A key untapped feature of game-based learning environments is their capacity to generate a rich stream of fine-grained learning interaction data. The learning behaviors captured in these data provide a wealth of information on student learning, which stealth assessment can utilize to unobtrusively draw inferences about student knowledge to provide tailored problem-solving support. In this paper, we present a long short-term memory network (LSTM)-based stealth assessment framework that takes as input an observed sequence of raw game-based learning environment interaction data along with external pre-learning measures to infer students’ post-competencies. The framework is evaluated using data collected from 191 middle school students interacting with a game-based learning environment for middle grade computational thinking. Results indicate that LSTM-based stealth assessors induced from student game-based learning interaction data outperform comparable models that required labor-intensive hand-engineering of input features. The findings suggest that the LSTM-based approach holds significant promise for evidence modeling in stealth assessment.
Improving Stealth Assessment in Game-Based Learning with LSTM-Based Analytics
A key affordance of game-based learning environments is their potential to unobtrusively assess student learning without interfering with gameplay. In this paper, we introduce a temporal analytics framework for stealth assessment that analyzes students' problem-solving strategies. The strategy-based temporal analytic framework uses long short-term memory network-based evidence models and clusters sequences of students' problem-solving behaviors across consecutive tasks. We investigate this strategy based temporal analytics framework on a dataset of problem solving behaviors collected from student interactions with a game-based learning environment for middle school computational thinking. The results of an evaluation indicate that the strategy-based temporal analytics framework significantly outperforms competitive baseline models with respect to stealth assessment predictive accuracy.
- Publication Date:
- NSF-PAR ID:
- Journal Name:
- International Conference on Educational Data Mining
- Page Range or eLocation-ID:
- 208 - 218
- Sponsoring Org:
- National Science Foundation
More Like this
Although prior research has highlighted the significance of representations for mathematical learning, there is still a lack of research on how students use multimodal external representations (MERs) to solve mathematical tasks in digital game-based learning (DGBL) environments. This exploratory study was to examine the salient patterns problem solvers demonstrated using MERs when they engaged in a single-player, three-dimensional architecture game that requires the acquisition and application of math knowledge and thinking in game-based context problem solving. We recorded and systematically coded the behaviors of using MERs demonstrated by 20 university students during 1.5 hours of gameplay. We conducted both cluster and sequential analyses with a total of 2654 encoded behaviors. The study indicated that the maneuverable visual-spatial representation was most frequently used in the selected architecture game. All of the participants performed a high level of representational transformations, including both treatment and conversion transformations. However, compared to the students in the second cluster who were mostly non-game players, students in the first cluster (composed of mainly experienced video game players) displayed a higher frequency of interacting with various MERs and a more cautious and optimized reflective problem-solving process.
To improve student's class experience, the use of mobile devices has been steadily increasing. However, such use of mobile learning environments in the class is mostly static in nature through content delivery or multiple choice and true/false quiz taking. In CS courses, we need learning environments where students can interact with the problem in a hands-on-approach and instructor can assess their learning skills in real-time with problems having different degree of difficulty. To facilitate such interactive problem solving and real-time assessment using mobile devices, a comprehensive backend system is necessary. This paper presents one such system, named Mobile Response System (MRS) software, associated interactive problem-solving activities, and lessons learned by using it in the CS classrooms. MRS provides instructor with the opportunity of evidence-based teaching by allowing students to perform interactive exercises in their mobile devices with different learning outcomes and by getting an instant feedback on their performance and mental models. MRS is easy-to-use, extensible and can render interactive exercises developed by third-party developers. The student performance data shows its effectiveness in increasing student understanding of difficult concepts and the overall perception of using the software was very positive.
Examining Gaze Behaviors and Metacognitive Judgments of Informational Text Within Game-Based Learning EnvironmentsAbstract. Game-based learning environments (GBLEs) are often criticized for not offering adequate support for students when learning and problem solving within these environments. A key aspect of GBLEs is the verbal representation of information such as text. This study examined learners’ metacognitive judgments of informational text (e.g., books and articles) through eye gaze behaviors within CRYSTAL ISLAND (CI). Ninety-one undergraduate students interacted with game elements during problem-solving in CI, a GBLE focused on facilitating the development of self-regulated learning (SRL) skills and domain-specific knowledge in microbiology. The results suggest engaging with informational text along with other goal-directed actions (actions needed to achieve the end goal) are large components of time spent within CI. Our findings revealed goal-directed actions, specifically reading informational texts, were significant predictors of participants’ proportional learning gains (PLGs) after problem solving with CI. Additionally, we found significant differences in PLGs where participants who spent a greater time fixating and reengaging with goal- relevant text within the environment demonstrated greater proportional learning after problem solving in CI.
Rapid technological advances and the increasing number of students in Southeast Asian nations present a difficult challenge: how should schools adequately equip teachers with the right tools to effectively teach Computational Thinking, when the demand for such teachers outstrips their readiness and availability? To address this challenge, we present the SAGE reference architecture: an architecture for a learning environment for elementary, middle-school and high-school students based on the Scratch programming language. We synthesize research in the domains of game-based learning, implicit assessments, intelligent tutoring systems, and learning conditions, and suggest a teacher-assisting instructional platform that provides automated and personalized machine learning recommendations to students as they learn Computational Thinking. We discuss the uses and components of this system that collects, categorizes, structures, and refines data generated from students’ and teachers’ interactions, and also facilitates personalized student learning through: 1) predictions of students’ distinct programming behaviors via employment of clustering and classification models, 2) automation of aspects of formative assessment formulations and just-in- time feedback delivery, and 3) utilization of item-based and user-based collaborative filtering to suggest customized learning paths. The proposed reference architecture consists of several architectural components, with explanations on their necessity and interactions to foster future replications ormore »