Validation of a computerized version of the MNREAD chart for children.

Abstract

We aimed the develop and validate a computerized version of the MNREAD for the assessment of reading speed for children. A computerized version of the MNREAD was built in Psykinematix software using the same parameters as the physical chart. We measured the reading velocity of 104 children (n = 44 of the 3rd grade; n = 60 of the 5th grade). Bland-Altaman analysis was used to quantify agreement between two chart measurements. Comparison between the 3rd and 5th-grade children for physical and computerized MNREAD versions showed statistical differences in Maximum Reading Speed (F = 2669.6; p < 0.001), Critical Print Size (F = 17.49; p < 0.001), and Reading Acuity (F = 14.19; p = 0.002) with huge effect size (η2 = 0.930). No differences were found between the versions within grades. Bland-Altman analysis showed 95% of the data points within ± 2s of the mean difference, suggesting a similarity between versions of the MNREAD parameters. Our reading speed values for children in the 5th grade were higher than previous studies, suggesting a better quality of the assessment in the computerized version. Tolerance Limits were calculated as normality references for clinical purposes.