Novice Designer Performance Comparison Between the Entity Relationship Event Network and the Event-Based Logical Relational Design Techniques

Abstract

Previous information systems (IS) research has established that novice database designers experience cognitive overload when faced with many interacting entities in modeling relationships. The authors contend that this problem occurs mainly when events are involved. Results of an initial study provide support that novice database designers indeed have difficulty recognizing and modeling events. They propose two techniques that can address the difficulties encountered by novices when modeling events using the entity-relationship model. Entity-relationship event network (EREN) is a top-down and template-driven technique. Event-based logical relational design (ELRD) is a bottom-up and heuristic-driven technique. Employing the cognitive load theory (CLT) to guide the hypotheses, the authors compare the usability of EREN and ELRD for novice designers. Results indicate that both techniques facilitate satisfactory designer performance when modeling events. Overall, the ELRD technique leads to better designer performance. There is an interaction effect between technique and task complexity as the significant performance advantage of ELRD at the lower-complexity task gets mitigated at the higher-complexity task. The two techniques do not differ significantly on the constructs of behavioral intention to use, perceived usefulness, perceived ease of use, and self-efficacy. Overall, the ELRD technique is recommended for modeling events.