Differences and trial-to-trial reliability of vertical jump heights assessed by ultrasonic system, force-plate, and high-speed video analyses

Abstract

Background: Jump analyses are frequently conducted in the sport and rehabilitation sector by different methods. This study aimed to compare a new ultrasonic system (US) measuring the distance between standing and the highest point of the jump with (a) the impulse-momentum (FP_IM), double integration (FP_DI), and flight-time (FP_FT) methods using a force plate and (b) the rise-time (VA_RT) and vertical distance (VA_VD) methods of the ankle using a high-speed video analysis. Additionally, the trial-to-trial reliability of each method was examined. Methods: Eighteen participants (25.8 ± 5.9 years) performed three countermovement jumps. The jump height was determined by the six methods. One-way repeated measures ANOVA with post-hoc comparisons were used to evaluate systematic bias. Pearson correlation coefficients and typical errors between all the methods were calculated to determine random errors. Trial-to-trial reliability was assessed with intraclass correlation coefficients (ICC). Results: A systematic bias existed between the US (≤ 15.4 cm) as well as FP_DI (≤ 14.0 cm) and all the other methods (all P < 0.05). Moreover, the vertical jump heights of FP_FT were 1.7 and 1.4 cm greater than the heights of FP_IM and VA_VD, respectively (all P < 0.05). Correlation coefficients and typical errors between all methods were ≥ 0.91 and ≤ 2.6 cm, respectively. The ICCs for the US, FP_IM, FP_DI, FP_FT, VA_VD, and VA_RT were 0.96, 0.96, 0.93, 0.95, 0.94, and 0.90, respectively. Conclusions: All methods showed a high trial-to-trial reliability, confirming their general usefulness. However, the systematic differences between the jump heights of the methods need to be considered. Consequently, regression equations allow the conversion of countermovement jump heights between the six methods with small typical errors of estimate.