Stuart Willett · Sioux Falls School Board Candidate · June 2, 2026
"This hypothesis was developed in the 70s and 80s as a second language acquisition theory, not a K–12 structured literacy curriculum. Current academic research describes this as conceptually flawed, empirically outdated, and practically insufficient. According to Frontiers in Psychology 2025, the science of reading, which is what UFly is built on, requires phonics and explicit instruction. Krashen moves in the opposite direction. And I just have to say homeschooling one family and running 38 schools serving 24,000 students including 3,000 ELL students and 4,400 IEP students should not focus on something that's from the 70s and 80s."
I appreciate DawnMarie raising this directly at the forum. This is exactly the kind of policy debate voters deserve. Her critique is not without merit in a narrow sense — but it fundamentally misunderstands what I am proposing, and what the research actually shows. Let me explain.
Krashen's approach does not emphasize phonics as the primary tool for teaching children to decode words. Phonics — the direct, systematic instruction of letter-sound relationships — is what the science of reading literature supports for early reading acquisition, and the research is unambiguous. Mississippi proved it. Starting in 2013, Mississippi retrained over 19,000 teachers in phonics-based instruction, required third graders to demonstrate reading proficiency before advancing to fourth grade, and climbed from 49th in the nation to 9th in fourth grade reading.1 That is a remarkable, documented result.
So Dawn is right that phonics works for beginning readers. South Dakota has recognized this as well, adopting new Science of Reading standards in April 2025 and investing $6 million in teacher retraining, with an additional $54 million federal grant to help local districts implement the approach.2 That is good policy, and I support it fully.
Where I respectfully disagree is the conclusion she draws from that fact.
Think of it as a two-stage rocket. The first stage gets the rocket off the ground. Without it, nothing moves. But the first stage alone will never reach orbit. You need the second stage to complete the mission.
Phonics is the first stage. It solves the decoding problem — teaching children that letters represent sounds, that sounds make words, and that words can be sounded out. This is the work of kindergarten through third grade.
Krashen is the second stage. Once children can decode, the challenge becomes vocabulary, fluency, comprehension, and the love of reading. That is where extensive, leveled reading takes over. This is the work of fourth grade and beyond.
Mississippi got it right: a child who cannot decode by the end of third grade is not ready for fourth grade.3 My proposal does not challenge that. It begins where that gate ends.
But here is what the research also shows: phonics instruction alone does not build reading comprehension. A landmark analysis of National Reading Panel data found that after first grade, intensive phonics training produces substantial gains in decoding ability but only modest gains in comprehension — an effect size of just .12 on comprehension measures compared to .49 and .52 on decoding measures.4 Children learn to sound out words — but they still struggle to understand increasingly complex texts, absorb academic vocabulary, and read independently with confidence.
| Stage | Grades | Problem Solved | Method | Research Basis |
|---|---|---|---|---|
| Stage One | K – 3 | Decoding words from print | Phonics / Science of Reading | National Reading Panel; Mississippi LBPA; SD HB 1022 |
| Stage Two | 4 – 6+ | Vocabulary, fluency & comprehension | Krashen's Input Hypothesis / Free Voluntary Reading | Krashen (1988–2011); Chall & Jacobs (1983, 2003); FVR replication studies |
South Dakota has invested in Stage One. My pilot proposal addresses Stage Two — precisely where the district's own data shows students falling behind.
Researchers Jeanne Chall and Vicki Jacobs identified this phenomenon decades ago, and it has been consistently confirmed since.5 Students who read adequately in third grade — including students who have been successfully taught to decode — often experience a sudden drop in reading performance in fourth grade. The reason is not that they forgot how to sound out words. The reason is that fourth grade is where reading shifts from learning to read to reading to learn.
The texts become more complex. The vocabulary becomes more abstract. Background knowledge becomes essential. Research shows the first and biggest gap to appear is in word meaning — students fall behind in comprehending abstract, academic, and literary vocabulary — and by seventh grade they can be more than two years behind grade norms.6
Phonics, which was exactly the right tool for the first three years, is simply not designed to address these challenges. This is not a criticism of phonics. It is a recognition that every tool has its purpose — and the right tool for Stage One is the wrong tool for Stage Two.
Krashen's research demonstrates that when students read material that is engaging and at an accessible but slightly challenging level — what he calls "comprehensible input" — they naturally acquire vocabulary, develop fluency, build background knowledge, and grow as readers in ways that direct instruction cannot replicate at scale.7 Free voluntary reading, structured around tiered, level-appropriate books, has been shown to produce significant gains in vocabulary acquisition, reading comprehension, spelling, and writing that persist over time.8
It is also worth noting: Krashen himself is not an opponent of phonics for beginning readers. He argues that once decoding is in place, extensive reading becomes the primary engine of literacy growth.9 That is exactly the sequence I am proposing. These two approaches do not contradict each other. They complete each other.
This argument, applied consistently, would disqualify most of what we know about reading. Jeanne Chall's landmark work on reading stages — which underpins the entire "learning to read / reading to learn" framework, including the rationale for Mississippi's third-grade gate — was published in 1983.10 The National Reading Panel report, the foundational document for the science of reading movement, was published in 2000. Good research is validated — or invalidated — by replication and evidence, not by age.
Krashen's core findings on free voluntary reading have been replicated across multiple languages, age groups, and settings over four decades.11 The question is not when the research was conducted. The question is whether it holds up — and it does.
I take the point seriously, and I will answer it directly. I did not homeschool "one family" as a casual experiment. I applied Krashen's Input Hypothesis to two children who arrived in America speaking no English at the start of sixth grade. By January — nine months later — they were reading Harry Potter in English. Both eventually graduated as valedictorians. I cited this experience not as a substitute for district-wide data, but as personal proof that the underlying theory works at the individual level when applied with discipline and intention.
The district-wide application is exactly what my pilot proposal is designed to test — carefully, at one school, in one semester, with measurable outcomes.12 That is what a responsible innovator does: not impose a theory on 24,000 students, but propose a structured, evidence-grounded pilot with clear success metrics. If it works, we expand it. If it does not, we learn from it.
That is how Mississippi got from 49th to 9th — not by defending the status quo, but by being willing to try something proven and measure the results. The scale of the district is an argument for a careful pilot — not an argument against ever trying something new.