Research & Efficacy
Scholar Ridge is built on published learning science — and we're honest about what we have proven, what we're measuring now, and what we haven't shown yet.
Where we stand on evidence
The U.S. ESSA framework rates education evidence on four tiers. The top three each require a completed study — a randomized trial (Tier 1), a matched-comparison study (Tier 2), or a controlled correlational study (Tier 3). The fourth, “Demonstrates a Rationale” (Tier 4), is a recognized evidence-based category in its own right: a program built on a well-defined logic model grounded in high-quality research, with an evaluation underway.
Scholar Ridge is built against Tier 4 — by design, not by default. Every core instructional choice is anchored to a specific, published line of research (below), and the pre/post measurement Tier 4 calls for is built into the product and already recording growth for our first families.
What we are not claiming.We are a new program, and our sample is still small — early enough that any percentage we printed today would say more about who happened to enrol than about whether the teaching works. So we are not publishing an effect size, and we are not calling ourselves “proven” or self-certifying a tier; that is a judgement for the people who read the data, not for us. What we will do is publish the number once it means something — and the measurement is already there to publish (see the roadmap below).
Our evidence roadmap
A research-based design is the start, not the finish. Most programs that sit at Tier 4 never show how they'll climb — here is our concrete, fundable path to higher, study-backed evidence, and exactly where we stand on it.
Demonstrates a Rationale
What this tier requires: A logic model grounded in high-quality research, plus an evaluation underway.
Our path: Where we are today: every design choice is anchored to published research, and the pre/post measurement is built into the product and recording growth for our first families. The sample is still small — we will not report a figure until it means something.
Promising Evidence
What this tier requires: A controlled correlational study showing a favorable association between using the program and student outcomes.
Our path: Analyze the pre/post growth data we are already collecting — with statistical controls — and have the analysis reviewed by an independent evaluator before we publish or claim it.
Moderate Evidence
What this tier requires: A quasi-experimental study comparing Scholar Ridge users to a matched group of similar non-users.
Our path: Partner with a school, district, or research group to run a matched-comparison study — the step that lets a district use it under ESSA Title evidence requirements.
Tier 1 (a randomized controlled trial) is the long-term goal once the earlier tiers are established. We will only ever claim a tier we have actually earned.
Our theory of action
The simple chain from what we do to the growth we expect — and measure.
Adaptive placement, mastery lessons with worked examples, a guiding AI tutor, structured phonics, and unlimited practice.
Each piece maps to a proven mechanism: explicit instruction, gradual release, retrieval, spacing, and immediate feedback.
Measurable growth in a child's mastery and placement level over weeks of consistent use.
The research behind the design
Each design choice is tied to a specific, published line of evidence.
Structured phonics (the science of reading)
What we do: K–2 reading is taught with explicit, systematic, cumulative phonics — real recorded speech sounds, blending, and decodable words — not guessing from pictures.
The evidence: Systematic phonics is one of the most strongly evidenced practices in education (National Reading Panel, 2000; What Works Clearinghouse). It is the core of structured literacy.
Gradual release (I do · we do · you do)
What we do: Sage, our AI tutor, models the method first, works a step with the child, then hands it over — and stops one step short so the child finishes it. It guides; it never gives the answer.
The evidence: The gradual release of responsibility model (Pearson & Gallagher, 1983) is a long-established framework for moving a learner from supported to independent work.
Mastery-based progression
What we do: Lessons teach a skill to mastery with worked examples before practice, and a child can keep getting fresh problems until the skill sticks.
The evidence: Mastery learning produces large gains over conventional instruction when feedback and correction are built in (Bloom, 1984).
Retrieval & spaced practice
What we do: Unlimited practice generates a new problem on demand, and skills resurface over time rather than being taught once and dropped.
The evidence: Repeated retrieval and spacing improve long-term retention far more than re-reading (Roediger & Karpicke, 2006).
Adaptivity & immediate feedback
What we do: An adaptive placement check meets each child at their level, and every question gives instant, specific feedback a child can act on.
The evidence: Targeted instruction at the right level and timely, specific feedback are consistently among the highest-impact instructional moves (Hattie & Timperley, 2007).
How we measure outcomes
Scholar Ridge measures growth from real use, not surveys:
- 1. A child takes an adaptive placement check — this is their baseline.
- 2. They learn through mastery lessons and practice over a defined window.
- 3. They take a comparable check again — the re-assessment — and we compute the change.
Families opt in before any of their child's progress is included in research, and all reported figures are anonymized and aggregated. We plan to have the analysis of our first pilot reviewed by an independent third party before publishing.
First pilot — in progress
Our first pre/post outcomes pilot is running now. When it's complete we'll publish the results here — the average growth, the sample size, the method, and the limitations — whatever the data shows. We report what we measure, and only what we measure.
What we're honest about
- • We have not yet run a randomized controlled trial, so we don't claim “proven to raise reading levels.” Our current evidence is a research-based design, with pre/post measurement in place and a pilot at an early stage.
- • That pilot is small. We will say so plainly until it isn't, and we won't publish a growth figure that a handful of children could swing.
- • Lessons are AI-generated and machine-verified for correctness; they are not a substitute for an accredited school or a clinician's evaluation.
- • A pre/post pilot without a control group shows growth, not proof of cause. We label it that way.
References: National Reading Panel (2000); Pearson & Gallagher (1983); Bloom (1984); Roediger & Karpicke (2006); Hattie & Timperley (2007); U.S. ESSA evidence tiers (IES What Works Clearinghouse).