|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Professional Perspectives |
CH Goldsmith, PhD, is Professor of Biostatistics, Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada L8N 3Z5, and Honorary Professor, Department of Physical Therapy, University of Western Ontario, London, Ontario, Canada N6G 1H1
This excerpt was created in the absence of an abstract.
Sim and Arnell have done a credible job of reviewing reliability and validity issues in physical therapy research. The examples that they use to illustrate various aspects of reliability, validity, and bias are useful to bring out the necessary nuances of developing a measurement procedure.
Although most of the key issues have been developed in the article, it would have been nice if the investigators had discussed the purpose of the measurement process, specifically, Is it diagnostic, discriminatory, or evaluative? A useful reference to describe these purposes is the article by Kirshner and Guyatt.1 It is possible that developers of measurement methods should look at responsiveness, particularly for evaluating therapies.
The context of the measurement does make the properties that are required to be illustrated for a measurement process slightly different....
Related Articles
![]()
CiteULike
Complore
Connotea
Del.icio.us
Digg
Reddit
Technorati What's this?
Physical Therapy 1993 73: 102-110.
Physical Therapy 1993 73: 110-113.
Physical Therapy 1993 73: 114-115.
Physical Therapy 1993 73: 115.
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |