McBride et al recently produced a study (JSCR, DEC, 2010) comparing the kinematics of both the box squat and back squat. The tested on 8 competitive male powerlifters using 1 rep at 60, 70 and 80% of 1 RM with each condition. The back squat was performed with a quick decent and ascent whilst the box squat required a 1 second pause at the bottom. See the results of the study in the figures presented Below.
They found little if now difference between the two back squats which would challenge quite a lot of the claims put forward by advocates of either squatting style. Interestingly they found no effect of the strength shortening cycle (SSC) on any of the squatting styles they did however report one trial with a turn over of 40ms where the SSC did contribute to the squat however all trials had longer turnover phases and resulted in no contribution.
This study does not support the use of either style of squat over the other it does not add nor distract from either style which would suggest that both are viable lower body exercises.
The EMG data is of particular interest to me as it does not support the notion that box squatting involves any more hip musculature than a back squat which is one of the biggest advantages often touted for this lift.
Having blogged recently and in the past on the problems of small populations sizes and EMG data I would caution people not to read too far into these results but it does make for very interesting reading.