Article

Reliability should be addressed for measures used in gait studies. Many protocols
designed for experimental or clinical research request subjects to walk
repeatedly at different steady-state walking speeds. Such data lend themselves to
reliability analysis, rendering a separate reliability test session superfluous.
Thus human and time resources are saved, and the reported reliability relates
directly to the study population and occasion at hand. In this report a procedure
is introduced where residuals from individual speed related curve estimates are
regarded as measurement error. If there is no speed related trend in data, the
procedure will yield identical results to a usual repeated measures reliability
analysis where all within-subject variability is regarded as measurement error.
CI - Copyright (c) 2010 Elsevier B.V. All rights reserved.

Langue : ANGLAIS