Original scientific paper
Non-Linear Approach in Kinesiology Should Be Preferred to the Linear – A Case of Basketball
Marko Trninić
; »Cedevita« Basketball club, Zagreb, Croatia
Mario Jeličić
; University of Split, Faculty of Kinesiology, Split, Croa
Vladan Papić
; University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Split, Croatia
Abstract
In kinesiology, medicine, biology and psychology, in which research focus is on dynamical self-organized systems,
complex connections exist between variables. Non-linear nature of complex systems has been discussed and explained by
the example of non-linear anthropometric predictors of performance in basketball. Previous studies interpreted relations
between anthropometric features and measures of effectiveness in basketball by (a) using linear correlation models, and
by (b) including all basketball athletes in the same sample of participants regardless of their playing position. In this
paper the signifi cance and character of linear and non-linear relations between simple anthropometric predictors (AP)
and performance criteria consisting of situation-related measures of effectiveness (SE) in basketball were determined and
evaluated. The sample of participants consisted of top-level junior basketball players divided in three groups according
to their playing time (8 minutes and more per game) and playing position: guards (N=42), forwards (N=26) and centers
(N=40). Linear (general model) and non-linear (general model) regression models were calculated simultaneously and
separately for each group. The conclusion is viable: non-linear regressions are frequently superior to linear correlations
when interpreting actual association logic among research variables.
Keywords
anthropometric predictors; performance indicators; elite junior men; non-linear multiple regression; sports science
Hrčak ID:
147622
URI
Publication date:
14.7.2015.
Visits: 938 *