統計的差異にもかかわらず実質的な同等性：セッション内パフォーマンスモニタリングのための反復的カウンター運動ジャンプシリーズの非劣性分析

The countermovement jump (CMJ) is commonly used to assess and monitor changes in neuromuscular performance. This study investigated how many CMJ series can be performed without causing a meaningful alteration in performance, comparing jump-height calculation methods and data-handling strategies to optimize reliable, practically relevant performance assessment within a single session.

One hundred three healthy and recreationally trained adults performed 9 × 3 CMJs (3-min rest between series; 15 s between jumps). Jump height was assessed using the flight-time and impulse-momentum methods, analyzed as mean, mean of 2 highest, and highest jump. Equivalence to baseline was tested for each time point using predefined irrelevant margins (differences between highest and lowest jumps at baseline) and classified as equivalent, nonequivalent, or uncertain.

For all data-handling methods, significant time effects were found in both the flight-time and the impulse-momentum method. Post hoc testing for the flight-time method revealed statistically significant reductions from the fourth (2 highest, highest) or sixth (mean) series onward, with series 2 to 4 (highest), 2 to 5 (2 highest), and 2 to 6 (mean) classified as equivalent to baseline. For the impulse-momentum method, statistically significant reductions occurred from the third (highest) or fourth series (2 highest, highest) onward compared with baseline, while series 2 to 4 (highest) and 2 to 5 (mean, 2 highest) were equivalent to baseline.

Averaging multiple CMJs improves accuracy and sensitivity. To avoid bias, no more than 6 series of 3 CMJs should be performed per session. Moreover, changes should be interpreted relative to baseline variability to differentiate statistical significance from practical relevance.