A framework for data collection, analysis and evaluation of the relationship between students’ computer interaction and course grades in laboratory courses

Big data is an emerging topic, with huge investments in IT and education worlds. Together with awareness of the knowledge discovery and education improvement progresses, big data concept has come with a growing consciousness. There are several educational data mining analysis methods created, evaluated and presented in the literature in this manner. But, how the educational big data for finding “intelligently” valuable results to benefit students, teachers and administrators will be collected, filtered and handled? In this study, a unique data collection, filtering and evaluation framework for educational data mining was designed, evaluated and presented. In this perspective, an IT infrastructure and process monitoring software for gathering client computers’ data of 3 laboratories in a public high school was developed. Time series big data was collected during 5 months from 62 computers with this back stage working software. After filtering this data with eligible methods, resultant data was evaluated with Pearson’s correlation analysis between students’ rate of interactions with computers and their exam grades on laboratory courses. Results showed that students’ computer interaction and their success in the courses are highly correlated.

View full Article in PDF

Series Navigation<< Towards a domain model for integrating competence frameworks into learning platformsThe ACADEMICA e-course: an example of good practice to train e-learning new users >>