Examining the Validity and Reliability of Evaluation in Ayres Sensory Integration Data Collected With Autistic Children: A Rasch Analysis.
Patricia Grady-Dominguez, Anita C Bundy, Zoe Mailloux and 1 others
The EASI sensory integration test battery is largely usable for autistic children.
Fourteen of its 19 tests produced consistent scores. Five (AL, PF, PJP, TPL, TPO) did not, so use those with other assessment data rather than in isolation.
Key findings
1Fourteen of 19 EASI tests met the study's thresholds for good or excellent reliability (PSR ≥ 0.70 and strata ≥ 2.0), and 16 of 19 met item-fit expectations. The authors state these support confidence that scores reflect the intended sensory integrative construct.
2Five tests (AL, PF, PJP, TPL, TPO) fell below the threshold for good reliability on both PSR and strata. The authors note that scores from these may not consistently distinguish among children with different ability levels and recommend corroborating them with clinical observations, parent report, or other standardized assessments.
3The PF test (Proprioception: Force) showed the most problems: two item clusters (crayon force-matching and bottle-rolling) appeared to measure different constructs, two items failed to fit the model, and it had the lowest reliability of any test (PSR 0.43, strata 1.48). Only 39 children completed this test, and the authors call these findings preliminary pending replication with a larger sample.
Still to come
How it was doneWhat they foundWhat it means for OTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
79% of the data (115 of 146 children) came from a single clinic in Australia, so the findings may not generalize to other settings or countries. Only 2 children were aged 3 and none were aged 4, so the results say little about the youngest children in the EASI's intended range. No data were collected on cognitive ability, adaptive functioning, or autism severity, all of which could influence performance on sensory and motor tests. The PF subtest was completed by only 39 children (compared with 92 to 145 for other tests), so its reliability and unidimensionality findings are less stable. The researchers did not formally test whether items functioned equivalently across the three countries or in different languages. Reasons for discontinuing tests or skipping items were not recorded, so it is unclear whether missing data were random or related to the child's difficulty with the task.
Declared interests
The study was funded by the Eagles Autism Foundation. Two of the authors (Dr. Mailloux and Dr. Schaaf) receive monetary compensation from the Collaborative for Leadership in Ayres Sensory Integration (CLASI) and from Thomas Jefferson University, both of which provide training in Ayres Sensory Integration. Dr. Mailloux also receives royalties for materials she created, and Dr. Schaaf is the director of research at CLASI. The EASI is the assessment tool used in ASI practice, so the authors have a financial and professional stake in the tool performing well.
The easy way to misread this
Do not read the overall finding as 'the EASI is fully validated for autistic children and every subtest can be used in isolation.' Five of the 19 subtests (AL, PF, PJP, TPL, TPO) fell below the study's own reliability threshold, and the authors explicitly recommend supporting scores from those with other assessment data. The PF subtest in particular may be measuring two different things rather than one, and its findings rest on only 39 children.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →