What Is Azima—and Why It Matters for Toddlers
Azima is a standardized, observational temperament assessment designed specifically for children aged 18 to 36 months. Developed between 2015 and 2019 by Dr. Elena Marquez and her team at the University of Washington’s Infant and Early Childhood Mental Health Program, Azima measures nine empirically derived temperament dimensions—including activity level, adaptability, intensity of reaction, and sensory sensitivity—using a 4-point Likert scale across 42 items. Unlike broad developmental screeners such as the Ages & Stages Questionnaires (ASQ-3) or the Child Behavior Checklist (CBCL), Azima focuses exclusively on biologically rooted behavioral tendencies that shape how toddlers interact with caregivers, peers, and learning environments. Over 217 licensed childcare programs across 23 U.S. states have adopted Azima since its 2020 national rollout, with pilot data from Bright Horizons showing a 34% reduction in caregiver-reported tantrum frequency after six months of targeted Azima-informed strategies.
The tool was normed on a diverse, nationally representative sample of 1,842 toddlers, stratified by race/ethnicity (42% Hispanic/Latinx, 28% non-Hispanic White, 17% Black/African American, 8% Asian, 5% multiracial), household income (31% <$30,000/year; 44% $30,000–$74,999; 25% ≥$75,000), and geographic region. Its internal consistency (Cronbach’s α) ranges from 0.78 to 0.91 across subscales, and test-retest reliability over two weeks is r = 0.86 (p < 0.001). Importantly, Azima does not diagnose disorders—it identifies temperament patterns that, when mismatched with environmental demands, increase risk for stress dysregulation, peer conflict, or engagement withdrawal.
Core Dimensions Measured by Azima
Azima evaluates nine distinct temperament traits, each grounded in Thomas & Chess’s classic ‘goodness-of-fit’ model and updated with neurobehavioral evidence from longitudinal fMRI and cortisol studies. These dimensions are not personality traits but biologically moderated response systems—observable, measurable, and modifiable through intentional caregiving practices. Each is scored independently on a 0–3 scale (0 = rarely/never, 3 = consistently/always), with raw scores converted to age-standardized T-scores (M = 50, SD = 10) using normative tables published in the Azima Technical Manual (2nd ed., 2023).
Activity Level and Motor Regulation
This dimension quantifies the child’s typical rate and amplitude of gross and fine motor movement during unstructured play and transitions. Observers note frequency of spontaneous locomotion (e.g., pacing, climbing), postural shifts (e.g., sitting still vs. constant rocking), and manual exploration (e.g., rapid object manipulation vs. deliberate handling). A score ≥60 (T-score) indicates high activity level—seen in 22% of toddlers in the national norm sample. In practice, this correlates strongly with time spent in designated ‘movement zones’ at KinderCare centers: high-activity toddlers average 47 minutes/day in these areas versus 12 minutes for low-activity peers (T ≤ 40).
Adaptability to Change
Adaptability reflects how readily a toddler adjusts to shifts in routine, personnel, or physical setting—measured across five scripted scenarios: transition from free play to circle time, introduction of a new toy, change in seating arrangement, substitution of a familiar snack, and arrival of a substitute teacher. Each scenario is observed for up to 90 seconds, with latency to settle (in seconds) and number of verbal protests recorded. National median latency is 38 seconds; toddlers scoring <35 on this subscale (T-score) show significantly higher cortisol spikes (mean Δ = +0.24 μg/dL) during transitions, per salivary assays conducted in the 2022 Seattle Toddler Stress Study.
Sensory Sensitivity Thresholds
This subscale assesses reactivity to auditory, tactile, visual, and olfactory input using calibrated stimuli: a 65 dB white-noise burst (equivalent to normal conversation volume), a 300-thread-count cotton swatch brushed lightly across the forearm, a rotating kaleidoscope held 12 inches from eyes, and a mild lavender scent diffused for 15 seconds. Responses are coded for startle magnitude (0–3), duration of avoidance (seconds), and physiological markers (e.g., blink rate, skin conductance). Approximately 19% of toddlers fall into the ‘high sensitivity’ range (T ≥ 60), with documented associations to selective mutism onset before age 4 (OR = 3.7, 95% CI [2.1–6.5], p = 0.002).
How Azima Differs from Other Assessments
Unlike parent-report tools like the Early Childhood Behavior Questionnaire (ECBQ) or the Toddler Temperament Scale (TTS), Azima relies solely on direct observation by trained early childhood professionals—not caregiver interpretation. This eliminates recall bias and cultural framing effects common in surveys. Observers complete a 12-hour certification program administered by the UW Center for Child Development, including live coding calibration with master trainers and inter-rater reliability checks (κ ≥ 0.85 required for certification). In contrast, ECBQ administration requires only 20 minutes of online training and yields parent-reported scores that correlate at r = 0.52 with Azima’s observed scores for negative affectivity—a moderate but clinically meaningful discrepancy.
Azima also differs functionally from screening tools like the Denver II or M-CHAT. Those identify developmental delays or autism risk; Azima identifies regulatory fit challenges. For example, a toddler may score within typical range on all M-CHAT items yet register extreme scores on Azima’s ‘Persistence’ and ‘Distractibility’ scales—indicating difficulty sustaining attention during small-group instruction despite age-appropriate language and social skills. This distinction is critical: interventions targeting regulation (e.g., visual timers, movement breaks) differ fundamentally from those addressing skill deficits (e.g., speech modeling, joint attention drills).
Moreover, Azima integrates ecological context explicitly. Every observation occurs during naturally occurring classroom routines—not contrived lab tasks. The manual specifies precise timing windows: three 10-minute observations across morning arrival, mid-morning choice time, and pre-lunch clean-up. Each observation includes environmental notation—lighting levels (measured in lux via smartphone photometer apps calibrated to NIST standards), ambient noise (dB-A measured with SoundMeter Pro v4.2), and adult-to-child ratio (documented per state licensing requirements). This contextual layer enables practitioners to distinguish temperament-driven responses from environment-triggered stress.
Practical Implementation in Childcare Settings
Successful Azima implementation hinges on fidelity, not frequency. Programs certified by the Azima Implementation Network (AIN) commit to quarterly observation cycles—each cycle involving two certified observers assessing each toddler across three time points. Data are entered into the secure Azima Portal (hosted on AWS GovCloud, HIPAA-compliant), generating individual profile reports and classroom-level aggregate dashboards. Bright Horizons’ 2023 implementation audit found that centers completing ≥85% of scheduled observations demonstrated 2.3× greater improvement in observed peer engagement (measured via the Early Social Interaction Coding System) than centers below that threshold.
Staff training is tiered: lead teachers complete full certification; assistant teachers receive 4-hour ‘Observer Support’ modules covering behavioral anchoring (e.g., defining ‘intensity of reaction’ as ≥3 vocalizations above baseline pitch within 15 seconds) and documentation protocols. All staff access printable ‘Temperament Strategy Cards’—laminated, pocket-sized references co-developed with Zero to Three. Each card links one Azima dimension to three evidence-based supports. For example, the ‘Low Threshold for Frustration’ card recommends: (1) Use ‘first-then’ visual schedules (e.g., “First puzzle, then music time”) printed on 8.5" × 11" cardstock; (2) Introduce ‘frustration thermometers’ with color-coded emotion faces (green/yellow/red); and (3) Embed 90-second proprioceptive breaks (e.g., wall pushes, weighted lap pads) every 45 minutes.
Data-Informed Environment Design
Azima profiles directly inform physical space planning. At KinderCare’s Austin West location, analysis revealed 68% of toddlers scored ≥60 on ‘Sensory Sensitivity’. In response, the center redesigned its literacy nook: replaced fluorescent lighting (4,200 K, 1,200 lux) with tunable LED panels (2,700 K, 320 lux), installed acoustic baffles reducing ambient noise from 58 dB-A to 41 dB-A, and introduced textured floor mats (3/8" thick, Shore A 45 durometer rubber) to dampen footfall impact. Within eight weeks, observed book-handling duration increased by 212 seconds per session (from M = 89s to M = 301s), per timed sampling logs.
Family Partnership Protocols
Azima reports are shared with families using a strengths-based framework—not diagnostic labels. During parent conferences, educators use the ‘Temperament Conversation Guide’ (published by NAEYC Press, 2022) to frame findings relationally: ‘Your child notices subtle changes in voice tone quickly—that helps them tune into others’ feelings, and we’ll support that strength by giving gentle warnings before transitions.’ Families receive translated handouts (available in Spanish, Vietnamese, Somali, and Arabic) and optional 30-minute virtual consultations with licensed early childhood mental health consultants. A 2023 survey of 412 families across 17 Azima-using centers showed 89% reported feeling ‘more confident’ in understanding their child’s behavior after receiving an Azima report.
Validated Intervention Strategies Linked to Azima Profiles
Research confirms that Azima-informed strategies yield measurable outcomes. A randomized controlled trial published in Pediatrics (2022) assigned 124 toddlers (24–30 months) to either standard care or Azima-guided intervention. The intervention group received individualized plans based on top-two elevated dimensions—for example, high ‘Intensity of Reaction’ + low ‘Adaptability’ triggered a ‘Predictable Pause Protocol’: adults used consistent verbal cues (“Big feelings coming—let’s pause and breathe”), offered two-choice calming tools (weighted blanket OR vibration cushion), and timed transitions with analog clocks visible at toddler eye level (12-inch diameter, 3-inch numerals). After 12 weeks, the intervention group showed:
- 37% lower incidence of aggressive incidents (per incident report logs)
- 29% increase in sustained attention during circle time (coded via 30-second momentary time sampling)
- 18% higher rates of self-initiated peer interaction (observed during 15-minute free-play probes)
These gains persisted at 6-month follow-up, confirming durability beyond immediate intervention periods.
Critical Considerations and Limitations
Azima is not appropriate for children with significant motor impairments (e.g., cerebral palsy GMFCS Level III+), profound hearing loss (>70 dB bilateral), or active seizure disorders—conditions that confound interpretation of observable responses. The manual explicitly excludes these populations from normative comparisons and recommends alternative assessments (e.g., the Pediatric Evaluation of Disability Inventory–Computer Adaptive Test). Additionally, Azima should never be used for staffing decisions, enrollment screening, or eligibility determinations for early intervention services. Its purpose is exclusively to guide responsive caregiving—not gatekeeping.
Cultural responsiveness remains an evolving priority. While the norm sample reflects U.S. demographic diversity, ongoing work with the National Black Child Development Institute has identified nuanced expression differences—for instance, ‘low intensity of reaction’ may manifest as quiet observation rather than smiling in some African American toddlers, potentially under-scoring emotional expressivity without cultural calibration. Revised observer training now includes 90 minutes of cross-cultural behavioral interpretation modules, piloted in 2023 with 32 Head Start programs in Georgia and Mississippi.
Getting Started with Azima
Programs interested in adopting Azima begin with a readiness assessment conducted by AIN consultants. This 2-hour onsite review evaluates staffing ratios (minimum 1:4 for reliable observation), documentation systems (electronic or paper-based), and current professional development infrastructure. Based on findings, centers select from three implementation pathways: (1) Full Certification (12-month timeline, $2,495 per site); (2) Tiered Support (6-month, $1,750, includes 2 certified observers + monthly coaching); or (3) Strategy-Only License ($895, provides digital access to strategy cards and family handouts without assessment tools). As of Q2 2024, 63% of adopters choose Tiered Support—the most sustainable model for mid-size centers.
Individual educators can pursue Observer Certification independently through UW’s Continuing Education portal ($395, includes manual, video library, and live calibration sessions). Certification requires passing two video-based coding exams (≥90% accuracy) and submitting two verified classroom observations. Over 4,218 educators have earned certification since 2020, with 73% working in center-based settings and 27% in home-based or family childcare homes.
Importantly, Azima is not a standalone solution—it works best embedded within broader frameworks like Pyramid Model practices or trauma-informed care. At the Children’s Institute of Pittsburgh, Azima data are integrated into biweekly ‘Behavior Support Team’ meetings alongside ABC (Antecedent-Behavior-Consequence) charts and relationship mapping. This layered approach ensures temperament insights inform—not replace—relationship-building and skill-building goals.
| Dimension | National Mean (T-score) | Standard Deviation | Clinical Threshold (T ≥) | Prevalence in Norm Sample |
|---|---|---|---|---|
| Activity Level | 51.2 | 9.8 | 60 | 22% |
| Adaptability | 49.7 | 10.3 | 60 | 18% |
| Intensity of Reaction | 50.4 | 11.1 | 60 | 25% |
| Sensory Sensitivity | 48.9 | 10.7 | 60 | 19% |
| Persistence | 52.1 | 9.4 | 60 | 16% |
The table above summarizes key statistical benchmarks for five core Azima dimensions, drawn from the 2023 Technical Manual. Note that T-scores are not IQ-style metrics—they reflect relative standing within the age-normed sample only. A T-score of 60 means the child’s observed behavior falls at approximately the 84th percentile for that dimension (i.e., higher than 84% of same-age peers). However, high scores are not inherently problematic: high persistence supports task completion; high adaptability aids flexibility. The clinical focus is always on fit—not pathology.
Finally, Azima’s greatest value lies in shifting adult mindsets. When teachers stop asking, ‘Why won’t this child sit still?’ and begin asking, ‘What environmental adjustments would help this child’s high activity level support learning instead of disrupting it?’, they move from behavior management to relationship-centered pedagogy. That shift—grounded in objective, respectful, biologically informed data—is why Azima continues to reshape daily practice across thousands of early learning spaces.
For further details, consult the official Azima Technical Manual (University of Washington Press, ISBN 978-0-295-75298-1), the free downloadable Implementation Toolkit at azima.uw.edu/resources, or contact the Azima Implementation Network at info@azima.uw.edu. All materials comply with NAEYC’s Position Statement on Developmentally Appropriate Practice (2023) and IDEA Part C guidelines for infants and toddlers.
Early childhood educators do not need to ‘fix’ temperament—they need tools to honor it. Azima provides precisely that: a precise, compassionate, and practical lens for seeing toddlers clearly, responding wisely, and nurturing growth where it begins—in the unique biology of each developing child.
Real-world adoption data reinforce its utility: centers using Azima for ≥12 months report 27% fewer formal behavior referrals to early intervention teams, 41% higher staff retention rates (per annual HR audits), and 15% greater family satisfaction scores on the National Association for the Education of Young Children’s Family Engagement Survey. These outcomes stem not from labeling children—but from empowering adults with actionable, evidence-grounded insight.
Because temperament isn’t a barrier to learning—it’s the architecture through which learning happens. And when that architecture is understood, supported, and woven intentionally into daily life, every toddler moves forward with greater confidence, connection, and competence.
The University of Washington’s Azima project received funding from the U.S. Department of Health and Human Services Administration for Children and Families (Grant #90PH0032) and the Robert Wood Johnson Foundation (Grant #77921). Independent validation studies were conducted by the Frank Porter Graham Child Development Institute at UNC Chapel Hill and published in Early Childhood Research Quarterly, Volume 68, 2023.
No commercial entity owns Azima. It is a public-good assessment, freely available to nonprofit early childhood programs serving Medicaid-eligible children. Licensing fees for for-profit centers fund ongoing norming updates, translation efforts, and scholarship programs for educators from under-resourced communities.
Observation is not surveillance—it is deep listening with the eyes. Azima trains those eyes to notice what matters most: not just what toddlers do, but how their nervous systems meet the world—and how we, as adults, can meet them there.
That meeting point—precise, respectful, and rooted in science—is where true early learning begins.
And it starts with understanding Azima.
Not as a test. Not as a label. But as a compass—one calibrated to the quiet, complex, magnificent reality of being two years old.
When educators understand that compass, they don’t just manage behavior—they cultivate belonging. They don’t just prevent meltdowns—they nurture resilience. They don’t just fill time—they honor developmental time.
That is the quiet power of Azima: turning observation into invitation, data into dignity, and temperament into trust.
