-
Experimental Design Types
- R Random Assignment
- X Intervention(s) (Subscripts indicating different interventions)
- O Observation (pretest or posttest)
- A,B,C,D Groups of subjects
-
Single-Group Posttest-Only Design
- Often called Pre-experimental
- - usually have inadequate control of extraneous variables
- - GROUP (A) --> Intervention (X) --> Posttest (O)
-
Single-Group Pretest-Posttest Design
- Group (A)--> Pretest(O)-->Intervention(X)-->Posttest(O)
- - possible affects from extraneous events
- - Regression is possible
- - pretest can affect attitudes
- - Attrition can be a problem
- - Maturation is a potential threat
-
-
-
Nonequivalent-Groups Posttest-Only Design
Has a comparison or control group, but no pre-test. One group receives the treatment, while the other group (B) acts as a control, receiving no treatment.
- Group (A)--> Intervention (X)--> Posttest (O)
- Group (B)--------------------------> Posttest (O)
- In some, groups recieve different treatments:
- A--> X1-->O
- B--> X2-->O
- C--> X3-->O
-
Nonequivalent-Groups Pretest-Posttest Design
AKA Quasi-experimental Design
Where one group will get pre and post test with Intervention. The other group will also get pre and post test, but with no Intervention, or different interventions.
- A--> O--> X--> O
- B--> O--------->O
or
- A--> O--> X1--> O
- B--> O--> X2--> O
- C--> O--> X3--> O
|
|