Table 3.

Simple statistics for PT on baseline and after training programme, and magnitude-based inferences for the change of the means.

	Baseline (n=12) Mean ± SD	After programme (n=12) Mean ± SD	Effect
			Mean; ±90%CL	Inferencea
Dominant Leg
H Con60	93 ± 18	109 ± 13	18.3%; ± 15.1%	moderate ↑**
Q Con60	184 ± 26	190 ± 32	2.8%; ± 11.0%	trivial ↑
H Con180	111 ± 16	130 ± 17	17.8%; ± 11.2%	moderate ↑***
Q Con180	139 ± 14	149 ± 13	7.4%; ± 7.8%	moderate ↑**
H Ecc60	113 ± 15	118 ± 16	4.4%; ± 11.2%	small ↑
H Ecc180	155 ± 17	155 ± 19	0.0%; ± 8.8%	none
Non-dominant Leg
H Con60	94 ± 15	107 ± 17	13.7%; ± 11.3%	moderate ↑**
Q Con60	180 ± 22	187 ± 25	3.7%; ± 11.0%	small unclear
H Con180	117 ± 13	133 ± 20	13.4%; ± 8.0%	moderate ↑***
Q Con180	129 ± 13	143 ± 14	10.7%; ± 11.5%	moderate ↑**
H Ecc60	111 ± 14	116 ± 17	4.4%; ± 11.1%	small ↑
H Ecc180	154 ± 16	156 ± 17	1.1%; ± 8.6%	trivial ↑

^aMagnitude thresholds (for change in means divided by baseline SD): <0.20, trivial; 0.20-0.59, small; 0.60-1.19, moderate. Asterisks indicate effects clear at the 5% level and likelihood that the true effect is substantial or trivial, as follows

^*possible

^**likely

^***very likely. Effects in bold are also clear at 0.5%.