Self-regulated learning (SRL) has been widely recognized as a critical skill for academic success in online and blended learning contexts. However, many students experience difficulty in effectively applying SRL strategies in the absence of structured instructional guidance. To address this challenge, this study developed and implemented a learning management system integrated with SRL support (SRL-LMS), specifically designed for the online component of a blended learning course. The SRL-LMS consisted of two sections: a conventional course content section and a SRL training section designed to support students in applying SRL strategies. A quasi-experimental design was adopted with 69 college students assigned to either an experimental group, with access to both course and SRL sections, or a control group, which accessed only the course section. Results indicated that students in the experimental group reported higher levels of self-regulation and showed more frequent and diverse application of SRL strategies compared to the control group. In terms of academic performance, the experimental group performed significantly better than the control group on the first exam, though no significant difference was observed on the second exam. These results highlight the effectiveness of structured SRL interventions within digital learning platforms for improving students’ self-regulatory behaviors. Future implementations should address cognitive load and incorporate strategic approaches to sustain student motivation. This study advances current SRL intervention designs and offers valuable insights for educators and instructional designers aiming to integrate targeted SRL supports in online and blended learning environments.

Early dropout prediction requires models that integrate heterogeneous educational records without information leakage. We propose multi-view neural risk augmentation (MVNRA), a cross-fitted framework that separates early-warning predictors into academic, contextual, and digital-engagement views. View-specific encoders and a gated fusion module produce four supervised risk scores. We evaluated MVNRA on a public longitudinal dataset from a Spanish technological university containing 464,739 student–course records and 81 predictors available by December. Within each outer student-level GroupKFold split, the neural risk generator was trained through inner student-grouped cross-fitting. Its out-of-fold fused, academic, contextual, and digital risk scores were then appended to the original predictors before six conventional classifiers were fitted. MVNRA increased mean area under the receiver operating characteristic curve (AUC) for all six classifiers. The largest gains occurred for AdaBoost (0.7717 to 0.9240), linear discriminant analysis (0.8701 to 0.9176), Gaussian Naive Bayes (0.7313 to 0.7786), and Logistic Regression (0.9139 to 0.9352). MVNRA + Random Forest achieved the highest absolute discrimination (AUC $=0.9577$), with increases of 0.0191 in precision–recall AUC and 0.0112 in F1 score. SHAP analysis ranked the fused and academic neural risk scores above the cumulative academic-progress indicators. These findings show that cross-fitted, view-specific neural risk scores can transfer nonlinear structure to conventional tabular classifiers while preserving a SHAP-compatible decision layer and strict student-level separation.

Background: The transition from doctoral student to independent scholar represents a critical developmental turning point in emerging adulthood. Knowledge production serves as the central vehicle for validating identity during this phase, yet achieving such output is an inherently social endeavor. From a social network perspective, academic ties function as essential developmental scaffolds. However, access to these relational resources is often stratified. Currently, limited empirical evidence exists regarding whether and how relational structures shape the stratification of knowledge production. Purpose: This interdisciplinary mixed‐methods study examines how the status and organizational attributes of alters in academic networks function as pathways to distinction. Specifically, we answer the following two questions: (1) How do diverse academic networks stratify doctoral students′ knowledge production? (2) How do the interactive mechanisms facilitate or constrain students′ knowledge production and developmental trajectories? Research Design: This study employed an explanatory sequential mixed‐methods design guided by Biglan′s widely cited framework, distinguishing between “soft” and “hard” disciplines. We first analyzed survey data applying zero‐inflated negative binomial regression to identify structural patterns, followed by semistructured interviews to unpack the mechanisms of network utility. Findings: The analysis reveals a “Matthew Effect” of cumulative advantage, where access to superior social capital creates widening gaps in academic advancement. Specifically, pathways to distinction are context‐dependent. In soft disciplines, both status and organizational heterogeneity significantly enhance knowledge production, primarily driven by external faculty ties. In contrast, in hard disciplines, only organizational heterogeneity demonstrates a significant positive effect, with external peer relationships serving as the key channel. Qualitative evidence illuminates the complex interactions among doctoral training environments, disciplinary epistemological traits, and the effectiveness of different relational ties. The disciplinary context determines which ties are important, but individual and group dynamics decide whether network resources translate into realized professional distinction. Conclusions: Grounded in social network theory, we argue that the stratification of knowledge production is significantly shaped by the unequal distribution of relational resources. Academic networks are not merely indicators of individual social competence but structural foundations for developmental trajectory. When institutions delegate network‐building solely to students, they may unintentionally reinforce structural isolation and create new forms of inequality amidst global doctoral enrollment expansion. These findings demand institutional investment in collaborative, justice‐oriented doctoral training frameworks.

Doctoral training occupies a liminal threshold between disciplinary initiation and the unfolding of scholarly autonomy. This formative terrain is shaped by a dense ecology of relations, including s...