Statistical comparison of movement amplitudes from groupings of normal geriatric speakers

An implicit assumption of many studies of articulatory kinematics is that all normal speakers are equivalent, at least in a statistical sense. The present report investigated this assumption by performing post hoc analyses on random groups of 8 normal, geriatric males. Results of these analyses suggest that, at least for the geriatric population, small numbers of subjects may not represent the parent population adequately, especially when the unit of statistical analysis is the individual trial. These findings have implications for the types of generalizations that can be made from studies with small samples and suggest the need for more comprehensive investigations of speech kinematics.