This study provides a decade-long bibliometric analysis of global research on educational robotics from 2015 to 2025, using the Web of Science Core Collection as its primary database. It examines over 1,000 peer-reviewed publications to trace the development, dissemination, and pedagogical applications of robotics in various educational settings. The analysis reveals exponential growth in publications since 2018, with significant surges during the post-pandemic digital transformation era. Key contributing countries include the United States, China, Spain, and South Korea, with growing participation from Latin America and Eastern Europe. Thematic mapping identifies dominant research clusters in areas such as robot-assisted language learning, STEM project-based robotics curricula, early childhood coding, and the integration of humanoid robots in special education. Citation and co-authorship networks reveal strong interdisciplinary collaboration among educators, computer scientists, and cognitive psychologists. Despite this growth, the study identifies critical gaps related to teacher training, curriculum integration, cost-effectiveness, and long-term learning outcomes. The findings underscore the need for policy frameworks that support inclusive access to robotics education, particularly in underserved regions. This bibliometric synthesis offers actionable insights for researchers, educators, and policymakers aiming to harness robotics to foster creativity, computational thinking, and problem-solving among 21st-century learners.

Keywords: Educational robotics, STEM education, computational thinking, robot-assisted learning, interdisciplinary teaching, bibliometric mapping.