Role of conceptual and mental models in motivating end users to learn new but related software packages
Conrad Shayo · Medical Entomology and Zoology · 1995
An increasing number of end users are faced with the task of learning to use new software from a particular family of packages. For example, they learn one database package and soon after must learn a different one. Although software training research has underscored the role of mental models in learning new software packages and the need to provide adequate mental models through formal training, it is not known how and to what extent adequacy of trainees' pre-training mental models, and the training approach used influence trainees' motivation toward learning new but related software packages. This research conducted two laboratory experiments. Experiment 1 focused on the relationship between the type of mental model trainees bring to a formal training session and their motivation toward learning a new but related software package, and whether the mental model influence on their motivation depended on: (a) the software to be learned, and/or (b), the task to be accomplished. Experiment 2 explored whether a mapped conceptual model vs. an un-mapped conceptual model would strengthen the motivation to learn a new but related software package differently for novices compared to proficient subjects. Although not statistically significant, Experiment 1 results indicate that subjects with more adequate mental models had higher pre-test perceptions of transfer than those with less adequate mental models. Main effects for adequacy of mental model on post-test RDBMS self-efficacy expectation and objective quiz performance were statistically significant. Subjects with low RDBMS pre-test self-efficacy expectation had the highest gain scores. Strong interaction effects between software and task context were observed. Task context significantly influenced the subjects' perceptions about the usefulness of the training. Experiment 2 results indicate that subjects more knowledgeable about RDBMSs had higher post-test performance, perceptions of transfer, and self-efficacy expectation. Subjects low in pre-test RDBMS perceptions of transfer and self-efficacy expectation had the highest gain scores. Statistically significant results for training method (joint task/software vs. disjoint task/software) on perceptions of transfer were obtained. The results provide some support for the predictions of mental model theories.