Reliability and validity of a novel Kinect-based software program for measuring a single leg squat

Table 2 Results from the test-retest reliability study

	Data^a			Differences^b	Relative reliability^c		Absolute reliability
	T1 Median (Q1, Q3)	T2 Median (Q1, Q3)	GRAND Median (Q1, Q3)	T1 vs. T2 p-value	Spearman (r) p-value	ICC (3.1) (95% CI)	SEM^d	SDC^e	Mean difference between T2 and T1 with 95% confidence interval (95% CI)^f
Left knee up (n = 64)	−6.34 (−16.93, 6.26)	0.66 (−8.40, 10.63)	−2.53 (−11.53, 9.20)	0.013*	0.53 (0.002)	0.50 (0.17 to 0 .72)	10.66	29.55	6.89 (1.98 to 11.81) *
Left knee down (n = 64)	1.07 (−6.50, 19.42)	0.80 (−9.50, 13.42)	1.07 (−6.76, 15.48)	0.5496	0.68 (< 0.001)	0.69 (0.45 to 0.83)	9.85	27.30	−2.15 (−7.19 to 2.89)
Right knee up (n = 61)	−6.48 (−16.59, 7.09)	−1.95 (−10.35, 7.87)	−3.20 (−12.06, 7.09)	0.2059	0.61 (< 0.001)	0.69 (0.45 to 0.84)	9.04	25.06	2.83 (−1.91 to 7.56)
Right knee down (n = 62)	0.59 (−11.71, 13.52)	2.75 (−10.04, 12.59)	2.27 (−10.04, 12.81)	0.5967	0.61 (< 0.001)	0.64 (0.38 to 0.81)	9.09	25.20	1.24 (−3.53 to 6.00)

*Denotes a statistically significant change; n = denotes the number of measurements done by Qinematic™ for each variable; A negative value (−) denotes a medial displacement for the left knee and a lateral displacement for the right knee, contrariwise a positive value (+) denotes a medial displacement of the right knee and a lateral displacement for the left knee
^a Test occasion 1 = T1, Test occasion 2 = T2, Q1: 1st quartile (25%), Q3: 3rd (75%). Grand Median = median of all measures from test occasion 1 and 2. All data are in degrees
^b Wilcoxon signed-rank test (paired test). A p-value of p < 0.05 was considered to be statistically significant and marked with*
^c Reliability based on a two-way mixed effect model, calculating the absolute agreement, based on single ratings. ICC (3.1):\( {ICC}_{agreement}=\frac{\sigma_p^2}{\sigma_p^2+{\sigma}_o^2+{\sigma}_{residual}^2},{\sigma}_{error}^2={\sigma}_o^2+{\sigma}_{residual}^2 \)
^d SEM: Standard error of the measurement: \( {SEM}_{agreement}=\surd \left({\sigma}_o^2+{\sigma}_{residual}^2\right) \)
^e SDC: Smallest detectable change: SDC = ± 1.96 ∗ √ 2 ∗ SEM_agreement.
^f Mean difference between the two-test occasion was calculated together with the 95% confidence interval (95% CI). *A 95% CI that does not include zero indicates a systematic change in the mean between T1 and T2.

ISSN: 2052-1847