When designing a resistance training (RT) program, numerous factors have to be taken into account for maximum performance benefit. Key variables to consider are rest period within or between sets (7,14), load, volume and intensity (35), type of muscle contraction (i.e., concentric, eccentric, or stretch-shortening cycle) (29,33), exercise order (3), and the velocity at which the load is lifted (10,12,17,30). Relationships among these variables and their appropriate management determines the type of adaptation and therefore, training effects. These training effects may be structural effects such as hypertrophy, or neuromuscular effects such as increases in force production in the form of strength and power. However, to ensure quality monitoring of the training process, tools are required to assist and quantify exercise intensity, both objectively (e.g., using assessment devices) and subjectively (e.g., using the rating of perceived exertion [RPE]). For this purpose, linear position transducers (LPT) and accelerometers are excellent tools to quantify training intensity in a given session. In addition, RPE scales, such as the OMNI-RES and the Borg CR-10, provide reliable and valid estimates of exercise intensity. In effect, good validity (6,27,28) and reliability (6,32) coefficients have been reported for these scales when used to monitor the intensity of training sessions involving external resistances in different populations.
Several RPE scales have been developed and validated for many different types of exercise such as those using resistance bands (4,5) or elliptical ergometers (20), kayaking (23), and team gym (21). In physiological terms, a clear relationship exists between perceived exertion and muscle activity for a given intensity, i.e., a greater subjective RPE score translates to more intense activation of the muscles involved (8). The main benefit of the use of effort perception scales is their good capacity to monitor training sessions. In a study examining the use of OMNI-RES to control exercise intensity during a training session, Naclerio et al. (22) detected a relationship between RPE and load or mechanical power. These authors found OMNI-RES to be a useful tool to assess the initial 1-repetition maximum (1RM) percentage and power fluctuations during an exercise set until fatigue in the bench press, and described the RPE as useful to control RT intensity. However, the main limitation of the study by Naclerio et al. (22) was that only the bench press was used. In another study, Day et al. (6) analyzed the reliability of RPE (CR-10) used between sessions to quantify the intensity (high [H], moderate [M] or low [L]) of the exercises back squat, bench press, overhead press, biceps curl, and triceps pushdown. Rating of perceived exertion was examined with both immediately after the end of each exercise and 30 minutes after the end of the session. Rating of perceived exertion scores showed a high intersession relative (intraclass correlation coefficient [ICC] = 0.88; 95% confidence interval = 0.70–0.96) and absolute reliability (coefficient of variation [CV] = 14%). Accordingly, these authors concluded that the RPE is a reliable method to quantify intensities between workouts involving different exercises.
Several factors may be used to calculate training intensity during an RT program; e.g., a given percentage of the 1RM (e.g., 80% 1RM) or a certain number of repetitions of the 1RM (e.g., 5RM). However, there are other ways to calculate RT intensities such as velocity of movement. One of the most important variables to consider when designing an optimal RT program is the execution velocity (17,25,26,30). Thus, the strength and power output gains produced through an RT program will be determined in part by the velocity at which the load is displaced (9). In a recent study by González-Badillo et al. (11), strength gains in response to 2 identical training programs were compared, the only difference being that one program was completed at the highest velocity of movement, whereas the other was conducted at half this velocity. Premeasurements and postmeasurements indicated significant strength gains in both training groups; yet, greater gains were produced in the maximum movement velocity group (18.2% vs. 9.7%). Thus, for a given load, the velocity at which it is displaced will determine training effects. In another recent study, Bautista et al. (2) validated a rating of perceived velocity scale applied to bench press exercise. This scale (Figure 2) consists of a range of numerical velocity values (0.10–1.60 m·s−1) and 5 qualitative velocity values (i.e., very fast, fast, power zone, slow, and very slow). Good partial (ICC = 0.88–0.96) and absolute reliability (standard error of measurement [SEM] = 0.07–0.13 m·s−1) results were obtained. However, to date, no study has examined the validity of the use of this scale of perceived velocity (SPV) in the full back squat. Compared with bench press, full squats require intense body mass mobilization, thus perception of velocity could vary.
This study was designed to assess the concurrent validity of the SPV for full back squat exercise in young elite skiers. The hypotheses tested were as follows: (a) whether the real velocity of execution (Velreal) is positively and linearly related to the perceived velocity (Velscale) determined using the perception velocity scale for each of the several intensities (20, 30, 40, 50, 60, and 70% of the estimated 1RM [1RMest]), (b) whether Velreal and Velscale are positively and linearly related for the whole range of intensities tested in a 2-session trial, and (c) whether the SPV is sufficiently sensitive to detect a small velocity change (e.g., for a 10% change in 1RM).
Experimental Approach to the Problem
A descriptive-correlational study was designed to validate the SPV for the concentric phase of the full back squat. The main purpose of the scale is to provide a method of quantifying resistance exercise intensity based on psychophysical perception. The validation process was performed in 2 stages, a familiarization and a final validation stage (Figure 1). In the first stage, subjects received theoretical information about the use of the scale and performed an incremental load test. Next, the 1RM was estimated from the load-velocity relationship (see Procedures for more detail). In the validation stage, the real velocity recorded in a test was compared with the subject's perceived velocity.
The subjects recruited for the study were 11 young international level skiers (n = 8 boys and 3 girls). All skiers belonged to the Andalusia Federation of Winter Sports. Mean ± standard deviation (SD) values of age, weight, height, and the ratio 1RMest/body weight were 15.4 ± 1.12 years, 63.6 ± 11.56 kg, 166.8 ± 8.83 cm, and 1.61 ± 0.40, respectively. All participants had at least 1-year of experience in RT involving the full back squat. The parents or guardians of the athletes signed an informed consent form outlining the risks and benefits of the study. The study protocol received University of Granada review board approval.
Tests were conducted at the High Performance Centre in Sierra Nevada (Granada, Spain) located 2,300 m above sea level. The experimental design consisted of 2 stages in line with the protocol conducted by Bautista et al. (2) (Figure 1). In the first stage, the subjects became familiarized with the SPV and undertook an incremental load test to determine 1RMest. In the second stage, the scale was validated in 2 sessions (on different days) by testing a range of work intensities in random order. The familiarization stage and both sessions of the validation stage were separated by rest periods of at least 72 hours (Figure 2).
The 2 objectives of this stage were to familiarize the athletes with the use of the SPV and to determine their 1RMest for the back squat. A linear velocity system was used (T-Force System; Ergotech, Murcia, Spain) to record the mean velocity attained using each load and the 1RMest variables. It has been recently shown that mean velocity can be use to precisely estimate loading intensity (12). This device consists of a cable extension connected to the barbell. Information on the movement velocity of the barbell is transmitted and transformed from analog-to-digital by a data acquisition board. These data are analyzed by customized software (T-Force: Dynamic Measurement System, version 2.28). Vertical instantaneous velocity was directly sampled by the device at a frequency of 1000 Hz and smoothed using a fourth order low-pass Butterworth filter with no phase shift and a 10 Hz cutoff frequency. Validity and reliability of this device were established by comparing the displacement measurements obtained by this device with a high-precision digital height gauge. The mean relative error in the velocity measurements was found to be <0.25%, whereas displacement was accurate to ±0.5 mm (30). Tests were conducted in a Smith machine (Technogym, Barcelona, Spain). All tests started with a general 15-minute warm-up consisting of runs and jumps, and this was followed by a specific warm-up of 2 sets of 10 repetitions of the back squat lifting a load of 16 kg. After the warm-up and 5 minutes of active recovery (i.e., ankle, knee, and hips mobility exercises), subjects executed the incremental load test. The initial load was 16 kg and this was increased by 20 kg until the displacement velocity of the barbell was 0.60 m·s−1. The subset of girls performed the same incremental load test but with increments of 10 kg. A range of 4–7 sets was performed to complete the full protocol. All participants performed a total of 4 repetitions in each evaluated set. For anchoring during each repetition, each subject was provided with barbell velocity feedback. At the end of each set, the average velocity of the barbell for the 4 repetitions was also provided. To avoid the rebound effect during the change from eccentric vs. concentric phase, the descent phase of the barbell was controlled by the verbal instructions from the observer; “down” (2 seconds), “stop” (1 second) and “go,” slightly varying the time interval between the last 2 commands. Next, each subject was encouraged to lift the barbell as quickly as possible during the concentric phase (i.e., explosively). Strong verbal encouragement was used to motivate the participants. Three minutes of rest were allowed between sets, except for the last 2 intensities when the rest period was 5 minutes.
For the validation protocol, subjects undertook the general and exercise-specific warm-up as described above, and then performed sets of the back squat at 20, 30, 40, 50, 60, and 70% of 1RMest in random order. To avoid the athlete identifying each intensity, a hat and ski glasses were worn to prevent lateral vision and blind the athlete to the load on the barbell. Four repetitions at each intensity were performed with 5 minutes of rest between sets. Velreal was established as the average velocity recorded for the 4 repetitions.
When subjects completed each set, they pointed on a printed version of the SPV to the speed at which they felt the load had been displaced. This velocity was recorded as Velscale.
All data are expressed as the mean ± SD. To identify outliers the data were transformed to typified Z scores. Extreme outliers were not observed. The normality of the data was confirmed using the Shapiro-Wilk test. Reliability in the 2 validation stage test sessions was measured using several statistical indices. Partial intersession reliability of variables was tested using the Intraclass Correlation Index model 2.1 (ICC2.1), whereas absolute reliability was calculated using the SEM and CV. The SEM was estimated using the following equation:
, where MSe is the mean square error derived by ANOVA (1,34). CV (%) was calculated as 100 (es − 1) (16). Pearson correlation coefficients were used to determine interrelationships between variables. The strength of correlation were defined as r = 0–0.1 (trivial), r = 0.1–0.3 (small), r = 0.3–0.5 (moderate), r = 0.5–0.7 (large), r = 0.7–0.9 (very large), and r = 0.9–1.0 (nearly complete) (15). A simple linear regression analysis was performed using Velreal as the criterion variable and Velscale as the predictor variable. Real and perceived velocities were assessed by 3-way RM ANOVA (scale  × session  × intensity ). When the sphericity assumption was violated, degrees of freedom were corrected using the Greenhouse-Geisser estimation. The Bonferroni post hoc procedure was used to test multiple comparisons. The effect size was calculated through partial eta squared (
). A Bland-Altman plot was constructed for the velocity of execution and analyzed for the presence of heteroscedasticity. Heteroscedasticity was determined by examining the correlation between absolute differences and mean values. Values between 0 and 0.10 were considered homoscedastic, and values >0.10 were taken as heteroscedastic (1). The level of significance was set at p ≤ 0.05. All statistical tests were performed using the software package SPSS version 20 (SPSS, Inc., Chicago, IL, USA).
Descriptive statistics and partial (ICC2.1) and absolute (CV and SEM) reliability of Velreal and Velscale across all intensities tested (20, 30, 40, 50, 60, and 70% 1RMest) in the 2 validation sessions are provided in Table 1.
Table 2 shows summarized Pearson correlation coefficients between Velreal and Velscale recorded at each intensity in the 2 validation sessions.
Figure 2 shows the linear regression analysis between Velreal and Velscale (all intensities analyzed together) recorded in the 2 validation sessions.
Real vs. Scale Velocity
The Mauchly test indicated that the sphericity assumption was violated for the main effect of intensity and the interaction session × intensity (χ2  = 47, p = 0.001 and χ2  = 30.77, p = 0.001, respectively). Accordingly, the degrees of freedom were corrected using the Greenhouse-Geisser estimation (ε = 0.294 and 0.352, respectively). Repeated measures ANOVA revealed significant differences for the main effect of intensity (F1.47,14.72 = 92; p = 0.0001,
= 0.90). Mean ± SE were 1.37 ± 0.06, 1.27 ± 0.03, 1.15 ± 0.03, 1.06 ± 0.02, 0.86 ± 0.01 and 0.72 ± 0.02 m·s−1 for 20, 30, 40, 50, 60, and 70% of 1RMest, respectively. Bonferroni post hoc comparisons indicated significant differences (p ≤ 0.05) for all possible comparisons. Furthermore, RM ANOVA revealed significant differences (F1,10 = 5.88; p = 0.0001,
= 0.37) for the interaction scale × intensity. Figure 4 shows all the post hoc comparisons.
In contrast, no significant differences were detected for the main effects of scale (F1,10 = 4.91; p = 0.051;
= 0.33), session (F1,10 = 0.41; p = 0.535;
= 0.04), session × intensity (F1.76,17.59 = 0.40; p = 0.648;
= 0.04), scale × session (F1,10 = 0.15; p = 0.705;
= 0.02) and scale × session × intensity (F5,50 = 1.56; p = 0.189;
For the scale factor, the mean ± SE of the displacement velocities were 1.08 ± 0.22 (1.03–1.13) m·s−1 and 1.07 ± 0.22 (1.02–1.12) m·s−1 for Velreal and Velscale, respectively. For the factor session, the mean ± SE of the displacement velocities were 1.07 ± 0.22 (1.02–1.12) m·s−1 and 1.08 ± 0.25 (1.02–1.13) m·s−1, for sessions 1 and 2, respectively. For the interaction scale × session, the mean ± SE of Velreal were 1.07 ± 0.23 (1.02–1.12) m·s−1 and 1.08 ± 0.23 (1.03–1.14) m·s−1, for sessions 1 and 2, respectively. For Velscale, the mean ± SE were 1.06 ± 0.02 (1.01–1.11) m·s−1 and 1.07 ± 0.03 (1.02–1.13) m·s−1, for sessions 1 and 2, respectively. Descriptive and inferential results of specific comparisons of scale × session × intensity are shown in Table 1.
The Shapiro-Wilk test showed that the velocity differences were normally distributed. No presence of heteroscedasticity was observed when absolute differences were plotted against an individual's mean Velreal values (R2 = 0.002; r = 0.05). The mean ± SD of the velocity differences was −0.011 ± 0.045 m·s−1. The bias and random error was −0.10 ± 0.09 m·s−1 (Figures 3–5).
In the present study, the concurrent validity of the use of SPV to monitor training intensity in the full back squat was examined. In a previous study (2), this scale was validated for use in the bench press. To validate the scale for back squat exercise, 11 international young elite skiers were tested. As a criterion variable to validate the scale, Velreal execution was measured using an LPT. In the validation process, it was tested whether (a) Velreal was positively and linearly related to Velscale for each of several intensities (20, 30, 40, 50, 60, and 70% of 1RMest) (Table 2), (b) Velreal and Velscale was distributed positively and linearly for the whole range of intensities tested in the 2-session trial (Figure 3), and (c) the SPV was sensitive enough to detect a small velocity/intensity change (Figures 3 and 4). The findings of this study support each of these hypotheses.
In this study, Velreal was used as the reference (i.e., gold standard) to validate the SPV. The Pearson correlation coefficients between Velreal and Velscale at each intensity ranged from large to nearly complete (Table 2). Several authors (11,25,30) have proposed the velocity of execution as an effective variable to quantify exercise intensity. The results obtained here indicate that perceived velocity scores correlate linearly and positively with the real velocity for each of a wide range of intensities, showing excellent Pearson correlation coefficients. It is important to highlight that despite obtaining low partial reliability (ICC2.1) at high intensities (50%–70% of 1RMest) absolute reliability measures (CV and SEM) were adequate. The main explanation for these lower coefficients was the lower within-subject variability in perceived and real velocity. In other words, when all subjects performed the back squat at the same absolute intensity, they accurately rated their perceived velocity. It should be noted that when variability within subjects is low, the ICC is not the best reliability statistic (1).
The validity of SPV was also examined for the whole range of intensities tested (i.e., 20, 30, 40, 50, 60, and 70% of the 1RMest) in both validation sessions. Thus, the analysis of simple linear correlation between Velscale and Velreal returned nearly complete correlation coefficients for both sessions (Figure 3). Participants accurately perceived the velocity for all the intensities analyzed. The RM ANOVA showed no significant differences for the scale effect. This means that regardless of the factors session and intensity, Velreal and Velscale were similar (Velreal vs. Velscale = 1.08 and 1.07 m·s−1, respectively). It should be underscored that for this specific comparison (i.e., the factor scale) p was close to 0.05. A plausible explanation for this is the overestimation and underestimation of the different loads (e.g., 40, 60, and 70% of 1RMest; Figure 4).
The sensitivity of this scale to discriminate intensities depending on the percentage of 1RMest was indicated by the significant differences observed between Velscale and Velreal ascribed to the different intensities, i.e., in the interaction scale × intensity (Figure 4). Participants were able to discriminate between Velscale for all the loads analyzed except 20% vs. 30% of the 1RMest. However, it is noteworthy that this phenomenon occurred for perceived velocity, i.e., when athletes lifted the barbell at 20% or 30% of the 1RMest no significant differences between perceived velocities were produced. Several studies (19,27,28,32) examining the use of the OMNI-RES scale, increased percentage loads were assigned increased RPE scores. The present findings are consistent with the results of these previous studies, in that higher loads lead to a lower perception of velocity. However, no significant differences were detected for the effects of session or the interactions session × intensity, scale × session, or scale × session × intensity. In other words, the scale works with the same accuracy regardless of the session or the intensity of exercise. However, it is important to note that when the relative intensity was approximately 40% of 1RMest, the perceived execution velocity tended to be overestimated. In contrast, for the higher relative intensities (i.e., 60% and 70% 1RMest) the perceived velocity was underestimated (Figure 4). These findings are consistent with those of the study by Bautista et al. (2) in which the perceived velocity tended to be higher than the actual velocity when light loads (>1.00 m·s−1) were lifted in the bench press, whereas velocity was underestimated for the heavy loads (<0.40 m·s−1). This observation may be explained by the short familiarization period completed by the athletes. Thus, to accurately perceive the velocity of movement across a wide range of work intensities, it seems that a longer process of familiarization with the use of the scale is required. Notwithstanding, the athletes tested were able to precisely estimate the velocity of execution at intermediate intensities of exercise, as indicated by the results obtained (Table 1). It should also be mentioned that these athletes were accustomed to this exercise modality and their technical execution of the back squat was near perfect. According to the present results, the use of the SPV in a population of young elite skiers resulted in low systematic biases and small random errors (Figure 5). The limits of agreement reported in this study quantify the reliability of Velreal measures. The range of random error for Velreal could be 0.09 m·s−1.
To the authors' knowledge, this is the first study to validate a scale of perceived velocity in the full back squat. In previous work, a moderate-high relationship was detected between the real and perceived velocity of bench press exercise using the SPV (2). Unlike when other subjective exertion scales have been validated (13,19,23,27), the criterion variable used in the present study was the actual execution velocity measured by LPT. The perception of velocity by subjects was highly accurate, indicating it is possible to discriminate exercise intensity using execution velocity in strength training. The present study has several limitations including the small population size of 11 elite athletes. Future studies should examine the influence of fatigue on perceived velocity values at different intensities in different training populations. In addition, SPV should be tested in a full session to assess how fatigue affects velocity perception. Finally, despite the known effects of the menstrual cycle phase (i.e., follicular and luteal) on responses to strength or resistance exercise (18,24,31), we could not control for this factor given that there were only 3 girls in our study population.
Perceived velocities in the full back squat were positively and linearly related to actual execution velocities in both relative and absolute terms. In addition, the results obtained suggest the sensitivity of the scale to differentiate perceived velocities across the full range of loads tested. Collectively, the present findings support the concurrent validity of perceived velocity to monitor the intensity of lower-body strength training.
The main practical application of the present findings is that training intensity can be quantified according to the perceived velocity of exercise execution without the need for an LPT. Although the present authors feel the use of an LPT is essential to monitor the effects of training, they are also aware that training tools such as LPT are not always available for regular use in large groups of athletes such as sport teams. The scale proposed here offers a simple and effective method to ensure training intensity is appropriate for the desired outcome, as it is not always possible to use expensive equipment such as LPT. Thus, regardless of when or where an athlete needs to train, the physical trainer can prescribe and monitor a strength training program. This will be especially useful when traveling for national or international competitions. Finally, based on our experience, the fact that the athlete knows the velocity of movement were mean that training is likely to help in the process of self-knowledge, self-control, and motivation.
We thank our participants and the assistance of Jesús Pérez in data collection. Also, we thank Federación Andaluza de Deportes de Invierno (FADI). This study received no external funding. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Atkinson G, Nevill AM. Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Med 26: 217–238, 1998.
2. Bautista IJ, Chirosa IJ, Chirosa LJ, Martín I, González A, Robertson RJ. Development and validity of a scale of perception of velocity in resistance exercise. J Sports Sci Med 13: 542–549, 2014.
3. Bellezza PA, Hall EE, Miller PC, Bixby WR. The influence of exercise order on blood lactate, perceptual, and affective responses. J Strength Cond Res 23: 203–208, 2009.
4. Colado JC, García-Masso X, Triplett N, Calatayud J, Flandez J, Behm D, Rogers ME. Construct and concurrent validation of a new resistance intensity scale for exercise with thera-bands elastic bands. J Sports Sci Med 13: 758–766, 2014.
5. Colado JC, Garcia-Masso X, Triplett TN, Flandez J, Borreani S, Tella V. Concurrent validation of the omni-resistance exercise scale of perceived exertion with thera-band resistance bands. J Strength Cond Res 26: 3018–3024, 2012.
6. Day ML, McGuigan MR, Brice G, Foster C. Monitoring exercise intensity during resistance training
using the session RPE scale. J Strength Cond Res 18: 353–358, 2004.
7. de Salles BF, Simao R, Miranda F, Novaes JDS, Lemos A, Willardson JM. Rest interval between sets in strength training
. Sports Med 39: 765–777, 2009.
8. Duncan MJ, Al-Nakeeb Y, Scurr J. Perceived exertion is related to muscle activity during leg extension exercise. Res Sports Med 14: 179–189, 2006.
9. Fielding RA, LeBrasseur NK, Cuoco A, Bean J, Mizer K, Singh MAF. High-velocity resistance training
increases skeletal muscle peak power in older women. J Am Geriatr Soc 50: 655–662, 2002.
10. Gonzalez-Badillo JJ, Marques MC, Sánchez-Medina L. The importance of movement velocity as a measure to control resistance training
intensity. J Hum Kinet 29: 15–29, 2011.
11. González-Badillo JJ, Rodríguez-Rosell D, Sánchez-Medina L, Gorostiaga E, Pareja-Blanco F. Maximal intended velocity training induces greater gains in bench press performance than deliberately slower half-velocity training. Eur J Sport Sci 15: 1–10, 2014.
12. Gonzalez-Badillo JJ, Sánchez-Medina L. Movement velocity as a measure of loading intensity in resistance training
. Int J Sports Med 31: 347–352, 2010.
13. Hackett DA, Johnson NA, Halaki M, Chow CM. A novel scale to assess resitance-exercise effort. J Sport Sci 30: 1405–1413, 2012.
14. Haff GG, Whitley A, McCoy LB, O'Bryant HS, Kilgore JL, Haff EE, Stone MH. Effects of different set configurations on barbell velocity and displacement during a clean pull. J Strength Cond Res 17: 95–103, 2003.
17. Izquierdo M, Gonzalez-Badillo JJ, Hakkinen K, Ibanez J, Kraemer WJ, Altadill A, Gorostiaga EM. Effect of loading on unintentional lifting velocity declines during single sets of repetitions to failure during upper and lower extremity muscle actions. Int J Sports Med 27: 718–724, 2006.
18. Janse de Jonge XAK. Effects of the menstrual cycle on exercise performance. Sprots Med 33: 833–851, 2003.
19. Lagally KM, Robertson RJ. Construct validity of the OMNI resistance exercise scale. J Strength Cond Res 20: 252–256, 2006.
20. Mays RJ, Cuss FL, Schafer MA, Kim KH, Nagle-Stilley EF, Robertson RJ. Validation of adult omni perceived exertion scales for elliptical ergometry. Percept Mot Skills 111: 848–862, 2010.
21. Minganti C, Capranica L, Meeusen R, Amici S, Piacentini MF. The validity of session rating of perceived exertion method for quantifying training load in teamgym. J Strength Cond Res 24: 3063–3068, 2010.
22. Naclerio F, Rodriguez-Romo G, Barriopedro-Moro MI, Jimenez A, Alvar BA, Triplett NT. Control of resistance training
intensity by the omni perceived exertion scale. J Strength Cond Res 25: 1879–1888, 2011.
23. Nakamura FY, Perandini LA, Okuno NM, Borges TO, Bertuzzi RM, Robertson RJ. Construct and concurrent validation of omni-kayak rating of perceived exertion scale. Percept Mot Skills 108: 744–758, 2009.
24. Oosthuyse T, Bosch AN. The effect of the menstrual cycle on exercise metabolism. Implications for exercise performance in eumenorrhoeic woman. Sport Med 40: 207–227, 2010.
25. Pereira MIR, Gomes ASC. Movement velocity in resitance training. Sports Med 33: 427–438, 2003.
26. Ramirez JM, Núñez VM, Lancho C, Poblador MS, Lancho JL. Velocity based training of lower limb to improve absolute and relative power outputs in concentric phase of half-squat in soccer players. J Strength Cond Res 29: 3084–3088, 2015.
27. Robertson RJ, Goss FL, Andreacci JL, Dube JJ, Rutkowski JJ, Frazee KM, Snee BM. Validation of the children's OMNI-resistance exercise scale of perceived exertion. Med Sci Sports Exerc 37: 819–826, 2005.
28. Robertson RJ, Goss FL, Rutkowski J, Lenz B, Dixon C, Timmer J, Andreacci J. Concurrent validation of the OMNI perceived exertion scale for resistance exercise. Med Sci Sports Exerc 35: 333–341, 2003.
29. Roig M, O'Brien K, Kirk G, Murray R, McKinnon P, Shadgan B, Reid WD. The effects of eccentric versus concentric resistance training
on muscle strength and mass in healthy adults: A systematic review with meta-analysis. Br J Sports Med 43: 556–568, 2009.
30. Sanchez-Medina L, Gonzalez-Badillo JJ. Velocity loss as an indicator of neuromuscular fatigue during resistance training
. Med Sci Sports Exerc 43: 1725–1734, 2011.
31. Sung E, Han A, Hinrichs T, Vorgerd M, Manchado C, Platen P. Effects of follicular versus luteal phase-based strength training
in young women. Springerplus 3: 888, 2014.
32. Tiggemann CL, Korzenowski AL, Brentano MA, Tartaruga MP, Alberton CL, Kruel LFM. Perceived exertion in different strength exercise loads in sedentary, active, and trained adults. J Strength Cond Res 24: 2032–2041, 2010.
33. Vikne H, Refsnes PE, Ekmark M, Medbo JI, Gundersen V, Gundersen K. Muscular performance after concentric and eccentric exercise in trained men. Med Sci Sports Exerc 38: 1770–1781, 2006.
34. Weir J. Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM. J Strength Cond Res 19: 231–240, 2005.
35. Wernbom M, Augustsson J, Thomeé R. The influence of frequency, intensity, volume and mode of strength training
on whole muscle cross-sectional area in humans. Sports Med 37: 225–264, 2007.