Online learning platforms have expanded access to education but also raise concerns about biased content, particularly in text-based learning materials such as textbooks, lesson plans, and course excerpts. Such biases can perpetuate discrimination, can harm student outcomes, and can often be difficult to detect, as identification typically relies on time-consuming human review. Learning analytics (LA) can enhance this process by supporting human reviewers through automated detection, offering a scalable solution while retaining human judgment for nuanced evaluations. Accordingly, this LA study explores two research questions: RQ1: Which features might support the identification of ethnic bias in text-based online learning materials? and RQ2: Which classification approaches might be suitable for identifying ethnic bias in text-based online learning materials? First, we identified features signalling potential ethnic bias (presence or absence) in textual content using a dataset (N = 345) labelled by 193 students from diverse ethnic backgrounds. Then, we evaluated multiple machine learning (ML) models for their effectiveness in bias classification. The results suggest significant correlations between perceived bias and content from social sciences. Additionally, through bootstrap analysis, support vector machines and random forest classifiers showed consistent performance in bias identification (with F1-scores of 0.71 and 0.70 on the test set, respectively). In contrast, the naive Bayes (NB) model demonstrated the highest precision (0.75 on the test set). We discuss these findings and their implications for LA, emphasizing the importance of quality and inclusive educational tools. As an initial step toward automated bias classification in education, this study provides a foundation for spotting ethnic bias in learning content, supporting fairer technologies for more inclusive learning environments.

Learning Analytics (LA) is increasingly applied to assess and foster creativity in educational settings. Whereas existing applications have shown promise in STEM contexts, less is known about the diversity of approaches across educational domains. Therefore, we conducted a scoping review that systematically mapped LA applications for creativity in educational contexts. Searches returned 278 articles, with 41 studies meeting eligibility criteria. Analysis revealed five fundamental mechanisms through which LA fosters creativity: process visualization, adaptive feedback, automated pattern recognition, behavioural analytics, and real-time intervention. Computational creativity (10 studies) was the most prevalent conceptualization, with log data as the primary source (12 studies) and automated assessment via platform-based metrics as the leading approach (10 studies). Programming platforms represented the main technological applications (11 studies), while collaborative learning was the most common pedagogical strategy (7 studies). Problem-solving emerged as the most frequently linked complementary skill (17 studies). However, research showed extensive STEM focus; methodological fragmentation, with 38 studies lacking specified study duration; and theoretical gaps, with nine studies missing explicit theoretical frameworks. These findings highlight LA’s transformative potential for creativity assessment and fostering while revealing opportunities for interdisciplinary expansion and methodological standardization.

This study investigates how individual differences related to creative and verbal fluencies can be differentiated and modelled through the linguistic features that manifest in student writing. Using computational linguistics, we analyzed samples of intrapersonal writing to extract linguistic features that align to specific dimensions of language. Creative fluency scores were calculated based on the number of unique ideas generated during the Alternate Uses Task (AUT); verbal fluency scores were calculated based on the number of unique semantic categories produced during the animal naming task. We then developed machine learning models to predict creative and verbal fluency scores based on the linguistic features of participant essays. Results indicate that creative fluency is more strongly predicted by linguistic features (particularly descriptive and cohesion indices) than verbal fluency. Key predictors of creative fluency include features that capture lexical diversity and the global connectedness of ideas. Importantly, our results align with theoretical frameworks related to convergent and divergent thinking. Findings highlight the potential for leveraging learning analytics to offer new insights into complex cognitive processes such as creativity. Implications for stealth assessments and personalized feedback within automated writing systems are discussed as paths for future research.

Despite the increasing availability of data used to inform educational policies and practices, concerns persist regarding its quality and accessibility. This study surveys quality education data from Brazil, Colombia, and Peru and evaluates their alignment with the FAIR principles and availability to support academic analytics (AA) and learning analytics (LA). We identified and analyzed 112 data sources, from which 93\% of the data sets originate from government repositories and open data platforms, with coverage of macro-level data relevant for AA but lack of granularity for LA. The FAIR assessment showed 50\% of compliance with findability (F), 33\% for accessibility (A), and less than 50\% for both interoperability (I) and reusability (R), which limits broader utility. Moreover, these diverse data sources present limitations in quality assurance metrics such as ``institutional development" and ``quality management." We conclude by offering recommendations, emphasizing the need for enhanced data frameworks that bridge macro- and micro-level data for AA and LA to enable data-driven decisions for improving educational quality in Latin America.

This paper presents a transmodal analysis (TMA) study that investigates the interplay between gaze and verbal interactions through regulation-triggering events within the context of socially shared regulation of learning (SSRL) in face-to-face collaborative settings with shared computer-mediated materials. In face-to-face collaborative learning environments, gaze interaction serves as a pivotal mechanism for both directing attention and communicating emotional states. This aspect of non-verbal communication is critical for aligning cognitive processes and establishing an emotional connection among learners, thereby improving the effectiveness of the learning environment. Despite substantial research on verbal interactions in SSRL, the role of non-verbal cues, particularly gaze, has been less studied. This study examines 3,523 gaze and verbal interactions from twenty-four high school students engaged in eight collaborative learning groups. This study demonstrates the application of TMA method, which reveals epistemic networks across various data modalities, each operating on different temporal scales. Furthermore, our study highlights the critical role of gaze interactions in conveying cognitive and emotional cues, enhancing verbal communication for SSRL.

Knowledge tracing (KT) is a method to evaluate a student’s knowledge state (KS) based on their historical problem-solving records by predicting the next answer’s binary correctness. Although widely applied to closed-ended questions, it lacks a detailed option tracing (OT) method for assessing multiple-choice questions (MCQs). This paper introduces a general OT method that can be seamlessly integrated into deep knowledge tracing (DKT) methods through data processing techniques and network output modules. Using a million-level assignment record of MCQs from a K–12 math learning platform, which includes two types of knowledge components (KCs), skill and misconception, we converted five different DKT models into deep option tracing (DOT) models. Performance metrics demonstrate that OT enhances KT performance and effectively identifies students’ future option selection tendencies. Furthermore, using the best OT model, we extracted students’ problem-solving sequence features and learning gains to analyze learning patterns. The results reveal that for beginners in middle school mathematics, consecutive errors in the same skill might lead to greater learning gains. Finally, we applied network analysis to reveal connections between skills based on students’ error tendencies. Our work contributes to KT methods and related empirical findings in learning analytics (LA) for knowledge assessment.

The focus of education is increasingly on learners’ ability to regulate their own learning within technology-enhanced learning environments. Prior research has shown that self-regulated learning (SRL) leads to better learning performance. However, many learners struggle to productively self-regulate their learning, as they typically need to navigate the myriad of cognitive, metacognitive, and motivational processes that SRL demands. To address these challenges, the FLoRA engine is developed to help students, workers, and professionals improve their SRL skills and become productive lifelong learners. FLoRA incorporates several learning tools that are grounded in SRL theory and enhanced with learning analytics (LA), aimed at improving learners’ mastery of different SRL skills. The engine tracks learners’ SRL behaviours during a learning task and provides automated scaffolding to help learners effectively regulate their learning. The main contributions of FLoRA include (1) creating instrumentation tools that unobtrusively collect intensively sampled, fine-grained, and temporally ordered trace data about learners’ learning actions; (2) building a trace parser that uses LA and related analytical techniques (e.g., process mining) to model and understand learners’ SRL processes; and (3) providing a scaffolding module that presents analytics-based adaptive, personalized scaffolds based on students’ learning progress. The architecture and implementation of the FLoRA engine are also discussed in this paper.

Socially shared regulation in learning (SSRL) contributes to successful collaborative learning (CL). Empirical research into SSRL has received considerable attention recently, with increasingly available multimodal data, advanced learning analytics (LA), and artificial intelligence (AI) providing promising research avenues. Yet, integrating these with traditional datasets remains a challenge in SSRL research due to the misalignment between theoretical constructs, methodological assumptions, and data structure. To address this challenge and expand our understanding of the nature of SSRL, the present research adopted a human–AI collaboration approach in a three-layer analysis to examine group interactions in response to cognitive and emotional regulation triggering events. Two-level theoretical lenses — macro-level (regulatory aspects) and micro-level (deliberative interactions) — were used to analyze 2,125 utterances from video-recorded tasks of ten groups of three Finnish secondary students (N=30). Results showed two types of deliberation patterns for SSRL, namely 1) the Plan and Implementation Approach (PIA) associated with adaptive patterns, and 2) the Trials and Failure Approach (TFA) associated with maladaptive patterns. Our findings revealed that groups often fail to recognize, or are ill-equipped to respond to, emerging regulatory needs. These findings advance SSRL theories and research methodologies by utilizing AI-enhanced LA to offer new insights into group dynamics and regulatory strategies.

Multimodal learning analytics (MMLA) integrates novel sensing technologies and artificial intelligence algorithms, providing opportunities to enhance student reflection during complex, collaborative learning experiences. Although recent advancements in MMLA have shown its capability to generate insights into diverse learning behaviours across various learning settings, little research has been conducted to evaluate these systems in authentic learning contexts, particularly regarding students' perceived fairness, accountability, transparency, and ethics (FATE). Understanding these perceptions is essential to using MMLA effectively without introducing ethical complications or negatively affecting how students learn. This study aimed to address this gap by assessing the FATE of MMLA in an authentic, collaborative learning context. We conducted semi-structured interviews with 14 undergraduate students who used MMLA visualizations for post-activity reflection. The findings highlighted the significance of accurate and comprehensive data representation to ensure visualization fairness, the need for different levels of data access to foster accountability, the imperative of measuring and cultivating transparency with students, and the necessity of transforming informed consent from dichotomous to continuous and measurable scales. While students value the benefits of MMLA, they also emphasize the importance of ethical considerations, highlighting a pressing need for the LA and MMLA communities to investigate and address FATE issues actively.

Providing automated feedback that facilitates the practice and acquisition of oral presentation skills has been one of the notable applications of multimodal learning analytics (MmLA). However, the closedness and general unavailability of existing systems have reduced their potential impact and benefits. This work introduces OpenOPAF, an open-source system designed to provide automated multimodal feedback for oral presentations. By leveraging analytics to assess body language, gaze direction, voice volume, articulation speed, filled pauses, and the use of text in visual aids, it provides real-time, actionable information to presenters. Evaluations conducted on OpenOPAF show that it performs similarly, both technically and pedagogically, to existing closed solutions. This system targets practitioners who wish to use it as-is to provide feedback to novice presenters, developers seeking to adapt it for other learning contexts, and researchers interested in experimenting with new feature extraction algorithms and report mechanisms and studying the acquisition of oral presentation skills. This initiative aims to foster a community-driven approach to democratize access to sophisticated analytics tools for oral presentation skill development.