Full vs. Half-Squats - Study Measures Actual Size Gains! A Tie for Quads, a Fail for Hams, and a Small Win for Glutes

The result, the full-squat is slightly superior, is not as clear as you may have expected it to be. Furthermore, the study suggests that you won't get away without an additional hamstring exercise if you want tree-trunk legs and a fabulous behind.
A new study (Kubo 2019) in the "European Journal of Applied Physiology" compared the effects of squat training with different depths on lower limb muscle volumes (that's news as previous studies measured way too often only the acute EMG response | Neto 2019)... with a realistic workout program and a sensible study duration of 10 weeks. over which twenty young men were randomly assigned to a full squat group (#FST, n = 10) or half squat group (#HST, n = 10 at study onset).
Are you trying to optimize your training for gains? Find inspiration in these articles:

What's the Latest on Failure?

36% Extra-Gains W/ Rest-Pause

Pyramid- and/or Drop-Setting!?

Training to Failure = Extra Gains?

Super-Setting - Yes, but How?

Eccentrics For Excellent Gains?
The subjects were matched to the two groups according to their average baseline physical characteristics and the one-repetition maximum (#1RM) of full and half squats between the two groups. The latter, i.e. the 1RMs were ~1.25x and ~1.5x the guys' body weight on the full and half-squat, respectively - an observation that tells you that all participants, who were in their early twenties, were not exactly powerlifters, but they were also no couch-potatoes. Still, theywere required not to  have participated in any organized resistance training program involving regular exercise for at least 1 year before testing.

Why would one use untrained subjects?

The authors explain the decision to use 'untrained' subjects with the argument that "the obtained results would be affected by the effects of training experiences before the experiment".
Full-squat group (n = 8)
Half-squat group (n = 9)
Age (years)
20.7 (0.4)
20.9 (0.8)
Height (cm)
173.6 (4.1)
172.3 (5.8)
Body mass (kg)
63.2 (6.6)
64.1 (6.1)
1RM of full squat (kg)
78.8 (14.6)
82.8 (15.2)
1RM of half squat (kg)
95.0 (16.0)
96.7 (15.0)
Table 1: Age, physical characteristics, and 1RM before training in both groups mean (sd) *1RM one repetition maximum.
As in every training study, the "subjects were instructed to maintain their normal diet and avoid taking any supplements during the experimental period." While that's standard practice, it's worth noting that the assessment of muscle gains which was not done by DXA or simple circumference measures, but via a series of cross-sectional images of the lower limb muscles using magnetic resonance imaging (FLEXART MRT-50GP, Toshiba Medical Systems, Tokyo, Japan). The muscle sizes the sizes of which the researchers measured were the...
Figure 1: Muscles of the lower limbs | Lumen Learning
  • knee extensor muscles: rectus femoris (RF), vastus lateralis (VL), vastus intermedius (VI), and vastus medialis (VM),
  • hamstring muscles: biceps femoris short head (BFs), biceps femoris long head (BFl), semitendinosus (ST), and semimembranosus (SM), and
  • adductor muscles: adductor magnus, adductor longus, and adductor brevis
The number of axial images obtained for each subject was the same before and after training and was 39.5 ± 2.3 for the knee extensor muscles: 37.2 ± 2.4 for the hamstring muscles, 29.4 ± 3.1 for the adductor muscles, and 28.5 ± 1.5 for the gluteus maximus muscle. Images obtained with magnetic resonance imaging were transferred to a computer and analyzed using Osirix DICOM image analysis software (Pixmeo, Geneva, Switzerland).
That's a half squat: Knee angle 90°
Full- vs. Half-Squat: Defined! I guess we are far from having a clearcut, universal definition of what constitutes a full- or half-squat, but what I can tell you is what the subjects in the study at  hand did:

#FST - The full squat was performed from complete knee extension to approximately 140° knee flexion and then immediately returned to the extended knee position.

#HST - The half-squat was performed at the half range of motion squat (from complete knee extension to approximately 90° knee flexion) and then immediately returned to the extended knee.
The subjects' actual training consisted of 2 workouts per week (for 10 weeks). For both groups, subjects were instructed that stance width was almost the same as shoulder width. The barbell was positioned across their shoulders on the trapezius. All subjects were allowed to use a lifting belt during squat training. All training sessions were monitored and supervised to ensure correct squat depth and form by at least one experienced investigator. The authors describe the exact protocol as follows (my emphasis):
"In order to become accustomed to training and acquire a correct form, subjects performed 3 sets of 60% 1RM × 10 repetitions in the first week, 3 sets of 70% 1RM × 8 repetitions in the second week, and 3 sets of 80% 1RM × 8 repetitions in the third week. [...] If subjects were able to perform 3 sets of 8 repetitions per set, the training load was increased by 5 kg for the next training session."
Another detail of the methodology section that's important is the way the scientists calculated the training volume as the arithmetic product of load × repetition × movement distance of the barbell - that's important, because it takes into account that, due to the greater range of motion (87.9 ± 2.1 cm in FST and 53.8 ± 1.8 cm in HST),  the full-squat will require significantly higher workloads on a per rep basis. This assumption does yet require that both train with identical weights. In view of the fact that the half-squat allowed for greater weights, it is still not surprising that - within the intra-group standard-deviations, there was ...
  • no significant difference in the total training volume between FST (186.4 ± 34.0 kg*rep*m) and HST (198.4 ± 19.9 kg*rep*m | p = 0.388, ES = 0.45) 
Unlike previous studies, which suggested slight benefits of partial vs. full squats for strength and power, the study at hand accounted for training specific specifications (keyword: you're getting better at the exact exercise you practice) by comparing the two groups on both, the half- and full-squat 1RM performance.
Figure 2: Relative changes in one repetition maximum in full (upper) and half (lower) squat exercises for full squat training (open) and half squat training (closed) groups. *Significantly different from before (**p < 0.01, ***p < 0.001). #Significantly different between the two groups (##p < 0.01) | Kubo 2019
In that, Figure 2 clearly indicates that trainees who are doing only half-squats progress well, but not better than their full-squatting peers on this very exercise (i.e. the half-squat | Figure 2, right). However, compared to their peers, who've been squatting "ass to the grass" over the whole study period, the partial squatters sucked on the unaccustomed full squat (Figure 2, left) - and the difference is both statistically and practically relevant (~30% w/ doing the actual full squat vs. ~10% increase in 1RM/10wk with the half-squat - that's worlds apart).

But who cares about 1RMs? Gainz is what ya want, right? 

The question: What's best for your gains has likewise been addressed before. In many cases, such as the previously cited 2016 study, with slight advantages (in terms of hypertrophy, not strength) for the full squat, i.e. "ass to the grass". The study at hand confirms that, but it adds that the size of the difference (pun intended) depends on the muscle group we're looking at (see Table 2). So, ...
  • the volumes of knee extensor muscles significantly increased by 4.9 ± 2.6% in FST (p < 0.001, ES = 0.34) and 4.6 ± 3.1% in HST (p = 0.003, ES = 0.43) - Note: This was not a significant effect-size difference (p = 0.812, ES = 0.11) favoring the full squat for a muscle group we often refer to as the "quads" as a whole or for the muscle volumes of VL, VI, and VM between the two groups (p = 0.497–0.892, ES = 0.02–0.34), individually;
  • Study leaves no doubt: For the biceps femoris, Romanian deadlifts rule.
    the volumes of each constituent of all hamstring muscles did not significantly change after training in either group (p = 0.129–0.911, ES = 0.01–0.07) - Note: For this muscle group, the often-heard superiority of "going all the way" down is absent in all submeasures, i.e. both heads of the biceps femoris, the semitendinosus and semimembranosus muscle 😮... and this is probably the actual (disappointing) surprise result for some of you;
  • the volumes of the adductor muscles significantly increased by 6.2 ± 2.6% in FST (p < 0.001, ES = 0.55) and 2.7 ± 3.1% in HST (p = 0.030, ES = 0.33) - Note: This time, however, with measurable, statistically and (potentially) practically relevant advantages for the full squat:  More specifically, the volume of the gluteus maximus muscle significantly increased by 6.7 ± 3.5% in FST (p < 0.001, ES = 0.35) and only 2.2 ± 2.6% in HST (p = 0.041, ES = 0.14); a difference that was also observed for the adductor and gluteus maximus muscle volumes which were significantly greater in FST than in HST (p = 0.026, ES = 1.23 for the adductor muscles, p = 0.008, ES = 1.50 for the gluteus maximus muscle).
So, there's some truth to the often-heard recommendation to squat deep to form a strong and muscular behind. What is more important to remember, IMHO, is the fact that scientists rightly point out that "[t]he main results of the present study were that 10 weeks of full and half squat training increased the volumes of the vasti muscles, but not rectus femoris or hamstring muscles". Essentially, this implies that you will have to train both separately - for example by doing stiff-legged deadlifts (learn more about the best exercises in this previous article). 
Table 2: Muscle volume of each constituent of knee extensor muscles before and after training mean (sd)
Compared to the unexpected shortcomings of squatting in terms of hamstring gains, the observation that "the volumes of the adductor and gluteus maximus muscles were greater with full squat training than half-squat training" will hardly come as a surprise most of you, I guess... don't worry, though, if you cannot squat ass-to-the-grass you can still do other exercises for your glutes - including the stiff-legged deadlift I've previously recommended as a hamstring adjunct to the squat.
Nice gimmick or must-have gym equipment? Answer: "It depends".
Bottom line: The study at hand confirms two advantages of having the full- vs half-squat exercise in your workout program: (a) You'll significantly increase your bragging rights (1RM) on the full-squat only by squatting over the whole range; (b) you will make greater (albeit suboptimal) gains on the adductor muscles and the glutes.

If we are honest, though, those are fewer and (if you look at the figures) also smaller benefits than broscience would dictate. Worst of all, though 😨, neither of the tested squat varieties targeted the hamstring muscles appropriately. So, even if you squat deep, a hamstring exercise such as stiff-legged deadlifts is a must.

No EMS/gainz disconnect for this, but older studies...

Unlike the study at hand, the previously cited "Romanian(stiff-legged) deadlift"-study by McAllister et al. and hip thrust studies such as Contreras et al. 2016 did not measure the actual hypertrophy response to the exercises (Neto 2019). Hence, we simply have to assume that an increased EMG activity will also yield increased size gains | Discuss!
  • Kubo, Keitaro, Toshihiro Ikebukuro, and Hideaki Yata. "Effects of squat training with different depths on lower limb muscle volumes." European journal of applied physiology (2019): 1-10.
  • McAllister, Matt J., et al. "Muscle activation during various hamstring exercises." The Journal of Strength & Conditioning Research 28.6 (2014): 1573-1580.
  • Neto, Walter Krause, Thais Lima Vieira, and Eliane Florencio Gama. "Barbell Hip Thrust, Muscular Activation and Performance: A Systematic Review." Journal of sports science & medicine 18.2 (2019): 198.
Disclaimer:The information provided on this website is for informational purposes only. It is by no means intended as professional medical advice. Do not use any of the agents or freely available dietary supplements mentioned on this website without further consultation with your medical practitioner.