’s eligibility for adapted physical education and physical therapy services. Instrument reliability is an important and requisite psychometric property of any research instrument. A high-quality measure should have acceptable reliability across time (i.e., test–retest), across individuals conducting the
Hyokju Maeng, E. Kipling Webster, E. Andrew Pitchford and Dale A. Ulrich
Nicole C. George, Charles Kahelin, Timothy A. Burkhart and David M. Andrews
equations to be useful tools for quantifying segment tissue composition, individuals taking the anthropometric measurements used in the equations must be as consistent as possible within and between measurement sessions. 15 Thus, assessing the reliability of the proposed measurements for the head, neck
Annabelle King, Mark Hancock and Joanne Munn
Functional strength measures correlate more closely with functional performance than non-functional strength measures.
To determine the reliability of the lateral step test as a measure of maximal strength.
Intertester repeated measures.
Twenty four healthy, pain free subjects.
Two protocols (A and B) were evaluated. The protocols were identical except protocol B involved a three second pause. Participants performed a one repetition maximum (1RM) for each protocol on two occasions separated by one week.
Main Outcome Measures:
Step height (nearest cm) representing
Both protocols demonstrated excellent reliability, protocol A: ICC = 0.94 (95% CI, 0.87 to 0.97), SEM = 1.47 cm. Protocol B: ICC= 0.94 (95%CI, 0.85 to 0.97).Percent close agreement within 2 cm was 83.3% for protocol A and 79.1% for protocol B.
Both protocols demonstrated excellent inter-tester reliability as measures of functional lower limb strength.
Luke J. Boyd, Kevin Ball and Robert J. Aughey
To assess the reliability of triaxial accelerometers as a measure of physical activity in team sports.
Eight accelerometers (MinimaxX 2.0, Catapult, Australia) were attached to a hydraulic universal testing machine (Instron 8501) and oscillated over two protocols (0.5 g and 3.0 g) to assess within- and between-device reliability. A static assessment was also conducted. Secondly, 10 players were instrumented with two accelerometers during Australian football matches. The vector magnitude was calculated, expressed as Player load and assessed for reliability using typical error (TE) ± 90% confidence intervals (CI), and expressed as a coefficient of variation (CV%). The smallest worthwhile difference (SWD) in Player load was calculated to determine if the device was capable of detecting differences in physical activity.
Laboratory: Within- (Dynamic: CV 0.91 to 1.05%; Static: CV 1.01%) and between-device (Dynamic: CV 1.02 to 1.04%; Static: CV 1.10%) reliability was acceptable across each test. Field: The between-device reliability of accelerometers during Australian football matches was also acceptable (CV 1.9%). The SWD was 5.88%.
The reliability of the MinimaxX accelerometer is acceptable both within and between devices under controlled laboratory conditions, and between devices during field testing. MinimaxX accelerometers can be confidently utilized as a reliable tool to measure physical activity in team sports across multiple players and repeated bouts of activity. The noise (CV%) of Player load was lower than the signal (SWD), suggesting that accelerometers can detect changes or differences in physical activity during Australian football.
Gavin L. Moir, Alberto Garcia and Gregory B. Dwyer
To investigate the intersession reliability of selected kinematic and kinetic variables during countermovement vertical jumps (CMJs).
Thirty-five men and 35 women performed CMJs on a force platform during four testing sessions each separated by 1 wk. Kinematic variables included time in the air (TIA), take-off velocity (TOV), total vertical displacement of the center of mass (TJH). and countermovement depth, whereas kinetic variables included positive impulse, negative impulse, vertical stiffness, and rates of force development. Systematic bias was assessed by calculating the 90% confidence interval of the change in the mean between consecutive testing sessions and between the first and final testing session for each variable. Coefficients of variation (CV) and intraclass correlation coefficients (ICC) were also calculated.
Systematic bias was observed only for peak rate of force development during the concentric phase of the movement. For TIA, TOV, and TJH, CV values ranged from 1.7% to 6.6%, with ICC values ranging from 0.82 to 0.97. The other variables showed greater variation (CV range: 1.7% to 39.9%; ICC range: 0.04 to 0.99). Only slight gender differences were found in the reliability statistics, and the reliability of most of the variables was diminished as the time between the testing sessions was increased.
Even though practitioners can expect good reliability for jump height measured from a force platform in men and women, other kinematic and kinetic variables often assessed during vertical jumps may not be reliable.
John R. Sirard, Ann Forsyth, J. Michael Oakes and Kathryn H. Schmitz
The purpose of this study was to determine 1) the test-retest reliability of adult accelerometer-measured physical activity, and 2) how data processing decisions affect physical activity levels and test-retest reliability.
143 people wore the ActiGraph accelerometer for 2 7-day periods, 1 to 4 weeks apart. Five algorithms, varying nonwear criteria (20 vs. 60 min of 0 counts) and minimum wear requirements (6 vs. 10 hrs/day for ≥ 4 days) and a separate algorithm requiring ≥ 3 counts per min and ≥ 2 hours per day, were used to process the accelerometer data.
Processing the accelerometer data with different algorithms resulted in different levels of counts per day, sedentary, and moderate-to-vigorous physical activity. Reliability correlations were very good to excellent (ICC = 0.70−0.90) for almost all algorithms and there were no significant differences between physical activity measures at Time 1 and Time 2.
This paper presents the first assessment of test-retest reliability of the Actigraph over separate administrations in free-living subjects. The ActiGraph was highly reliable in measuring activity over a 7-day period in natural settings but data were sensitive to the algorithms used to process them.
Jeremiah Tate, Hanna True, Barry Dale and Carrie Baker
The high rate of ACL injury has led to the need for large-scale injury risk screening to aid in prevention. Cost and time alternatives to three-dimensional assessments of knee injury risk factors are necessary for largescale, clinical use. The purpose of this study was to determine the expert versus novice interrater reliability and test-retest reliability of the frontal plane projection angle (FPPA) during the single-leg squat (SLS) test. Excellent interrater reliability was demonstrated for session 1 and session 2. Excellent intrarater (i.e., test-retest) reliability was demonstrated for both the novice and expert. The minimal training requirements and excellent reliability of the FPPA during the SLS test make it a key screening tool for large-scale assessment of injury risk.
Damien O’Meara, Benedicte Vanwanseele, Adrienne Hunt and Richard Smith
The purpose was to develop a foot image capture and measurement system with web cameras (the 3-FIS) to provide reliable and valid foot anthropometric measures with efficiency comparable to that of the conventional method of using a handheld anthropometer. Eleven foot measures were obtained from 10 subjects using both methods. Reliability of each method was determined over 3 consecutive days using the intraclass correlation coefficient and root mean square error (RMSE). Reliability was excellent for both the 3-FIS and the handheld anthropometer for the same 10 variables, and good for the fifth metatarsophalangeal joint height. The RMSE values over 3 days ranged from 0.9 to 2.2 mm for the handheld anthropometer, and from 0.8 to 3.6 mm for the 3-FIS. The RMSE values between the 3-FIS and the handheld anthropometer were between 2.3 and 7.4 mm. The 3-FIS required less time to collect and obtain the final variables than the handheld anthropometer. The 3-FIS provided accurate and reproducible results for each of the foot variables and in less time than the conventional approach of a handheld anthropometer.
Kate Anne McGirr, Thomas Kennedy, Carsten M. Mølgaard and Michael Skovdal Rathleff
Measurement of ankle evertor and invertor strength is important for assessment of ankle sprain risk and assessment of restoration of normal function after an ankle sprain.
To compare the intra-tester reliability of ankle evertor and invertor strength measurements derived from hand-held dynamometry (HHD) and strap-mounted dynamometry (SMD) for both make-test and break-test techniques.
25 healthy individuals.
Main Outcome Measure:
Repeated measurements of isometric evertor and invertor strength derived from both HHD and SMD devices using both make-test and break-test techniques.
High make-test reliability values were found for both inversion and eversion force measured by both the HHD and SMD devices (ICC = 0.80 to ICC = 0.88). The correlation coeffcients between HHD and SMD measurements for the inversion make-test, the inversion break-test, and the eversion make-test ranged from r = 0.74 to r = 0.87, but the correlation between HHD and SMD measurements for the eversion break-test was poor.
The fndings suggest that testers with limited experience can obtain reliable measurements of ankle evertor and invertor strength using either HHD or SMD.
David M. Wert, Jessie M. VanSwearingen, Subashan Perera and Jennifer S. Brach
The purpose of this study was to assess the relative and absolute reliability of metabolic measures of energy expenditure and gait speed during overground walking in older adults with mobility limitations. Thirty-three (mean age [SD] = 76.4 [6.6] years; 66% female) older adults with slow gait participated. Measures of energy expenditure and gait speed were recorded during two 6-min bouts of overground walking (1 week apart) at a self-selected “usual” walking pace. The relative reliability for all variables was excellent: ICC = .81−.91. Mean differences for five of the six outcome variables was less than or equal to the respected SEM, while all six mean differences fell below the calculated MDC95. Clinicians and researchers can be confident that metabolic measures of energy expenditure and gait speed in older adults with slow walking speeds can be reliably assessed during overground walking, providing an alternative to traditional treadmill assessments.