An Alternative Approach to Teaching Database Normalization: A Simple Algorithm and an Interactive e-Learning Tool

Journal of Information Systems Education, Fall 2006 by Kung, Hsiang-Jui, Tung, Hui-Lien

6. CONCLUSIONS AND FUTURE RESEARCH DIRECTIONS

An alternative normalization approach has been developed to enhance the teaching and learning of database normalization. The approach contains the easy-to-follow simple algorithm and the interactive e-learning tool available on the Web (http://www.georgiasouthern.edu/~hjkung/3NF). It was evaluated by students and found to be robust. Students' responses to the approach were mostly favorable. The students indicated that they had found the e-learning tool easy to use and noted that the step-by-step feature helped them gain an understanding of database normalization process. The approach had a positive impact on students' perceptions of the ease, confidence and performance.

These findings have many implications. Students can use the e-learning tool as a practice and drill tool to help them in learning database normalization. It is important to point out that we do not propose using the approach as the only coverage of normalization. Students can work on the normalization problems by themselves and use the e-learning tool to validate their answers. We believe that students need to understand the concept of normalization from the traditional approach and can practice the mechanical normalization steps using the simple algorithm and e-learning tool. Students can also use their normalization skills to validate their Entity-Relationship diagrams (ERD) that are error prone to novice designers (Batra and Wishart, 2004). The study suggests that the most common errors pertain to the partial dependency. Thus, IS/IT educators should spend a little more time explaining partial dependency.

We suggest several extensions to our research. Currently, the e-learning tool can handle small-size problems, e.g., a set of ten functional dependencies, which is adequate for teaching purposes. More features, e.g., 'Load,' 'Save,', and 'Print', are still under development. One extension of the future enhancement is to integrate normalization with ERD to show the connection between relational and ER models. Another extension of this study would be a longitudinal assessment of the normalization learning process from multiple institutions, so that we would have statistical tests for between- and within-subjects effect. The next extension would be to develop a perception instrument using multi-item scales to measure subjects' perceptions. Finally, we could refine the error rate computation technique to weigh objects by the level of difficulties.

7. ACKNOWLEDGEMENTS

We are indebted to the reviewers for their detailed comments on an earlier version of this paper. Especially, their recommended title is appropriate for this paper.

8. REFERENCES

Avison, D. E. and Fitzgerald, G., (2002), Information Systems Development: Methodologies, Techniques and Tools, 3rd Ed., Lodon, UK: McGraw Hill.

Batra, D., Antony, S. R., (1994), Novice errors in database design, European Journalof Information Systems, 3 (1), 57-69.

Batra, D., and Wishart, N. A., (2004), Comparing a RuleBased Approach with a Pattern-Based Approach at Different Levels of Complexity of Conceptual Data Modelling Tasks, International Journal of HumanComputer Studies, 61, 397-419.