The Relationship between Anthropometric Variables and Isokinetic Strength
in a Women’s Collegiate Soccer Team
Paul A. Burkett, Shawn D. Felton,
Mitchell
L.
Cordova,
FACSM
Sports Medicine Research Laboratory, Department of Rehabilitation Sciences, Florida Gulf Coast University, Fort Myers, FL USA
Abstract
Knee isokinetic strength has been reported to be correlated with body fat, fat free mass, and BMI in
college students. It is unclear if these anthropometric variables are related to isokinetic performance
in relatively homogeneous groups of athletes. PURPOSE: To investigate the relationship between
common anthropometrics and average torque production of the knee extensors in a women’s
collegiate soccer team. METHODS: Twenty-eight healthy female collegiate soccer athletes aged
18-22 years participated. The participants had no history of significant lower leg injuries. Participants
were screened using standard anthropometric measurements that included: height, weight, and
skinfold measures of the triceps, suprailiac, and thigh areas. The measurements allowed for
calculation of the BMI, lean body mass, lean body mass index (LBMI), and body fat percentage.
Isokinetic strength of knee flexion and extension was measured through three angular velocities of
60, 180, and 300 degrees/sec. The relationships between the anthropometric measurements (height,
weight, BMI, lean body mass, LBMI, body fat percentage) and average peak torques at the three
angular velocities were assessed utilizing hierarchal linear regression and bivariate correlation
coefficients. RESULTS: Hierarchal linear regression revealed a significant relationship for average
peak torque at 300 degrees/sec with BMI, lean body mass index, and body fat as the predictors, (F
(3, 24)=4.0, P = .019), . Further analysis utilizing Pearson’s bivariate correlation coefficient matrix
found moderate correlations between average peak torque and BMI (.408 to .557; P<.05), lean body
mass (.404 to .425; P<.05), and LBMI (.376 to .413; P<.05). CONCLUSION: The results found in this
athletic population differ from previous research involving physically active non-athletes of a similar
age. While anthropometric measures have been reported to be related to isokinetic knee strength in
non-athletes, in this athletic population the relationship varied depending on the angular velocity. The
results suggest that anthropometric measurements such as height, weight, BMI, lean body mass,
LBMI, and body fat percentage may not be strong predictors of isokinetic knee muscle strength
across angular velocities in an athletic population.
Introduction
Isokinetic testing is widely utilized by clinicians to evaluate lower extremity
strength and as a tool to assist in treatment decisions. 1 Previous
research has demonstrated its efficacy and reliability in muscle training
and clinical evaluation of muscular performance. 2 Research has also
suggested that isokinetic force production is associated with sports
related performance. For example, It has been suggested that isokinetic
force production in the legs could be used to evaluate and predict on-ice
skating speed and skating power in men’s intercollegiate ice hockey
players. 3 Prior research also has investigated the relationship between
strength output and anthropometric measurements such as BMI and body
fat percentage. The conclusions concerning these relationships have
been inconsistent. Knee isokinetic strength was reported to be
significantly negatively correlated with the percentage of body fat and
positively correlated with fat free mass in college students who were not
competitive athletes. 4 In young male competitive soccer players ages 10
to 17 years, certain anthropometric measures were related to isokinetic
strength for both knee extensors and flexors, with 73-93% of the variance
explained by using combinations of age, body mass, percentage of body
fat, and hours training per week. 5 Body mass was the main independent
variable that explained variance, which was in agreement with a previous
study. 6 Fat free mass, but not percent body fat, was reported to be
related to anaerobic power in male and female elite young wrestlers. 7
However, percent body fat was found to be related to on-ice skating
speed and skating power in men’s intercollegiate ice hockey players. 3 In
contrast, lean body mass, percentage of fat tissue, and percentage of
muscle tissue were reported to be poor predictors of sprinting
performance in well-conditioned males. 8 Percent body fat and percent
skeletal muscle mass were found to be strong predictors of jumping
performance in female non-athletes, but were not as strongly related to
jumping performance in elite female athletes. 9 The inconsistent findings
concerning the relationship between strength output and anthropometric
measurements such as BMI and body fat percentage suggest that the
strength of the relationship may vary depending on the population
studied. The purpose of this study was to investigate the relationship
between common anthropometric variables and average torque
production of the knee extensors measured by isokinetic testing at three
angular velocities in a women’s collegiate soccer team.
Methods
Table. 2 Pearson Correlations Between Anthropometric Measurements and Quadriceps Isokinetic Torque Production
Subjects: Twenty-eight healthy female collegiate soccer athletes (mean: age
19.57, height 166.04 cm ± 20.88 cm, mass 62.10 kg ±1.16 kg ) volunteered
for this study. The anthropometric measurements of interest and isokinetic
strength measures were collected during routine pre-season athlete
evaluations. No informed consent was needed because all athletes signed a
medical release document releasing medical professionals to examine and
share information while protecting their specific anonymity. This study was
approved by the University institutional review board.
Height
Methods: Anthropometric measures were obtained including height, mass,
and skinfold thickness. Skinfold measures were taken at three sites: triceps,
suprailiac, and thigh. The sum of the skinfolds was used to estimate percent
body fat and lean body mass (LBM) using equations specific for gender and
age (2). Body mass was measured to the nearest 0.01 kg with participants
clothed in shorts and tee shirts using digital scales (Healthometer). Height
was measured to the nearest 0.01 cm with participants barefoot using a wallmounted stadiometer (Heightronics, QuickMedical, Issaquah, USA). Body
mass index (BMI, kg/m2) and lean body mass index (LBMI, kg/m2) were
calculated. The athletes participated in a lower extremity warm-up and then
Isokinetic strength of knee flexion and extension was measured through
three angular velocities of 60°, 180°, and 300°/sec. Figure 1.
Weight
Age
Body Fat
-.223
-.223
1
Body Fat
.107
.107
.223
1
BMI
.040
-.199
.282
.342
1
Lean Mass
.226
.413*
-.028
.171
.318
1
LBMI
.220
-.433*
.209
-.194
.821**
.296
1
60 Avg. Peak Torque - R
.566**
.059
-.002
-.061
.362
.425*
.413*
1
180 Avg. Peak Torque - R
.364
.129
.253
.262
.408*
.404*
.265
.791**
1
300 Avg. Peak Torque - R
.143
.157
.195
.192
.105
.280
.011
.520**
.790**
1
60 Avg. Peak Torque - L
.198
.131
.020
-.069
.308
.246
.312
.610**
.497**
.314
1
180 Avg. Peak Torque - L
.138
.107
.374
.115
.449*
.347
.377*
.570**
.650**
.437*
.874**
1
300 Avg. Peak Torque -L
.164
.109
.528**
.332
.557**
.423*
.376*
.512**
.697**
.594**
.623**
.782**
9.32
28
Avg Peak Torque -300 R 47.46
6.63
28
Avg Peak Torque 60 L Avg Peak Torque 180 L 98.57
15.89
28
66.40
8.97
28
Avg Peak Torque -300 L 48.86
5.85
28
0.01
* sig at
0.05
1
Summary
Figure 1. Isokinetic Knee Testing
Table. 3. ANOVA for Predictors of Avg. Peak Torque at 300 degrees/sec
Sum of
Sqaures
Table. 1 Means and SD of Observed Variables
Avg Peak Torque 180 R 66.78
** sig at
This study examined the relationship between common anthropometric
variables and average torque production of the knee extensors measured by
isokinetic testing at three angular velocities in a women’s collegiate soccer
teamThe subjects tested were members of a NCAA Division I women’s soccer
team. This group went on to win the conference regular season and conference
tournament championships, and advance to the second round of the NCAA
national tournament. These outcomes suggest that the subjects were highly
skilled athletes in their sport. Previous studies involving highly skilled female
athletes suggested that measures such as percent body fat 7,9 and skeletal
muscle mass 9 were not strong predictors of anaerobic power and jumping
performance. The same may be true of the relationship between
anthropometric measures and isokinetic knee strength in highly skilled female
athletes. The results found in this group of highly skilled female athletes differ
from previous research involving physically active female non-athletes of a
similar age. Anthropometric measures have been reported to be related to
isokinetic knee strength in physically active female non-athlete. 4 The subjects
in this study represent a more heterogeneous athletic population than found in
the general population.
Model
28
180 Avg. Peak
300 Avg.
Torque -- L Peak Torque
-- L
Discussion
•  The level of significance was established at P <0.05.
16.34
60 Avg. Peak
Torque -- L
Age
•  Hierarchical linear regression analysis was used to test if anthropometric
measurements predicted isokinetic average peak torque at 300°/sec. The
results of the regression indicated the three predictors (BMI, LBMI, & %
Body Fat explained 33.3% of the variance (R2 = .33, F(3, 24) = 3.994, P
= .019. Table 3. However the t-test for each predictor was not significant
due to the highly correlated variables.
101.97
300 Avg. Peak
Torque - R
1
•  Pearson Bivariate Correlations presented in Table. 2
Height (cm) Mass (kg) Age % Body Fat BMI Lean Mass LBMI Avg Peak Torque 60 R 180 Avg. Peak
Torque - R
.525**
Results
N
28
28
28
28
28
28
28
60 Avg. Peak
Torque - R
Weight
•  Hierarchical linear regression utilized for prediction due to the hierarchical
nature of the data set.
Std. Deviation
20.88
1.16
1.23
2.98
1.58
6.12
1.05
LBMI
1
•  Descriptive Statistics Table 1 were calculated for all anthropometric
measurements and average peak torques of the bi-lateral thigh
musculature at 60°, 180°, & 300°/sec.
Mean
166.04
62.10
19.57
24.41
22.25
103.29
16.86
Lean
Mass
Height
Statistical Analysis:
•  A significant moderate positive correlation between average peak torque
and BMI (.408 to .449 P < 0.5 and .557; P <.001), Lean body mass (.404
to .425; P < 0.5), and LBMI (.376 to .413; P <.05)
BMI
1
2
3
Regression
Residual
Total
Regression
Residual
Total
Regression
Residual
Total
285.91
636.96
922.86
304.5
618.37
922.86
307.33
615.53
922.86
df
Mean
Square
1
26
27
2
25
27
3
24
27
285.91
24.5
11.67
.002
152.25
24.74
6.16
.007
102.44
25.65
a. 
Dependent Variable: Avg. Peak Torque 300
b. 
Predictors: (Constant), BMI
c. 
Predictors: (Constant), BMI, LBMI
d. 
Predictors: (Constant): BMI, LBMI, Body Fat
F
3.99
Sig.
.019
•  The results of this study were inconsistent in that the strength of the
relationship varied depending on the angular velocity and limb being tested.
Significant relationships were found only at 300°/sec, with BMI, lean body
mass index, and body fat as the predictors of average peak torque of the
quadricep muscles.
•  The results further suggested anthropometric measurements such as height,
weight, BMI, lean body mass, LBMI, and body fat percentage may not be
strong predictors of isokinetic knee muscle strength across angular velocities
in highly skilled women’s soccer athletes.
References
1. 
Skrzek A, Ignasiak Z, Kozieł S, Sławińska T, Rożek K. Differences in muscle strength depend on age, gender and muscle functions. Isokinetics & Exercise Science. 2012;20(3):
229-235.
2. 
Tsiros MD, Grimshaw PN, Shield AJ, Buckley JD. Test-retest reliability of the Biodex system 4 isokinetic dynamometer for knee strength assessment in paediatric populations. Journal
of Allied Health. 2011;40(3): 115-119.
3. 
Potteiger, JA, Smith, DL, Maier, ML, Foster, TS. Relationship between body composition, leg strength, anaerobic power, and on-ice skating performance in Division 1 men’s hockey
athletes. J Strength Cond Res. 2010;24(7):1755-1762.
4. 
Lue YJ, Chang JJ, Chen HM, Lin RF, Chen SS. Knee isokinetic strength and body fat analysis in university students. Kaohsiung J Med Sci. 2000;16(10):517-24.
5. 
Kellis S, Kellis E, Manou V, Gerodimos V. Prediction of knee extensor and flexor isokinetic strength in young male soccer players. Orthop Sports Phys Ther. 2000;30:693-701.
6. 
Gross MT, Brugnolotti, JC. Relationship between multiple predictor variables and normal Biodex eversion-inversion peak torque and angular work. JOSPT. 1992;15(1):24-31.
7. 
Vardar, SA, Tezel, S, Ozturk, L, Kaya, O. The relationship between body composition and anaerobic performance of elite young wrestlers. J Sports Sci Med. 2007;6(CSSI-2): 34-38.
8. 
Kukolj, M, Ropret R, Ugarkovic, D, Jaric, S. Anthropometric, strength, and power predictors of sprinting performance. J Sports Med Phys Fitness. 1999;39(2):120-122.
9. 
Copic, N, Dopsaj M, Ivanovic, J, Nesic, G, Jarig, S. Body composition and muscle strength predictors of jumping performance: Differences between elite female volleyball competitors
and nontrained individuals. J Strength Cond Res. 2014;28(10):2709-2716.
10.  Jackson, AW, Pollock, ML. Practical assessment of body composition. Phys Sportsmed. 1985; 13(5):76, 80, 82-90