Secondary Logo

Journal Logo

Original Research

Test-Retest Reliability, Criterion-Related Validity, and Minimal Detectable Change of the Illinois Agility Test in Male Team Sport Athletes

Hachana, Younes1; Chaabène, Helmi2; Nabli, Mohamed A.1; Attia, Ahmed1; Moualhi, Jamel1; Farhat, Najiba3; Elloumi, Mohamed4

Author Information
Journal of Strength and Conditioning Research: October 2013 - Volume 27 - Issue 10 - p 2752-2759
doi: 10.1519/JSC.0b013e3182890ac3
  • Free

Abstract

Introduction

The ability to repeat sprint and change direction in response to a stimulus is a major determinant of performance in field sports (30,35). Team sports players need to be proficient multidirectional movers typically in a very small area (6). Communally, this ability is often termed “agility.” In fact, agility comprises both perceptual decision-making process and the outcome of this process, a change of direction (COD) (8).

The COD can be described as a movement where no immediate reaction to a stimulus is required, and thus, the COD is preplanned. Actually, it is well established in the literature that COD ability has been considered as a prerequisite in the majority of field sports and an important performance variable for predicting success (12,15). Accordingly, a major interest exists for developing field tests and specific training programs that can effectively measure and improve COD. The majority of previous field tests used to assess COD are based on COD speed such as the cone T-test (Ttest) (33), the COD IAGT (20), the 505 test (13), the L-run test (30), and the zigzag test (29). Among these tests, the Ttest is the most commonly used and is considered as the gold standard test to assess COD (14,31,32,34). The COD IAGT is also a commonly used test. It has been developed and proposed as a tool for assessing speed of COD (1,13,14,20,25,38). It is a timed task involving straight sprinting and multiple direction changes around obstacles. The advantage of this test compared with the Ttest and the Hexagon test is the incorporation of generic cues that closely replicate the majority of those of the movement patterns of field team sports such as soccer, rugby, and handball. Considering the COD IAGT's portability and its widespread use in strength and conditioning professions, an investigation into the test reproducibility is warranted. Reliability of a test is defined as “the consistency of performance when an athlete performs the test repeatedly.” Although the validity (a test is valid if it measures what it claims to measure) of any speed of COD is somewhat difficult to prove because of the lack of “Gold Standard,” researchers and practitioners should use highly reliable tests, because these are the only tests that can also have high validity (22). An appropriate method for assessing reliability is via the SEM (22,27,36). The SEM is a contemporary tool of reliability analysis and this value represents the variation in subjects' performance from measurement to measurement. Moreover, the SEM is important to establish the usefulness of the test by estimating the smallest worthwhile change (SWC) so that one may determine if a real change has occurred over time for subsequent testing periods. Furthermore, reliability data allow one to calculate the minimal individual change that can be interpreted as real with an acceptable probability level, that is, the minimal detectable change (MDC). Consideration of the MDC of a particular test is of great interest for monitoring the progress of athletes because intertrial variation may incorrectly suggest a change that has not exceeded the threshold of error. To the author's knowledge, only Beekhuizen et al. (4) have attempted to calculate the MDC for a speed COD, that is, the Hexagone speed COD test.

Thus, the aims of this study were therefore to explore and quantify measurement reliability of the COD IAGT, its relationship to the Ttest and to the power of legs evaluated by sprint and vertical jump performances. We hypothesized that the IAGT would provide stable test-retest scores, and low MDC95. It would have a stronger relationship with strength and speed.

Methods

Experimental Approach to the Problem

The ability to speed and to change direction is crucial for achieving high-level performance in team sport athletes. Therefore, COD IAGT represents a commonly used test that has been developed to evaluate an athlete's COD speed. However, this test lacks information about his absolute and relative reliabilities and validity. Then, the purpose of this study was threefold: (a) to establish whether COD IAGT is a reliable protocol, (b) to determine the relationship between COD IAGT with another COD test, which is the Ttest, and (c) to establish the association between and countermovement jump (CMJ) performed with an arm swing and with the 30-m sprint test performance. Reliability and measurement error were analyzed by repeating the testing over 3 sessions, each separated by 1 week and analyzing scores using intraclass correlation coefficients (ICCs), SEMs, and SWCs, and 95% confidence intervals (CIs). The MDC95 was reported in conjunction with the between-day test-retest results. All the procedures for each test were administered by the same individual.

Subjects

One hundred five male sports sciences students volunteered to participate in this study (mean [±SD] age: 20.82 ± 1.31 years; height: 180 ± 7 cm; body mass: 72.33 ± 8.75 kg; fat mass: 20.18%). All the participants were holders in their team and were involved in various team sports (football, rugby, and handball) completing 5–6 training sessions per week (∼11–12 h·wk−1) plus 1 official match in the weekend. They have an average experience of 7.4 ± 3.7 years. None of the participants reported any current or ongoing neuromuscular diseases or musculoskeletal injuries specific to the ankle, knee, or hip joints, and none of them were taking any dietary or performance supplements that might be expected to affect performance during the study. All the athletes were at the competitive phase of their periodization. Before the study commenced, each subject had the risks and benefits of the investigation explained to them and provided written informed consent. The university ethics committee approved the study protocol.

Procedure

The present investigation consisted of 3 separate phases: During the first phase were analyzed the absolute and relative reliabilities of the COD IAGT test in a random group of 89 athletes (mean [±SD] age: 20.33 ± 0.83 years; height: 179 ± 6.3 cm; weight: 73.11 ± 5.65 kg) out of the 105 volunteers. During this session, each subject performed the COD IAGT test twice on separate days, at the same time of the day. On each day, the COD test was performed in triplicate. The best trial was retained for statistical analyses. A minimum of 3 minutes of rest was provided between trials.

During the second phase, we analyzed the criterion-related validity of the COD IAGT by looking for its correlation with the Ttest using the “Pearson” moment correlation in 105 athletes (mean [±SD] age: 20.82 ± 1.31 years; height: 180 ± 7 cm; weight: 72.33 ± 8.75 kg; fat mass: 20.18%). During this phase, performances were assessed in a single session. A minimum of 3 minutes of rest was provided between trials and 30 minutes of rest between tests to reduce the likelihood of fatigue.

In the last phase, we analyzed the relationships between the COD IAGT and acceleration, maximum speed, and the vertical jump performance (peak height for lower limbs power) in 105 athletes (mean [±SD] age: 20.82 ± 1.31 years; height: 180±7 cm; weight: 72.33 ± 8.75 kg). In this phase, all the tests were performed in a single session as follows: 30-m linear sprint, CMJ, and COD IAGT. All tests were performed in triplicate, and a minimum of 3 minutes of rest was allocated between trials, and the best trial was retained for statistical analysis. Approximately, 30 minutes of rest was provided between tests to reduce the likelihood of fatigue. In all sessions, the tests were performed on an indoor synthetic track. All the tests were performed at the same time (8–10 PM) of the day (±1 hour). Before each testing session, the subjects were instructed to follow their normal diet, consume a light meal with a caffeine-free beverage >3 hours before testing, and to avoid vigorous activity for 24 hours before the tests. They were also asked to wear the same training shoes on each testing session so as to negate the effect of different designs of shoe and support they provide on individual performance.

All the 3 phases of our study were preceded by a 15-minute standardized warm-up replicated at the beginning of each testing session. Ambient temperatures of 22 ± 1.2° C and relative humidity of 63 ± 2% have been recorded, using a digital environmental station (VaisalaOyj, Helsinki, Finland).

TTest

The Ttest has been performed according to the protocol proposed by Pauole et al. (33) where the athlete begins with both feet behind the starting line A (Figure 1). At his own discretion, he sprints forward to cone B and touches its base with the right hand. Facing forward and without crossing feet, he shuffles to the left to cone C and touch its base with the left hand. After that, he shuffles to the right to cone D and touches its base with the right hand. He shuffles back to the left to cone B and touches its base. Finally, the athlete runs backward as quickly as possible and returns to line A. The test is repeated if the athlete crosses one foot in front of the other, does not touch the base of the cone and fails to face forward throughout. The best time obtained from 3 trials was retained for analysis. The ICC and the SEM for test-retest reliability for the Ttest were 0.95 (95% CI, 0.81–0.98) and 0.18 seconds, respectively. Performances of the Ttest were recorded using an electronic timing system (Brower Timing System, Salt Lake City, UT, USA) positioned at the start line at a height of 1.00 m.

F1-15
Figure 1:
Schematic representation of the change of direction speed T test.

Illinois Agility Test

The COD IAGT is set up with 4 cones forming the agility area (Figure 2). On command, the athlete sprints 9.20 m, turns, and returns to the starting line. After returning to the starting line, he swerves in and out of 4 markers, completing two 9.20-m sprints to finish the agility course (38). Performances were recorded using an electronic timing system (Brower Timing). The infrared timing gates were positioned at the start and the finish lines at a height of approximately 1.00 m. The best performance of the 3 trials was recorded for statistical analyses.

F2-15
Figure 2:
Schematic representation of the Illinois change of direction speed test.

Sprint Performance

Linear sprinting was evaluated over 30 m. Infrared timing gates (Brower Timing) were placed at the start line at 10 and at 30 m at a height of approximately 1 m. The participants were instructed to run at maximal speed from a 2-point start through the final pair of sensors. The subjects performed 3 test trials with the best performance used for statistical analyses. Acceleration was assessed as the time from 0 to 10 m, whereas maximum speed was assessed as the time from 10 to 30 minutes. The ICC and the SEM of the acceleration phase were 0.89 (95% CI, 0.79–0.94) and 0.03 seconds, respectively, and the ICC and SEM of maximal speed were 0.90 (95% CI, 0.81–0.95) and 0.01 seconds, respectively.

Vertical Jump Performance

Neuromuscular power of the lower extremity was assessed using a CMJ performed with an arm swing (i.e., hands were free to move). The subjects were asked to leave the floor with the knees and ankles extended and land in a similarly extended position. The participants performed 3 trials with the best retained for analysis. Jump height during the CMJ was measured by the Opto-jump system (MicrogateSARL, Italy). It has been established that the Opto-jump system presented a strong concurrent validity when compared with force plate and excellent test-retest reliability for the estimation of vertical jump height (16). The Opto-jump photoelectric cells consist of 2 parallel bars (1 receiver and the other transmitter unit, each measuring 100 × 4 × 3 cm), were positioned at 1 m apart. The transmitter contains 32 light emitting diodes, which are placed 0.3 cm from ground level at 3.12-cm intervals. By his connection to a computer and via the software (Opto-jump Software, version 1.3), this system allows jump quantification (16). The Opto-jump system is based on the measurement of flight time of vertical jumps with an accuracy of 1/1,000 seconds (1 kHz). Then, the jump height was estimated as 9.81 × flight time2/8 (7). Because the calculation of jump height can be affected by body position when assessed using this system, the participants were thoroughly familiarized to the correct technique, and this was strictly monitored during testing. If a trial contained incorrect technique (i.e., if a participant banded their knees before ground contact after the flight phase), the trial was discarded and the participant completed another trial. The CMJ test used in this study had good reliability as indicated by ICC = 0.96 (95% CI, 0.89–0.98) and SEM = 2.9 cm.

Statistical Analyses

Data analyses were performed using statistical software (SPSS version 19.0 for Windows). Normality of distributions was verified using the Kolmogorov-Smirnov test. Systematic bias was investigated using a dependent t-test to evaluate the hypothesis that there was no significant difference between tests and retest sample means. Estimates of effect size were calculated to assess meaningfulness of differences. Effect sizes of >1.2, between 1.2 and 0.6, between 0.6 and 0.2, and 0.2 have been considered as large, moderate, small, and trivial, respectively (21). The relative reliability of the COD IAGT was determined by calculating ICC. Absolute reliability of the COD IAGT was analyzed using the SEM (22,27,36) and 95% limits of agreement (LOAs) (6). To establish the usefulness of the COD IAGT, the SWC was determined (2). The sensitivity of the test was assessed by comparing the SWC and SEM, using the thresholds proposed by Liow and Hopkins (28). If the SEM is smaller than the SWC, the ability of the test to detect a change is “good”; if the SEM equals SWC, then the test is “satisfactory,” but if the SEM is greater than the SWC, then the test is rated as “marginal.” Heteroscedasticity was assessed using a zero-order correlation coefficient between the absolute residuals and predicted scores for each participant. Knowledge of the SEM allows the calculation of the MDC95. The MDC reflects the 95% CI of the difference in score between paired observations, calculated as MDC95 = SEM x2.77 (3,19,27). Data from both COD IAGT and the Ttest were correlated for criterion-related validity of COD IAGT. Pearson's product-moment correlation coefficients were used to examine correlations between variables. The magnitude of the correlations was also determined using the modified scale by Hopkins (21): r < 0.1, trivial; 0.1–0.3, small; 0.3–0.5, moderate; 0.5–0.7, large; 0.7–0.9, very large; >0.9, nearly. Significance for all statistical tests was accepted at p ≤ 0.05. Results are reported as mean ± SD and 95% CI.

Results

The COD IAGT was observed to have acceptable relative and absolute reliability. Residual data for the Illinois test and retest trials comparison were normally distributed (p = 0.36). Mean scores (SD) of the COD IAGT, mean difference ± SDdiff., ICC, and SEM values between test and retest are given in Tables 1 and Table 2. The heteroscedasticity coefficient was r = 0.06 (95% CI, 0.02–0.09; p = 0.51). The mean difference (bias) ± the 95% LOAs was 0.07 ± 0.55 seconds. Heteroscedasticity diminished when test and retest data were Log transformed (r = −0.31 [95% CI, −0.23 to −0.41; p < 0.05]). Log transformed data were normally distributed (p = 0.62). There was no significant bias (dz = 0.01) between Log transformed mean scores for the test (1.213) and the retest (1.211). The mean difference (bias) ± the 95% LOAs of the Log transformed data were of 0.002 ± 0.015 seconds. Taking antilog of these LOAs gave a mean bias of 1.002 with an agreement component of ×/÷ 1.015. Mean difference performance scores at different tests are given in Table 3.

T1-15
Table 1:
Performance characteristics and results of relative reliability of the Illinois agility test for male team sports athletes (n = 89).*
T2-15
Table 2:
Performance characteristics, MDC, and results of absolute reliability of the Illinois agility test for male team sports athletes (n = 89).*
T3-15
Table 3:
Mean performance scores of the Illinois agility test, T test CMJ, acceleration, and speed (n = 105).*

The COD IAGT is significantly correlated to agility Ttest and maximal speed (Table 4). A low correlation was found between the Illinois COD test and acceleration (i.e., 10-m sprint time).

T4-15
Table 4:
Correlation between the Illinois agility test, the T test, the CMJ test, acceleration, and speed (n = 105).*

Discussion

The purpose of this investigation was to assess the reliability and validity of the COD IAGT and to examine the relationships between this test and both sprint and vertical jump performances. The main findings of this study demonstrated that the IAGT is a reliable and valid COD test in male team sport athletes. Likewise, we showed that the COD IAGT performance was significantly associated with CMJ and maximal speed performances but not with acceleration.

Usually, reliability is investigated by 2 common indices: the ICC values and the 95% LOA method. These 2 indices have been considered as the most appropriate for reliability assessment (9–11,17,26). In our study, the ICC across the 2 trials was 0.96 (95% CI, 0.85–0.98). This value was in the same range of relative reliability value indices reported in other agility tests (17,18,33,37).

The relative reliability of the COD IAGT has been also confirmed in our study by the 95% LOAs. In our study, the bias ± the 95% LOA of the COD IAGT was 0.07 ± 0.55 seconds. The antilog of these LOAs could be expressed as the mean bias of 1.002 ×/÷ 1.015. Thus, the 95% of the ratios for the log transformed test score divided by log transformed retest score should be contained between 0.987 (1.002 ÷ 1.015) and 1.017 (1.002 × 1.015). As practical considerations, when an athlete from the experimental group performed 16 seconds on the COD IAGT, on the retest he could perform a score as high as 16 × 1.017 = 16.27 seconds, or as low as 16 × 0.987 = 15.79 seconds (11). According to Atkinson and Nevill (2), it is important to use the MDC95 as a criterion to determine whether a real change has occurred between test and retest. In this study, the MDC95 for the COD IAGT has been of 0.52 seconds. When a change in the COD IAGT-retest score is ±0.52 seconds, a true change can be associated. Consequently, the observed LOA in our study can be considered as acceptable.

We have used a range of other reliability measures in our study, to provide a comparison between previous studies and our results. In fact, the stability reliability of the COD IAGT has been also analyzed in our study by the SEM as recommended by Weir (41) and Atkinson and Nevill (2). The SEM provides useful information to examine the effectiveness of researchers’ and practitioners’ intervention (training, nutrition, etc.) (39). The difference observed between the 2 COD IAGT test trials could be related to the error measurement, which is of (0.19 seconds) 1.72%. Recent studies recommend extending the analyses of absolute reliability by the use of the SWC (23,24). The SWC is defined as the minimal individual change that can be interpreted as real with an acceptable probability level (23). In this context, we have also calculated the likelihood that the true value of estimated difference in the COD IAGT test performance has been substantial (i.e., larger than the SWC). It was found that the SEM of the COD IAGT has been less than its respective SWC (0.19 vs. 0.20), which indicates that the usefulness of this test could be rated as “good” and that the COD IAGT test has a good ability to detect real changes in the ability to change direction in male team sports athletes (Table 2). The results of our study fit with those of previous investigations. Accordingly, all the tests performed to assess COD ability have shown similar reliability (ICC 0.8–0.96; measurement error 1–5%) regardless of their duration (1.65–135 seconds), their complexity (the number of direction) or the type of movements and forces used throughout the test (8).

The correlation analysis established between the COD IAGT and the Ttest revealed a significant correlation (r = 0.31 [95% CI, 0.24–0.39]; p = 0.002). Although the correlation was significant in this study, the coefficient of determination (R2 = 0.07) was considered weak. In such circumstances, this could be because of differences in the direction of force application (in the Ttest the COD is preceded by shuffling movements, which are absent in the COD IAGT) and energetic requirement (the duration of the COD IAGT is approximately 3 times greater than that of the Ttest and the number of changes of direction is 4 times higher in the COD IAGT). However, a defined consensus about the number of changes of direction and energetic requirement that a COD test must contain should be determined to better understand speed COD.

Pearson product-moment correlations have been also calculated between the COD IAGT and vertical jump and straight sprint tests (Table 4). According to Brughelli et al. (8), the study of the relationship between COD and straight running speed and vertical jump tests is of a high importance, because it might explain in part if running speed and vertical jump is a determinant factor or parameter of COD. However, these 3 qualities (running, speed, and jumping) could be evaluated within the same test as there should be a great deal of shared variance between the tests. Consequently, we would assume improving COD ability through improving speed and power of lower limbs. The relationships between the COD IAGT, straight running speed, and vertical jump tests could be described as small to moderate. Our results are consistent with previous studies, which have analyzed the relation between COD tests and vertical jump and straight sprint tests (17,18,29,40,42). Haj Sassi et al. (17) have found that vertical jumps are significantly correlated with repeated agility test consisting of 10 × 20-m maximal running performances (moving in forward, lateral, and backward) with approximately 25-second recovery between each run, whereas Little and Williams (29) have reported a low significant correlation between agility (zigzag test, 20 m) and acceleration (10 m), and maximum speed (20 m). On the other hand, Young et al. (42) have revealed a low and nonsignificant correlation between the CMJ test and 20-m COD test. Similar results have been reported by Webb and Lander (40). They have reported a low and nonsignificant correlation between the “L” COD run test and vertical jump. Recently, Haj Sassi et al. (18) have shown that a modified agility Ttest was not correlated to free CMJ and 10-m straight sprint in male athletes.

Given the variety in effort patterns of COD tests, it is difficult to certainly disentangle if vertical jump and straight sprint are determinants of speed COD. Additionally, it is important to note that correlation analysis is of a limited value in identifying the causal relationship between variables. The strength and the significance of correlation outcomes provide no insight into whether the relationship between 2 variables is causal. Thus, the data of this study highlighted a partial correlation to analyze to which extent the observed significant correlation (−0.39; p < 0.001) observed in our study between speed COD evaluated by the speed COD IAGT and leg power (jump height) is real. When partial correlation controlled for the effect of speed the correlation lost its significance (−0.22; p = 0.06). This implies that measurement of speed, to a large extent, may differentiate the variance in agility and leg power (jump height) profiles between male team sport athletes in our study.

In summary, the COD IAGT is a reliable test for assessing speed COD in male team sport athletes. Its performance is independent of acceleration and power of the legs but significantly related to speed.

Practical Applications

The most important findings of the present study have shown that the COD IAGT is a reliable and valid protocol in male team athletes. The COD IAGT can be routinely included by Sports scientists, strength and conditioning practitioners, and sports coaches within an assessment battery for male team sport athletes to evaluate the athlete's ability to speed and to change direction. Moreover, it is important for coaches administering this test to know that a change of >0.52 seconds occurred, for example, after a training program with male team athletes is necessary to be 95% certain that the change in performance highlights improvements and exceeds measurement error.

Furthers studies are needed to investigate other factors that will affect COD speed performance and to evaluate the contribution of coordination quality in COD speed performance.

Acknowledgments

The authors would like to thank the subjects for their enthusiastic participation.

References

1. Amiri-Khorasani M, Sahebozamani M, Tabrizi OG, Yusof AB. Acute effect of different stretching methods on Illinois agility test in soccer players. J Strength Cond Res 24: 2698–2704, 2010.
2. Atkinson G, Nevill A. Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Med 26: 217–238, 1998.
3. Beckerman H, Roebroeck ME, Ankhorst GJ, Becher JG, Bezemer PD, Verbek AL. Smallest real difference, a link between reproducibility and responsiveness. Qual Life Res 10: 571–578, 2001.
4. Beekhuizen KS, Davis MD, Kolber MJ, Cheng MS. Test-retest reliability and minimal detectable change of the hexagon agility test. J Strength Cond Res 23: 2167–2171, 2009.
5. Bland JM, Altman DG. Comparing two methods of clinical measurements: A personal history. Int J Epidemiol 24: 7–17, 1995.
    6. Bloomfield J, Polman R, O’Donoghue P. Physical demands of different positions in FA premier league soccer. J Sports Sci Med 6: 63–70, 2007.
    7. Bosco C, Luhtanen P, Komi PV. A simple method for measurement of mechanical power in jumping. Eur J Appl Physiol Occup Physiol 50: 273–282, 1983.
    8. Brughelli M, Cronin J, Levin G, Chaouachi A. Understanding change of direction ability in sport. Sports Med 38: 1045–1063, 2008.
    9. Buckley JP, Sim J, Hession R, Fox R. Reliability and validity of measures taken during the Chester step test to predict aerobic power and to prescribe aerobic exercise. Br J Sports Med 38: 197–205, 2004.
    10. Cooper SM, Baker JS, Eaton ZE, Matthews N. A simple multistage field test for the prediction of anaerobic capacity in female games players. Br J Sports Med 38: 784–789, 2005.
    11. Cooper SM, Baker JS, Tong RJ, Roberts E, Hanford M. The repeatability and criterion related validity of the 20 m multistage fitness test as a predictor of maximal oxygen uptake in active young men. Br J Sports Med 39: 19–26, 2005.
    12. Davis DS, Barnette BJ, Kiger JT, Mirasola JJ, Young SM. Physical characteristics that predict functional performance in Division I college football players. J Strength Cond Res 18: 115–120, 2004.
    13. Draper JA, Lancaster MG. The 505 test: A test for agility in the horizontal plane. Aust J Sci Med Sport 17: 15–18, 1985.
    14. Gabbett T, Georgieff B. Physiological and anthropometric characteristics of Australian junior national, state and novice volleyball players. J Strength Cond Res 21: 902–908, 2007.
    15. Gil SM, Gil J, Ruiz F, Irazusta A, Irazusta J. Physiological and anthropometric characteristics of young soccer players according to their playing position: Relevance for the selection process. J Strength Cond Res 21: 438–445, 2007.
    16. Glatthorn JF, Gouge S, Nussbaumer S, Stauffacher S, Impellizzeri FM, Maffiuletti NA. Validity and reliability of Opto-jump photoelectric cells for estimating vertical jump height. J Strength Cond Res 25: 556–560, 2011.
    17. Haj Sassi R, Dardouri W, Gharbi Z, Chaouachi A, Mansour H, Rabhi A, Haj Yahmed M. Reliability and validity of a new repeated agility test as a measure of anaerobic and explosive power. J Strength Cond Res 25: 472–480, 2011.
    18. Haj Sassi R, Dardouri W, Haj Yahmed M, Gmada N, Mahfoudi MH, Garbi Z. Relative and absolute reliability of a modified agility T-test and its relationship with vertical jump and straight sprint. J Strength Cond Res 23: 1644–1651, 2009.
    19. Haley SM, Fragala-Pinkham MA. Interpreting change scores of tests and measures used in physical therapy. Phys Ther 86: 735–743, 2006.
    20. Hastad DN, Lacy AC. Measurement and Evaluation in Physical Education and Exercise Science. (2nd ed.). Scottsdale, AZ: GorsuchScarisbric, Hoolahan, P. Agility. NSCAJ 12: 22–24, 1990.
    21. Hopkins WG. Measures of reliability in sports medicine and science. Sports Med 30: 1–15, 2000.
    22. Hopkins WG. A new view on statistics: Log transformation. In: Sportscience. Available at: http://sportsci.org/resource/stats/index.html/Log. Accessed March 5, 2005.
    23. Impellizzeri FM, Marcora SM. Test validation in sport physiology: Lessons learned from clinimetrics. Int J Sports Physiol Perform 4: 269–277, 2009.
    24. Impellizzeri FM, Rampinini E, Castagna C, Bishop D, Ferrari Bravo D, Tibaudi A, Wisloff U. Validity of a repeated-sprint test for football. Int J Sports Med 29: 899–905, 2008.
    25. Jarvis S, Sullivan LO, Davies B. Interrelationships between measured running intensities and agility performance in sub elite rugby union players. Res Sports Med 17: 217–230, 2009.
    26. Lemmink KAPM, Elferink-Gemser M, Vissher C. Evaluation of the reliability of two field hockey specific sprint and dribble tests in young field hockey players. Br J Sports Med 38: 138–142, 2004.
    27. Lexell JE, Downham DY. How to assess the reliability of measurements in rehabilitation. Am J Phys Med Rehabil 84: 719–723, 2005.
    28. Liow DK, Hopkins WG. Velocity specificity of weight training for kayak sprint performance. Med Sci Sports Exerc 35: 1232–1237, 2003.
    29. Little T, Williams AG. Specificity of acceleration, maximum speed, and agility in professional soccer players. J Strength Cond Res 19: 76–78, 2005.
    30. Meir R, Newton R, Curtis E, Fardell M, Butler B. Physical fitness qualities of professional rugby league football players: Determination of positional differences. J Strength Cond Res 15: 450–458, 2001.
    31. Melrose DR, Spaniol FJ, Bohling ME, Bonnette RA. Physiological and performance characteristics of adolescent club volleyball players. J Strength Cond Res 21: 481–486, 2007.
    32. Miller MG, Herniman JJ, Ricard MD, Cheatham CC, Michael TJ. The effect of a 6-week plyometric training program on agility. J Sports Sci Med 5: 459–465, 2006.
    33. Pauole K, Madole K, Garhammer J, Lacourse M, Rozenek R. Reliability and validity of the T-test as a measure of agility, leg power, and leg speed in college-aged men and women. J Strength Cond Res 14: 443–450, 2000.
    34. Peterson MD, Alvar BA, Rhea MR. The contribution of maximal force production to explosive movement among young collegiate athletes. J Strength Cond Res 20: 867–873, 2006.
    35. Reilly T, Williams AM, Nevill A, Franks A. A multidisciplinary approach to talent identification in soccer. J Sports Sci 18: 695–702, 2000.
    36. Schreuders TA, Roebroeck ME, Goumans J, Van Nieuwenhuijzen JF, Stijnen TH, Stam HJ. Measurement error in grip and pinch force measurements in patients with hand injuries. Phys Ther 83: 806–815, 2003.
    37. Sheppard JM, Young WB. Agility literature review: Classifications, training and testing. J Sports Sci 24: 919–932, 2006.
    38. Vescovi JD, McGuigan MR. Relationships between sprinting, agility, and jump ability in female athletes. J Sports Sci 26: 97–107, 2008.
    39. Wang CY, Sheu CF, Protas EJ. Test-retest reliability and measurement errors of six mobility tests in the community-dwelling elderly. Asian J Gerontol Geriatr 4: 8–13, 2009.
    40. Webb P, Lander J. An economical fitness testing battery for high school and college rugby teams. Sports Coach 7: 44–46, 1983.
    41. Weir JP. Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM. J Strength Cond Res 19: 231–240, 2005.
    42. Young WB, Hawken M, McDonald L. Relationship between speed, agility, and strength qualities in Australian Rules football. Strength Cond Coach 4: 3–6, 1996.
    Keywords:

    change of direction; criterion validity; power of the legs; velocity; fidelity

    Copyright © 2013 by the National Strength & Conditioning Association.