Improving the Utility of Large-Scale Assessments in Canada - Issues, Solutions, and Multidimensional Scheduling

Oct 15, 2026

Principals and teachers often avoid using large-scale assessment results because subtests are not distinct or reliable, results arrive too late to inform instruction, and educators need support to translate findings into teaching improvements. The recommendation is to first confirm that the assessed domain is multidimensional, then revise scheduling so each subject is assessed in non-consecutive years while keeping the same number of sittings. Assistance should help educators use results effectively. Three assessment cycles are proposed to strengthen subtest reliability and provide at least two years for instructional change.

LanguageEnglish

CategoryExam

FormatPDF

Pages22

Reading Time55 min

AccessFree to download

Abstract

Introduction

Beneficial Effects of Large-Scale Assessments

Detrimental Effects of Large-Scale Assessments

Recommended Assessment Design and Scheduling

Multidimensional Domain Establishment

Non-Consecutive Year Scheduling

Proposed Assessment Cycles and Reliability Improvements

Improving the Utility of Large-Scale Assessments in Canada W. Todd Rogers University of Alberta Abstract Principals and teachers do not use large-scale assessment results because the lack of distinct and reliable subtests prevents identifying strengths and weaknesses of students and instruction, the results arrive too late to be used, and principals and teachers need assistance to use the results to improve instruction so as to improve student learning. Therefore, it is recommended that the first assessment activity should be to clearly establish that the domain to be assessed is multidimensional. Given this, the assessment schedule should be changed so that a given subject area is assessed in non-consecutive years but the number of sittings remains the same each year. Assistance should be provided to principals and teachers so as to increase their understanding of how to use large-scale assessment results. Three suggested assessment cycles are presented, each of which increases the reliability of subtests and provides principals and teachers with at least two years to make changes in instruction. Keywords: large-scale assessments, subtests, issues, solutions, multidimensional Canadian Journal of Education / Revue canadienne de l’éducation 37:3 (2014)©2014 Canadian Society for the Study of Education/ Société canadienne pour l’étude de l’éducation [www.cje-rce.ca](www.cje-rce.ca) Résumé Les directeurs et enseignants n’utilisent pas les résultats d’évaluations à grande échelle car le manque de sous-tests fiables et distincts empêche l’identification des forces et faiblesses des étudiants et de l’instruction, les résultats arrivent trop tard pour être utiliséset les directeurs ainsi que les enseignants ont besoin d’aide pour exploiter les résultatsafin d’améliorer l’instruction, de manière à améliorer l’apprentissage des élèves. Il est donc recommandé que la première activité d’évaluation consiste à établir clairement que le domaine à évaluer est multidimensionnel. Ceci étant établi, le programme d’évaluation devrait être modifié de manière qu’un domaine donné soit évalué lors d’années non consécutives, mais que le nombre de séances demeure identique chaque année. Une aidedoit être apportée aux directeurs et enseignants afin d’accroître leur compréhension dela façon d’utiliser les résultats d’évaluations à grande échelle. Trois cycles d’évaluation suggérés sont présentés, chacun augmentant la fiabilité des sous-tests et donnant aux directeurs et enseignants au moins deux ans pour apporter des modifications aux méthodes d’instruction. Mots-clés : évaluations à grande échelle, sous-tests, problèmes, solutions, multidimensionnel This article was first presented as a Presidential Address delivered to members of the Canadian Educational Researchers’Association (CERA) during the annual conference of the Canadian Society for the Study of Education (CSSE), in Victoria, June 3, 2013. Cet article fut présenté pour la première fois sous la forme d’un message du président aux membres del’Association canadienne de chercheurs en éducation (ACCE) lors de la conférence annuelle de la Société canadienne pour l’étude de l’éducation (SCÉÉ), à Victoria, C.-B., le 3 juin 2013. Canadian Journal of Education / Revue canadienne de l’éducation 37:3 (2014) [www.cje-rce.ca](www.cje-rce.ca) Introduction Government policy makers and officials in Canada believe public evidence of student performance drawn from sound and credible large-scale assessments will help focus educators’ attention on improving curriculum and instruction and, as result, enhance student learning and performance. While there are many who agree that data-informed decision making is a way to improve curriculum and instruction so that student learning is improved, others question the perceived value of large-scale assessments as a way to improve curriculum and instruction so that student learning is improved. The purpose of this article is to briefly review

FAQs

Why do principals and teachers not use large-scale assessment results in the article?

They do not use the results because subtests are not distinct and reliable, results arrive too late to affect instruction, and educators need help to use the results to improve teaching and student learning.

What is the first recommended step before redesigning assessment activities?

Clearly establish that the domain being assessed is multidimensional.

How should the assessment schedule be changed to address the concerns?

Assess each subject area in non-consecutive years while keeping the number of sittings the same each year.

Oct 15, 2026

Uploaded by Well Done