This study investigates the structural relationships among sixth-grade students' number sense, understanding of fractions and decimals, and mathematics achievement through Structural Equation Modeling (SEM). The sample consisted of 272 sixth-grade students (aged 11-12) from public schools in western T & uuml;rkiye. Number Sense Test (NST), Fraction Conceptual Understanding Test (FCUT), and Decimal Conceptual Understanding Test (DCUT) were used. Data were analyzed using SPSS and SEM path analysis. Results showed that students' overall number sense performance was low and differed across components. Students performed better in numerical meaning and magnitude but struggled with interpreting operations and judging result reasonableness. SEM results showed NST significantly predicted FCUT and DCUT, which in turn predicted GPA. NST had indirect effects on GPA through FCUT and DCUT and a direct effect in the partial mediation model, explaining 46% of GPA variance. Overall, number sense influences achievement directly and indirectly through rational number understanding.
Over the last decade, educational policies have recommended incorporating data science literacies into the K-12 education curriculum. At the same time, policies have called for educators to support all learners using frameworks such as Universal Design for Learning (UDL). However, data science instruction is largely understudied at the elementary level, and few studies have explored the concept of UDL within data science instruction (specifically computational thinking). Utilizing an exploratory case study approach, we evaluated the effectiveness of a professional learning and development experience with educators in grades three through five. We analyzed survey data, student data, unit plans, and journal prompts. There were statistically significant increases in the participants' perceived knowledge and usefulness related to UDL. Similarly, there were statistically significant increases in student gain scores between pre- and post-tests for students with and without disabilities. Additionally, qualitative findings illustrate how one group of educators integrated UDL into data science instruction at the elementary level.
The present study examined the longitudinal interrelationships among homework motivation (expectancy and value) and homework behaviors (effort, time management, and procrastination) in fourth-grade primary school students in Turkey. Data were collected at two time points and analyzed using a cross-lagged panel model. The results revealed strong autoregressive patterns for all constructs, indicating that students' motivational and behavioral tendencies remained stable over time. In contrast, cross-lagged paths were generally weak. The only significant cross-lagged association indicated that higher levels of homework procrastination at the initial measurement were negatively prospectively associated with subsequent task value, reflecting a temporal association. Expectancy and value did not significantly exhibit associations with later effort, time management, or procrastination, and no reciprocal temporal association emerged between expectancy and value. Time management also showed no significant associations with any variable, indicating the absence of statistically detectable temporal associations within the present sample. Overall, the findings highlight the developmental distinctiveness of motivational processes in childhood and underscore the need for early interventions aimed at sustaining task value and preventing the consolidation of procrastination patterns.
Active learning enhances online teaching but faces challenges like superficial interaction and misalignment with learners' cognitive realities in synchronous settings. This study proposes the "Articulation-Elicitation-Calibration-Negotiation (AECN)" framework by integrating a two-tier diagnostic-based cognitive assessment, aiming to enhance active learning effectiveness and facilitate deeper synchronous online learning experiences. A quasi-experimental study was conducted with 67 Grade 4 students, where the experimental group received instruction via the AECN framework and the control group via conventional lecturing. Results suggest that the AECN group significantly outperformed the control group in both academic performance and higher-order cognitive skills. The findings indicate that the AECN framework can facilitate active knowledge construction without imposing additional cognitive load, highlighting the potential of prior knowledge diagnosis-driven structured guidance for optimizing synchronous online active learning.
