Large-scale assessments and their effects: The case of mid-stakes tests in Ontario.

ABSTRACT

This paper analyzes the nature and perceived effects of mid-stakes testing (known as the EQAO) in Ontario, Canada. Ontario's mid-stakes tests were meant to ensure accountability and transparency, and assure system-wide improvement, while avoiding the negative effects and perverse incentives of their high-stakes counterparts. The paper provides new evidence from two projects covering almost a 10-year time-span in 10 of Ontario's 72 school districts. It shows that even though mid-stakes testing is milder in its manifestations and effects than high-stakes testing, concerns remain about the need for and side effects of such testing. The findings concern two periods of Ontario educational reform. In the first period, with a specific focus on improving performance in literacy and mathematics, administrators and special education support staff felt that the assessments raised teachers' expectations and sense of urgency leading to steady improvements in measured achievement, but that there was also evidence of negative effects, especially on paying undue attention to "bubble" students just below the threshold for minimum proficiency. In the second reform period focused on broad excellence, well-being and equity as inclusion, mid-stakes tests were perceived as having more widespread negative effects. These included teaching to the test, cultural bias, avoidance of innovation, dilemmas of whether to include highly vulnerable students in the testing process or not, and emotional ill-being among students and teachers. The paper concludes that Ontario's twentieth century system of large scale, mid-stakes assessment has not kept pace with its twenty first century commitments to deeper learning and stronger well-being.