Comparison of different regression models to fit the force–velocity relationship of a knee extension exercise

Abstract

[Abstract]: The aims of this study were to compare the goodness of fit and theconcurrent validity of three regression models of the force–velocityrelationship in a unilateral knee extension exercise. The force–velocityrelationship and the one-repetition-maximum load in the dominantand non-dominant leg were obtained in 24 male sports sciencesstudents by a progressive protocol. Additionally, the maximumvoluntary contraction (MVC) of the knee extensor muscles wasrecorded. Individual force–velocity relationships were obtained bythe linear, quadratic polynomial and exponential regression models.Although the adjusted coefficients of determination of all threemodels were high, the polynomial model’s coefficient was slightlybut significantly higher than the rest of the models (p < 0.05), whilethe standard error of estimate was slightly higher for the linearthan for polynomial model (p = 0.001). MVC was underestimatedby F0 calculated from the linear and polynomial models, while themaximum power was accurately estimated by the linear model. Insummary, while the polynomial model revealed somewhat betterfit, the linear model more accurately estimates the maximum powerand provides the parameters of apparent physiological meaning.Therefore, we recommend using the linear model in research androutine testing of mechanical capacities of knee extensors.