Skip to main content

Table 4 β -coefficients of the logistic regression for study B, using the relative difference between non-Stroop and Stroop results as measure

From: Shotgun approaches to gait analysis: insights & limitations

Measure

p 1

p 2

β 13

β 23

β 12

|| β||

Force plate SE of sample entropy (y)

0.02

0.03

-2.8

-0.6

-2.1

3.5

EMG sample entropy (TA)

0.006

0.005

0.3

-1.8

2.1

2.8

Force plate DFA- α of stride duration

0.09

0.1

1.2*

-0.7

1.9

2.4

EMG SE of sample entropy of phase (VM)

0.05

0.7

-1.2

0.6

-1.8

2.3

Metabolism mean (HR)

0.07

0.08

-1.5

-0.2

-1.3

2.0

EMG PCA eigenvalues (5)

0.07

0.02

-0.8

-1.3

0.5

1.6

Force plate mean template (z)

0.07

0.4

0.7

0.8

-0.1

1.1

  1. The 7 measures with the highest p-value were included. The second column (p1) shows the p-value for the one-way ANOVA using the alternative measure, the third column (p2) shows the p-values for the interaction effect of the two-way ANOVA. Note that in both cases, the significance level was 0.0004 after the Bonferroni correction. Significant coefficients (p<0.05) are indicated with an *. For β12 and ||β|| no significance could be indicated because these were calculated after the regression. Note that for the β-coefficients, 1 indicates the transtibial amputee group, 2 the transfemoral group and 3 the control group.