March 11, 2021
There has been a lot of discussion around the science of reading. As a result, organizations have published reports and commentary on the issue one example being the Council of Chief State School Officers’® (CCSSO®) recently released A Nation of Readers report calling for states to take concrete steps to help every child learn to read. The focus on reading is based on the simple observation that literacy is “the foundational skill for success at every school level and in postsecondary education, work, and citizenship.” While literacy is critical for success in both school and life, far too many children lack access to high quality opportunities to learn foundational literacy skills. Reading performance is unacceptably low on key literacy indicators, with large and persistent gaps in achievement between Black and White students. As educators, we must do better.
The good news is that there is an emerging consensus on how to effectively teach reading. As noted in A Nation of Readers, the “evidence base is strong and settled regarding how children learn early foundational reading skills and how teachers can teach those skills.” The education field already knows what to do. The fundamental challenge is to help states, schools and especially teachers provide effective reading instruction for all students.
A Nation of Readers identifies a number of actions that states need to take to improve reading outcomes. These include implementing state policies that support coordination among key resources such as curriculum materials and reading standards, engaging college and university partners to ensure that teachers develop the core knowledge and skill needed to teach reading effectively, and keeping a focus on equity so that all students, regardless of background, have access to high-quality reading instruction.
All components of this agenda depend on “a coherent vision for improving literacy that is grounded in evidence-based practices.” To succeed, this agenda needs to identify and develop tools that can provide relevant, concrete, and actionable feedback on both evidence-based teaching practices and student learning outcomes. In short, both teacher and student assessments have an important, and arguably necessary, role to play in ensuring the success of this vision.
The first step is to develop assessments that focus directly on the knowledge and skill actually used to carry out effective reading instruction. While teacher knowledge provides a critical foundation, too often teacher assessments focus on competencies that have little to no direct connection to the evidence-based practices that are actually used to carry out the work of teaching reading. To advance the agenda laid out in A Nation of Readers, teacher assessments need to be designed for a variety of uses. For example, formative assessments can support prospective teachers as they try out and learn new practices or provide feedback to practicing teachers working to improve their reading instruction. Practice-based licensure assessments help policymakers and administrators ensure that all students have access to highly qualified teachers. While it is important that assessments focus on teachers’ own knowledge about reading and best practice, it is critical to assess the skills that are actually used when teaching reading.
The science of reading not only applies to assessments of teacher knowledge and practice, but also to students. A key component of student assessment is to provide an opportunity to measure what students know and can do, not just what they can’t do. Reading is a complex process that involves a number of key skills. In order for assessments to be maximally effective, they need to measure the full range of skills, including foundational (e.g., decoding) and higher-level skills (e.g., critical thinking). Foundational reading skills represent a barrier to allowing many students to demonstrate what skills and knowledge they do have. For example, English-language learners may have vocabulary knowledge but difficulty decoding words in English. Likewise, a student with dyslexia may have good reading comprehension skills using text to speech support but struggle with decoding or reading fluency.
The National Center for Educational Outcomes (PDF) reports that only 10 states prohibit the use of text to speech accommodations on the state reading assessment. While this is helpful to allow some students to demonstrate their comprehension skills, it may result in a reduced instructional focus on the foundational reading skills. ETS researchers have proposed models for how to measure both comprehension skills and foundational skills at scale using adaptive testing, automated estimation or oral reading fluency, and diagnostic assessment models. The diagnostic approach helps to ensure weaknesses are identified early and that appropriate interventions are implemented before students start to fall behind. But assessments can do more than just measure what students know and can do. When designed carefully, student assessments also model good habits of mind, support strategic processing, and facilitate student development. Advances in technology and the learning sciences need to be leveraged to both measure and support student learning.
While many good assessments are already available, gaps exist. Filling these gaps will require new development that draws on innovations and new assessment technologies. Aligning student and teacher assessment to focus directly on evidence-based teaching and learning can provide a powerful set of tools to support the actions recommended in A Nation of Readers. Assessments grounded in the science of reading will help make concrete both what teachers need to know and what students need to learn. Carefully aligned student and teacher assessments can support learning and equity by providing the types of actionable evidence that help to ensure all students learn to read.
Geoffrey Phelps is a senior research scientist at ETS. Tenaha O’Reilly and Cara Laitusis are managing principal research scientists at ETS.