Test-retest reliability and comparison of single- and dual-beam photocell timing system with video-based applications to measure linear and change of direction sprint times

Abstract

This study aimed to assess the test-retest reliability of untested single- and dual-beam timing gates and compare them with previously validated video-based applications to measure linear and change of direction sprint (CODS) times. Twenty-three participants were concurrently assessed for 30 m linear sprint and CODS time using single- and dual-beam timing gates and the MySprint and COD Timer applications. Interclass correlation coefficient (ICC), Pearson correlation, independent t-test and Bland-Altman plots were used for comparison between instruments. ICC, Cronbach’s alpha and coefficient of variation (CV) analyses were used to assess the test-retest reliability. Excellent ICC was noted for test-retest reliability (0.982–0.984 [sprint], 0.940–0.942 [CODS]), with a high Cronbach’s alpha (all 0.997 [sprints], 0.988–0.989 [CODS]) and acceptable CV (1.296–1.946%) for all the timing systems. Similarly, excellent ICC (0.989–0.994 [sprint], 0.998–0.999 [CODS]) and very high correlation (r = 0.990–0.994 [sprints] and r = 0.998–1.000 [CODS]) were reported between the single- and dual-beam timing gates, and the MySprint and COD Timer applications, with non-significant differences between the measurements (p = 0.754–0.960). However, the Bland-Altman plots represented that values measured with the three instruments were inconsistent with most values away from the mean of the difference between instruments. In conclusion, both photocell timing systems are reliable instruments for measuring linear sprint time and CODS time. However, the timing systems should not be used interchangeably to interpret findings. Furthermore, it is suggested that similar timing systems with an identical setup should be used for the measurement of timings for interpretations.