Bayley scores are standardized so that 100 is the average, with most children falling between 85 and 115. A score of 100 means a child performed right at the middle compared to same-age peers in the test’s reference group, and each 15 points above or below represents one standard deviation from that average. But those numbers carry more nuance than a single composite figure suggests, and the version of the test used, the child’s gestational history, and even the cultural context of the assessment can shift what any given score actually tells you.
What the Bayley Actually Measures
The Bayley Scales of Infant and Toddler Development assess children from ages 1 to 42 months across five broad areas: cognitive, language, motor, social-emotional, and adaptive behavior. The first three are assessed directly by a trained examiner who presents the child with tasks and objects. The last two come from a caregiver questionnaire rather than direct observation of the child.
Within each directly assessed domain, the examiner scores individual items and converts those raw scores into scaled scores for subtests. Language, for instance, breaks into receptive communication (what the child understands) and expressive communication (what the child can say or gesture). Motor splits into fine motor (grasping, stacking) and gross motor (sitting, walking). Those subtest scaled scores are then combined into composite scores for each domain, which is the number most parents encounter on the report.
Reading the Composite Score
The composite score is the headline number, and it sits on a scale where 100 is average and the standard deviation is 15. The Bayley-III Technical Manual assigns qualitative labels to score ranges to help clinicians and families interpret them:
- Very Superior: 130 and above
- Superior: 120 to 129
- High Average: 110 to 119
- Average: 90 to 109
- Low Average: 80 to 89
- Borderline: 70 to 79
- Extremely Low: 69 and below
Alongside the composite, reports typically include a percentile rank. A child with a composite score of 95 falls at roughly the 37th percentile, meaning the child scored at or above 37 percent of same-age children in the normative sample.1Bayley-III Clinical Use and Interpretation. Bayley Scales of Infant Development – Section: Percentile Ranks, Composite Scores, and Confidence Intervals That can feel underwhelming to a parent who hears “below average,” but 95 sits squarely within the Average classification. The difference between the 37th and 50th percentile is small in practical terms.
Reports also include confidence intervals, usually at the 90% or 95% level. A composite of 95 might come with a confidence interval of 90 to 100, meaning the child’s true ability likely falls somewhere in that band. Children rarely perform identically on two separate test days, and the confidence interval captures that natural wobble. Parents who fixate on a single number miss the point; the range matters more than the midpoint.
Why the Version of the Test Matters Enormously
One of the most consequential things to know about Bayley scores is that the version used can shift a child’s number by a dramatic margin. The Bayley-III, published in 2006, has been widely used for nearly two decades. But evidence has accumulated that it tends to produce scores that are higher than a child’s actual developmental level, particularly in high-risk populations. The newer Bayley-4 was designed to correct for this.
A recent study comparing the two versions in very preterm children at two years of corrected age found that cognitive scores on the Bayley-4 were on average about 10 points lower than on the Bayley-III. Language scores dropped by about 6 points, and motor scores by about 4. The practical consequence was stark: roughly 13% of children scored more than two standard deviations below the mean on the Bayley-4 cognitive scale, compared to only 4% on the Bayley-III. That means the Bayley-III was missing a substantial fraction of children with meaningful cognitive delay.2PubMed. The Bayley-4 Versus the Bayley-III in Very Preterm Children at 24 Months’ Corrected Age
If your child was assessed with the Bayley-III, a composite of 90 may not mean quite the same thing as a 90 on the Bayley-4. This is not a flaw unique to the Bayley; it is a known phenomenon in standardized testing called the Flynn effect and its reverse, where norms drift over time and newer editions reset the baseline. The takeaway for parents: always note which version was used, and understand that scores from different editions are not directly comparable.
Prematurity and Corrected Age
For children born preterm, how the examiner handles age correction has a measurable effect on the score. Standard practice is to use “corrected age,” which subtracts the weeks of prematurity from the child’s calendar age, so that an infant born two months early and tested at 14 months of calendar age would be compared to the norms for a 12-month-old. This adjustment gives preterm children credit for the development time they missed in the womb.
The timing of the assessment within the corrected-age window also matters. A large study of extremely preterm infants found that those assessed at 21 to 24 months corrected age scored an average of 3 to 4 composite points lower than those assessed at 18 to 20 months, and they were more likely to be classified as delayed. About 20% assessed in the later window met criteria for significant delay, compared to roughly 13% in the earlier window.3PubMed. Corrected Age at Bayley Assessment and Developmental Delay in Extreme Preterms This suggests the assessment window is not just a scheduling detail; it actively shapes the results.
There is also an ongoing debate about whether full age correction is appropriate for every domain. One study of preterm infants found that correction for gestational age made sense for the cognitive domain but that motor scores were better interpreted using the child’s actual calendar age.4PubMed. Age correction in cognitive, linguistic, and motor domains for infants born preterm In practice, most clinical settings apply the same correction across all domains, which means motor scores in preterm children may look more favorable than they should.
What Bayley Scores Can and Cannot Predict
Parents naturally want to know whether a score at age two tells them anything about their child’s abilities at school age and beyond. The honest answer is: somewhat, and it depends on the age at testing and on how low the score is.
For preterm children tested at age two, one study found strong correlations between Bayley-III cognitive and language scores and IQ measured at age four, with about 81% of children retaining the same broad developmental classification across that two-year gap.5PubMed. Predictive validity of the Bayley, Third Edition at 2 years for intelligence quotient at 4 years in preterm infants That is a reasonably strong signal, particularly for children who score consistently in the average or above-average range.
But the picture is messier at the low end, which is exactly where families worry most. In extremely preterm children, Bayley-III cognitive scores accounted for about 38% of the variance in later IQ. Using a cutoff score of 85 to flag potential problems, the test correctly identified only 44% of children who later had IQ scores below 85, while producing a relatively low false-positive rate of 7%.6PubMed. The ability of Bayley-III scores to predict later intelligence in children born extremely preterm In other words, a low Bayley score is more reliably informative than a normal one: a score in the average range is reassuring but does not guarantee that development will stay on track.
Age at testing also shapes predictive power. A longitudinal study in Colombia found that below 19 months, Bayley-III scores had very weak correlations with later IQ. Between 19 and 30 months the correlations improved but were still modest, and they only reached moderate strength for assessments done after 30 months.7PLOS ONE. Predictive validity in middle childhood of short tests of early childhood development used in large scale studies compared to the Bayley-III, the Family Care Indicators, height-for-age, and stunting So a Bayley score obtained at 12 months is much less stable than one obtained at 30 months, and clinical decisions based on very early assessments should be held loosely.
Instability of Delay Classifications in the Early Months
One of the most underappreciated realities of infant developmental testing is that young children change rapidly and unevenly. A child who appears delayed at six months may look perfectly typical at 18 months, or vice versa. A study tracking both low-risk and high-risk infants across seven assessments between 3 and 24 months found that Bayley-III delay classifications were highly unstable over time, with low sensitivity and poor ability to predict which children would still qualify as delayed later on.8PubMed Central. Instability of delay classification and determination of early intervention eligibility in the first two years of life The authors concluded that clinicians should weigh risk factors and clinical judgment at least as heavily as the test score when deciding about early intervention eligibility.
This does not mean early Bayley scores are useless. They can flag children who need closer monitoring and provide a structured snapshot of where a child stands at one moment. But they are snapshots, not prophecies, and a single score should never be treated as a verdict about a child’s future.
How Environment and Socioeconomic Factors Shape Scores
Bayley scores do not measure a child’s innate capacity in isolation. They reflect how development has unfolded in a particular environment up to the moment of testing. Research consistently shows that household income, parental education, and the quality of the home learning environment all influence where a child lands on the scale.
A Brazilian study found that children from the lowest socioeconomic bracket had the highest rates of suspected delay in receptive communication, and that parental unemployment was independently associated with lower cognitive and communication scores.9PubMed Central. The effect of poverty on developmental screening scores among infants A separate cohort study modeled the interaction between socioeconomic status, maternal IQ, and the home environment, finding that a stimulating home environment that promoted the child’s autonomy was a major mediator of cognitive and language scores, accounting for roughly a third of the total socioeconomic effect on those domains.10PLOS ONE. The Complex Interaction between Home Environment, Socioeconomic Status, Maternal IQ and Early Child Neurocognitive Development
What this means practically: a child growing up in a resource-poor household who scores an 85 is not necessarily less capable than a child from a wealthier home who scores a 100. The score reflects the interaction between the child and their environment. Interventions that enrich the home learning environment, such as early literacy programs and parent coaching, can shift these scores and, more importantly, shift the developmental trajectory underlying them.
Cultural Context and Norm Differences
The Bayley was normed on a sample of 1,700 children in the United States, and those American norms are the default reference for scoring worldwide.11ScienceDirect. Bayley Scales of Infant Development – Section: Screening and Assessment Tools That creates a problem: developmental milestones are influenced by cultural practices, caregiving routines, and even which skills are prioritized in a given community. A child in Nairobi and a child in New York may both be developing normally, but the test may score them differently because it was built around one set of cultural expectations.
When researchers compared Bayley-III scores using Flemish versus American norms on the same group of typically developing children, they found significant differences across almost every domain.12PubMed. Validity and cross-cultural differences of the Bayley Scales of Infant and Toddler Development, Third Edition in typically developing infants A cultural adaptation study in Kenya found that even carefully modified versions of the test produced scores that were difficult to interpret against U.S. norms, because changes made to reduce cultural bias (like removing pronoun-specific language items that did not translate) affected the scoring in ways that no longer reflected the child’s true ability.13PubMed Central. Cultural Adaptation of the Bayley Scales of Infant and Toddler Development, 3rd Edition for use in Kenyan Children Aged 18–36 Months
For families outside the U.S. or from cultural backgrounds underrepresented in the normative sample, these differences are not academic. A child scored against inappropriate norms may be flagged as delayed when their development is perfectly on track, or may be missed when they genuinely need support. Researchers have repeatedly called for population-specific reference values rather than blanket use of U.S. norms.
What Happens During Testing Affects the Score
Because the Bayley is administered live, in a room, with an unfamiliar adult asking a toddler to perform tasks, the child’s temperament and mood on test day are part of the equation. Research dating back to the original Bayley Scales found that infant temperament shares a small but meaningful amount of variance with test performance, and that this relationship grows stronger as children get older. More active babies tended to score higher, while a negative relationship emerged between smiling and laughter and the mental development index at 8 and 12 months, possibly because more socially engaged infants spent more time interacting with the examiner rather than attending to task demands.14Infant Behavior and Development. Infant temperament and performance on the Bayley Scales of Infant Development at 4, 8, and 12 months of age
A tired child, a hungry child, or a child terrified of strangers will almost certainly underperform their true ability. Experienced examiners know this and will try to reschedule or note behavioral observations in the report. Parents should feel empowered to mention if their child did not seem like themselves during the session, and to ask whether the examiner’s behavioral observations suggest the scores might underestimate the child’s abilities.
Examiner Training and Scoring Consistency
The reliability of any standardized test depends partly on whether different examiners arrive at the same score for the same child. Across multiple studies and cultural settings, the Bayley has shown strong inter-rater reliability when examiners are properly trained. A Persian-language validation found test-retest correlations above 0.98 across all domains, and inter-rater correlations above 0.99.15PubMed Central. A Psychometric Study of the Bayley Scales of Infant and Toddler Development in Persian Language Children A Nepalese study found inter-rater agreement ranging from 0.95 to 0.99.16PubMed Central. Acceptability and Reliability of the Bayley Scales of Infant and Toddler Development-III Among Children in Bhaktapur, Nepal
A study of the newer Bayley-4 found that high scoring consistency held across a large team of 18 clinicians from different professional backgrounds, most of whom had never administered a Bayley before, as long as they completed accredited mastery training, supervised practice, and ongoing quality checks.17Pediatric Research. Inter-rater reliability and agreement of the Bayley-4 in a multidisciplinary team The upshot is that the test itself is mechanically reliable when administered correctly. The variability that concerns families is less about measurement error and more about the contextual factors discussed above.
Accommodations for Children with Physical or Sensory Impairments
Standard Bayley administration assumes a child can see the test materials and manipulate objects with their hands. Children with significant motor impairments, such as cerebral palsy, or with visual impairments may be unable to complete tasks even if they cognitively understand what is being asked. The Bayley-III introduced a Low Motor/Vision accommodated version designed to reduce this kind of impairment bias without changing what the test fundamentally measures.18PubMed. Accommodating the Bayley-III for motor and/or visual impairment: a comparative pilot study
However, accommodations have their own interpretive complexity. A study of children with multiple disabilities resulting from congenital Zika syndrome found a strong negative correlation between the number of accommodations used and the resulting cognitive score: the more accommodations needed, the lower the score, because children requiring many accommodations tended to have more severe impairments overall.19PubMed Central. Accommodations provided and used to assess the cognitive performance of children with multiple disabilities resulting from Congenital Zika Syndrome This means accommodated scores should be read with the understanding that the accommodation partially disentangles motor ability from cognitive ability, but cannot fully eliminate the interaction between the two in a child with severe impairments.
The Bayley Versus Parent-Completed Screeners
Many families first encounter developmental assessment through a parent-completed screening tool like the ASQ-3 (Ages and Stages Questionnaire), which pediatricians often hand out at well-child visits. The Bayley is a much longer, clinician-administered test and is considered the more authoritative measure. But how well do the two agree?
A study comparing the ASQ-3 to the Bayley-III in preterm infants found a moderate correlation of 0.63 between the two, with the ASQ-3 catching about 47% of children the Bayley flagged as delayed while correctly identifying 94% of those who were on track.20PubMed Central. ASQ3 and/or the Bayley-III to support clinicians’ decision making That means the ASQ-3 is good at reassuring families whose children are developing normally but misses about half the children who would score low on the Bayley. This is by design: parent-completed screeners are meant to be quick, cheap first-pass tools, not replacements for a full assessment. A normal ASQ score does not rule out problems that a Bayley might catch, and a concerning ASQ score is a reason to pursue formal testing, not a diagnosis.
Research also shows that parent-reported and clinician-observed social-communication skills have only weak associations in infancy, underscoring the value of direct clinical observation alongside caregiver input.21Frontiers in Psychology. Agreement Between Parental Report and Clinician Observation of Infant Developmental Skills
Parental Stress and the Emotional Weight of a Number
Clinicians sometimes underestimate how a Bayley score lands on a family. A number below 85 or a phrase like “borderline” or “low average” can trigger significant anxiety, especially for parents already navigating the stress of prematurity or other medical complexity. Research has found that developmental delays do not directly increase parenting stress in a simple cause-and-effect way; instead, delays tend to reduce children’s interactive engagement behaviors (things like eye contact, shared attention, and social responsiveness), and it is that reduction in engagement that drives parental stress.22PubMed Central. Parenting stress and interactive engagement behaviors in children with developmental delay
This finding has a practical implication for how scores should be communicated. When a clinician shares results, framing the score as one data point in an ongoing process, and pointing toward specific intervention strategies rather than just a classification label, helps parents channel their energy productively. A “low average” motor score, for example, might lead to a referral for physical therapy that meaningfully changes the child’s trajectory. The score is useful precisely because it can open a door to services, not because it defines a child’s potential.
When the Bayley Is Used to Detect Cerebral Palsy
Some clinicians use low motor scores on the Bayley as part of the diagnostic workup for cerebral palsy, though the test was not specifically designed for this purpose. An early study found that the Bayley Motor Scale at four months corrected age was less than half as sensitive as a dedicated movement assessment tool at detecting early signs of cerebral palsy, though it produced fewer false positives.23PubMed. Early detection of cerebral palsy: sensitivity and specificity of two motor assessment tools A later study using the Bayley-II found that when a motor score below 85 was used as the cutoff, sensitivity for cerebral palsy was reasonably high at 0.88, but specificity dropped to 0.41, meaning many children without cerebral palsy also scored below 85.24Annals of Rehabilitation Medicine. The Usefulness of Bayley Scale of Infant Development II in the Diagnosis of Cerebral Palsy
The Bayley is a developmental assessment, not a diagnostic tool for specific neurological conditions. A low motor score should prompt further specialized evaluation rather than being treated as a diagnosis in itself. Parents who receive a low motor score and worry about cerebral palsy should know that additional assessments, neuroimaging, and clinical observation are all part of a proper diagnostic pathway.