A major problem faced by instructors post-implementation of unsupervised online assessments is that they may lack real-time access to the students' actual learning behaviors. Limitations in student-feedback, limited know-how of accessing and analyzing log data, and large class sizes could restrict instructors' access to learners' behaviors. This study investigated how learning analytics (LA) can identify learners' actual behaviors within low-stake unsupervised online quizzing, the relationship between behaviors and performance in exams, and how the results can inform pedagogy. To achieve these goals, the present study used LA methods to analyze quiz-logs and qualitative interviews with instructors. Findings show that data-driven methods informed by learning theories can become a valuable tool in providing real-time insights into students' actual learning behaviors. Seven pedagogically meaningful variables related to learners' quiz-taking behaviors were designed and extracted from the quiz-logs. These variables provide evidence that if unsupervised, all students may not self-regulate their learning effectively to engage in productive learning behaviors and hence may need additional guidance from instructors. The instructors were actively involved in the study to interlink the implemented learning design and quiz-log analytics. We conclude that LA methods, when taken into account with instructors' input, may help plan timely pedagogic interventions such as providing the students meaningful and timely feedback, redesigning the existing quizzes, and educating students on the benefits of effective learning strategies. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]