| |
| | CIOS - (Site not responding. Last check: 2007-10-02) |
 | | As a simple example, using a Q sort as data: The = formula M =3D Sum(X) / N tells us where the mean is located (usually at = zero for forced data), but it tells us nothing about the dispersion, = which may be leptokurtic, platykurtic, bimodal, or distributed in some = other way. |
 | | Michael Pepperday has purposely = created Q sorts that have means in certain locations (e.g., a high mean = for the A-liberal statements but low means for the B, C, and D = statements), but correlation and factor analysis do not respond to mean = differences, but to dispersion. |
 | | As a simple = example,=20 using a Q sort as data: The formula M =3D Sum(X) / N tells us = where the=20 mean is located (usually at zero for forced data), but it tells us = nothing=20 about the dispersion, which may be leptokurtic, platykurtic, bimodal, = or=20 distributed in some other way. |
| www.cios.org /mailboxes/Q-method/09055191.001 (1947 words) |
|