Shoelace Learning Whitepaper

How Shoelace Determines and Tracks Reading Comprehension Level (RCL)

A research-grounded, adaptive, and educator-aligned approach to measuring adolescent literacy growth

Executive Summary

Shoelace Learning provides an adaptive reading comprehension engine that identifies each learner’s Reading Comprehension Level (RCL)—a holistic, dynamic indicator of a student’s ability to understand increasingly complex texts. Unlike traditional readability scores that measure decoding difficulty alone, Shoelace’s RCL represents the full comprehension experience, integrating vocabulary, syntax, task complexity, and sub-skill performance into a single, continuously updated measure.

RCL is a proprietary metric, validated across millions of learning moments inside Shoelace’s game-based platform and refined through years of research and educator feedback. It enables schools, teachers, and tutors to:

reliably determine a learner’s approximate reading level,
assign texts at the appropriate challenge level, and
track growth over time with actionable granularity.

This paper provides an overview of how RCL is constructed, updated, and used within Shoelace’s learning engine—ensuring transparency about the research and methodology, while protecting Shoelace’s underlying intellectual property.

1. The Philosophy Behind Shoelace’s Adaptive Reading Model

Shoelace is built on three core educational principles:

1.1 Comprehension Is More Than Decoding

Traditional readability scores (e.g., Flesch-Kincaid, SMOG, Coleman-Liau) estimate how difficult a text is to decode, but do not measure how difficult it is to comprehend. Shoelace uses readability formulas as a baseline, then layers in comprehension complexity through question skills, cognitive load, and inferential difficulty.

1.2 Students Learn Best at the Edge of Their Ability

The learning engine primarily delivers content “at level,” what we call the Goldilocks Zone of just right content, but intentionally mixes in easier items for confidence and harder ones for stretch opportunities—reflecting well-established learning theory.

1.3 Growth Comes From Continuous, Responsive Feedback

RCL is updated only when students complete meaningful reading tasks—longer passages with both literal and inferential questions—ensuring the measure reflects real comprehension, not isolated skill probes.

Together, these design principles ensure RCL is not a static number, but a living, evidence-based profile of a student’s reading comprehension skills.

2. What Is the Reading Comprehension Level (RCL)?

RCL is a numeric scale from 1.0 to 8.9 representing the approximate comprehension level of the student.

The whole number approximates the student’s grade-level comprehension (e.g., 4.x ≈ Grade 4).
The decimal value represents progress within that range with 0.1 being the equivalent to one calendar month (4.6 means the learner is over halfway to the next level).

Crucially:

RCL is not simply Fountas & Pinnell, Lexile, or readability—it is a composite comprehension measure built from:

passage readability,
question skill difficulty, and
real-time student performance on literal and inferential tasks.

This allows Shoelace to detect nuances traditional leveling systems often miss.

3. The Research Foundations of RCL

3.1 Readability Science

Shoelace incorporates readability formulas including:

Flesch-Kincaid
SMOG Index
Coleman-Liau
Automated Readability Index
Gunning Fog
Linsear Write

These establish the minimum RCL for each passage before comprehension complexity is considered.

3.2 Evidence-Based Comprehension Skills

Shoelace questions are built around ~130 sub-skills, spanning:

Literal retrieval
Inferencing
Predicting
Figurative language
Vocabulary and morphology
Structure and genre understanding
Critical reading and reasoning

Skills are organized by Starting Skill Level (SSL) and three tiers of difficulty (Learning → Practicing → Mastery).

3.3 Adaptive Assessment Principles

Shoelace follows adaptive testing best practices:

challenge is dynamically adjusted,
mastery is evaluated based on recent performance only,
sub-skills update separately from overall comprehension,
growth is detected through patterns—not single responses.

This combination ensures Shoelace can accurately detect emerging strengths and hidden misconceptions.

4. How Shoelace Levels Every Passage

Shoelace passages receive an RCL using three converging factors:

4.1 Algorithmic Readability Baseline

Each passage is analyzed for vocabulary difficulty, sentence length, syntax, and text structure. This sets the minimum possible RCL.

4.2 Question Skill Difficulty

If a passage contains questions tied to higher-level comprehension skills, its RCL is raised to match the cognitive demand. A passage’s RCL will never be lower than its readability score, but it may be significantly higher if comprehension demands warrant it.

4.3 Expert Educator Review

Pedagogical experts review all passages to ensure:

alignment with state standards,
age-appropriate themes,
interdisciplinary learning opportunities,
culturally relevant content.

They adjust RCLs when human judgment finds nuance the formulas cannot.

This combination yields highly reliable comprehension levels aligned to real-world expectations for Grade 3–9 learners.

5. How Shoelace Determines a Student’s RCL

Shoelace determines each learner’s RCL through a multi-stage adaptive process:

5.1 Initial Estimate

Students begin with a grade-based estimate and a brief diagnostic to calibrate their starting point.

5.2 Continuous Adaptive Evaluation

RCL updates only when students complete sufficiently robust comprehension tasks (i.e., passages with >30 words and multiple literal and inferential questions).

The learning engine examines:

accuracy
difficulty of the passage
difficulty of the questions
whether performance was above, at, or below level.

5.3 Dynamic RCL Adjustment

The RCL algorithm (proprietary details withheld) applies incremental changes based on meaningful performance patterns:

Incorrect answers on easier texts → possible RCL decrease
Strong accuracy at level → stability or growth
Correct answers above level → accelerated RCL growth

These adjustments occur in small increments (typically 0.1), producing a smooth, trend-based reading level trajectory rather than sudden jumps.

6. How the Learning Engine Uses RCL to Assign Content

After each question or passage, the learning engine chooses the next task using probabilities grounded in educational best practices.

6.1 Choosing Passage Difficulty

Most passages fall within the student’s “instructional zone”—a range that balances challenge with achievability:

75% passages at level
15% below
10% above

6.2 Choosing Skills to Target

When assigning independent skill questions, the engine selects skills:

80% at the learner’s current level
15% below
5% above

6.3 Balancing Literal & Inferential Questions

Question selection is tied to passage length and follows a structured progression from literal → inferential. If a student cannot achieve basic comprehension thresholds, the engine ends the passage early and adjusts RCL accordingly.

This produces a comprehensive, research-aligned feedback loop.

7. Skill-Level Tracking vs. RCL Tracking

Shoelace maintains two parallel systems:

7.1 Skill Scores (130 Sub-Skills)

Represent micro-skills (figurative language, summarizing, character traits, etc.) and update on every question. Scores reflect the last 10 questions for that skill exclusively.

7.2 RCL (Holistic Comprehension)

Represents the global comprehension level across texts and skills. Updates only after completing meaningful passages.

Together, they give educators:

a big-picture reading level (RCL),
and granular diagnostic information (skills).

8. Why Educators Can Trust RCL

8.1 It Aligns With Multiple Curricula and Standards

Expert educators ensure every passage and skill reflects the expectations of North American and global literacy standards (Ontario 2023, TEKS, CCSS, Alberta ELAL, Florida B.E.S.T. The Cambridge curriculum, etc.).

8.2 It Is Built From Millions of Learning Moments

RCL is validated continuously with data collected in classrooms and across large-scale deployments.

8.3 It Uses Adaptive Testing Principles Adopted by Leading Assessments

Methods mirror the philosophy behind adaptive assessments—while being optimized for daily use inside a game-based practice environment.

8.4 It Balances Algorithmic Reliability With Human Expertise

Algorithms identify patterns; expert educators ensure the content and progression remain instructionally sound.

9. What We Do Not Share (IP Protection)

Shoelace’s competitive advantage relies on details that cannot be publicly disclosed, including:

proprietary scoring algorithms,
exact weighting of skill vs. passage difficulty,
formulas governing RCL increases/decreases,
content-selection decision trees,
thresholds for early passage termination,
normalization and scaling methods,
full skill taxonomy and metadata architecture.

What we provide partners is transparency about the logic, research, and outcomes, not exposure of the algorithmic source code.

10. Summary

Shoelace’s Reading Comprehension Level (RCL) is a scientifically grounded, field-tested, and instructionally aligned measure designed to give educators and tutors the clearest possible picture of a learner’s reading comprehension ability.

By blending:

rigorous readability science,
a deep skill taxonomy,
adaptive psychometric principles,
expert educator review, and
millions of real-world data points.

Shoelace delivers a reading engine that gives partners confidence that they are investing in a system proven to accelerate adolescent literacy growth.