A lot of research into the use of recorded lectures has been done by using surveys or interviews. We will show that triangulation of multiple data sources is needed. We will discuss how students use recorded lectures according to their self-report and what actual usage of the recorded lectures can be derived from the data on the system. We will present the data collections and cover areas where the data can be triangulated to increase the credibility of the results or to question the students' responses. The triangulation shows that we lack data for a number of areas. We will need high-quality surveys and interviews combined with the log data to get a complete picture. We need to be able to link data sets together based on the identification of the individual students, which might raise privacy issues.
Recorded lectures provide an integral recording of live lectures, enabling students to review those lecture at their own pace and whenever they want. Most research into the use of recorded lectures by students has been done by using surveys or interviews. Our research combines this data with data logged by the recording system. We will present the two data collections and cover areas where the data can be triangulated to increase the credibility of the results or to question the student responses. The results of the triangulation show its value, in that it identifies discrepancies in the students' responses in particular where it concerns their perceptions of the amount of use of the recorded lectures. It also shows that we lack data for a number of other areas. We will still need surveys and interviews to get a complete picture.
Abstract Despite the numerous business benefits of data science, the number of data science models in production is limited. Data science model deployment presents many challenges and many organisations have little model deployment knowledge. This research studied five model deployments in a Dutch government organisation. The study revealed that as a result of model deployment a data science subprocess is added into the target business process, the model itself can be adapted, model maintenance is incorporated in the model development process and a feedback loop is established between the target business process and the model development process. These model deployment effects and the related deployment challenges are different in strategic and operational target business processes. Based on these findings, guidelines are formulated which can form a basis for future principles how to successfully deploy data science models. Organisations can use these guidelines as suggestions to solve their own model deployment challenges.
