Early literacy intervention: what works for K-2 struggling readers

The cost of waiting on early literacy intervention compounds across a child’s entire trajectory. A first grader six months behind in decoding is a fixable problem. The same child in fourth grade — now with two years of failed text exposure, eroded reading identity, and a vocabulary gap that grew because they stopped reading enough to build one — is much harder, and the school will spend many times more to address it.

The research on this is unusually unambiguous for an education topic. Early and targeted beats delayed and comprehensive. The intervention that would have worked in kindergarten still works in fourth grade — it just works less, costs more, and runs into a student who has already learned that reading is a thing other kids do.

This article is a practical map of K-2 intervention done well: when to start, what to screen with, which evidence-based programs the field trusts at Tier 2 and Tier 3, how to structure groups, how often to check progress, and when to escalate.

When intervention should start

The answer is kindergarten, not third grade.

The traditional “wait to fail” model — flag struggling readers in second or third grade after they’ve fallen behind — is the model the research has argued against for thirty years. By the time a third grader is identified as struggling, the cognitive cost of catching up has grown, the student’s reading identity has already taken hits, and the window when foundational skills are most teachable has narrowed.

The current standard is universal screening starting in kindergarten, administered three times per year (fall, winter, spring). Students below grade-level benchmarks are flagged for diagnostic assessment and, where indicated, Tier 2 intervention — often within the first month of the school year. Several state literacy laws now require K-3 screening on this cadence (Florida, Mississippi, Ohio, Texas, Tennessee, North Carolina, and others have varying versions), accelerating adoption nationally.

Early identification plus early intervention has replaced wait-to-fail as the field’s default. Buildings that haven’t operationalized that shift are working off an outdated playbook.

What the evidence says about intervention timing

Across decades of National Reading Panel, IES, and What Works Clearinghouse syntheses, one finding shows up repeatedly: the same intervention produces substantially larger effects when delivered in kindergarten or first grade than when delivered in third or fourth grade. Specific effect-size estimates vary across studies and meta-analyses, but the directional finding is consistent — early delivery materially outperforms late delivery on the same content.

Three mechanisms are usually cited. First, foundational decoding skills are most malleable in the early years, when phonological processing is still developing rapidly — older students still benefit, but the slope tends to be shallower. Second, the Matthew effect — the tendency for early reading advantages to compound and early disadvantages to widen — means a year of inadequate reading instruction in first grade is not “made up” later. The student who couldn’t decode in first grade read less text, encountered fewer vocabulary words, and built less background knowledge than peers. By third grade, the decoding gap is real but it’s no longer the only gap. Third, late-elementary students still struggling with decoding have usually developed compensatory habits — guessing from context, skipping words, avoiding reading — that intervention has to undo before new instruction can land.

The practical implication is the entire premise of K-2 intervention: catch the gap early enough that you’re teaching skills, not also undoing years of avoidance.

Universal screening: the entry point to MTSS

Universal screening is the data layer that drives the whole intervention system. Without it, intervention decisions are based on teacher referral, which is unevenly calibrated and tends to lag the data by a year or more.

The widely used screeners in US K-2 buildings:

• DIBELS 8th Edition (University of Oregon). The most established CBM system for early literacy. Subtests for letter-name fluency, phoneme segmentation, nonsense-word fluency, and oral reading fluency map to grade and time-of-year benchmarks.
• Acadience Reading (formerly DIBELS Next). A close cousin of DIBELS with overlapping authorship and similar structure; many districts use one or the other.
• mCLASS (Amplify). A digital screening and progress-monitoring platform built around DIBELS 8 subtests, with classroom dashboards layered on top.
• FAST / FastBridge (Illuminate Education). A widely adopted screening suite with earlyReading and CBMreading components for K-2.
• i-Ready Diagnostic (Curriculum Associates). An adaptive computer-based assessment used as both screener and diagnostic; broader diagnostic depth, but it doesn’t replace a fluency-based CBM for progress monitoring.

The screener doesn’t decide whether a child needs intervention — it flags the candidate pool. The diagnostic follow-up determines what kind. A child below benchmark on phoneme segmentation needs a different intervention than one below benchmark on nonsense-word fluency. The screener says “look here.” The diagnostic says “this is the skill gap.”

State literacy laws increasingly mandate one of these screeners on a defined schedule. Districts that haven’t standardized on one will need to — both to comply and to make tier-movement decisions defensibly.

Evidence-based Tier 2 programs for K-2

Tier 2 is the supplemental, small-group layer for students whose screening data shows they aren’t responding adequately to core instruction alone. For K-2, the programs most commonly used in well-implemented buildings:

• UFLI Foundations (small-group configuration). The University of Florida Literacy Institute’s structured-literacy program adapted for small-group Tier 2 use. Strong fit for K-2 decoding gaps and one of the most-adopted free curricula in the country.
• Wilson Just Words and Wilson Step-by-Step. Wilson’s intervention programs — Just Words for grades 4-12, Step-by-Step for early elementary — designed for students with persistent decoding gaps.
• SIPPS (Systematic Instruction in Phonological Awareness, Phonics, and Sight Words, Collaborative Classroom). Used at Tier 2 across K-5 with placement at the level matching the student’s current skills.
• Heggerty intervention curriculum. Targeted phonemic-awareness intervention, especially common when the screening data points at the PA strand specifically.
• REWARDS. Multisyllabic-word decoding for grades 4-12 — relevant for older K-2 transitions and for the upper grades these students will move into.

When federal funds (Title I, IDEA, ESSER) are part of procurement, ESSA evidence tiers become a constraint on which programs qualify. ESSA defines four tiers — strong, moderate, promising, and demonstrates a rationale — and districts using federal funds typically have to document the tier of the chosen program. This is a separate concept from MTSS instructional tiers, and the two are easy to conflate. ESSA “Tier 2” means moderate evidence. MTSS “Tier 2” means small-group intervention. Confusing one for the other is one of the easiest ways to fail a federal audit.

Evidence-based Tier 3 programs for K-2

Tier 3 is the most intensive layer — for students whose response to Tier 2 was inadequate. The programs that show up most often:

• Wilson Reading System. Multi-year scope and sequence, certified-teacher delivery, decodable text matched to each step. The de facto Tier 3 standard in many districts.
• IMSE Orton-Gillingham. Comprehensive O-G training and curriculum widely adopted by reading specialists and dyslexia interventionists.
• Lindamood-Bell LiPS (Lindamood Phoneme Sequencing). Explicit phonemic-awareness intervention emphasizing articulatory features of speech sounds; appropriate when phonological processing is the bottleneck.
• Take Flight. The Texas Scottish Rite Hospital program for elementary students with dyslexia; successor to the Dyslexia Training Program.

All four are Orton-Gillingham-aligned: explicit, systematic, cumulative, diagnostic, multi-sensory. They differ in scope-and-sequence detail and certification model, but the underlying instructional logic is the same.

The Tier 3 question is rarely which program is best in the abstract. It’s which one your district can staff, train, and deliver with fidelity. A well-implemented Tier 2 program will beat a poorly implemented Tier 3 program every time.

Group structure: who, how many, how long

The structural parameters that distinguish the tiers are not arbitrary. They reflect what the research on small-group intervention has consistently shown about dose and feedback.

Tier 2

• Group size: 3-5 students with similar skill profiles
• Frequency: 3-5 days per week, in addition to Tier 1 core instruction
• Session length: 20-30 minutes
• Cycle length: 8-12 weeks before formal team review
• Delivered by: classroom teacher (during protected intervention block), reading specialist, Title I interventionist, or trained instructional aide

Tier 3

• Group size: 1-3 students (often 1-on-1)
• Frequency: 5 days per week (some districts double-dose with two daily sessions)
• Session length: 30-60 minutes
• Cycle length: 8-12 weeks with weekly progress monitoring
• Delivered by: reading specialist, dyslexia specialist, or special-education teacher with structured-literacy training

Two principles hold across both tiers. First, group size has to be small enough that every student gets enough teacher feedback per session to actually accelerate. A “small group” of eight is functionally a class. Second, intervention time has to be protected on the master schedule. Tier 2 that gets cancelled for assemblies isn’t Tier 2 — it’s a study group with optimistic naming.

Progress monitoring cadence

The data layer is what makes intervention a system rather than a guess.

At Tier 2, the standard is curriculum-based measurement weekly or bi-weekly, graphed against an expected growth slope. DIBELS, Acadience, AIMSWeb, and FAST all support this pattern. The teacher sets a target slope from the student’s starting point and grade-level benchmark, takes a weekly probe, and watches whether the trend line is on, above, or below target.

At Tier 3, monitoring is typically weekly with a shorter data review cycle. Students stuck on a flat trajectory at Tier 3 are the ones the team needs to look at most carefully — that pattern often signals a deeper issue.

Two failure modes recur. The first is collecting probes without using them. Weekly DIBELS data that sits in a binder isn’t progress monitoring, it’s documentation. The probes have to drive instructional adjustment and tier-movement decisions, or they’re administrative theater. The second is tier movement that ignores the data. A student who has been at Tier 2 for two full cycles without adequate response should not be at Tier 2 in a third cycle. Either the intervention is wrong, the dose is wrong, or the student needs the next tier. Letting students stagnate in a tier they aren’t responding to is the most common way well-intentioned MTSS systems fail their hardest cases.

Decision rules and when to escalate

Tier movement isn’t a vibe — it’s a documented decision rule that a building writes once and applies consistently.

A defensible decision rule looks something like this:

• Entry to Tier 2. Screener score below the season benchmark, confirmed by a brief diagnostic to identify the specific skill gap.
• Continuation at Tier 2. Progress monitoring shows the student is closing the gap or on track to reach benchmark by the next screening window.
• Move to Tier 3. After one or two 8-12 week Tier 2 cycles, progress monitoring shows the student isn’t on track to reach benchmark even with the supplemental instruction.
• Special-education evaluation. Intensive, evidence-based Tier 3 intervention delivered with fidelity has not produced adequate response.

The IDEA 2004 regulations permit states to use response-to-intervention data as part of identifying specific learning disabilities. Many districts treat inadequate response to Tier 3 intervention — delivered with fidelity, documented through progress monitoring — as one of the strongest indicators that a full special-education evaluation is warranted. The pattern of inadequate response, combined with a comprehensive evaluation, supports an SLD eligibility decision in many jurisdictions.

Tier 3 and special education are not synonymous. A student can receive Tier 3 entirely in general education; a special-education student may receive services at any tier intensity. The overlap is real but not automatic. What is consistent is that inadequate Tier 3 response is the most common trigger for SPED evaluation, and a building that lets students sit in Tier 3 for years without escalation is denying those students timely access to the legal protections an IEP provides.

Where Storytime fits

Storytime is built to sit alongside whichever Tier 1, Tier 2, and Tier 3 programs a building has adopted. It’s the practice and progress-monitoring layer around the structured-literacy core — not a replacement for the small-group curriculum itself.

• Skill Tree analytics flags Tier 2 candidates. The classroom view groups students by mastery distribution across the six SoR pillars, so the at-risk cluster on phonemic awareness or decoding surfaces without manual roster review.
• Per-student journey overrides. Teachers and specialists can build a targeted small-group or 1-on-1 journey focused on the pattern the screener surfaced — CVC, blends, R-controlled, multisyllabic — without disrupting Tier 1.
• Decodable library matched to curriculum lesson. The book inventory mirrors UFLI, Wilson, IMSE, Amplify CKLA, and LMW scope-and-sequences, so practice text uses only patterns the student has been taught.
• On-demand generation for targeted patterns. When the specialist teaches a new pattern, the system generates fresh decodable text at that exact pattern for concentrated independent practice.
• ORF scoring as the weekly probe. Built-in oral reading fluency assessment with automatic WCPM scoring provides the CBM data many districts use as the progress-monitoring metric.
• Recovery actions for stuck students. If a student isn’t responding, teachers can reset the item, repeat the lesson, or release to the next lesson directly from the classroom management panel.

Storytime doesn’t replace UFLI, Wilson, IMSE, SIPPS, REWARDS, or any other intervention curriculum. It makes practice tighter, data faster, and tier-movement decisions easier to defend.

Bottom line

Early literacy intervention works when the building runs it as a system, not a program. Universal screening three times a year flags candidates. Diagnostic assessment names the specific skill gap. An evidence-based Tier 2 program delivers small-group instruction at the right dose. Weekly progress monitoring drives tier-movement decisions. Tier 3 escalates intensity for students who didn’t respond. Inadequate Tier 3 response triggers SPED evaluation.

The buildings that do this well don’t have a secret. They started in kindergarten, screened universally, picked one or two programs and trained staff on them, protected intervention time on the master schedule, wrote the decision rules down, and actually used the data. None of those steps are expensive. All of them are easier in K and 1 than they will ever be again.

Wait until third grade and the same instruction works less, costs more, and runs into a student who has already learned to avoid the activity it’s trying to teach. Start in kindergarten and most of the problem solves itself.