Assessment: We Cannot Improve at Scale What We Cannot Measure

The title of this post is borrowed from a chapter in an inspiring new book, Learning to Improve: How America’s Schools Can Get Better. How does this apply to social-emotional and character development (SECD)? It relates to the question of whether schools should administer standardized tests for SECD. Should they? Would we want an EQ (emotional intelligence) to accompany a child’s IQ in his or her permanent record?

Assessing School Assessment

The authors of the book make a vital distinction between assessment for research and evaluation (assessment designed primarily to provide information for researchers or policy makers concerning theoretical questions or general trends, usually using standardized procedures) and assessment for improvement (assessment designed to provide specific feedback to implementers about particular programming, using tailored procedures). They assert that the high-stakes testing movement has been harmful because they advocate assessment for research and evaluation over improvement.

They also point out that it has been an elusive task to create standardized metrics that are fair, developmentally and culturally sensitive, and scalable. The tests currently in use in most states do not get passing grades for test construction.  

In some cases, these tests have led to more harm than good for students and teachers by providing flawed evaluation metrics under the guise of feedback for improvement. We should be highly skeptical of heading down this road for social-emotional and character development.

A Better Way to Assess

The book authors propose that school professionals in a building or district get together and define the skills and virtues they believe to be important and the indicators that make the most sense to them in everyday practice. What emerges from this kind of process is a tailored, shared, and relevant metric that points the way to student improvement.  

This point was reinforced in the commentary section of Education Week (June 8, 2016) which asserted that teacher observation of and reflection on students’ social competencies is powerful, appropriate, and accurate, more so than any student-completed assessment yet devised.

Further, when professionals form networked improvement communities, they share highly relevant best practices and help one another refine and improve their assessments more quickly than doing so in relative isolation. 

For more schools to accomplish this sort of collaboration, districts need to be investing more money in the professional development of their personnel and less in the machinery that supports the testing industry. There is no justification for annual high-stakes testing that does not contain the mechanisms of improvement.

The Purpose of Assessment? To Guide Improvement

When assessment is connected directly to the focal skills that schools and districts want to enhance most, a positive, supportive cycle is created between assessment and action that does not exist with the current standardized testing system.

Student work habits, social-emotional behavior, attitudes toward learning, and character attributes are strongly related to student academic outcomes and are amenable to intervention and improvement.  

Right now, our assessment of these areas needs improvement if our students' skills are to improve.

In Ann Brown’s Presidential Address for the American Educational Research Association (AERA) more than two decades ago, she advocated for assessment procedures that are “authentic, transparent, and aligned with the curriculum.”

If we are to improve students’ social-emotional and character development, observations and rating by teachers need to be aligned to local SECD goals and curricula. Once this is accomplished regionally, it is then possible to implement it to scale in districts across the country -- and even internationally.