View Full Version : Testing Validity Responses

Steve Lindley
05-04-2005, 11:09 PM
Thank you for all of the responses that I received on my posting “Testing
Validity”. Following is the original query and the responses:

This query is about test validation.

We have developed a sitting reach test with individuals with spinal cord
injury to measure trunk control and have had high test/re-test reliability
with the 3D kinematics. Now we want to validate this test. We began this
quest by sitting the patients directly on a force plate to measure the COP
excursion in the direction of the reaching. The subject is positioned in an
unsupported sitting position with feet on the floor. A problem has now
arisen. If the patient leans far enough, the COP reaches the edge of the
force plate, thus truncating the excursion.

Are there any suggestions for remedying this problem or alternative for
validating this test?

Stephen Lindley MS, PTA


Have you tried sitting the subject on a Swiss Ball, BOSU, or Dynadisk? I've
been thinking about doing something like this for years for entirely
different reasons, but this might be applicable in your scenario as I
understand it. I'm admittedly not clear on how the force plate will "read"
movement occuring on such a device, and specifically wonder if one will lose
measurement sensitivity. Nonetheless, at first glance, the Dynadisk might
give some valuable measures on center of pressure while simultaneously
limiting the excursion. Please summarize and post the responses to your
query: I'll be among those very interested in learning about the suggestions
from around the world.

Don Don Hoover dhoover@uindy.edu
Add a force plate under their feet & use the COP of their entire contact
area rather than just under their rear. It would move through 3D space, but
could never pass beyond the edge of a force plate.

Brian Schulz, Ph.D.
Program Specialist
VISN 8 Patient Safety Research Center (118M)
James A. Haley Veterans' Hospital
11605 North Nebraska Ave.
Tampa, FL 33612-5738
Phone: (813) 558-3944
Fax: (813) 558-3990
Perhaps a metal plate, bigger in dimension than the surface of the force
plate, could be clamped or bolted to the top of the force plate. The
plate would need to be re-calibrated to get a new coordinate system
center after the attachment. Then the change in center of pressure
could occur over the dimensions of the metal plate. I am not sure if
this would work or not.

Ray Burdett Burdett, Ray rgb@pitt.edu
If you measure the COP with the force plate, the COP can't really go beyond
edge. Unless you have tensile forces acting on the plate.

Is the chair placed on the force plate? If so, what you may be seeing is
COP reaching the outline of a rectangle formed by the legs of the chair.
would cause the signal to look truncated.

van den Bogert, Ton

what about building a construction (like a seat) and mount that on the force
plate, so that the patient has his COP near to the rear edge of the plate
while sitting in an upright position? The feet of the patient could rest on
the platform. The construction must not touch the ground outside the

Best regards

Lars Dr. Lars Janshen lars.janshen@rz.hu-berlin.de
Dean et al. (Sitting balance I: trunk-arm coordination and the
contribution of the lower limbs during self-paced reaching in sitting.
(1999) Gait and Posture, 10(2): 135 - 146) used two force plates, one
under the feet and one under the buttocks, to capture CoP displacement
during seated reach. This allowed them to look at the transfer of loading
from the seat to the feet, and could give you characterisation of CoP
outside the buttock-only base of support.

Hope this helps,
Jon Doan, M.Sc., P.Eng.
Balance Research Laboratory
University of Lethbridge
Please call me at the numbers listed below, I have a couple of suggestions
that may help.
Gary Blanchard
Product Manager
Tel: 617-926-6700, mobile: 617-216-5397
Fax: 617-926-5045

Once again, thank you all for your responses
Stephen D Lindley
Senior Research Associate
The Miami Project to Cure Paralysis
University of Miami – Medical Campus