Date Presented 4/21/2018
Internal consistency reliability and item analyses using the Rasch model were examined for the Sensory Processing 3-Dimensions Scale, a performance-based tool for measuring sensory processing in children. The scale shows promise as reliable measure for both research and practice.
Primary Author and Speaker: Shelley Mulligan
Additional Authors and Speakers: Sarah Schoen, Lucy Miller
PURPOSE AND BACKGROUND: This study gathered psychometric data for a new performance-based scale for assessing sensory processing abilities and challenges in children aged 3–13. Internal consistency reliability of the Sensory Processing 3-Dimensions Scale (SP-3D) was examined, and item response theory using the Rasch model was applied to examine item fit and item difficulty to assist in refining the scale.
The SP-3D measures sensory modulation and discrimination and sensory-based motor abilities and is based on the sensory processing disorder (SPD) patterns and subtypes described by Miller et al. (2007). The SP-3D includes Visual, Auditory, Vestibular, Proprioceptive, and Tactile domains for evaluating skills and behaviors associated with sensory discrimination and modulation behaviors, as well as two sensory-based motor domains, Posture and Praxis. The Sensory Integration and Praxis Tests (Ayres, 1989) is one of the only other performance-based assessments available for measuring sensory processing and integration, and there is a significant need for a more updated, comprehensive performance-based instrument for assessing sensory processing in children (Schaaf et al., 2015). The SP-3D is intended to address this need.
METHOD: A convenience sample of 103 participants aged 4–13 (M = 7.5), including 64 boys and 39 girls, were recruited. Forty-one children were identified with SPD and 62 were typically developing. Children with SPD were recruited from clinics where they had been referred for occupational therapy because of sensory processing concerns. Typically developing children were recruited from local communities. Cronbach’s alphas were obtained to examine internal consistency reliability for the sensory discrimination, praxis, and postural subtests. Rasch analysis was conducted with all subtests with multiple items scored dichotomously (pass–fail; n = 18) to examine item fit by considering mean-square infit and outfit values and item difficulty level. Items with mean-square values outside 0.6–1.4 were targeted for further study.
RESULTS: Results revealed that internal consistency reliability for the sensory discrimination subtests is fair to strong, with coefficients ranging from .58 to .96 and domain coefficients ranging from .60 (Proprioception) to .91 (Vestibular). The reliability coefficients for the Praxis and Postural domains and associated subtests were fair to strong, ranging from .67 to .88. Reliability for the atypical sensory modulation ratings suggests fair internal consistency for overresponsivity, underresponsivity, and sensory craving, with Cronbach’s alphas ranging from .52 to .89 (M = .74). Item difficulty values revealed some redundancies and assisted in ordering items from easiest to most difficult. For most subtests, <5% of the items failed to fit the Rasch model, suggesting that subtests represent unified constructs.
CONCLUSION: The SP-3D shows promise as a reliable measure of sensory processing to assist in identifying children with specific patterns and subtypes of SPD. The results suggest that items within subtests and subtests within domains represent unified, well-defined constructs for defining sensory processing abilities and deficits. Furthermore, Rasch analyses were helpful in identifying some redundancies related to item difficulty and in identifying some ill-fitting items to assist in refining the scale. The SP-3D appears to be a valuable tool for guiding occupational therapy intervention planning for children with SPD and for use as an outcome measure to examine the effects of occupational therapy using a sensory integrative approach. Normative data collection and further studies of validity and reliability are needed.
References
Ayres, A. J. (1989). Sensory Integration and Praxis Tests. Torrance, CA: Western Psychological Services.
Miller, L. J., Anzalone, M. E., Lane, S. J., Cermak, S. A., & Osten, E. T. (2007). Concept evolution in sensory integration: A proposed nosology for diagnosis. American Journal of Occupational Therapy, 61, 135–140. https://doi.org/10.5014/ajot.61.2.135
Schaaf, R. C., Burke, J. P., Cohn, E., May-Benson, T. A., Schoen, S. A., Smith Roley, S., . . . & Mailloux, Z. (2015). The Issue Is—State of measurement in occupational therapy using sensory integration. American Journal of Occupational Therapy, 68, e149–e153. https://doi.org/10.5014/ajot.2014.012526
