The purpose of this study was to determine which set of selected kinematic variables affected the speed of visually impaired residential youth in the sprint run. The subjects were 27 students, 16 males and 11 females, between the ages of 9.4 and 16.4 years. Film data were collected during two trials of the 50-m dash. A Fortran computer program produced nine variables from these digitized data. A multiple regression analysis was performed on the variables using running speed as the dependent variable. Results of a correlation matrix yielded five variables with significant bivariate correlations to running speed. Results of a regression analysis indicated that the cycle length and hip joint range of motion had significant effects on running speed. Implications for an increase in sprinting speed include increasing stride length via the generation of greater hip extension during the drive phase and a greater hip flexion during the recovery phase of sprint running.