How would you structure observation protocols to inform literacy instruction in an EMC classroom?

• A. Observe only test scores.
• B. Observe student engagement, strategy use, accuracy, fluency, independence; collect data on interactive behaviors; use to pivot instruction and supports.
• C. Focus exclusively on compliance.
• D. Ignore data for instruction.

Answer: (B)

When guiding literacy instruction in an EMC classroom, you want observation protocols that capture how learning happens, not just what the final result looks like. This means watching a blend of observable factors: how actively students engage with reading tasks, what strategies they attempt or use spontaneously, their accuracy and fluency during reading, and how independently they work. It also involves noting interactive behaviors—how students talk about texts, ask and answer questions, seek help, and collaborate with peers or respond to feedback.

Why this approach works is that it gives you a real-time, nuanced picture of each learner’s strengths and needs. If a student is engaged but struggles with decoding, you can pivot to targeted mini-lessons or guided practice focusing on that strategy. If another student reads with good fluency but shows limited comprehension, you can adjust supports like explicit modeling of prediction, questioning, and summarizing. Data on independence helps you decide when a student is ready for more challenge or when they still need scaffolds. Observing interactive behaviors lets you see how effectively students are applying strategies with guidance, and whether your feedback is helping them move forward.

With this information, you tailor instruction on the fly—select appropriate texts, adjust grouping, design prompts, and provide specific supports to build independence and strategy use. That formative, data-driven approach stands in contrast to relying solely on test scores, which don’t reveal day-to-day progress or the processes students use to decode and comprehend, and it avoids focusing only on compliance or ignoring data altogether.