Student achievement trajectories in Ontario: Creating and validating a province-wide, multi-cohort and longitudinal...

Area of Science:

• Education
• Public Health
• Data Science

Background:

• Longitudinal student data is vital for understanding learning and informing policy.
• Ontario lacks large-scale, longitudinal data on student achievement.
• Linking diverse datasets requires robust protocols and partnerships.

Purpose of the Study:

• To develop and validate a linkage protocol for creating a longitudinal student achievement database in Ontario.
• To merge administrative data on child development, educational assessments, and neighborhood/school characteristics.
• To assess the generalizability of the linked database to the broader student population.

Main Methods:

• Linked Early Development Instrument (EDI) data with Educational Quality and Assessment Office (EQAO) assessments using deterministic methods.
• Integrated school-level and neighborhood-level datasets.
• Examined differences between linked and unlinked cases to validate data representativeness.

Main Results:

• Successfully linked 50% of EDI cases, creating a database tracking achievement from kindergarten to grade 10.
• The linked database includes developmental, demographic, neighborhood, and school covariates.
• Negligible differences were found between linked and unlinked cases for most demographic measures.

Conclusions:

• The developed linkage protocol successfully created a representative longitudinal database for Ontario students.
• This approach addresses critical gaps in sustainable research capacity.
• Recommendations include implementing linkage protocols and data verification for representative data creation.