86% of the Students We Assessed Had Holes in Their Phonics Knowledge. Here’s What That Means for Your Classroom.

When you assess 1,000 students on every foundational reading skill, the patterns that emerge are hard to ignore.

Here’s a scenario most reading teachers know well: a student is making reasonable progress, participating in class, maybe even reading grade-level texts with some support. Then they hit a wall. A new pattern trips them up and suddenly the progress stalls. Reteaching doesn’t seem to help. The root cause isn’t obvious.

We’ve seen this pattern play out with hundreds of students. And this year, when we analyzed assessment data from 1,000 K–8 students across six Indianapolis schools, the data told us exactly why it happens—and what to do about it.

The Foundation Has More Holes Than You Think

We assessed every student on all 105 lessons in the UFLI Foundations curriculum—not just grade-level skills, but everything from basic letter sounds through advanced multisyllabic patterns. When we ran the numbers, 86% of students showed what researchers call a “swiss cheese” pattern: gaps in earlier skills they had technically progressed past.

That’s not a small finding. It means that for the overwhelming majority of students, knowledge of foundational reading skills isn’t a clean, continuous progression. It’s patchwork. Students have learned to compensate—using context clues, word recognition, and memory—in ways that mask the gaps underneath.

Among students who had progressed past Lesson 34, the most common lingering gaps were CVC-e patterns (41% still failing), basic digraphs like ‘sh’ and ‘ch’ (31%), and short vowel CVC words (23%). These aren’t complex skills. They’re the foundation. And for a third of students, those foundational bricks are missing.

One Gap Becomes Five

The swiss cheese problem gets sharper when you look at how gaps compound. We analyzed conditional probabilities across all 105 skills—if a student fails Skill A, what’s the probability they also fail Skill B?

The chains are long. Students who fail short vowel encoding almost universally fail CVC-e patterns. That failure predicts failure at r-controlled vowels. Which predicts failure at vowel teams. Which predicts failure at multisyllabic patterns. Each link in the chain is a student who may look fine in class—until the next phonics cliff arrives.

We found multiple failure chains where a single foundational gap predicted failure in four or more downstream skills with greater than 90% probability. That’s not a coincidence—it’s the architecture of how reading works. Skills build on each other. When the base is cracked, the structure above it is unstable.

The Grade 3 Number That Surprised Us Most

If we had to pick one data point that gave us pause, it would be this: 39% of 3rd graders had gaps in basic letter sounds—UFLI Lessons 1–4. That’s almost double the rate for 2nd graders (21%). And it’s higher than 4th grade (23%).

Why? In most schools, the intensity of foundational reading instruction drops in Grade 3 as the curriculum shifts toward comprehension. Students who had gaps in K–2 but kept up through memorization and context clues suddenly find those strategies aren’t enough. And because foundational skill assessment often stops after 2nd grade, the gaps aren’t caught until the wheels come off.

The same dynamic—in a less dramatic form—shows up in middle school. About 1 in 7 middle school students in our sample still had basic letter-sound gaps after 6–8 years of schooling. Without explicit assessment, those gaps are nearly invisible in a regular classroom.

What “Group by the Gap” Actually Looks Like

One of the most practical things this data tells us is that traditional reading groups—organized by grade level or Lexile score—are the wrong tool for foundational skills instruction. Two students at the same reading level can have completely different phonics profiles. Grouping them together and teaching toward the same gap is inefficient for both.

The more effective approach is to group by the gap: cluster students by their placement (the first skill they haven’t mastered) and by the pattern of gaps in their profile. Instruction then targets exactly what each group needs. Students aren’t sitting through lessons on skills they’ve already mastered, and they’re not being asked to consolidate advanced patterns when the foundation is still shaky.

The Takeaway

Data-driven instruction isn’t about looking at test scores and adjusting general priorities. It’s about knowing each student’s specific profile, grouping by that profile, sequencing instruction from the earliest gap, and reassessing regularly to see what’s changed.

That loop—assess, group by gap, sequence from the root, reassess and regroup—is replicable. The findings in this analysis aren’t unique to our students or our sites. The patterns of swiss cheese gaps, failure cascades, and the Grade 3 wall are features of how reading development works, not quirks of one particular school.

You don’t need a partner to start. But if you’re interested in how TILT implements this framework in Indianapolis classrooms, the full analysis is available at theindylearningteam.org.

Download Full White Paper

Now Open 2026-2027 TILT Adira Reads Implementation Grants. Find out more or contact sappel@theindylearningteam.org