TABLE 4.
Differences in the mean SD of power from the Reference System across the stages for Peach, EmPower, Weba, and the Concept2. Data are mean (%), ±90% compatibility limits, with observed magnitude and p-values for inferiority and superiority tests (p–/p+).
| Peach | EmPower | Weba | Concept2 | |
| 30-s test | –16, ±11; Small** 0.85/0.005 | 57, ±27; Moderate**** <0.001/0.998 | 139, ±110; Large*** 0.01/0.98 | 177, ±46; Large**** <0.001/>0.999 |
| Stage 1 | –3.0, ±4.0; Trivial0000 0.004/<0.001 | 14, ±16; Small*0 0.02/0.64 | 61, ±74; Moderate** 0.03/0.93 | –27.8, ±3.6; Small**** >0.999/<0.001 |
| Stage 2 | –4.4, ±3.6; Trivial000 0.009/<0.001 | 9.7, ±5.9; Trivial0* <0.001/0.35 | 73, ±84; Moderate** 0.03/0.94 | –27.9, ±4.6; Small**** >0.999/<0.001 |
| Stage 3 | –6.5, ±4.0; Trivial00 0.07/<0.001 | 20, ±13; Small** <0.001/0.88 | 88, ±82; Moderate*** 0.02/0.97 | –24.3, ±8.2; Small*** 0.99/<0.001 |
| Stage 4 | –6.4, ±5.1; Trivial00 0.11/<0.001 | 15.5, ±8.4; Small** <0.001/0.81 | 91, ±78; Moderate*** 0.02/0.97 | –20.8, ±9.6; Small*0 0.96/<0.001 |
| Stage 5 | –5.2, ±5.4; Trivial00 0.07/<0.001 | 29, ±26; Small** 0.09/0.90 | 126, ±95; Large*** 0.01/0.99 | –13.1, ±11.4; Small*0 0.68/0.003 |
| Stage 6 | –3.9, ±5.3; Trivial000 0.03/<0.001 | 17, ±14; Small** 0.001/0.77 | 136, ±148; Large*** 0.02/0.97 | –13.8, ±10.9; Small*0 0.90/0.01 |
| Stage 7 | –9.7, ±6.9; Trivial0* 0.47/<0.001 | 15, ±29; Small 0.05/0.60 | 109, ±65; Large*** 0.01/0.99 | –7.2, ±8.1; Trivial0* 0.27/0.002 |
Scale of magnitudes: <11 and >−10%, trivial; ≥11 or ≤−10%, small; ≥43 or ≤−30%, moderate (mod); ≥100 or ≤−50%, large; ≥230 or ≤−70%, very large (v.large); and ≥900 or ≤−90%, extremely large (e.large).
Reference-Bayesian likelihoods of substantial change: *possibly; **likely; ***very likely; and ****most likely.
The symbols *** and **** indicate rejection of the non-superiority or non-inferiority hypothesis (pN– or pN+ <0.05 and <0.005, respectively).
Reference-Bayesian likelihoods of trivial change: 0possibly; 00likely; 000very likely; and 0000most likely.
The symbols 000 and 0000 indicate rejection of the superiority and inferiority hypothesis (p– and p+ <0.05 and <0.005, respectively).
Effects in bold have adequate precision at the 99% level (p < 0.005).