Reliability and validity of the instrument.

The COVID-19 pandemic disrupted global education systems, forcing rapid shifts in teaching practices, technology integration, and assessment methods. However, little is known about how teacher efficacy, job satisfaction, and digital adoption vary across economic contexts. Insufficient research examines how income levels influence these factors, hindering equitable support for educators in post-pandemic recovery. This study examines variations in teacher efficacy (TE), job satisfaction (JS), assessment practices (AP), and technology adoption (UIT/UDT) across low-, upper-middle-, and high-income countries (LMICs, UMICs, HICs) before and after the COVID-19 pandemic. Guided by Bandura’s Social Cognitive Theory, the analysis explores how these factors interact and shift in response to pandemic-related disruptions in educational systems. The study utilizes PISA 2018 (pre-pandemic) and 2022 (post-pandemic) data from 128,866 teachers across 24 countries, employing structural equation modeling and machine learning as primary analytical techniques. Results indicate that job satisfaction significantly affects teacher efficacy but has minimal direct impact on the use of instructional technology tools. Teacher efficacy demonstrates a significant positive effect on both technology adoption (UIT) and assessment practices, while the use of digital learning and communication tools similarly influences assessment practices. These findings suggest that teacher efficacy and digital tool integration are key determinants of assessment practices. The study highlights how economic contexts shape teacher development, proposing targeted approaches for equitable post-pandemic education. HICs benefit from institutional support reinforcing the JS-TE relationship, while LMICs require solutions addressing resource gaps that impede consistent technology implementation. These evidence-based findings support context-specific policy interventions to enhance teacher support and digital integration globally.