Learning Analytics and other branches of Educational Research such as Computing Education Research (CER) implicitly assume that students, especially college students, have no barriers to access learning platforms or software packages. This assumption may be attributed to such pervasive beliefs such as "everyone has a device", or "everyone can access internet". However, if students don't use a laptop or other computing resources (i.e., phones, tablets, computer labs, loaner laptops, printers) as we expect them to, these studies are missing a critical element to validate their findings. Consequently, Learning Analytics research often overlooks the usage of computing resources as a factor of student success. An explicit investigation of whether undergraduate student usage of computing resources are factors of academic success closes a crucial feedback loop for Learning Analytics.
In this thesis, we extract actual usage of student computing resources by developing a method using operational data sets already collected by the university. This method also addresses oversights from traditional Learning Analytics studies, which lack systematic and continuous observation of the students.
The observed behaviors are triangulated by two studies: (1) we administer a survey to collect students' perceptions and motivations for using computing resources, and compare observed vs. reported behaviors; and (2) we introduce a taxonomy of extrinsic parameters to differentiate behavioral patterns caused by external environmental factors vs. students' inner motivations.
Finally, we use Structural Equation Modeling to model the intensity of usage of computing resources as a factor of student success. Results indicate that students' intensity of usage significantly impacts academic success for CS students.