Birmingham T B, Kramer J F
School of Physical Therapy, Faculty of Health Sciences, University of Western Ontario, London, Canada.
Percept Mot Skills. 1998 Dec;87(3 Pt 2):1183-91. doi: 10.2466/pms.1998.87.3f.1183.
The present investigation examined the reliability of a derived strength measurement and also how variability between test sessions affects the utility of this variable for judging an individual's muscular effort. 31 young healthy men (M age = 25 yr.) completed three isometric and isokinetic concentric contractions of the knee extensors, using maximal and self-selected submaximal efforts, on each of two test sessions. Difference scores between isometric and isokinetic measurements were calculated by subtracting the mean of the three isokinetic peak torques from the mean of the three isometric peak torques for each individual subject for maximal and submaximal efforts performed on both test sessions. For the group of subjects, difference scores were significantly greater during maximal (33 +/- 29 Nm) than submaximal (13 +/- 30 Nm) efforts, suggesting subjects could not maintain the same relationship between isometric and isokinetic muscular actions across the maximal and submaximal conditions. However, the test-retest reliability of the difference scores was only modest even when data were averaged over two test sessions (intraclass correlation coefficients were .82 for maximal, and .58 for submaximal). As a result, the range of scores within which an individual's true score might be expected to lie was large (+/- 25 Nm for maximal, and +/- 37 Nm for submaximal). Although derived strength parameters like difference scores may be effective in distinguishing submaximal from maximal efforts completed by groups of subjects, the test-retest reliability of the present scores suggests that their use in judging an individual's performance may be limited.
