Rethinking K-12 Assessment 

Alternative Assessments in Practice: Early Lessons from District Implementation

In the first post in this series, Rethinking K-12 Assessment, we examined how states are designing, piloting, and supporting alternatives to traditional summative assessments. But state policy tells only part of the story. To understand what these models look like in practice, EduDream conducted a landscape scan of 21 districts across eight states. Eleven districts were implementing through-year assessments (TYAs), while 10 were implementing other innovative approaches. The scan examined how districts were implementing the assessments, early successes and challenges, and what was known about testing time, instructional time, and outcomes for different subgroups. 

For the district-level scan, we brought together state and district materials, assessment documentation, case studies, student performance data, and other publicly available sources. The clearest finding was that public information is much more readily available on what assessment models are intended to do than on how they work in practice. Publicly available documentation commonly described assessment design, structure, purpose, and administration requirements, but much less information was available about actual classroom implementation, teacher experience, student testing burden, or district adaptation. 

Early Lessons from Districts 

In our scan, we found some early successes emerging alongside many practical challenges.

Lesson 1: Through-year assessments may provide faster feedback, but usefulness is not guaranteed. 

Early examples of through-year implementation suggest that students may prefer shorter yet more frequent testing sessions to a single end-of-year exam. A survey of more than 35,000 students in Montana revealed that students preferred the more frequent, yet shorter assessment format of the Montana Aligned to Standards Through-year (MAST) assessment over a single end-of-year exam. 

Some Montana educators expressed that the TYA model provides quicker feedback that helps inform instruction throughout the school year. In Indiana’s Carmel Clay schools, leaders have used checkpoint results from the state’s through-year model, Indiana Learning Evaluation and Assessment Readiness Network (ILEARN), to inform daily classroom instruction, as well as placement in high-ability courses. These early examples illustrate a common argument for through-year approaches: results may reach educators early enough to inform instructional decision-making.  

Nonetheless, more frequent data are only useful when the results are clear and credible. Educators and leaders in Montana and Indiana have raised concerns about score reliability, reporting, and instructional alignment. In Carmel Clay, some educators noted that score reports can be difficult to interpret and questioned how overall performance levels related to individual test questions. In Billings, district leaders wondered why the state’s through-year proficiency results did not align with other nationally measured assessments. 

Lesson 2: Innovative assessment models can broaden what counts as evidence of learning. 

Innovative assessments offer a different vision of student learning and skill demonstration. In California, many local districts offer students Linked Learning Pathways. At Los Angeles Unified, students participating in these pathways complete portfolios and defend their work before a panel, while at Oakland Unified, students complete capstone projects connected to their academies. 

Teachers and students alike have expressed that these opportunities have helped students develop and demonstrate critical-thinking, problem-solving, and communication skills. Rather than simply selecting answers on an exam, students must be able to explain their thinking, synthesize information, and present what they have learned. 

Other districts have linked assessment more directly to college and career readiness. For example, New Mexico’s Santa Fe and Gallup-McKinley school districts have expanded their career-connected pathways, while districts in Colorado allow students to demonstrate graduate competencies through state-approved options such as projects, presentations, capstones, and other performance-based opportunities. 

Some districts, like Denver and Pueblo in Colorado and Santa Fe in New Mexico, have reported upward graduation trends concurrent with the implementation of these innovative assessment approaches. While these rates may be promising indicators, further research is still needed to isolate the role played by these innovative assessments. 

Lesson 3: Educator capacity may determine whether innovative assessments can be sustained. 

Our scan revealed that many districts identified educator capacity as a central implementation challenge. Santa Fe, Gallup-McKinley, and Los Angeles Unified reported difficulty maintaining qualified staff to implement innovative assessment approaches. Oakland Unified students described inconsistent teacher guidance during the capstone process, while Los Angeles educators also voiced concerns about additional work, lack of compensation, and the need for dedicated on-site coordinators. 

These innovative assessments require educators to do more than just administer a single test. These innovative, performance-based assessments require educators to engage with the process by helping students plan long-term projects, develop scoring criteria, provide iterative feedback, coordinate presentations, and evaluate more complex student work. 

Lesson 4: The effect of alternative assessments on testing and instructional time remains unclear. 

Despite reduced testing time being one of the most lauded promises associated with through-year assessment, our scan did not produce meaningful evidence that these models indeed reduce total testing time. Although Louisiana reported a reduction in testing time across state assessments, the available documentation did not clearly indicate whether this testing time reduction included the state’s through-year assessment. Meanwhile, Montana estimates that the average student will spend 6-10 hours on the through-year MAST assessment over the school year, which amounts to less than 1% of instructional time. However, our scan did not find evidence of how much time students are actually spending on these exams. 

Additionally, some teachers in Montana expressed concern about the timing of TYA administration and whether the exam content aligned with what students had learned up until that point. Nearly 73% of Montana teachers felt that they were unable to schedule TYA administration to align with their syllabi. This finding suggests that assessment burden goes beyond measuring in hours; shorter assessments can still interrupt instruction if students are tested on material not yet taught. 

For innovative assessment models, we found no public evidence documenting the effects on student testing burden or instructional time.

Learning Implications 

Taken together, these early lessons suggest that changing the structure or format of assessment is only one part of meaningful assessment reform. The success of these alternative assessment methods requires more than just strong design. Whether these models ultimately deliver on their promise depends on how they are implemented, how educators are supported, and if (and how) their effects on students and instruction are measured. Without this, these assessment approaches simply add another layer to existing assessment systems and increase the testing burden on students. In the final post in this series, we explore what these findings suggest states, districts, researchers, and funders should prioritize as they continue to reimagine K-12 assessment.