At 12:12 PM 10/21/99 -0400, Ken wrote:
>Campbell & Stanley use the term One-Shot as a description of a
>nonexperimental design.
And to elaborate a little further, the following is adapted from Isaac &
Michael (1995). Handbook in research and evaluation 3rd edition, pp
70-71... And they reference back to Campbell & Stanley (1966)..
A "POOR" research design (no control): The "One-Shot Case Study"
Expose subjects to treatment
Posttest to measure effects (belief that standardized test could be used
and standardization sample serves as a "control"- see disadvantage D below)
Advantage: useful for finding potential research problems. Not a basis from
which we can draw defensible conclusions. Danger that we will try and
justify things on impressionistic evidence alone.
Disadvatages:
a) Complete absence of control and no internal validity. The "quick and
easy" nature of this approach, often used as a basis for a change or
innovation in education, is entirely misleading.
b) no provision for comparison exists except implicity, intuitively, and
impresionistically. (NOTE: basic to science is the process of comparison,
and securing scientific evidence involves at least on comparison.)
c) This approach usually involves the "error of misplaced precision" - a
great deal of care given to the colletion of data about which our
conclusions can only be impressionistic and imprecise.
d) The attempt to use standardized tests in lieu of a control group is
misguided, since rival sources of difference other than X are so numerous
as to render the standardization group useless as a "control" group.
So, it is a design... just a really bad one...
- Marc
G. Marc Turner, MEd
Lecturer & Head of Computer Operations
Department of Psychology
Southwest Texas State University
San Marcos, TX 78666
phone: (512)245-2526
email: [EMAIL PROTECTED] or [EMAIL PROTECTED] or [EMAIL PROTECTED] or ...
