Skip to main content

Table 4 Effect size and statistical significance of features

From: Predicting functional performance via classification of lower extremity strength in older adults with exergame-collected data

Feature Not fit-fit effect size (g) Not fit-fit significance (p) Feature Not fit-fit effect size (g) Not fit-fit significance (p)
\({Com}_{{Avg}_{Up},x}\) 0.3168 0.1998 \({Avg}_{y}\) − 0.0056 0.9817
\({Com}_{{Avg}_{Up},y}\) − 0.4678 0.0595 \({Max}_{x}\) − 0.3507 0.1528
\({Com}_{{Avg}_{Down,x}}\) 0.2790 0.2692 \({{\varvec{M}}{\varvec{a}}{\varvec{x}}}_{{\varvec{y}}}\) − 0.5279 0.0310
\({Com}_{{Avg}_{Down,y}}\) − 0.1171 0.6238 \({Min}_{x}\) 0.3565 0.1472
\({Com}_{{Avg}_{Right},x}\) − 0.2328 0.3404 \({Min}_{y}\) − 0.1036 0.6630
\({Com}_{{Avg}_{Right},y}\) − 0.4073 0.1028 \({{\varvec{S}}{\varvec{t}}{\varvec{d}}}_{{\varvec{x}}}\) − 0.6665 0.0068
\({Com}_{{Avg}_{Left},x}\) 0.2661 0.2756 \({Std}_{y}\) − 0.3789 0.1247
\({Com}_{{Avg}_{Left},y}\) − 0.3234 0.1863 \({{\varvec{I}}{\varvec{f}}}_{{\varvec{A}}{\varvec{v}}{\varvec{g}}}\) − 0.7478 0.0035
\({Com}_{{Std}_{Up,x}}\) − 0.0323 0.8966 \({{\varvec{I}}{\varvec{f}}}_{{\varvec{M}}{\varvec{a}}{\varvec{x}}}\) − 0.6337 0.0119
\({{\varvec{C}}{\varvec{o}}{\varvec{m}}}_{{{\varvec{S}}{\varvec{t}}{\varvec{d}}}_{{\varvec{U}}{\varvec{p}},{\varvec{y}}}}\) − 0.4913 0.0461 \({{\varvec{I}}{\varvec{f}}}_{{\varvec{T}}{\varvec{h}}{\varvec{r}}{\varvec{e}}{\varvec{s}}{\varvec{h}}{\varvec{o}}{\varvec{l}}{\varvec{d}},0.5}\) − 0.7452 0.0024
\({Com}_{{Std}_{Down,x}}\) 0.2234 0.3490 \({If}_{Threshold,1}\) − 0.2411 0.2873
\({Com}_{{Std}_{Down,y}}\) 0.4173 0.0860 \({If}_{Threshold,1.5}\) 0 0
\({Com}_{{Std}_{Right,x}}\) 0.0318 0.8977 \({If}_{Threshold,2}\) 0 0
\({Com}_{{Std}_{Right,y}}\) 0.0753 0.7621 \({{\varvec{I}}{\varvec{f}}}_{{{\varvec{S}}{\varvec{u}}{\varvec{m}}}_{{\varvec{A}}{\varvec{v}}{\varvec{g}}}}\) − 0.7387 0.0038
\({Com}_{{Std}_{Left,x}}\) − 0.3164 0.1959 \({If}_{{Sum}_{Max}}\) − 0.2107 0.3938
\({Com}_{{Std}_{Left,y}}\) − 0.0237 0.9217 \({{\varvec{I}}{\varvec{f}}}_{{{\varvec{S}}{\varvec{u}}{\varvec{m}}}_{{\varvec{O}}{\varvec{v}}{\varvec{e}}{\varvec{r}}0.5}}\) − 1.5261 < 0.0001
\({Balance}_{Up}\) 0.1598 0.5215 \({{\varvec{I}}{\varvec{f}}}_{{{\varvec{S}}{\varvec{u}}{\varvec{m}}}_{{\varvec{O}}{\varvec{v}}{\varvec{e}}{\varvec{r}}1}}\) − 0.9196 0.0003
\({Balance}_{Down}\) 0.1623 0.4988 \({If}_{{Sum}_{Over1.5}}\) − 0.2206 0.3477
\({Balance}_{Right}\) − 0.3628 0.1429 \({If}_{{Sum}_{Over2}}\) − 0.2062 0.3762
\({{\varvec{B}}{\varvec{a}}{\varvec{l}}{\varvec{a}}{\varvec{n}}{\varvec{c}}{\varvec{e}}}_{{\varvec{L}}{\varvec{e}}{\varvec{f}}{\varvec{t}}}\) 0.6306 0.0091 \({{\varvec{S}}{\varvec{t}}{\varvec{e}}{\varvec{p}}}_{{\varvec{A}}{\varvec{v}}{\varvec{g}}}\) 1.2260 < 0.0001
\({Avg}_{x}\) 0.1925 0.4267 \({{\varvec{S}}{\varvec{t}}{\varvec{e}}{\varvec{p}}}_{{\varvec{S}}{\varvec{t}}{\varvec{d}}}\) 0.8446 0.0020
  1. All values are presented as not fit vs. fit, meaning that a negative effect size indicates the parameter has lower values in the not fit group. According to Cohen’s rule, g > 0.8 indicates a large effect size. Bold emphasis indicates statistical significance (p < 0.05)