Secondary Logo

Journal Logo

Original Research

The Effect of Squat Depth on Multiarticular Muscle Activation in Collegiate Cross-Country Runners

Gorsuch, Joshua1; Long, Janey1; Miller, Katie1; Primeau, Kyle1; Rutledge, Sarah1; Sossong, Andrew1; Durocher, John J.1,2

Author Information
Journal of Strength and Conditioning Research: September 2013 - Volume 27 - Issue 9 - p 2619-2625
doi: 10.1519/JSC.0b013e31828055d5
  • Free



Concurrent resistance and endurance training programs are commonly employed by runners (6,11,12,15,16,19,21,23) in an effort to maximize fitness and performance, while minimizing the risk of injury. Specifically, benefits of concurrent training include improvements in running economy (11,16,21), anaerobic running velocity (15), and 5-km performance (17,19). Squats are a frequently used resistance exercise within a runner’s concurrent training program (6,12,15,21,23), but it remains unclear how deep runners should squat to yield the greatest benefits. Squats of different depths have been shown to alter muscle activity when examined with electromyography (EMG) in male weightlifters (3). However, multiarticular muscle activation during parallel and partial squats has not been reported for male and female cross-country runners. Although weightlifters have a primary focus on increasing muscular size and strength, most runners are concerned with goals such as increasing muscular endurance, improving performance, avoiding injury, and preventing the early onset of fatigue. The squat is a regularly performed closed-chain lower body resistance exercise that has been shown to effectively target the multiarticular rectus femoris, biceps femoris, gastrocnemius (5,10), and erector spinae (18), which are all instrumental in the general wellbeing and performance of runners.

The biceps femoris and rectus femoris are often the first muscles to exhibit signs of fatigue in runners (8); a trend that has been attributed to their biarticular orientation (9). Furthermore, faulty running technique and poor posture may also lead to injuries of the hamstrings or lower back. For instance, hamstring strains and tears have been noted to occur as a result of overstretching as a runner leans forward in an attempt to accelerate (9). Excessive forward leaning in runners has been attributed to possible weakness of postural stabilizers such as the multiarticular erector spinae. In fact, running is shown to activate the erector spinae muscles to a greater extent than isometric prone back extensions (1). Finally, the gastrocnemii are important muscles for running, especially when traveling uphill (20).

Given the abundance of evidence in favor of concurrent training, but the relative lack of research regarding the most effective squat depth for cross-country runners, it is important to determine the extent of multiarticular muscle activation during this common resistance exercise. Therefore, the purpose of this study is to determine the most effective squat depth in regard to muscle activation for both male and female collegiate cross-country runners. This may help athletes and coaches to determine the most effective squat depth (i.e. partial or parallel) to utilize when formulating concurrent training programs. We hypothesized that the parallel squat would increase extensor muscle activation (i.e. hamstrings and erector spinae) when examined with surface EMG. Furthermore, we sought to determine if changes in muscle activation were different between male and female runners. A tertiary purpose of the present study was to probe for deficiencies in hamstrings-to-quadriceps ratios through isokinetic testing.


Experimental Approach to the Problem

Our study utilized a randomized crossover design with repeated measures to examine muscle activation during partial and parallel squats. The independent variable of interest was squat condition. Dependent variables included muscle activation determined via surface EMG, squat loads, and hamstrings-to-quadriceps ratios. Squat loads, joint angles, and hamstrings-to-quadriceps ratios were determined at a preliminary testing session that preceded EMG testing by about 1 week. This design allowed us to perform EMG testing during partial and parallel squats on 1 day in a randomized order without needing to move surface electrodes between conditions. Because squat loads were predetermined and the order of conditions was randomized, the EMG testing sessions were brief and the potential confounding effects of fatigue were minimized.


Twenty Division I cross-country runners, 10 males (mean ± SE; age = 19.2 ± 0.4 years; height = 176.8 ± 1.5 cm; body mass = 66.2 ± 2.5 kg; body fat percentage = 9.0 ± 1.1%) and 10 females (age = 19.9 ± 0.4 years; height = 166.7 ± 1.5 cm; body mass = 55.9 ± 1.4 kg; body fat percentage = 19.7 ± 1.3%) volunteered to serve as participants for this study. All runners had at least 4 years of significant running and resistance training experience. Each participant abstained from the squat exercise for a minimum of 6 weeks before the study. Testing was completed early in the spring semester between the cross-country and track seasons. All testing sessions occurred in the late afternoon or early evening. Participants refrained from exercise the day before testing and fasted for a minimum of 3 hours. The Saint Francis University Institutional Review Board approved this study before any data collection, and each participant provided his/her informed consent before participating in the study.


Approximately 1 week before EMG testing, each participant attended a preliminary orientation and measurement session. Participants provided their informed consent and completed a brief running history questionnaire. Height, weight, and age were recorded, and then body composition was evaluated using air displacement plethysmography (BodPod, COSMED USA, Inc., Concord, CA, USA). Each subject then performed isokinetic testing on a Biodex system (System 4 Pro, Biodex Medical Systems, Shirley, NY, USA) to determine his/her hamstrings-to-quadriceps ratio. Five repetitions were performed at 60º and 180º per second, and results are reported as average values over the 5 repetitions. Participants then walked to an adjacent building to perform 10 repetition maximum (10 RM) squat testing. Before testing their 10 RM, participants performed a warm-up of light cycling for 5 minutes on a stationary bike followed by 2 minutes of rest. Participants then performed 2 warm-up sets of 10 repetitions at 50 and 75% of their estimated parallel squat 10 RM with 3 minutes of rest between sets. Each participant’s 10 RM was then determined for the partial squat at 45º angle of knee flexion and the parallel squat at 90º angle of knee flexion in a randomized order, with joint angles measured via standard goniometer. Participants determined their 10 RM for each condition within 4 sets. Parallel and partial squat knee joint angles utilized were consistent with a previous study by Caterisano et al. (3). Hip and ankle joint angles were also recorded for each squat condition during the preliminary testing session to help quantify differences between conditions.

A Biopac (Biopac Systems, Inc., Goleta, CA, USA) EMG system was used to measure skeletal muscle activity. Surface electrode sites were prepared using an abrasive pad and cleaned with alcohol on the right rectus femoris (4), biceps femoris (4), lumbo-sacral erector spinae (7), and lateral head of the gastrocnemius (10). Impedance was verified to be less than 5,000 Ohms between each electrode pair. Data were recorded at 2,000 Hz, and all data were integrated using the root mean square method and averaged over 100 samples. High-pass and low-pass filters were set at 30 and 500 Hz, respectively. Ground electrodes were placed on the iliac crest, head of the fibula, and the anterior crest of the tibia. Table 1 provides a detailed description of recording electrode placement.

Table 1
Table 1:
Electrode placement.

EMG testing occurred 7–10 days after the preliminary session. On the day of testing, participants performed a warm-up of light cycling for 3 minutes on a stationary bike. Using the parallel squat, participants completed 2 warm-up sets of 10 repetitions at 50% and 75% of their parallel squat 10 RM, with 3 minutes of rest between sets. Then the order of EMG testing trials was randomized by a coin toss, with heads indicating the performance of partial squat first and tails indicating parallel squat first. A trained examiner used a standard goniometer to ensure proper squat depth via knee joint angle and provided verbal cues to maintain proper technique (Figure 1). The use of the goniometer during the EMG testing was strictly used to verify the correct knee joint angle for each squat condition and not to measure any other joint angle. After the warm-up, participants performed 6 repetitions with their respective 10 RM load for each squat condition, with 5 minutes of rest between sets. Repetitions were paced by a metronome set at 60 Hz. Cadence was 1 second down and 1 second up for the partial squat and 2 seconds down and 2 seconds up for the parallel squat. Data from the middle 4 repetitions were utilized for EMG analysis, and values are reported as mean activity during those 4 repetitions. After the completion of each squat condition, participants reported their ratings of perceived exertion (RPE) using a standard 10-point scale (2). On the EMG testing day, participants were also asked whether they preferred the parallel or partial squat. Choices for responses included: parallel, partial, or no preference.

Figure 1
Figure 1:
A) Partial squat depth of 45° angle at the knee joint measured with a standard goniometer; B) Parallel squat depth of 90° angle at the knee joint; C) Partial squat depth during electromyography (EMG) analysis; and D) Parallel squat depth during EMG.

Statistical Analysis

Data were analyzed with SPSS version 17.0 statistical software (SPSS, Chicago, IL, USA). Repeated measures analysis of variance (ANOVA) (4 × 2; muscle group by squat condition) was used with gender as a between-subjects factor for EMG analyses. Likewise, a repeated measures ANOVA (3 × 2; joint angle by squat condition) with gender as a between-subjects factor was used to examine the joint angles from the preliminary testing session. Significant differences were determined when p < 0.05. When appropriate, post hoc analyses were performed with paired t tests. All results are presented as means ± SE.


The average 10 RM for the partial squat (78.4 ± 4.6 kg) was significantly greater than that of the parallel squat (51.2 ± 3.1 kg; p < 0.01). The knee joint angles were measured at 45° angle for the partial squat and 90° angle for the parallel squat. Hip joint angles were measured to be 50.0 ± 2.8º for the partial condition and 94.6 ± 3.6° for the parallel condition. Mean ankle joint angles were 77.7 ± 1.8° for the partial squat, and 69.7 ± 1.8° for the parallel squat. Each of these joint angles was determined to be significantly different between conditions (p < 0.01). Examples of each squat condition and measurement of joint angles are depicted in Figure 1. There were no significant differences between males and females in regard to joint angles (interaction p values > 0.05 for each joint × sex × squat condition).

Rectus femoris activity was higher during the parallel squat (0.18 ± 0.01 mV) than it was during the partial squat (0.14 ± 0.01 mV; p < 0.05). Similarly, erector spinae activity was higher during the parallel squat (0.16 ± 0.01 mV), as opposed to the partial condition (0.13 ± 0.01 mV; p < 0.05), as shown in Figure 2. In contrast, biceps femoris activity was similar between the parallel (0.08 ± 0.01 mV) and partial (0.07 ± 0.01 mV) squats. Likewise, no significant differences were found for gastrocnemius activity between parallel (0.05 ± 0.00 mV) and partial (0.05 ± 0.00 mV) squats. Muscle activity was not different between males and females (interaction p values > 0.05 for each muscle × sex × squat condition). RPE was similar between conditions for males (5.0 ± 0.5 vs. 5.2 ± 0.5 a.u.) and females (4.3 ± 0.5 vs. 4.1 ± 0.5 a.u.) during the parallel and partial squats, respectively. On the EMG testing day, all but 1 participant replied to a survey question asked whether they preferred the parallel or partial squat. Among the 10 males who responded, 7 reported preferring the parallel squat, 2 reported preferring the partial squat, and 1 had no preference. Of the 9 females who responded, 4 preferred the parallel squat, 1 preferred the partial squat, and 4 had no preference.

Figure 2
Figure 2:
Millivolts of muscle activity via electromyography. Rectus femoris (RF) and lumbar erector spinae (ES) activity were significantly (p <0.05) higher during the parallel squat, as opposed to the partial squat. Biceps femoris (BF) and gastrocnemius (GN) activity were similar for both the partial and parallel squat conditions. *Significantly different than half-squat condition (p < 0.05).

According to Biodex Medical Systems, males are recommended to have a hamstrings-to-quadriceps ratio of 61% at a rate of 60° per second and 72% at a rate of 180° per second during isokinetic leg extension/flexion exercise. Females are recommended to have a hamstrings-to-quadriceps ratio of 62% at a rate of 60° per second and 76% at a rate of 180° per second. In this study, men and women did not differ in their respective hamstrings-to-quadriceps ratio deficiencies, when compared with the Biodex standards set for each gender. However, both male and female runners were deficient in the right lower extremity at a rate of 180° per second, and they were deficient in the left lower extremity at both 60° and 180° per second, as shown in Table 2.

Table 2
Table 2:
Hamstrings-to-quadriceps ratio deficiencies vs. Biodex standards.


This study measured muscle activation during parallel and partial squats in male and female collegiate cross-country runners. The aim of the present study was to determine which squat depth most effectively activated the rectus femoris, biceps femoris, lumbar erector spinae, and gastrocnemius and to examine if muscle activation was different between male and female runners. Hamstrings-to-quadriceps ratios were also measured with an isokinetic dynamometer. Results of this study reveal that the parallel squat improved activation of the rectus femoris and lumbar erector spinae when compared with the partial squat condition. In contrast, biceps femoris and gastrocnemius activity were similar during each squat condition. Likewise, muscle activation during each of the squat conditions was similar in male and female runners. Finally, both male and female runners had deficient hamstrings-to-quadriceps ratios. Although the EMG findings indicate that neither the partial nor the parallel squat is likely to help improve muscle imbalances between the hamstrings and quadriceps, the parallel squat may be superior to the partial squat for runners due to increased range of motion and muscle activation while using less resistance.

Despite using a significantly lighter load during parallel squats, rectus femoris and erector spinae activity still increased. Parallel squats could benefit runners by reducing compressive forces on the spine, while maintaining or increasing muscle activity, when compared with partial squats. Given that the lower extremity is a kinetic chain, the increase in rectus femoris and erector spinae activity during the parallel squat may be attributed to greater ranges of motion at the hip and knee joints. Because the rectus femoris fatigues early (8) and the erector spinae aids in maintaining an upright posture (9) while running, these findings are relevant to both collegiate cross-country runners and the coaches responsible for conditioning these athletes.

Casterisano et al. (3) determined that there were no significant differences in muscle activation between the biceps femoris and 2 quadriceps muscles (vastus medialis and vastus lateralis) at various squat depths in weightlifters. However, based on our results, the parallel squat better activated the rectus femoris than partial squats. Therefore, parallel squats might benefit the multiarticular rectus femoris in runners by potentially increasing muscular stamina to avoid early fatigue and also aiding in joint stabilization during ground contact (13). Mann and Hagy (13) also reported that the body’s center of gravity is lowered as gait increases with running and sprinting, which increases the amount of hip and knee flexion, and increases the workload on the rectus femoris. Thus, runners should consider inclusion of parallel squats in their concurrent training regimens to better activate the rectus femoris which could help to avoid poor running technique and premature fatigue (8).

By increasing erector spinae activation during the parallel squat, runners could benefit by conditioning the back extensors to help maintain more upright postures while running, possibly improving their ventilatory capacity (9). In contrast, weakness in erector spinae can exacerbate the risk of hamstring injury during the terminal swing phase of running by contributing to excessive trunk flexion during knee extension (9). Furthermore, Behm et al. (1) reported that running activates the lumbar erector spinae more than prone isometric back extensions. It is evident that conditioning the core musculature in runners is vital for maximizing performance and preventing injury. The results of the present investigation demonstrates that parallel squats could be an important component of a runner’s training program for conditioning the muscles of the lower back. Because erector spinae activity was increased during the parallel squat, although using less resistance, it is likely the safer and more efficient option for targeting these important postural stabilizers in runners.

Given the increased amount of knee flexion necessary to assume the parallel position and the increased hip extension necessary to return from the deeper parallel squat, it was hypothesized that biceps femoris activity would be significantly higher when compared with the shallow partial squat. In contrast to our hypothesis, biceps femoris activity was not increased during the parallel squat. This is consistent with Caterisano et al. (3) who demonstrated that male weightlifters did not increase hamstring activity during the parallel squat when compared with the partial squat. Other studies have also previously demonstrated that the squat does not provide a great training stimulus to the hamstrings (4,22). Given that the squat is not the best hamstring exercise, coupled with the fact that the cross-country runners in this study had deficient hamstrings-to-quadriceps ratios, an increased focus on hamstring training is advised. Recently, Ebben (4) determined that Russian curls and seated leg curls yielded the best hamstrings-to-quadriceps muscle activation ratios, while the squat yielded the lowest ratio of 6 exercises that were tested. Thus, cross-country runners and coaches should consider an increased focus on resistance exercises that specifically target the hamstrings and not limit lower body resistance training to the squat exercise.

Finally, the gastrocnemius was tested because of its multiarticular nature and key role in one’s ability to run, but its activity was not significantly different between the parallel and partial squats. However, the parallel squat increases the extent of dorsiflexion at the ankle joint, which may help to maintain mobility while running on uneven surfaces, which is a common requirement for cross-country runners. The parallel squat condition in the present study yielded a similar degree of dorsiflexion to that required while running (14). An optimal length–tension relationship of the gastrocnemius can also be achieved by increasing the amount of dorsiflexion at the ankle joint, which is necessary for a forceful push-off when running. Ultimately, it remains possible that, if a greater number of repetitions were performed during the squat exercise until approaching volitional fatigue, a significant change in gastrocnemius muscle activity between conditions could become evident (8).

In the current study, the amount of knee flexion and cadence of the squat were the 2 components of the subjects’ squat exercise performances controlled by the researchers. This was done to allow the participants to utilize their customary squat exercise techniques. However, the lack of uniformity in relation to the degrees of movement at the hip and ankle could, in some ways, limit the results. Another potential limitation may have occurred due to the use of a metronome with the squat exercises. A set cadence for the timing of each repetition was difficult at first for a few of the participants, but the orientation session was beneficial in minimizing this effect. Finally, we have determined that the parallel squat increases activity of the rectus femoris and lumbar erector spinae, but we have not tested the effects of parallel squats versus partial squats on measures of running performance, such as economy. Future studies may elect to employ a longitudinal protocol in an attempt to determine the effect of partial and parallel squat resistance training on running economy and other measures of endurance and performance.

In summary, the parallel squat can be beneficial for cross-country runners and/or coaches when designing a resistance-based training program to maximize development of the rectus femoris and erector spinae. The rectus femoris is most active during the last 20% of the swing phase during the running gait pattern in order to stabilize the knee (13), but has been found to fatigue early, which could potentially decrease performance and increase the likelihood of injury (8). Thus, properly conditioning the rectus femoris is an important consideration for runners. The erector spinae are known to aid in stabilization of the spine through trunk extension (9). Therefore, strengthening the erector spinae is imperative for improving running posture, for lowering the risk of hamstring strains or tears (due to excessive trunk flexion coupled with knee extension), and for possibly enhancing ventilatory function (9). The parallel squat seems to be more effective than the partial squat at targeting sport-specific muscles for running (i.e. rectus femoris and erector spinae) while requiring a lighter load and larger range of motion.

Practical Applications

Based on the deficient hamstrings-to-quadriceps ratios in our participants, isokinetic testing may be warranted for cross-country runners. As a result of these findings, coaches should develop appropriate strategies to promote balanced muscular development of the thigh and lessen the risk of athletic injuries, such as those to the anterior cruciate ligament of the knee. A recent article by Ebben (4) provides some practical considerations for coaches and athletes seeking to specifically improve hamstring development and decrease the risk of injury.

For cross-country runners, the parallel squat seems to be superior to the partial squat for 3 practical reasons: (a) the range of motion during the exercise is greater; (b) lighter loads are required during the exercise; and (c) activation of the rectus femoris and lumbar erector spinae is enhanced. These inherent qualities may help cross-country runners to remain flexible, prevent back or knee injuries, and improve the strength and endurance of sport-specific musculature (i.e. rectus femoris and lumbar erector spinae). In contrast, partial squats may contribute to limited range of motion, excessive loads, and limited muscle activation. Based on our findings, coaches should consider including parallel squats as part of a runner’s concurrent training program.


The authors are grateful to the Saint Francis University cross-country runners who volunteered for testing, and appreciate the assistance of Gordon Thomson, Jesse Webber, and Dr John Miko during data collection and analyses. The authors also thank Dr Kay Malek and the Saint Francis University Department of Physical Therapy for providing funding to support this project. Finally, the authors appreciate the contributions of Freya Moran, Jenna Edwards, and Marie Mets towards revisions of this manuscript.


1. Behm DG, Cappa D, Power GA. Trunk muscle activation during moderate- and high-intensity running. Appl Physiol Nutr Metab 34: 1008–1016, 2009.
2. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 14: 377–381, 1982.
3. Caterisano A, Moss RF, Pellinger TK, Woodruff K, Lewis VC, Booth W, Khadra T. The effect of back squat depth on the EMG activity of 4 superficial hip and thigh muscles. J Strength Cond Res 16: 428–432, 2002.
4. Ebben WP. Hamstring activation during lower body resistance training exercises. Int J Sports Physiol Perform 4: 84–96, 2009.
5. Escamilla RF. Knee biomechanics of the dynamic squat exercise. Med Sci Sports Exerc 33: 127–141, 2001.
6. Esteve-Lanao J, Rhea MR, Fleck SJ, Lucia A. Running-specific, periodized strength training attenuates loss of stride length during intense endurance running. J Strength Cond Res 22: 1176–1183, 2008.
7. Hamlyn N, Behm DG, Young WB. Trunk muscle activation during dynamic weight-training exercises and isometric instability activities. J Strength Cond Res 21: 1108–1112, 2007.
8. Hanon C, Thepaut-Mathieu C, Vandewalle H. Determination of muscular fatigue in elite runners. Eur J Appl Physiol 94: 118–125, 2005.
9. Hoskins W, Pollard H. The management of hamstring injury—Part 1: Issues in diagnosis. Man Ther 10: 96–107, 2005.
10. Isear JA Jr, Erickson JC, Worrell TW. EMG analysis of lower extremity muscle recruitment patterns during an unloaded squat. Med Sci Sports Exerc 29: 532–539, 1997.
11. Johnston RE, Quinn TJ, Kertzer R, Vroman NB. Strength training in female distance runners: Impact on running economy. J Strength Cond Res 11: 224–229, 1997.
12. Kelly CM, Burnett AF, Newton MJ. The effect of strength training on three-kilometer performance in recreational women endurance runners. J Strength Cond Res 22: 396–403, 2008.
13. Mann RA, Hagy J. Biomechanics of walking, running, and sprinting. Am J Sports Med 8: 345–350, 1980.
14. Mann RA, Moran GT, Dougherty SE. Comparative electromyography of the lower extremity in jogging, running, and sprinting. Am J Sports Med 14: 501–510, 1986.
15. Mikkola J, Rusko H, Nummela A, Pollari T, Hakkinen K. Concurrent endurance and explosive type strength training improves neuromuscular and anaerobic characteristics in young distance runners. Int J Sports Med 28: 602–611, 2007.
16. Millet GP, Jaouen B, Borrani F, Candau R. Effects of concurrent endurance and strength training on running economy and.VO(2) kinetics. Med Sci Sports Exerc 34: 1351–1359, 2002.
17. Nummela AT, Paavolainen LM, Sharwood KA, Lambert MI, Noakes TD, Rusko HK. Neuromuscular factors determining 5 km running performance and running economy in well-trained athletes. Eur J Appl Physiol 97: 1–8, 2006.
18. Nuzzo JL, McCaulley GO, Cormie P, Cavill MJ, McBride JM. Trunk muscle activity during stability ball and free weight exercises. J Strength Cond Res 22: 95–102, 2008.
19. Paavolainen L, Hakkinen K, Hamalainen I, Nummela A, Rusko H. Explosive-strength training improves 5-km running time by improving running economy and muscle power. J Appl Physiol 86: 1527–1533, 1999.
20. Sloniger MA, Cureton KJ, Prior BM, Evans EM. Lower extremity muscle activation during horizontal and uphill running. J Appl Physiol 83: 2073–2079, 1997.
21. Storen O, Helgerud J, Stoa EM, Hoff J. Maximal strength training improves running economy in distance runners. Med Sci Sports Exerc 40: 1087–1092, 2008.
22. Wright GA, DeLong TH, Gehlsen G. Electromyographic activity of the hamstrings during performance of the leg curl, stiff-leg deadlift, and back squat movements. J Strength Cond Res 13: 168–174, 1999.
23. Yamamoto LM, Lopez RM, Klau JF, Casa DJ, Kraemer WJ, Maresh CM. The effects of resistance training on endurance distance running performance among highly trained runners: a systematic review. J Strength Cond Res 22: 2036–2044, 2008.

concurrent training; EMG; resistance training; running; strength

Copyright © 2013 by the National Strength & Conditioning Association.