The main purpose of constructing an educational database is to find the key factors that affect students’ learning outcomes in view of the educational issues that are of great concern by all sectors of society, so as to improve the educational system and teaching site problems, to predict future educational trends, and to implement an educational decision-making mode that is “evidence-based”.
The purpose of this study is to explore how to construct and promote the longitudinal educational database of students’ learning process. Since the United States has already obtained fruitful results from its construction of the P-20W longitudinal educational database, this study takes it as the research object, collects relevant laws and practical cases, and uses the document analysis approach to explore and explain the findings.
This study notes that P-20W has taken a series of actions to achieve its strategic objective at improving students’ learning effectiveness, which include building personal data on learning for each student from pre-school through workforce entry, capturing information sufficient to effectively achieve its strategic objective, analyzing and providing this information to help stakeholders make educational decisions, and improving problems in the classroom. To implement the strategy and ensure data quality, P-20W has invested a lot of resources, formulated relevant bills, set up a helpful environment, removed related barriers, and supported it with various activities.
Based upon the perspective of education and information technology and referring to the planning practice of an information system, this study presents the research results through the sequence of working out data strategy, identifying the educational needs and problems, establishing data governance structure and processing, setting up privacy, security, and confidentiality policies and facilities for students’ data, and finally proposing some suggestions as a reference for the establishment of such databases in Taiwan.
Keywords: educational big data, longitudinal data, data strategy, data mining, quality assurance