China
Africa
Explore chapters and articles related to this topic
A Review of the Basic Concepts
Published in Ramin Rostamkhani, Mahdi Karbasian, Quality Engineering Techniques, 2020
Ramin Rostamkhani, Mahdi Karbasian
Descriptive statistics provides characteristics of sample data (for instance, means and standard deviations). However, these tools are contingent upon limitations such as sample size, and sampling method. These quantitative tools are considered valid when considered in relation to statistical assumptions.
The relationship between trunk rotation and shot speed when performing ice hockey wrist shots
Published in Journal of Sports Sciences, 2021
Shawn M. Robbins, Philippe J Renaud, Neil MacInnis, David J Pearsall
Hierarchical linear models were constructed to address the primary objective. Separate analyses were conducted for skating and stationary shots. Dependent variables were puck and blade speed and separate models were run for these variables. Individual shooting trial data were entered into the models, opposed to ensemble averages, and data were clustered with-in participants. This allowed for more accurate partitioning of between and within participant variability (Tirrell et al., 2018). The first step of the model construction was entering the intercept and control variables including age, body mass index, shooting trial number, shooting skate (i.e. skate that remained on the ice), and skating speed. Next, group (elite vs. recreational) was entered, which was followed by peak trunk rotation. Potential interactions between peak trunk rotation and group were explored and only retained in the final model if it was statistically significant. Analyses were repeated with trunk rotation ROM replacing peak trunk rotation. These variables were not entered together because of concerns of multicollinearity. In total, eight hierarchical linear models were constructed that varied shot type (skating or stationary), dependent variable (puck or blade speed), and independent variable (peak trunk rotation or trunk rotation ROM). Shooting skate (0 = lead, 1 = trail) and group (recreational = 0; elite = 1) were entered as categorical variables and remaining variables were entered as continuous. Different stages of model development were assessed with −2 log-likelihood (−2LL) and critical values for the chi-square statistic. The regression coefficients (i.e. slope) with 95% confidence intervals were reported. Statistical significance was set at p = 0.05. For the models, the covariance structure was variance components, full maximum-likelihood was chosen, and degrees of freedoms were calculated with the Kenward-Roger method (Singer & Willett, 2003; Wang, Xie, & Fisher, 2011). The following diagnostics were examined to ensure statistical assumptions were met: linearity, normality, homoscedasticity, multicollinearity, and influential cases. Statistical analyses were performed with SPSS version 24 (IBM Corp., Armonk, USA).