ch 10 self text
Terms in this set (3)
why is it said that nonequivalent control group design has a "built-in confound"?
the groups are nonequivalent; in addition, one group gets one type of treatment and the other group gets a different treatment (or none at all).
If nonequivalent groups are being used and the groups are matched on some pretest score, the results can be distorted by a _________ effect
Time series designs sometimes include "switching replications." How does thing design differ from the basic interrupted time series design?
in a switching replication, the treatment program is implemented in a different place and at a different time.