1. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Chapter 10 Using Specialized Research Designs

2. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. The Mixed Design Includes a between-subjects and a within- subjects factor in the same design Allows you to evaluate the effects of variables that cannot be manipulated effectively within- subjects Complex mixed designs would include more than two factors, with any combination of between- subjects and within-subjects factors

3. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. The Nested Design Combines between-subjects and within-subjects factors Different levels of within-subjects factors (e.g., A and B, C and D) are “nested” under different levels of a between-subjects factor (level 1 and level 2) Types of nested designs Nesting tasks Nesting groups of subjects

4. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Example of a Nested Design Between-Subjects Level 1 Between Subjects Level 2 A B C D S1S1 S6S6 S2S2 S7S7 S3S3 S8S8 S4S4 S9S9 S5S5 S 10 S6S6 S 11

5. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Combining Experimental and Correlational Designs Including a covariate in an experimental design A covariate is a correlational variable (e.g., self- esteem) in an experimental design “Subtracting out” the influence of the covariate reduces error variance Makes your design more sensitive to the effects of the independent variable

6. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Including a quasi-independent variable in an experimental design A quasi-independent variable is a correlational variable (e.g., gender) that looks like an experimental variable Resulting design looks like a factorial experimental design The quasi-independent variable must not be interpreted as causing changes in the dependent variable

7. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Quasi-Experimental Designs Time Series Design Make several observations of behavior before and after introducing your independent variable Interrupted Time Series Design Make several observations before and after some naturally occurring event Equivalent Time Samples Design Repeatedly introduce the treatment condition, alternated with periods of observation without the treatment

8. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Nonequivalent Control Group Design Include a time series component and a control group that is not exposed to the independent variable

9. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. The Pretest-Posttest Design Pretest administered before exposure to experimental treatment Unlike quasi-experimental designs, this is a true experimental design Used to assess the impact of some change on performance There is a problem with pretest sensitization Taking the pretest may alter the way a person performs in an experiment

10. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. A Mixed Design With Pretest-Posttest as the Within-Subjects Factor Group 1:Pretest-Treatment-Posttest Group 2:Pretest Posttest

11. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. The Solomon Four-Group Design Variation on the pretest-posttest design Allows you to evaluate the impact of a pretest on posttest performance Adds two groups to the basic pretest-posttest design A treatment-posttest group A posttest only group

12. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. The Solomon Four-Group Design Group 1:Pretest-Treatment-Posttest Group 2:Pretest Posttest Group 3: Treatment-Posttest Group 4: Posttest

13. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Developmental Designs The Cross-Sectional Design Participants from different age groups are run through a study at the same time Creating “cohort” groups based on participants’ ages Allows you to collect developmental data in a short period of time May not be appropriate studies using widely ranging age groups Generation effects may be a problem

14. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. The Longitudinal Design A single group of participants is measured several times over some period of time (e.g., months or years) Avoids the generation effect that may plague a cross- sectional study May still have a cross-generational problem Results from a longitudinal study on one generation may not generalize to another Problems with the longitudinal design Subject mortality Multiple observation effects

15. © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. The Cohort-Sequential Design Combines a cross-sectional and longitudinal component in the same design Allows you to test for, but not eliminate, generation effects