Voices from the Field: Performance Assessments in State Accountability as discussed at the National Conference on Student Assessment

In June 2017, the National Conference on Student Assessment offered opportunities to delve into the possibilities for flexibility and innovation in assessment and accountability presented under the federal Every Student Succeeds Act (ESSA). The conference, sponsored by the Council of Chief State School Officers, provided deep dives into a range of topics, including discussions related to the use of performance assessments for school accountability.

Stuart Kahl of Measured Progress outlined a rich history of state-led initiatives to adopt performance assessments in the 1980s and ‘90s, such as the Connecticut Assessment of Educational Progress – Science.¹ In response, Susan Tave Zelman of the Ohio Department of Education offered that despite great interest in successfully maintaining a performance-based system of assessments and a long history of attempting to do so, “states are still struggling with how to integrate performance tasks in a system with standardized tests, . . . and no state has yet been successful in creating a state-wide system integrated into state accountability.” Zelman identified three sources of obstacles to implementing performance assessments for this purpose:

1. Design challenges, such as a lack of consensus on the role of states, alignment of performance assessments to states’ standards, integration into state accountability systems, and opportunities to learn;
2. Resource challenges, such as limits on money, time, state capacity, vendors, technology platforms, and political will; and
3. Measurement challenges, such as validity, reliability, timeliness of results, comparability, and scalability.

John Weiss of the Pennsylvania Department of Education and Joyce Zurkowski from the Colorado Department of Education echoed the challenges related to political will.² Zurkowski recommended introducing performance assessments into a balanced assessment system, but avoiding immediate association of results with accountability. Following this model, Colorado has had early success experimenting with performance assessments locally to improve instruction and student learning before making the transition to accountability.

Though implementing performance assessments at the classroom level may be a pathway for states to build towards innovative systems of accountability, this type of work at the district level is not without its own challenges. Giselle Martin-Kniep of Learner-Centered Initiatives works with teachers to design and implement performance assessments.³ Martin-Kniep explained that designing authentic assessment tasks is a challenge for teachers in every subject and grade. Teachers also need guidance on how to provide formative assessment opportunities prior to administering summative tasks. However, early evidence suggests that once successfully implemented, school districts value the use of performance-based assessments to inform instruction. Jacob Mishook of Achieve, Inc. commented that his work with districts to inventory the suites of assessments administered to students throughout the year revealed that “tests that are the most valued [by districts] measure real learning [and] assess problem solving and critical thinking.”

Alternate assessments for students with significant cognitive disabilities may offer a more readily-accessible entry point for the practice of using performance tasks for accountability. Jan Sheinker of Sheinker Educational Services warned states that “portfolio assessments are not going to fly anymore. . . . States need to start thinking about complex performance tasks that can be developed with integrated accommodations.”⁴ Sheinker suggested that performance assessments can be designed to measure student proficiency with scaffolding that could be added or removed based on student individual needs. While Sheinker touted consistent scoring and comparability as benefits of performance assessments over portfolio-based alternate assessment systems, performance assessment design for accountability still presents significant technical challenges, as Zelman previously noted.

The tenor of this conference suggested great interest shared among education stakeholders in using performance assessments as part of state accountability systems. That being said, overcoming the real technical and political barriers will require additional research and evaluation – such as those results presented by Art Thacker of HumRRO at the session titled, “Formative Evaluation of New Hampshire’s Performance Assessment of Competency Education (PACE)”⁵ – and a widespread effort to improve assessment literacy, not only among teachers, but also for policy makers, parents, and even students.