Identifying Key Features of Student Performance in Educational Video Games and Simulations through Cluster Analysis

Learning analytics for enhancing the usability of serious games in formal education: A systematic literature review and research agenda

“…According to the selected studies, applying LA to SEGs has several objectives including: Understanding how students learn using games and modeling their behaviors (affective, social and cognitive states) during gameplay (Hou, 2012 ) (Kerr & Chung, 2012 ) (Cheng et al, 2015 ) (Minović et al, 2015 ) (Israel-Fishelson & Hershkovitz, 2020 ); Implementing teaching support to improve the process of teacher inquiry (Rodríguez-Cerezo et al, 2014 ) (Sergis & Sampson, 2017 ) (Kiili & Ketamo, 2018 ); Conducting formative and summative assessments of students’ learning based on in-game data (stealth assessment) (Serrano-Laguna et al, 2014 ) and game evaluation (Freire et al, 2016 ) (Abdellatif et al, 2018 ); Predicting learning results based on students’ interactivity (Loh & Sheng, 2015 ) (Kouraki et al, 2017 ) (Rowe et al, 2017 ) (Hernández-Lara et al, 2019 ); In-game personalization and adaptation by providing an appropriate level of challenge according to learners’ competencies level (Kiili et al, 2018 ) (Mostefai et al, 2019 ); and Validating the actual educational and game designs or finding possible improvements in game design as well as the cost efficiency of using games in education (Freire et al, 2016 ). …”

“…Understanding how students learn using games and modeling their behaviors (affective, social and cognitive states) during gameplay (Hou, 2012 ) (Kerr & Chung, 2012 ) (Cheng et al, 2015 ) (Minović et al, 2015 ) (Israel-Fishelson & Hershkovitz, 2020 );…”

Examining Students’ Behavior in a Digital Simulation Game for Nurse Training

“…These indicators comprise of multiple parameters such as a number of clicks, frequency of tool use, and duration of interaction, which can be interpreted as a specific indicator of player's behavior. For instance, Kerr et al [31] have identified the main features of learners' proficiency by logging actions via mouse clicks during gameplay. Denden et al have suggested that the learners personalities can be identified based on their actions and choices in the games and examine such indicators as earned during the game score and places that the learner visited while exploring the game environment [32].…”

“…There are numerous options for clustering methods, but all of them attempt to merge similar objects in the same cluster, and split dissimilar objects into different clusters. For instance, Kerr and Chung in [31] have applied fuzzy feature cluster analysis to identify key features of student performance in log data collected from a mathematical game for sixth grade students; Slimani et al in [43] used Expectation-Maximization and K-means clustering approaches for students' performance in a game-simulation of the biological room; Martin et al in [62] have performed hierarchical clustering in order to investigate student success study of fractions.…”