As a former principal, I had the privilege of witnessing what’s possible when schools truly commit to empowering multilingual learners (MLLs). At Berry Elementary in the South Bay Union School District, our dedicated team of teachers helped move the percentage of MLLs meeting proficiency from just 15% to over 55% in only three years. That success was not about quick fixes or test prep—it was about creating a culture of belonging, leveraging formative assessment, and honoring the languages and identities our students brought with them each day. I remain deeply grateful to the incredible educators at Berry who proved what’s possible when we center students’ strengths and invest in their growth.

Today, across the nation, schools are grappling with the persistent and growing challenge of Long-Term English Learners (LTELs). These students, after spending six or more years in U.S. schools, have not yet achieved English proficiency—a barrier that limits their access to rigorous coursework, higher education, and full participation in civic life (WIDA, 2019). To reverse this trend, we must shift away from compliance-driven assessments toward holistic practices that affirm identity, honor home languages, and provide timely, actionable feedback.

 

Moving from Testing to Learning with Formative Language Assessment

Formative language assessment is not a one-time event—it is a continuous, embedded process that helps multilingual learners (MLs) give and receive feedback, refine their skills, and set goals. Research highlights its power:

  • Accelerates language acquisition through targeted feedback (Heritage, 2010).
  • Reduces anxiety by focusing on growth, not punishment (Gottlieb, 2016).
  • Empowers students through self- and peer-assessment (WIDA Consortium, 2020).
  • Supports differentiated instruction by giving teachers real-time data (Stiggins, 2005).
  • Improves academic outcomes when paired with content instruction (Black & Wiliam, 1998).

Practical Classroom Examples

  • WIDA “Can Do” Descriptors: Students self-assess using friendly rubrics (“At Level 2, I can share ideas using short phrases. At Level 3, I can explain my thinking in sentences.”).
  • Peer Feedback Circles: After a science presentation, peers use sentence frames like “I understood when you said…” or “A next step could be…”.
  • Performance Tasks: Students write lab reports or retell stories using both English and their home language, showing depth of understanding beyond English-only responses.
  • Progress Journals: Learners track growth with weekly reflection prompts: “One new word I used this week was…” or “A strategy that helped me was…”.

 

Goal Setting and Learner Identity

Goal setting transforms language acquisition from a compliance-driven milestone into a student-owned journey. Research shows students who set specific, challenging goals achieve more and build greater self-efficacy (Shi, 2021).

Practical Goal Setting Strategies

  1. Initial Orientation: Use WIDA proficiency descriptors with visuals to help MLs understand their current level.

Co-
  2. Construction of Success Criteria: Define “what good work looks like” across listening, speaking, reading, and writing.
  3. SMARTIE Goals: Guide students to set measurable goals, e.g., “I will write a science paragraph using hypothesis, data, and conclusion vocabulary.”


  4. Tracking Tools: Use charts, portfolios, or digital logs for students to visualize growth.
  5. Family Integration: Share progress in multiple languages, inviting caregivers into the goal-setting process.

This process fosters learner identity, helping students see themselves as capable, multilingual contributors. Home languages are reframed as assets, not barriers, through translanguaging practices that encourage MLs to use their full linguistic repertoire (Bloomberg, Fairchild, Trower, Mascorro & Wells, 2025).

 

How AI-PowerED Protocols Can Help

At The Core Collaborative, our AI-Powered Protocols extend the power of formative assessment and goal setting by giving teachers and students real-time, actionable supports:

  • Analysis of Evidence (AOE) Protocol: The AI-AOE protocol allows teachers to upload student work samples, which AI then analyzes against WIDA-aligned rubrics. Within minutes, teachers receive strengths-based insights on how each learner is progressing in reading, writing, speaking, and listening. This feedback can be shared directly with students in student-friendly language, so they know exactly what they are doing well and what their next step could be.
    • For teachers: AI synthesizes patterns across the class, showing where scaffolds are needed and which strategies are working.
    • For students: AI generates just-right feedback paired with sentence stems and exemplars that match their WIDA level, turning feedback into immediate action.
    • For differentiation: The protocol helps group students by strengths and needs, ensuring targeted instruction without stigmatizing learners.
  • Scaffolds for Differentiation: AI generates leveled sentence stems, vocabulary banks, and exemplars tailored to each learner’s WIDA level. For example, a Level 1 student might receive “The experiment shows…” while a Level 4 student is prompted with “The evidence suggests that…”.
  • Personalized Goal Suggestions: AI recommends next-step goals aligned with both language growth and content mastery, allowing students to co-create SMARTIE goals with teachers.
  • Equity in Feedback: AI ensures every student receives timely, just-in-time scaffolds—something nearly impossible for a single teacher to manage in a linguistically diverse classroom.

Quick AI-Generated Scaffolds for Reading Literature Standards

RL.2 (Determine the central message, lesson, or moral):

  • Level 2 Stem: “The story is about ____.”
  • Level 3 Frame: “The central message is ____ because the character ____.”
  • Level 4 Frame: “One lesson from this story is ____; the author shows this through ____.”

RL.3 (Describe characters and how their actions contribute to events):

  • Level 2 Stem: “The character is ____.”
  • Level 3 Frame: “The character is ____ and this shows when they ____.”
  • Level 4 Frame: “The character’s choice to ____ affects the story because ____.”

These AI-generated scaffolds give teachers quick, level-appropriate entry points for students to access grade-level standards while growing in English proficiency.

 

Confronting the LTEL Challenge

Embedding formative language assessment + goal setting + AI-powered supports into daily practice is essential to addressing the LTEL challenge. Too often, students remain stuck in linguistic limbo because they lack individualized feedback and opportunities to use language authentically. These practices break the cycle, accelerate language growth, and affirm student identity (Umansky & Reardon, 2014).

 

A Call to Action

This is not just about assessment—it is about justice. Every child deserves to see their culture, language, and identity reflected in the classroom. By combining formative assessment, goal setting, and AI-powered scaffolds, we can reduce the LTEL population, nurture confident multilingual citizens, and ensure every learner has the tools to thrive.

Gind out more about how we can unlock the brilliance of multilingual learners together.

👉 Download our White Paper: Empowering Multilingual Learners

References

  • Black, P., & Wiliam, D. (1998). Assessment and classroom learning. Assessment in Education: Principles, Policy & Practice, 5(1), 7–74.
  • Bloomberg, P. J., & Wells, I. (2025). Empowering multilingual learners: Holistic approaches to advance learner agency. Mimi & Todd Press.
  • Gottlieb, M. (2016). Assessing English language learners: Bridges from language proficiency to academic achievement. Corwin.
  • Heritage, M. (2010). Formative assessment: Making it happen in the classroom. Corwin.
  • Shi, H. (2021). Examining college-level ELLs’ self-efficacy beliefs and goal orientation. Journal of Comparative & International Higher Education, 13(2), 65–82.
  • Stiggins, R. J. (2005). From formative assessment to assessment for learning: A path to success in standards-based schools. Phi Delta Kappan, 87(4), 324–328.
  • Umansky, I. M., & Reardon, S. F. (2014). Reclassification patterns among Latino English learner students. American Educational Research Journal, 51(5), 879–912. https://doi.org/10.3102/0002831214545110
  • WIDA Consortium. (2020). WIDA Can Do Descriptors. University of Wisconsin–Madison, Wisconsin Center for Education Research.
  • WIDA. (2019). Long-term English learners across 15 WIDA states. University of Wisconsin–Madison, Wisconsin Center for Education Research.