The development of trustworthy learning analytics technology has become a central focus for researchers globally.Both domestic and international scholars have explored its connotations and underlying value logic extensively.However,several chal-lenges remain,including unclear core issues,ambiguous theoretical boundaries,and vague practical applications.This study aims to address these challenges by constructing a theoretical framework and exploring the practical potentials of trustworthy learning analytics in education.Practically,this study identifies the key barriers to implementing trustworthy learning analytics,such as algorithmic in-accuracies due to data drift,biases from imbalanced sample distribution,and the loss of trust caused by the"black box"nature of al-gorithms.Theoretically,the study combines analyses of theoretical and practical concerns to establish a comprehensive framework for trustworthy learning analytics,focusing on four key elements:learning environments,data governance,modeling,and intervention feedback.Building on this framework,the study proposes practical solutions,including ethical adaptation through value-sensitive de-sign,to systematically bridge the gap between theory and practice in applying trustworthy learning analytics within educational re-search.
万方数据
维普
