Test-retest reliability of jump execution variables using mechanography: a comparison of jump protocols

Authors: FITZGERALD, JOHN - University Of North Dakota; JOHNSON, LUANN - University Of North Dakota; TOMKINSON, GRANT - University Of North Dakota; STEIN, JESSE - University Of North Dakota; Roemmich, James

Submitted to: Journal of Sports Sciences
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/20/2017
Publication Date: 6/26/2017
Publication URL: http://handle.nal.usda.gov/10113/5852208
Citation: Fitzgerald, J.S., Johnson, L., Tomkinson, G., Stein, J., Roemmich, J.N. 2017. Test-retest reliability of jump execution variables using mechanography: a comparison of jump protocols. Journal of Sports Sciences. https://doi.org/10.1080/02640414.2017.1346818.

Interpretive Summary: Testing measures of the force and power produced when people jump may enhance the ability to diagnose functional deficits that reduce the ability to stand, walk, or exercise. However, there are few data of how reproducible these force measures are from one test to another. If they are not reproducible then it is difficult to know the true force or power or measure their changes with exercise training. In collaboration with scientists at the University of North Dakota, ARS scientists at the Grand Forks Human Nutrition Research Center tested how reproducibility of the jump measures when people performed a squat jump with squat depth self-selected by the person and when starting at a depth consistent for all subjects. Jump measures were very reproducible and the reproducibility of the results was not affected by whether subjects self-selected the starting position or if the starting position was assigned by the scientist. To simplify testing, a self-selected jump protocol can be used to assess jump measures with little impact on measurement error.

Technical Abstract: Mechanography during the vertical jump may enhance screening and determining mechanistic causes for functional deficits that reduce physical performance. Utility of jump mechanography for evaluation is limited by scant test-retest reliability data on force-time variables. This study examined the test-retest reliability of eight jump execution variables assessed from mechanography. Thirty two women (mean±SD: age 20.8±1.3 yr,) and 16 men (age 22.1±1.9 yr,) attended a familiarization session and two testing sessions, all one week apart. Participants performed the squat jump with squat depth self-selected and controlled using a goniometer to 110º knee flexion. Test-retest reliability was quantified as the systematic error (using effect size between jumps), random error (using coefficients of variation), and test-retest correlations (using intra-class correlation coefficients). Overall, jump execution variables demonstrated acceptable reliability, evidenced by small systematic errors (mean±95%CI: 0.2±0.07) , moderate random errors (mean ±95%CI: 17.8 ±3.7%), and very strong test-retest correlations (range: 0.73-0.97). Differences in random errors between controlled and self-selected protocols were negligible (mean ±95%CI: 1.3 ±2.3%). Jump execution variables demonstrated acceptable reliability, with no meaningful differences between the controlled and self-selected jump protocols. To simplify testing, a self-selected jump protocol can be used to assess force-time variables with negligible impact on measurement error.