The Reliability and Validity of SARD: A Pilot Study in UAE Primary-Grade Native Arabic Readers

Early identification of students at risk for reading difficulties is essential for successful effective intervention, yet validated reliable screening tools for assessment of reading in Arabic are scarce.Therefore, we developed the Smart Arabic Reading Diagnostic (SARD) tool. SARD is a web-based assessment that measures both accuracy and response time across 16 tasks measuring multiplecomponents of reading. This pilot study aims to prove SARD’s reliability, convergent and divergent validity. We administered SARD to 354 native Arabic speaking students in Grades 2–6 from United Arab Emirates schools. We computed grade specific means and standard deviations, reliability indices, partial correlations, and component analysis. Cronbach’s α for the seven decoding tasks was 0.91, and item–total correlations ranged from 0.76 to 0.84. Principal component analysis showed a dominant factor explaining 67% of variance, and hierarchical clustering separated letter level from word level tasks. We conclude that SARD is a reliable tool for Arabic phonics skills assessment with excellent internal consistency showing convergent and divergent validity. Future studies will explore SARD’s test–pretest reliability, criterion validity, and expand population to include different countries.