetes to discuss methods of dealing with the disease.
One possible solution to these constraints would be to use computer assisted training with computer video and sound. However, recent attempts to implement computer assisted training for general diabetes information have not met with success (Gesteland, et al., 1989), and a search turned up no software specifically designed for in-service teacher training (Computer Programs, 1993). Part of the reason for this lack of success is the fact that most current computer assisted training techniques are machine dependent. Training modules implemented on one type of computer cannot be easily moved to another type of computer. Most current diabetes training software is written for use on a standalone computer and those that are developed for use on a computer network are designed for a specific brand of network. Such software is difficult to update and often the content soon becomes obsolete.
One possible solution to the problems outlined above would be to develop a computer assisted training system with HyperText Markup Language (HTML). HTML would allow the training to include text, sound, and computer videos. Because HTML is machine independent, the training module could be used, without change, on a variety of different computer systems. Additionally, HTML can be easily used on a standalone computer, a local area network, and through the Internet. The current research discusses how such a system could be developed and whether the system would compare favorably to standard paper-based diabetes instruction.
Subjects
Forty-two College of Education students at the University of North Florida
were randomly divided into two research groups: Paper-based information and
HTML-based information. Both research groups were given a demographic
questionnaire (on paper) before testing began. Both research groups
contained subjects who were currently teaching school and subjects who
were completing their education in preparation for a teaching career.
Twenty-nine females and thirteen males participated. All subjects had
vision that was naturally 20/20 or corrected to 20/20. Three of
the subjects had once had students who were diabetic and 30 had a family
member or a close friend with diabetes. Thirty-eight of the subjects had
some computer experience; both research groups had subjects with little
computer experience.
Research Procedures
The material content for both research groups was based on a booklet provided
by the American Diabetes Association. Every effort was made to insure that
the two research presentations (paper-based and HTML-based) were equivalent.
However, the HTML research group had supplemental material in the form of
vocal material from children with diabetes and computer videos of
children discussing diabetes. This supplemental material was not part of the
primary presentation and could only be accessed by having the subjects
explicitly select the material. The diabetes material was separated into
chapters based on the paper version. At the end of each chapter, the subjects
could either click on an icon to take them to the next chapter or return to
the last chapter.
All subjects were given an open-ended time period to view the material. No subject took longer than 45 minutes to go through the material. Both groups completed the tasks in group settings: the paper-based group in a class room and the HTML-based group in a computer lab. The HTML group was allowed to adjust the keyboard, mouse, and monitor to comfortable positions. The computer equipment was selected for its minimal nature. Sound was produced through the computers' small speakers. The monitors were set for VGA resolution. The CPU in each machine was a 33 Mhz Intel 80486.
After completing the diabetes material, each group was given the Diabetes Knowledge Test Questionnaire. This questionnaire was administered in a paper format. The tests were scored by summing the number of correct answers. These scores were used as the dependent variable in the analysis. Once the subjects had completed the test, informal comments were recorded.
Analysis
Pearson Product Moment correlation analyses showed that all groups were independent with respect to age, computer experience, prior diabetes knowledge, and class room experience. A Student's t-test showed that the HTML-based group had statistically higher scores (t = 2.22, p < .033) than the paper-based group (20.3 correct answers for the HTML group versus 18.6 correct answers for the paper group).
This research project compared two training modules about children with diabetes, one administered on paper and one administered using an HTML computer system. The material for the training modules was intended for school teachers who might have children with diabetes in their classes. The primary finding of this research was that subjects who used the HTML training module retained significantly more information about diabetes than the subjects who read the diabetes information from paper.
This finding is quantified, direct evidence that HTML-based training modules could be an effective tool for in-service teacher diabetes training. HTML can provide a viable solution to many of the problems mentioned in the Introduction. However, a number of potential problems with HTML training modules were discovered during the testing.
Any questions that the subjects had about diabetes that weren't answered in the HTML module, would have to be referred to a diabetes education professional. By tracking these questions, a section of the HTML module could include a Frequently Asked Questions (FAQ) chapter. There were no other subject comments about the training content.
The lab setting presented several issues that need discussion. The computer speakers produced sound that was low quality but still understandable. Prolonged use of such poor quality sound would eventually irritate users and could produce headaches. Having a number of subjects, all hearing sound from the computer speakers at the same time, made concentration difficult. Using an inexpensive sound card and small headphones should eliminate these issues. Subjects who did not have much computer experience tended to interrupt other users to ask questions or to see what the others were doing. Having teachers use the HTML system in a private location would keep them from being disturbed and would allow the teachers to set their own in-service training schedules.
The computer video files were in Indeo AVI format. Pilot work by the authors had found that some other video formats were not as good (jerky, bad coordination between video and sound). This format worked well in the test computers and subjects commented about the positive aspects of the computer videos.
Several subjects commented on the navigation of the HTML training system. They felt that a larger HTML document would have caused problems as there was no mechanism to easily move with the document without going back to the Table of Contents. None of the subjects stated that they had trouble with navigation, just that it was an issue of concern.
The current study has provided evidence that further research is worthwhile. The HTML material was based on existing paper documents. Study should be done to take further advantage of HTML's Internet capabilities, such as links to diabetes sites on the Internet, Internet phone connections to diabetes education professionals, and on line testing and evaluation from a central location. While an HTML diabetes training module would allow the teachers to repeat the module at their convenience, long term retention of diabetes information should be tested.
In conclusion, this research has shown that an HTML system can present information about diabetes more effectively than having teachers read the material. Using HTML removes many of the long-term problems associated with current computer assisted training modules.
Updates to this paper will be housed on one of the author's computers.
Bradbury, A. J. and Smith, C. S. (1983) An assessment of diabetic knowledge of school teachers. Archives of Diseases in Children, 58, 692-696.
Computer Programs for Diabetes Management. (1993) Bethesda, MD: National Diabetes Information Clearinghouse.
Gesteland, H. M., Sims, S. and Lindsay, R. (1989) Evaluation of two approaches to educating elementary school teachers about insulin-independent diabetes mellitus. The Diabetes Educator, 15, 510-513.
Hodges, L. C. and Parker, J. (1987) Concerns of parents with diabetic children. Pediatric Nursing, 13, 22-24.
Lindsay, R., Jarrett, L., and Hillam, K. (1987) Elementary school teachers' understanding of diabetes. The Diabetes Educator, 13, 312-314.
Name: F. Layne Wallace
Title/Position: Associate Professor
University/College/Affiliation: Department of Computer and Information Sciences, University of North Florida
Postal Address: Dept. of CIS
4567 St. Johns Bluff Rd., S.
Jacksonville, FL 32224
e-mail: lwallace@unf.edu
N.A.WEB 96 - The Second International North America World Wide Web Conference http://www.unb.ca/web/wwwdev/ University of New Brunswick.