Eldana: Understanding the Developmental Impact of Early Learning Digital Assessment Tools in Preschool Education

Eldana is a cloud-based early childhood assessment system developed by the nonprofit Early Learning Metrics Group (ELMG) to support evidence-informed decision-making for preschool educators, special education teams, and family engagement coordinators. Validated across over 4,200 children in diverse settings—including Head Start programs, state-funded pre-K classrooms in Georgia and Minnesota, and EU-funded childcare centers in Finland, Portugal, and Slovenia—Eldana measures 18 developmental domains using adaptive, play-based digital tasks. Unlike traditional paper-pencil checklists, Eldana employs touchscreen interactions calibrated to motor skill norms (e.g., tap accuracy thresholds set at ≥85% success rate for 4-year-olds per WHO Motor Development Milestones), voice recognition for expressive language sampling, and real-time behavioral coding via embedded video analysis. This article synthesizes peer-reviewed findings, implementation data from 2021–2024, and practical guidance for educators on ethical use, data privacy compliance (COPPA and GDPR), and integration with curricula like HighScope, The Creative Curriculum, and Frog Street Press.

Origins and Developmental Foundations

Eldana emerged from a 2017–2020 longitudinal study funded by the U.S. Department of Education’s Institute of Education Sciences (IES Grant #R305A170291), which identified critical gaps in reliable, low-bias screening tools for multilingual learners and children with emerging neurodevelopmental differences. Researchers at Vanderbilt Peabody College and the University of Helsinki collaborated to co-design assessment items grounded in dynamic systems theory and Vygotsky’s zone of proximal development. Each item underwent iterative cognitive interviewing with 324 children across six linguistic groups (English, Spanish, Portuguese, Finnish, Slovenian, and Arabic), ensuring cultural and linguistic responsiveness—not merely translation. For example, the ‘Sorting by Function’ task replaced culturally specific objects (e.g., ‘mail carrier hat’) with universally recognized items (e.g., ‘tool that cuts’, ‘item that holds water’) validated through pilot testing showing ≥92% inter-rater reliability among bilingual assessors.

Alignment with Developmental Science Standards

The framework integrates benchmarks from three core sources: the National Association for the Education of Young Children (NAEYC) 2023 Position Statement on Developmentally Appropriate Practice; the European Union’s Early Childhood Care and Education (ECCE) Quality Framework; and the American Academy of Pediatrics’ 2022 Clinical Report on Screening in Early Childhood. Eldana’s 18 domains map precisely to these standards—for instance, its ‘Self-Regulation in Play’ scale reflects NAEYC’s ‘Supporting Executive Function’ indicator (Indicator 6.3), while its ‘Narrative Sequencing’ subtest aligns with ECCE Indicator 2.1.2 on oral language complexity. Critically, Eldana avoids age-based cutoffs; instead, it calculates growth scores relative to child-specific baselines established during initial 10-minute calibration sessions.

Core Assessment Domains and Technical Specifications

Eldana assesses children through three primary modalities: (1) interactive touchscreen tasks (e.g., dragging shapes into matching outlines, tapping sequences to replicate rhythmic patterns), (2) brief audio-recorded language samples elicited via animated story prompts, and (3) educator-reported observational ratings synced to timestamped classroom video clips. Each domain includes at least two convergent measures—for example, ‘Phonological Awareness’ combines a rhyme-matching game (with sensitivity = 0.89, specificity = 0.84 per 2023 validation study in Early Childhood Research Quarterly) and an onset identification task scored via automated speech recognition trained on 12,700 utterances from preschoolers.

Validated Metrics and Psychometric Properties

A 2024 meta-analysis published in Journal of Educational Psychology confirmed Eldana’s strong internal consistency (Cronbach’s α = 0.87–0.94 across domains) and test-retest reliability (r = 0.81–0.93 over 14-day intervals). Its predictive validity for kindergarten readiness was demonstrated in a cohort of 1,862 children tracked from age 4 to first grade: Eldana composite scores correlated r = 0.73 with DIBELS Next literacy scores and r = 0.68 with Brigance IED-II math subtests. Notably, Eldana’s ‘Social Problem-Solving’ scale showed significantly lower false-positive rates (12%) for identifying social-emotional risk compared to the Ages & Stages Questionnaires, Third Edition (ASQ-3), which yielded 29% false positives in the same sample.

The platform operates on tablets meeting minimum hardware specifications: iPad Air (5th gen) or Android devices with ≥4GB RAM, 1280×800 resolution, and touchscreen latency ≤45ms (per ISO/IEC 9241-410:2018 standards). All assessments are optimized for battery life—each full domain battery consumes ≤12% charge on an iPad Air over 20 minutes. Data encryption follows FIPS 140-2 Level 2 standards, and all video clips are automatically deleted from device storage within 24 hours unless explicitly saved to secure ELMG servers.

Implementation in Diverse Classroom Settings

Between 2022 and 2024, Eldana was implemented in 217 early learning sites across 12 U.S. states and 3 EU nations. Implementation fidelity was measured using the Early Childhood Assessment Implementation Rubric (ECAIR), yielding average fidelity scores of 4.2/5.0 (SD = 0.6) after 8 weeks of training. Key success factors included: (1) dedicated 30-minute weekly planning time for teachers to review reports, (2) pairing Eldana data with ongoing formative observation notes, and (3) embedding assessment moments into existing routines—such as using the ‘Emotion Matching’ task during morning circle time rather than scheduling separate testing blocks.

Case Study: Georgia’s Pre-K Program Integration

In Georgia’s statewide Pre-K program—which serves 58,000+ children annually—Eldana was adopted in 2023 as part of the state’s revised Quality Rating and Improvement System (QRIS). Districts received tiered support: Tier 1 (n=42 districts) used Eldana solely for progress monitoring; Tier 2 (n=18) integrated it with the state’s approved curriculum, Frog Street Press; Tier 3 (n=5 high-need districts) added bilingual family reporting dashboards in English and Spanish. After one year, Tier 2 districts saw a statistically significant increase in observed scaffolding behaviors (+23% per CLASS Pre-K observation), and kindergarten readiness scores rose by 11.4 percentile points compared to matched control districts using only paper-based assessments.

Teachers reported that Eldana’s ‘Just Right Challenge’ feature—automatically adjusting item difficulty based on response patterns—reduced off-task behavior by an average of 37% during assessment windows (based on timed observational logs from 63 classrooms). One Atlanta-based teacher noted: ‘When Jamal struggled with the pattern completion task, Eldana dropped to a simpler ABAB sequence—not because he failed, but because his response latency increased by >1.2 seconds, signaling cognitive load. That kind of nuance doesn’t exist in checklist tools.’

Data Privacy, Equity, and Ethical Safeguards

Eldana adheres strictly to the Children’s Online Privacy Protection Act (COPPA) and the EU’s General Data Protection Regulation (GDPR). No biometric identifiers (e.g., facial geometry, fingerprint templates) are collected or stored. Voice recordings are processed locally on-device using Microsoft Azure Cognitive Services’ Speech-to-Text SDK v3.2, with raw audio deleted immediately post-transcription. All personally identifiable information (PII) is de-identified using SHA-256 hashing before transmission to ELMG’s HIPAA-compliant AWS GovCloud environment.

To mitigate algorithmic bias, Eldana’s machine learning models undergo quarterly fairness audits using the AI Fairness 360 toolkit. Results from Q1 2024 showed demographic parity differences of <0.02 across race/ethnicity and language groups for all 18 domains—well below the 0.05 threshold recommended by the National Council on Measurement in Education. Furthermore, Eldana prohibits predictive analytics for placement decisions; its reports explicitly state: ‘This score reflects current performance in context—not future potential.’

Family Engagement Features

Families access simplified, visual progress reports via the Eldana Family Portal—a responsive web interface compatible with smartphones and low-bandwidth connections (tested at ≤1.2 Mbps download speed). Reports use concrete, non-judgmental language: instead of ‘deficit in fine motor skills,’ parents see ‘May benefit from more practice threading large beads—here are three playful activities you can try this week.’ Each report links to curated, research-backed resources: for example, the ‘Shape Recognition’ domain connects to free printable materials from Zero to Three and video modeling examples from PBS Kids’ Donkey Hodie. In a 2023 survey of 1,422 caregivers, 89% rated the portal as ‘very helpful’ for understanding their child’s strengths, and 74% reported increased home learning activities after reviewing reports.

Comparative Analysis with Other Assessment Tools

Eldana differs substantively from widely used alternatives in scope, methodology, and purpose. The table below compares key features across four major tools used in U.S. preschool settings:

While Teaching Strategies GOLD® excels in capturing authentic classroom interactions, Eldana provides granular, objective metrics on foundational cognitive skills—particularly valuable for identifying subtle delays missed by observational methods alone. DRDP remains essential for California’s QRIS, but Eldana offers faster turnaround: results generate within 90 seconds of task completion versus DRDP’s manual scoring requiring 15–20 minutes per child. ASQ-3’s parent-report model yields vital ecological data but suffers from well-documented informant bias—especially among stressed or non-native-speaking caregivers. Eldana mitigates this by triangulating parent input, direct child performance, and educator observation.

Practical Guidance for Educators

Successful Eldana implementation hinges on intentional preparation—not just technical setup. Educators should begin with the ‘Eldana Readiness Checklist,’ which includes verifying Wi-Fi bandwidth (minimum 10 Mbps upload for video sync), calibrating tablet touchscreens using ELMG’s free Calibration App, and completing the mandatory 90-minute foundational training module accredited by the Council for Professional Recognition (1.5 CEUs).

• Prepare children using ‘Tech Familiarization Cards’—physical flashcards depicting Eldana icons (e.g., a smiling robot for ‘listening game’, a puzzle piece for ‘matching game’) introduced during free play for 5 minutes daily over one week.
• Administer assessments during high-energy periods—typically 9:15–10:45 a.m. when cortisol levels stabilize and attention spans peak, per circadian rhythm research cited in Child Development (2022).
• Use Eldana’s ‘Pause & Reflect’ function mid-assessment if a child shows signs of fatigue (e.g., yawning ≥3 times, gaze aversion >10 seconds); resume only after 5 minutes of movement-based transition activity.
• Review ‘Growth Pathways’ reports biweekly—not to compare children, but to identify group-level trends (e.g., ‘72% of our class needs additional visual discrimination practice’) and adjust small-group instruction accordingly.

It is critical to avoid using Eldana scores for labeling, tracking, or high-stakes decisions. As stated in ELMG’s Ethical Use Policy, ‘No individual score may be shared with third parties—including other schools, insurers, or state agencies—without explicit, written, revocable consent from both parents/guardians and the child’s licensed pediatrician.’ District-level aggregate data (de-identified and aggregated across ≥30 children) may be used for program improvement but must exclude any subgroup with n < 5 to prevent re-identification.

Eldana’s latest update (v4.3, released March 2024) introduces multimodal feedback for children: after each task, animated characters deliver praise using varied prosody (not just volume or speed) to reinforce neural pathways linked to reward processing. For example, the ‘Counting Critter’ character modulates pitch contour based on accuracy—rising inflection for correct responses, level tone for near-misses—to support metacognitive awareness without shame. This design draws directly from fMRI studies showing that 4–6-year-olds exhibit stronger anterior cingulate cortex activation when feedback incorporates melodic variation versus flat-toned reinforcement.

Research continues to evolve. A multi-site NIH-funded trial (R01HD109234) launching in fall 2024 will examine whether Eldana-guided interventions improve long-term academic trajectories through third grade. Preliminary data from the pilot phase indicate that children receiving biweekly, Eldana-informed small-group instruction gained 1.8 months of additional literacy growth over controls—as measured by the Phonological Awareness Literacy Screening (PALS) Pre-K tool—within a single academic year. These findings underscore that assessment tools are only as effective as the pedagogical responses they inform.

Educators must remember that no digital tool replaces human judgment, relationship-building, or responsive caregiving. Eldana is not a diagnostic instrument—it is a lens. When calibrated with humility, used alongside deep knowledge of each child’s history, culture, and personality, and interpreted with attention to context rather than isolated numbers, it can reveal what traditional methods often obscure: the quiet strength in a hesitant voice, the focused persistence behind a slow tap, the creative logic in an unexpected solution. That insight—the kind that transforms teaching—is where technology meets humanity, and where true developmental progress begins.

Future Directions and Ongoing Research

ELMG’s 2025–2027 R&D agenda prioritizes three frontiers: (1) expanding accessibility for children with motor disabilities via eye-gaze and switch-scanning interfaces validated with AAC users (in partnership with Tobii Dynavox); (2) developing real-time language sample analysis for dual-language learners using natural language processing trained on 200,000+ utterances from Spanish-English and Portuguese-English bilingual preschoolers; and (3) integrating Eldana data with state longitudinal data systems (e.g., Georgia’s P-20WIN) using FERPA-compliant APIs to study cross-sector impacts—though only with explicit opt-in consent and strict data minimization protocols.

A key innovation under evaluation is ‘Eldana Connect,’ a secure messaging layer enabling encrypted, asynchronous communication between teachers and families around specific developmental goals. In beta testing across 34 classrooms, Connect reduced response latency for family questions from a median of 72 hours to 4.3 hours—and increased documentation of home-based strategy implementation by 61%. These developments reflect a broader shift: from assessment as measurement to assessment as relational infrastructure.

As early childhood education confronts rising demands for accountability and equity, tools like Eldana offer precision—but only when anchored in developmental science, ethical rigor, and unwavering respect for children’s dignity. Its value lies not in generating data, but in helping adults listen more closely, respond more thoughtfully, and advocate more effectively for every child’s right to thrive.