(x) is a presentation of X irrelevant to the argument

0 The design is appropriate where one cannot randomly segregate subgroups
for differential treatments. Designs 12, 12a, 12b, and 12c (see Table 2) '
are superior to the "true" experiment because they put so little demand
upon the respondents for cooperation.

13. Separate-Sample Pretest-Posttest Control Group Design
O (X)

R

R X 0
RO 0

R (0)

-In addition to the group used in Design 12, if there are comparable
groups from which X can be withheld then this design is appropriate.

14. Multiple Time-Series Design
O 0 0 0 X 0 0 0 0

O 0 0 0 0 o 0 0

This design is appropriate if there is a similar group (or institution)
8 not undergoing X, from which to collect a similar "control" time series.

15. Recurrent Institutional Cycle Design: A "Patched-Up" Design

(For diagram see Table 3)

This design illustrates a strategy for field research in which one starts
out with an inadeguate dešign and then adds a specific features to control
for one or another of the recurrent sources of invaidity. The design is
appropriate to those situations in which a given aspect of an institutional
process is continually being presented to a new group of respondents. This
design can combine the longitudinal and cross-sectional approaches in order
to more effectively prevent validity problems.

16. Regression-Discontinuity Analysis

When people or groups are given awards or those in special need are given
extra help, one would like to discover the conseguences of such provisions,
This design is appropriate for these situations. A treatment effect is de-
tected if there is a discontinuity between the regression lines derived
from the scores of pretest and posttest of the two groups -- one with X
and one without X. The ANCOVA is the most efficient test of significance,

0 Ouestions
1. Do experimental researches in social science fully evaluate and regard
importantly the validity treats in their designs ?
2, How freguently are the guasi- experiments used in business research ?

Why?

NM