Data & Methodology
The Research Behind Pitch Rise
Our approach is grounded in two foundations: authoritative data on the adult literacy crisis (PIAAC) and decades of learning science research. This page provides deep detail on both.
Understanding PIAAC
The Programme for the International Assessment of Adult Competencies is the gold standard for measuring adult skills globally.
What is PIAAC?
The Programme for the International Assessment of Adult Competencies (PIAAC) is the most comprehensive international study of adult skills. Coordinated by the Organisation for Economic Co-operation and Development (OECD), it measures literacy, numeracy, and problem-solving in technology-rich environments among adults aged 16-65. PIAAC is administered in over 40 countries, enabling cross-national comparisons of adult skill levels. It uses sophisticated adaptive testing methodology, adjusting question difficulty based on responses to provide precise measurements across the skill spectrum. In the United States, PIAAC is administered by the National Center for Education Statistics (NCES), a division of the U.S. Department of Education.
Why PIAAC Matters
PIAAC provides the most reliable data on adult foundational skills available. Unlike self-reported surveys, it directly measures what adults can actually do through standardized assessments. The 2023 results were particularly alarming, showing declines rather than the expected stability or improvement. The US dropped from middle-of-the-pack among developed nations to below average in both literacy and numeracy. This data informs policy decisions about adult education funding, workforce development programs, and educational priorities. It also validates what many educators have long suspected: foundational skill gaps are widespread and growing.
How Skills Are Measured
PIAAC uses a 0-500 scale divided into proficiency levels: Below Level 1 (0-175): Very limited skills Level 1 (176-225): Basic functional skills Level 2 (226-275): Moderate skills, can handle routine tasks Level 3 (276-325): Good skills, can handle complex tasks Level 4 (326-375): Advanced skills Level 5 (376-500): Expert skills The threshold for "functional" literacy/numeracy—skills needed for full participation in modern society—is generally considered Level 3 (score of 276+). Disturbingly, only 40% of US adults reach Level 3 or above in literacy, and only 35% in numeracy.
Literacy Proficiency Levels
Distribution of US adults across PIAAC literacy proficiency levels (2023)
| Level | 2017 | 2023 | Change | Description |
|---|---|---|---|---|
| Below Level 1 | 4% | 8% | +4% | Can read brief texts on familiar topics, locate a single piece of specific information identical in form to information in the question or directive. |
| Level 1 | 15% | 20% | +5% | Can read relatively short digital or print texts to locate a single piece of information. Limited vocabulary knowledge and understanding of sentence structure. |
| Level 2 | 33% | 32% | -1% | Can make matches between text and information, paraphrase, make low-level inferences. Navigate within digital texts to access and identify information. |
| Level 3 | 35% | 30% | -5% | Can understand and respond appropriately to dense or lengthy texts. Construct meaning across larger chunks of text, perform multi-step operations. |
| Level 4/5 | 13% | 10% | -3% | Can integrate, synthesize, and evaluate information from multiple sources. Engage with abstract ideas and apply sophisticated reasoning. |
Key Finding: The percentage of US adults at the lowest levels (Below Level 1 and Level 1) increased from 19% to 28% between 2017 and 2023—a 47% increase in the proportion of adults struggling with basic literacy.
Numeracy Proficiency Levels
Distribution of US adults across PIAAC numeracy proficiency levels (2023)
| Level | 2017 | 2023 | Change | Description |
|---|---|---|---|---|
| Below Level 1 | 9% | 12% | +3% | Can perform basic arithmetic like addition and subtraction with whole numbers. Count, sort, and compare concrete quantities. |
| Level 1 | 20% | 22% | +2% | Can complete one-step or simple processes involving counting, sorting, basic arithmetic. Understand simple percents like 50%. |
| Level 2 | 32% | 31% | -1% | Can identify and act on mathematical information in common contexts. Apply two-step processes involving calculation with whole numbers and common decimals. |
| Level 3 | 27% | 25% | -2% | Can complete multi-step calculations using order of operations. Analyze data, understand proportions, interpret data in graphs and charts. |
| Level 4/5 | 12% | 10% | -2% | Can understand arguments based on complex mathematical information. Analyze relationships in complex situations using reasoning and insight. |
Key Finding: The percentage of US adults at the lowest levels increased from 29% to 34%— meaning more than one-third of American adults now struggle with basic math concepts like percentages, measurements, and simple calculations.
Learning Science Foundations
Pitch Rise is built on proven research from cognitive psychology and learning sciences. Here's the evidence behind our methodology.
The Forgetting Curve & Spaced Repetition
Hermann Ebbinghaus (1885) — Memory: A Contribution to Experimental Psychology
The Research
Ebbinghaus discovered that memory retention decays exponentially over time—the "forgetting curve." Without review, we forget approximately 70% of new information within 24 hours and up to 90% within a week. However, each review "resets" the curve and makes subsequent forgetting slower. This led to the development of spaced repetition: reviewing material at gradually increasing intervals for optimal retention. Modern research confirms that spaced practice leads to retention rates of 80-90% compared to 20-30% with massed practice (cramming). The spacing effect is one of the most robust findings in learning science.
How We Apply It
Our system tracks when each skill was last practiced and schedules reviews at optimal intervals. Skills showing decay are automatically prioritized. This ensures long-term retention rather than temporary memorization.
💡 Key Insight: Space your learning over time. Review before you forget completely, but after some forgetting has occurred.
Zone of Proximal Development (ZPD)
Lev Vygotsky (1978) — Mind in Society: The Development of Higher Psychological Processes
The Research
Vygotsky identified that learning happens most effectively in the "Zone of Proximal Development"—the gap between what a learner can do independently and what they can do with guidance. Tasks too easy provide no learning opportunity. Tasks too hard lead to frustration and disengagement. The ZPD is the sweet spot where challenge meets capability with appropriate support. This concept revolutionized education by shifting focus from what students already know to what they're ready to learn next. It also emphasized the role of scaffolding—temporary support that's gradually removed as competence grows.
How We Apply It
Our adaptive algorithm continuously calibrates to each learner's ZPD. After every answer, we update our model of what that learner can do and select the next problem to maintain optimal challenge—approximately 70-85% success rate.
💡 Key Insight: The right difficulty level is crucial. Too easy = boredom. Too hard = frustration. The ZPD = growth.
Mastery Learning & The 2 Sigma Problem
Benjamin Bloom (1984) — The 2 Sigma Problem: The Search for Methods of Group Instruction as Effective as One-to-One Tutoring
The Research
Bloom's research demonstrated that students receiving one-on-one tutoring performed two standard deviations (2 sigma) better than students in conventional classrooms—meaning the average tutored student outperformed 98% of classroom students. This "2 Sigma Problem" challenged educators to find scalable methods approaching tutoring's effectiveness. Bloom also showed that mastery-based learning—where students must demonstrate competence before advancing—produces significantly better outcomes than time-based advancement. Traditional education advances students by age, not competence, leading to accumulated "educational debt" as gaps compound. Mastery learning prevents this by ensuring solid foundations before building higher.
How We Apply It
Our AI attempts to solve the 2 Sigma Problem by providing personalized, 1-on-1 tutoring at scale. Students must demonstrate mastery before progressing. The system identifies and addresses knowledge gaps rather than advancing past them.
💡 Key Insight: Mastery, not time. Ensure competence before advancing. Fill gaps instead of building on shaky foundations.
The Impact of Feedback
John Hattie (2009) — Visible Learning: A Synthesis of Over 800 Meta-Analyses Relating to Achievement
The Research
Hattie's massive synthesis analyzed over 800 meta-analyses covering 50,000+ studies and millions of students. He calculated "effect sizes" for various educational interventions, with 0.40 considered the threshold for meaningful impact. Feedback emerged as one of the most powerful interventions, with an effect size of 0.73—nearly double the threshold. However, not all feedback is equal: • Immediate feedback outperforms delayed feedback • Specific feedback outperforms generic praise • Feedback on the task (what's wrong and why) outperforms feedback on the person • Feedback that guides next steps is most effective The "feedback loop" of attempt → feedback → adjustment → attempt is central to effective learning.
How We Apply It
Every problem receives immediate, specific feedback. When answers are incorrect, the AI explains what went wrong and why. Learners never wonder if they got something right—they know instantly and can adjust immediately.
💡 Key Insight: Immediate, specific feedback is one of the highest-impact interventions in education.
Growth Mindset
Carol Dweck (2006) — Mindset: The New Psychology of Success
The Research
Dweck's research distinguished between two mindsets about intelligence and ability: Fixed Mindset: Belief that abilities are innate and unchangeable. Leads to avoiding challenges (might reveal limitations), giving up easily, ignoring feedback, and feeling threatened by others' success. Growth Mindset: Belief that abilities can be developed through effort and practice. Leads to embracing challenges, persisting through setbacks, learning from criticism, and being inspired by others' success. Crucially, mindset can be influenced by how we frame learning. Praising effort ("You worked hard on that") rather than ability ("You're so smart") promotes growth mindset. Viewing mistakes as learning opportunities rather than failures does the same. Students with growth mindsets show greater resilience, higher achievement, and more enjoyment in learning.
How We Apply It
Our gamification system celebrates effort, progress, and improvement—not just correct answers. XP is earned for trying. Streaks reward consistency. Messaging emphasizes that skills grow with practice. Mistakes are framed as part of learning, not failure.
💡 Key Insight: Mindset matters. Believing you can improve leads to actually improving.
Desirable Difficulties & Interleaving
Robert Bjork (1994) — Memory and Metamemory Considerations in the Training of Human Beings
The Research
Bjork coined "desirable difficulties"—learning conditions that make initial performance harder but enhance long-term retention and transfer. These include: Spacing: Distributing practice over time (vs. massing/cramming) Interleaving: Mixing different topics/problem types (vs. blocking by type) Testing: Retrieving information rather than re-reading Variation: Practicing in varying conditions Interleaving is particularly counterintuitive. Students feel like they're learning less because performance during practice is lower, but retention and transfer are significantly better. The challenge is that blocked practice feels easier and more productive, leading learners to incorrectly believe it's more effective.
How We Apply It
We interleave problem types rather than presenting all problems of one type together. Math, reading, and other subjects are mixed within sessions. This feels harder but produces deeper learning and better transfer.
💡 Key Insight: Make learning appropriately difficult. Easy practice produces poor retention.
References & Further Reading
PIAAC Data Sources
- • National Center for Education Statistics. (2024). PIAAC Results 2023. nces.ed.gov/surveys/piaac/
- • OECD. (2024). Survey of Adult Skills (PIAAC). oecd.org/skills/piaac/
Learning Science References
- • Ebbinghaus, H. (1885). Memory: A Contribution to Experimental Psychology. Teachers College, Columbia University.
- • Vygotsky, L.S. (1978). Mind in Society: The Development of Higher Psychological Processes. Harvard University Press.
- • Bloom, B.S. (1984). The 2 Sigma Problem: The Search for Methods of Group Instruction as Effective as One-to-One Tutoring. Educational Researcher, 13(6), 4-16.
- • Hattie, J. (2009). Visible Learning: A Synthesis of Over 800 Meta-Analyses Relating to Achievement. Routledge.
- • Dweck, C.S. (2006). Mindset: The New Psychology of Success. Random House.
- • Bjork, R.A. (1994). Memory and Metamemory Considerations in the Training of Human Beings. In Metcalfe, J. & Shimamura, A. (Eds.), Metacognition: Knowing About Knowing. MIT Press.
- • Roediger, H.L. & Karpicke, J.D. (2006). The Power of Testing Memory: Basic Research and Implications for Educational Practice. Perspectives on Psychological Science, 1(3), 181-210.
- • Kornell, N. & Bjork, R.A. (2008). Learning Concepts and Categories: Is Spacing the “Enemy of Induction”? Psychological Science, 19(6), 585-592.
Research-Backed Learning for Everyone
Now you understand the science behind Pitch Rise. Experience it yourself.