Given the growing use of grit in sales research, this study addresses challenges in its operationalization and offers recommendations for more effective application Data from 321 B2B salespeople across industries are used to compare the incremental predictive validity of domain-specific and domain-general grit, each modeled as unidimensional and multidimensional constructs. Results from confirmatory factor analysis, structural equation modeling, and hierarchical regression indicate that sales-contextualized grit is psychometrically distinct and exhibits stronger contextual relevance than its domain-general counterpart. Consistent with theory, grit (conceptualized as perseverance of effort and consistency of interest) emerges as an important antecedent to sales self-efficacy, which in turn relates to salesperson performance and turnover intentions. Exploratory analyses indicate these relationships vary based on personal resources, suggesting boundary conditions. These findings challenge scholars to conceptualize grit as multidimensional and prioritize domain-specific measures in sales contexts, while offering practical insights for managers seeking to enhance performance and reduce turnover.
Nuclear factor erythroid 2-related factor 2 (NRF2) is a central transcription factor in the cellular antioxidant stress response, whose activity exhibits marked context-dependence. Under normal physiological conditions and in precancerous lesions, timely and moderate activation of the NRF2 pathway establishes a multi-layered, synergistic cellular defense network. Through mechanisms including efficient detoxification of exogenous carcinogens, clearance of excess reactive oxygen species (ROS), suppression of chronic inflammation, maintenance of genomic stability, regulation of autophagic homeostasis, and promotion of immune surveillance to eliminate precancerous cells, NRF2 exerts potent chemopreventive effects, thereby preventing malignant transformation. In established malignancies, however, constitutive activation of the NRF2 pathway—driven by mutations in KEAP1/NRF2 or persistent oncogenic signaling—fundamentally reverses its role. Under these conditions, NRF2 becomes a central hub that promotes tumor progression and therapy resistance by remodeling redox homeostasis to sustain low ROS levels favorable for proliferation, reprogramming metabolism to support biosynthetic demands, upregulating multidrug resistance-associated proteins, and shaping an immunosuppressive tumor microenvironment. This review examines the molecular basis of the dual roles of NRF2 at different stages of carcinogenesis and summarizes the distinct biological effects of natural and synthetic modulators of this pathway.
Graph Neural Networks (GNNs) have emerged as effective models for analyzing complex and structured data in software engineering tasks, particularly issue categorization and bug triage. This study investigates the performance of Prior Structural Information GNN (PSI-GNN) on the open-source TAWOS dataset containing over 500 000 Agile project issues. Our pipeline includes an ETL (Extract–Transform–Load) process and semantic descriptor extraction using Term Frequency–Inverse Document Frequency (TF–IDF) and Porter Stemmer to transform issue descriptions into rich feature representations. PSI-GNN is optimized using the Taguchi orthogonal-array method, achieving exceptional eco-efficiency with an F1 score of 94.86
Serbia's agricultural budget has shifted repeatedly between production-linked market support and direct payments (MSDPs) and investment-oriented rural development support (RD). Using harmonized annual data for 2015-2025 from final amended incentive regulations, food-price indices, and yield statistics, we examine whether the composition of agricultural support is associated with short-run productivity and technical-efficiency dynamics. Descriptive analysis documents partial decoupling through 2017-2023, followed by an abrupt recoupling in 2024-2025 when MSDP absorbed over 92% of the incentive budget. Econometric results from OLS models with Newey-West standard errors indicate that food-price inflation is strongly associated with lower productivity growth, while the dominance of coupled instruments does not predict improved short-run performance. The political-importance indicator is not statistically significant once inflation is controlled for. The findings suggest that the structure and stability of agricultural spending matter at least as much as its volume, underscoring the importance of safeguarding investment-oriented measures as Serbia pursues EU policy alignment.
Recent research in learning analytics has increasingly focused on identifying “at-risk” students to enable timely and meaningful interventions. While predictive analytics has played a central role in this effort, the integration of explainable artificial intelligence (xAI) has emerged more recently as a way to enhance the transparency and interpretability of predictive models. This paper proposes a framework that integrates clustering, classification, and xAI techniques to support the generation of transparent and personalized recommendations aimed at enhancing student learning progress. The approach consists of four sequential phases: (i) clustering (ii) classifying (iii) generation and evaluation of explainable rules, and (iv) intervention generation based on the associated rules. A key contribution of the proposed framework is its ability to provide transparency and learner-friendly generated insights. To achieve this, a rule-based system was developed to transform xAI-based rules into clear, and human-readable feedback for each student. The proposed framework was empirically validated using real, longitudinal learner data, examining learner achievement patterns over time, and visualizing temporal dynamics. This validation shows that interpretable analytics can provide deep insights into student learning progression and offer practical value for improving teaching strategies and learning outcomes.
