This paper describes a project, which goal is to develop a powerful, but easy to use, data recording system for the development of teaching materials. The paper focuses on data collection algorithms and problems of measurement handled trough research questions about specific experiment.

Introduction
The World Wide Web (WWW) offers a good opportunity to produce and deliver learning materials. The WWW platform makes possible to support different constructive learning theories by many kinds of interactions, which can be used in quite free and dynamic environments. The WWW also promises to be very a rich environment for educational research, as many powerful data collection algorithms can easily be added to WWW-based learning materials. With these algorithms researchers can record every mouse or keyboard input with an observation system that is invisible for the user of learning materials.

According to Loomis et al. (1999) the traditional relationship between experimental control and ecological validity of research is negative: when the experimental control of research is high then the ecological validity is low, and when ecological validity of research is high, then experimental control is low. By using a virtual environment as a research and observation environment, the ecological validity and experimental control of research can both be quite high (Loomis & Blascovich, 1999). Although the WWW is not a virtual environment in the full meaning of that word, we can assume that hidden observation in WWW-based learning materials can give more ecologically valid data from using the material than some traditional observation. Naturally, we can only have information about the learning materials' use. When comparing this kind of code-embedded observation and traditional observation in quantitative research, we cannot find any major differences in quality of data. The implementation of a test is not important, most important and meaningful is that we know what kind of information we want to have with the test (Buchanan & Smith, 1999). In quantitative research, we can even have a good benefit from code-embedded observation when code is written so that data is written direct to a raw matrix.

Research questions and method
Our research contains two groups of 6-year old children: experiment group (n=30) and control group (n=30). Both groups were pre- and post-tested considering their geometric skills. Only the experimental group uses the learning materials about geometry (45 min); the control group did not get any educational effect. Measures of the study are scheduled for completion in April 2000. The research questions of the study are summarized as follows:

• What kind of general differences in use of material can be found?
• Is there any connection between use of materials and learning results?

References

1. Buchanan, T.; Smith, J.L. (1999). Research on the Internet: Validation of a World Wide Web mediated personality scale. Behavior Research Methods, Instruments & Computers, 31, 565-571.
2. Loomis, J.M.; Blascovich, J.J.; Beall, A.C. (1999). Immersive virtual environment technology as a basic research tool in psychology. Behavior Research Methods, Instruments & Computers, 31, 557-564.