In yoga research, the most common research designs is to compare pre and post measures of a group of people completing a yoga programme. The second most favourite design is to compare a yoga group to an inactive group. Both these designs are unreliable. The first one usually fails to produce similar results, thereby causing serious issues for generalisation of those results. And the second one carries the potential to exaggerate the results in favour of yoga and ignore the fact that other types of exercise have similar benefits.
Thus it is important to randomised control trial where the yoga group is compared to both another type of exercise and an inactive control group in order to be able isolate the unique benefits of yoga in comparison with other types of exercises.
Schuver and Lewis (2016) “examined the efficacy of a mindfulness-based yoga intervention for the reduction of depressive symptoms among adult women. … Women with depression were randomized to a mindfulness-based yoga intervention or a walking control group. … Consistent with previous studies, in the current study there were no between mindfulness-based yoga group and daily walking group on depression scores at post-intervention or one-month follow-up. … The increase in both yoga and walking is promising given physical activity has the dual benefit of improving overall health and alleviating depression and anxiety.”
Schuver, K. & Lewis, B. (2016) Mindfulness-based yoga intervention for women with depression. Complementary Therapies in Medicine Vol. 26. http://dx.doi.org/10.1016/j.ctim.2016.03.003