Today there is an increasing demand on educational institutions to "produce" more people that have a higher education. Interestingly enough, the resources for educational institutions in Sweden are at the same time diminishing, forcing them to look into new more cost effective ways of teaching. By using computer technology many hope to find a more efficient way of teaching without sacrificing the quality.

We have concerned ourselves with one aspect of this technology – what tools are needed to produce online course material and how should they work. As a first step to begin to understand these requirements we have made a number of informal interviews with teachers at Luleå University of Technology.

The interview indicates that an authoring tool for teachers should be able to handle different versions of documents, searching, reuse and patterns. We are now in the initial phase of a project that aims to explore this functionality further.

Background

Today it is often necessary to have good education to find a job, but it is interesting to note that the resources for education have become more limited. At Luleå University of Technology the time reserved for one course has decreased from one-fifth of a teachers annual duties in 1986 to about one-eighth today. There has been an increasing demand on universities to "produce" more students but the resources given have not increased proportionally.

That together with the demand for decentralized teaching in Sweden has made the use of computer technology an interesting approach. Today many universities use different kinds of computer tools to save resources and at the same time make it possible for students, that are off-campus, to participate in courses.

We are concerned with the tools that the teacher uses to produce these online courses since the tools offered today are both complex and complicated to use. In order to begin understanding the needs of the teachers, we decided to make a few informal interviews with five teachers at Luleå University of Technology.

Conducting informal interviews

We decided not to ask the teachers to fill in forms or to ask them a number of prepared questions; instead we tried to make the interviews an informal conversation about courses, teaching and teaching-aids. During the conversation we tried to cover a number of subjects and questions that we had prepared beforehand. We did our best to avoid asking the questions, but instead wrote down the answers as they occurred during the conversation. We only asked a question directly when we felt that the teachers were not going to touch the subject at all.

The interviews were made in each informant’s office. Notes were written down during the conversation and transcribed directly after finishing the interview.

The informants

The informants were all teachers at Luleå University of Technology; they work in different departments teaching psychology, structural mechanics, computer programming and chemistry. Their age ranged from their thirties to their sixties with a teaching experience ranging from three years to over thirty. They all seemed to enjoy being a teacher saying that one of the biggest rewards of their work was to successfully explain a subject to a group of students. Depending on their subject they used computers to a various degree, ranging from once-a-day to the whole day; most of them used their machines for word processing and similar tasks.

Teaching

Below follows a summary of what the informants talked about during the interviews.

Creating a new course

During the interviews we discovered that the informants all followed a certain pattern when developing new courses. The first step was to decide if there was a need for a new course or not. This was done by looking at courses given at other educational institutions, talking to friends, colleges and companies. The second step was usually to select a text book that would form the basis for the course; the informants emphasized that it was important to choose a good text book. The third step was to make a detailed course plan, most often closely following the text book with only minor adjustments to make sure that the content would fit the time frame of a lecture.

Teaching an existing course

All courses were updated each time they were offered. The updates could be minor, such as rearranging the order in which the material was presented during a lecture, or major, such as a change of programming language. These revisions can be a major part of a teachers workload in certain departments. One notable example of this is the Computer Science department that is forced to do major updates to many of their courses every year. This forces the teachers to spend a lot of time doing revisions, correcting mistakes in the material and adding new material. This does not leave much time for improving the pedagogical side of the material. In other departments, the courses had looked almost the same for the last hundred years. So, a lot of time is available to polish the course.

If a course was given in parallel by the same teacher, there was a difference between each presentation of a certain lecture. As someone put it: "If you have the chance to choose between two presentations on the same subject by the same presenter, choose the second … the quality of the presentation is much better".

While the course material is to a large extent determined by the text book, the students have a big influence on the presentation of the material. If the students are active, asking questions and questioning things said by the teacher, the presentations are generally better; at least several of the informants told us this. Most of the informants said that they preferred to have an informal atmosphere during the lecture, giving room for discussions and questions.

The structure of courses and lectures

When asked if there were some similarities concerning the structure of a course all the informants said "yes" and meant that there were strong similarities among the courses each of them taught. Since we only asked the informants to judge the courses they were teaching, we cannot say if these similarities were common for all courses.

Each informant also seemed to follow a common pattern when he/she designed the structure for each lecture. It seems that there is some "standard" patterns for how teachers create different kinds of lectures, but here it was noticeable that each teacher had their own standard or style. In fact it seems that these individual patterns can cause some friction between teachers (see "Working together") but all seemed intent on giving the lectures "their own touch".

Working together

We found large differences between different departments when we asked how teachers worked together. At some departments, working together is common practice, and it is done either by the teaching several courses in parallel by different teachers or by having a set of courses that build on each other with the teachers required to teach every course once in a while.

Other departments have tried working together, letting several teachers teach a course in parallel. The problem has been that one or more teachers have refused to follow the guidelines laid down for teaching the course. This has resulted in that the department have not been able to let the students write the same exam since the content for the different classes have varied strongly. The solution to this has been to let one teacher teach all classes.

All the informants mentioned that they used material created by other teachers, for example by borrowing a binder and copying exercises from it or by copying a computer file. But, some mentioned that copying files often gave rise to format conversion problems. That is, the material is available in format A, but the teacher can only read format B. The solution to this problem seems to be:

• Define a lowest common format, that is, something that everybody can read and edit. Examples of these formats are plain text and HTML. Usually these formats lack the ability for advanced formatting. If advanced formatting is available, the task of doing the formatting becomes daunting.
• Define a common software platform, everybody agrees to use certain software which makes it possible to exchange documents. This works fine as long as everybody keeps using this platform but there is usually someone who wants to use an older/newer version of a program or a completely different program.
• Use paper. This usually works well but there are problems concerning slow distribution of documents and some incompatibility problems, like different page sizes and punch holes.

Evaluation

All the informants thought it was important to get feedback from the students and they all collected feedback information when a course was finished (it is mandatory to do this at our university). However, there is a great difference in what happens with this information after it has been collected. At one department the evaluation forms stay with the teacher, and no follow-up from the department is done. At another all the forms are collected by the Director of Undergraduate Studies who reads them and offers the teachers feedback. One even summarizes the replies , presents this to the students, and discusses suggestions for how to improve the course.

Some of the informants noted that this feedback was not enough; it was both too late and too imprecise. A few of the informants tried to remedy this by talking to the students in small groups or on a one-to-one basis, asking for their opinions about the course. Another method was to hand out one or two small feedback forms during the course and evaluate those.

It could also be noted that no one seemed to have any method for remembering ideas, comments, student feedback from one time a course was given to another. All of the informants seemdd to rely on their ability to remember this the next time the course was given.

Using the computer as a teaching aid

In general the preferred tools were the blackboard and/or transparencies with only a few of the informants actively using computers as a teaching aid during lectures. Some did not know how to use computers in this way. Some thought it was too complicated. Others preferred not to use a computer since the blackboard and transparencies gave them more freedom and better contact with the students.

When asked if they thought about using computers as a teaching aid most of the informants were hesitant while the single informant who actually used the computer during lectures was more enthusiastic. The informants seemed to think that using a computer would make them more "rigid", something confirmed by the teacher who used it, and loose the contact with the students. It should be noted that this "rigidity" sometimes could be an advantage; there is no risk of forgetting to mention something during the lecture. Others could not see how computers could help the students understand a certain subject better — "if they don’t understand this math using the blackboard they won’t understand it on a computer screen either".

Many of the informants thought that computers could be useful for showing simulations and complex processes or visualizing information dynamically. There also seemed to be a underlying notion that "computers can be good for information gathering" by many of the teachers, but there seemed to be no systematic way for doing it — "I just browse around the Internet trying to find something".

Given the explicit question if computers would be useful in distance education they seemed to think that it would be most useful, especially for keeping the contact between students/students and students/teacher, allowing quick feedback even if the group members were located in different places.Some of the teachers also pointed out that using computer networks for distributing course material was very nice, allowing the teachers to add material during the course and give extensive background references.

Being a teacher

When asked they all said that they liked being a teacher and that they felt that their work was rewarding. They found the contact with students to be one of the best parts of their occupation. Not only were they given the chance to closely interact with other people, but they could also see that they made a difference. That the students actually learned something new. Some of them also mentioned that the contact with old students calling up asking for help with a certain problem was very rewarding.

All the informants complained about a "lack of time", this was especially pronounced at the Department of Computer Science where the teaching load seemed to be higher than other departments. They all wanted to "make that little extra" but very seldom found the time to actually do it.

Discussion

Before continuing with a discussion of our findings we would like to emphasize that the interviews have been informal leaving much to be desired when it comes to scientific formality. The reason for choosing this method have been

• It is inexpensive. There is no need for costly experimental setups, etc.
• It is quick. The interviews have been easy to schedule and have usually had a duration of between 45 minutes to 2 hours.
• The purpose of the interviews have been to give an indication of how teachers think about their work situation, and the tools they are using. This report is not intended to describe how computer supported teaching should be done. For an overview and discussion on different approaches to computer supported teaching see Broberg (1997).

Despite these considerations we believe that these interviews form a basis for further investigations as well as a basis for a prototype implementation of an authoring environment.

Is an authoring tool needed?

As can be seen above, most of the informants were negative about using a computer during the lectures, but they also seem to think that it can be a useful tool for simulations, distance education, etc. It might seem that this indicates that there is no need for an authoring environment but we should remember the following:

• Few teachers have actually used these tools so there could be some natural resistance against new technologies.
• The technology we can find today are the first stepping stones towards the technology we may find in 10-20 years.
• The pressure for more education will most probably continue in the future.
• It seems improbable that universities will get more resources in the future.

Based on this we have decided to continue our research of an authoring environment aimed to help teachers produce online course material and both students and teachers in communicating with each other.

As a first step towards this system we have decided to look into the authoring part and concentrate on the areas outlined below.

Patterns

During the interviews we have seen that teachers tend to work according to patterns and that these patterns can be specific to a specific person, as in the case of organizing a lecture, or form a more general pattern that is specific to a certain subject, as in the case of designing a course. Some of these patterns might be of a more general nature, but further research is needed to be able to determine if that is the case. In view of these investigations, it is clear that these patterns exist and that they are useful tools for a teacher. Therefore, it is natural that an authoring environment should support these patterns by allowing the teachers to use template documents or other methods for pattern support.

A simple example of templates might be the templates that are included with any commercial word processor; there are often numerous templates that let the user be able to create professional looking letters/memos/reports without knowing the details of that specific word processor. More advanced templates have the ability to interact with the user asking questions of how the document should be formatted.

Reuse and modifications

It is clear from the interviews that material is reused/modified very often and any authoring environment should support this. The system should make it easy to share documents across work groups and organizations. It should be noted that investigations in other application areas (Mackay 1990; Nardi 1993; Nardi and Miller 1993) have shown similar patterns of sharing.

There are several ways of supporting both reuse and modifications to existing documents. Let us for a moment concentrate on the subject of reuse.

What is reuse?

We can specify reuse in several ways, for example the above mentioned templates. By using a template we reuse an already defined structure/layout by copying. The teacher takes an existing document, duplicates it, and makes the desired changes to the newly created copy. No links/references will show the relationship between the original and the copy, and changes made to any of the documents will not be reflected in the other.

Another method of reuse is inclusion of existing material into different documents, this is exemplified in Figure 1 where a picture is used in two different presentations.

Figure 1: Inclusion.

A third method is referencing; here the reused document is not a part of the presentation but rather something that provides some background material, see Figure 2.

Figure 2: Reference.

The last method we will mention here is inheritance. In Figure 3 we see how the two dice in box A form the basis, allowing other documents to extend the content by adding a gray square, box B, or a pattered rectangle, box C. Box B is furthered extended by adding a white box to it, box D. The interesting thing is that changes made to any of the boxes propagate to all boxes that use it as a basis. For example if a third die is added to box A, the three other boxes will also get a third die. But if the third die is added to the box B only the boxes B and D will get it.

Figure 3: Inheritance.

We will not describe any other methods of reuse, but instead discuss the suitability of reuse methods in an authoring environment. The requirements for reuse should be that is it simple to understand and use. After some discussion we have decided not to recommend inheritance since making full of use inheritance might confuse non-experts. The reason we think that inheritance is confusing is that changes might propagate in ways that cause unexpected results for the user.

Modifications

Related to reuse but slightly different is the ability to make modifications to existing documents. This means that a new version of the document is created with major or minor changes compared to the original document. This new version is most often used to replace the original, in other words it is an update to an existing document. It is also possible to create versions of a documents that are targeted for different audiences.

Searching

Some of the informants mentioned that they used the computer to search for material but there was little help from the computer in doing these searches. A common method when searching for information on the Internet is to use a "search engine" like AltaVista™. These systems allows the user to search for one or more expressions by entering them in a text field and then clicking on a "search" button. Unfortunately, the results from such a search is either overwhelming, Figure 4, or by making some minor changes no documents are found, Figure 5.

Figure 4: A search giving a large number of hits.

Figure 5: Some minor changes results in no hits at all.

It can be very difficult to know how to improve the results doing this kind of searching, and users can get quite frustrated when refining their questions to the database. A common way to allow the users to refine their questions is to use logical expression, for example: find documents containing "newton" and "technology" and not "apples". Unfortunately, these kind of expressions are difficult to formulate for most users making this feature "under used".

An authoring system intended for teachers should explore different methods of searching like Guided Information Access (Nisus Software Inc) , latent semantic analysis, and full text searching.

Group tools

One important role of computers in teaching seems to be the ability to support "work groups" that can work together and exchange information regardless of the place where the different group members are. Other uses mentioned by the informants are tools for distributing documentation, registering for courses, handing in assignments, etc.

If we were to speculate about the future in the Swedish academic world we see a large demand for teachers, and it seems probable that this demand will increase. Unfortunately, most universities find it difficult to fill all the vacant positions. This, in turn, is forcing the universities to find new methods for teaching. One such method would be to let one experienced teacher take the responsibility for a course and to assign a number of less experienced undergraduate staff members to help out. This would probable mean significant increase in coordination – something that is problematic already now. To cite one of the informants "The problem isn’t in agreeing on what to do, it’s in finding the time to agree". If this scenario becomes true, it is even more important that there are flexible and powerful ways in which people can work together even when not co-located in time and space.

Import of existing material

All experienced teachers have created material that is reused during teaching, some of this material is available as physical artifacts, for example models and printed documents, but there is already a large body of material that is electronically stored, for example simulations and word processing documents. If a new authoring environment is to be used it either has to be able to import existing documents or know how to handle the documents in their native format.

Result

During the interviews we have seen that the majority of the informants are against using computers during lectures but that they are considered useful for other tasks. Despite this, we believe that in order to meet future demand for education there has to be systems that support the process of creating online course material and that the availability of such tools will change the way teaching is done. If this is a good thing or not, we can not judge. But, there is a need for research in the usability aspects of these tools.We have decided to start a project to look into a few of the areas that we have mentioned above. We will concentrate our efforts on the following areas:

• Version control. The interviews as well as our own experiences as teachers show that course documents are constantly being revised. There can be several versions of one document in use and sometimes the coordination of these different versions can become overwhelming. It is also important that the system supports several users tracking what file is being modified by whom making sure that only that person change it. This is a problem that has existed in software engineering for a long time and is usually solved with the help of a version control system. We plan to incorporate such a system into our authoring tool and explore what properties the version control system should have.
• Reuse. Since all the informants complained about "a lack of time" it is important to ease the burden by adding the possibility to reuse material that has already been created. To be truly useful each piece of information has to associated with meta-data, for example creation date, author, content, making it easier to find the desired information. This is a fundamental property of our system and will be implemented early in development stage.
• Searching. In order to implement reuse we have to find efficient ways of finding existing information. We will implement our system in a way that makes it possible to quickly add new search methods making it possible to compare different methods.
• Patterns. Our informants all seem to have certain work patterns that help them in their work. We want our tool to support the teachers by supporting their use of patterns. We plan to incorporate support for this by implementing reuse and other technologies.

Acknowledgements

We wish to thank Nils-Erik Gustafsson and the interviewed teachers for their help and their time.

References

1. Broberg, A. (1997). Cognitive Tools for Learning. Department of Computer Science. Umeå, Umeå University: pp 187.
2. Mackay, W. E. (1990). Pattern of Sharing Customizable Software. pp 209-221. Proceedings of the Conference on Computer-Supported Cooperative Work CSCW'90, ACM Press.
3. Nardi, B. A. (1993). A Small Matter of Programming; Perspectives on End User Computing, MIT Press:.
4. Nardi, B. A. and J. R. Miller (1993). Twinkling lights and nested loops; distributed problem solving and spreadsheet development. Readings in Groupware and Computer-Supported Cooperative Work, Morgan Kaufmann: pp 260-271.
5. Nisus Software Inc "Guided Information Access – GIA." http://www.infoclick.com/.

Footnote

Current address: Department of Computer Science, Linköping University, 581 83 Linköping, Sweden.
Email: davca@ida.liu.se