Analyzing the use of adaptive learning in a flipped classroom for preclass learning

Abstract

Preclass learning, an obstacle in the success of a flipped classroom, is addressed via placing lessons on an online adaptive platform. The lessons combine the power of video lectures, textbook content, simulations, and assessments while using personalized paths for each student. This article describes the development of the adaptive lessons for a course in Numerical Methods, and the interpretation of the analytic data collected via the adaptive lesson platform and student focus groups over a two‐semester period with 146 students. Analytical data includes student metrics, such as the lesson scores and the time spent and lesson metrics, such as the percentage of students who completed the lesson and the percentage of possible adaptive paths used by students. The focus groups were conducted for two different demographic groups—students who are “white males” (comprise the majority of students in public engineering schools in the USA) and “other than white males”—to compare their perspectives on adaptive learning. Students in the focus group of the “other than white male” pupils demonstrated more favorable and positive perspectives towards the adaptive learning compared with the “white males”, although both groups identified benefits with the adaptive platform. Final examination scores were found to be correlated with the raw score of the adaptive lessons.

Learning Analytics in Flipped Classrooms: A Scoping Review

“…Stemming from the previous category, we also identified 13 studies that specifically focus on online performance leading to an evaluation or assessment. Metrics such as the completion rate and success rate in the course are taken into account (Kaw, et al 2019;Yang, Wu and Cao, 2016), as well as summative evaluations (Smallhorn, 2017). A single study integrated peer evaluation (Fernández, Merino and Kloos, 2019).…”

“…Some articles focused on other student metrics, such as student self-reported satisfaction in the experience (Lei et al, 2017;Van Leeuwen, 2018), student self-assessment (Corrias and Hong, 2015), or self-regulation in learning (Hwang and Chen, 2019). Two studies investigated the students' adaptive learning pathway in the FC experience (Kaw and al., 2019;Reidsema et al, 2017). One study (Xiao, Pham and Wang, 2015) introduced a rare example of a mobile MOOC, where learning was adapted via implicit cognitive states inference that integrated a tangible video control, and implicit heart rate sensing based on heartbeat waveforms.…”

“…Such studies investigated the participants' experience through interviews (e.g. Hui et al, 2018;Kravchenko and Cass, 2017) or focus groups (e.g, Ayres et al, 2018;Kaw et al, 2019), or in-class observation based on the action research methodology (Isomöttönen and Tirronen, 2017). Some studies also investigated the lived experiences of students through self-reported opinion or motivation questionnaires (Marasco et al, 2018;Lau et al, 2018;Lei et al, 2017;Chu, Wang and Wang, 2019 However, although mixed methods are applicable to the observation of the FC experience, the core of the data was quantitative and used to apply LA algorithms.…”

“…Finally, a smattering of studies reference other educational theories to analyse the efficiency of the FC. These studies reference for example the principle of active learning (Kaw et al, 2019;Hui et al, 2018), Bloom's taxonomy (Hwang and Chen, 2019;Giannakos, Chorianopoulos, and Chrisochoides, 2015), and Game-Based Learning (GBL) (Chu, Wang and Wang, 2019).…”

Contrasting Classical and Machine Learning Approaches in the Estimation of Value-Added Scores in Large-Scale Educational Data

“…This is both a challenge and an opportunity, as it is becoming feasible to use the strength of interdisciplinary approaches and combine expertise from the domains of Education, Psychology, and Computational Sciences to apply machine learning methods to estimate VA scores. While machine learning seems to be promising for practices within the classroom (see, e.g., Kaw et al, 2019 ; Moltudal et al, 2020 ), the focus of the present study is on their potential use for the estimation of school VA scores.…”