What better school tests would look like

The predominant rating system in the United States urgently needs to be updated. The country needs to move to assessment systems that provide a more robust overview of what students know and can do without distracting and taking away from learning time.

Jay McTighe, the renowned writer and educational consultant with 50 years of experience in education, has created a vision of what this could look like that offers both hope and feasibility.

McTighe’s vision is based on his observation of the impact of assessments, especially standardized ones, used for accountability purposes.

Like McTighe told me recently“In many cases, the evaluation tail seemed to be wagging the dog, the instruction and the school dog. … [So] let’s focus our attention on the tail. Let’s think about trying to better understand assessment and how to get the dog’s tail to wag in a better direction.

This best direction for testing that McTighe envisions is described in his white paper entitled “Measuring what matters: It’s time for a redesign of assessment.

In it, he details a bold but achievable three-part plan for re-doing assessment that pays attention to three learning goals: acquisition, comprehension, and transfer.

Acquisition focuses on the knowledge and skills we want students to acquire. Comprehension refers to understanding big ideas, such as key concepts and principles in major disciplines, as well as conceptualizing key processes, such as research methods, conceptual thinking, or effective writing. And transfer refers to a learner’s ability to take something learned in one context and apply it in another.

The assessment system proposed by McTighe addresses these varied learning goals through a three-pronged infrastructure to produce not just a single snapshot of learning, but a comprehensive photo album to capture the full picture of education in the schools.

The first step

The first “leg” of this evaluation stool would be “similar to what we have now,” in some ways, McTighe said. “Content-based testing to test important knowledge and skills students need to acquire, usually in subject areas.”

McTighe does, however, propose some adaptations to the current system.

For example, rather than having 50 different state tests that are not comparable, the National Assessment of Educational Progress (NAEP) could be used to provide reliable measures of student achievement in the content areas of mathematics, of reading and writing.

A second change would use a sampling system, similar in many ways to NAEP’s, so that every student wouldn’t need to be tested on the same things at the same time. For example, a representative sample of children at each grade level tested would pass the math section, while a separate sample would take the reading test, and so on. Such a system could obtain reliable data to assess student achievement at the school, district, and state levels, while saving significant testing costs.

The return match

The second stage is a set of performance tasks built into the curriculum that challenge students to apply their learning in realistic situations. The aim, according to McTighe, is to give “evidence of understanding and transfer”.

Rather than interrupting the flow of instruction with a traditional standardized test, these performance tasks would be based on curriculum standards and therefore integrated into the natural flow of the classroom. A single performance task would also be able to gather evidence of achievement in multiple subjects, McTighe said. For example, a science task might include a math and writing component, while assessing important skills like critical thinking and creativity.

In McTighe’s vision, student performance on these tasks would then be graded in a manner similar to advanced placement exams in writing and the arts.

“Three times a year schools in an area would close and teachers from those schools and administrators would meet at grading sites to review the work children were producing on performance tasks and grade them against rubrics. and well-developed anchor documents, following interrater. reliability protocols,” McTighe said.

Such a system would, as McTighe acknowledges, be difficult to convince some to implement. But he argues that the challenge of implementation is not one of know-how, because our education system already has experience with such grading systems.

He also argues that because the group grading process provides invaluable professional development for teachers, the costs of implementing this system could come from both training and assessment budgets, making it more affordable and feasible. And the professional development offered would be better than traditional teacher training.

The third leg

The third step would be “local assessments integrated into the program”.

In his design, McTighe describes these assessments “including the traditional course exams that many high schools now have, but also things like genius hour or personal projects, passion projects that kids are interested in. This could involve This could involve a variety of more authentic learning that is otherwise outside the mold of traditional testing.

Unlike the other two streams, these local assessments would be unique to each school or district and therefore would not provide comparability data like the other two streams.

Instead, this third element would allow local schools and districts to participate in and invest in a more comprehensive assessment system, as well as “filling in the things that slip through the cracks in the first two stages,” said McTighe. “[This is] aims to complete the assessment table and ensure that we assess all the things we claim to value, not just the things that are easiest to test and quantify.

This more broadly reflects McTighe’s view that an assessment system should not only produce accountability data on the most easily measured learning outcomes. It should serve to enrich and enhance learning of the most important skills that will prepare today’s children to thrive in an increasingly complex world.

Norma A. Roth