Conducting

1
Chapter 8

• Experimental Research

2
Steps in Experimental Research

1. Stating the research problem
2. Determining if experimental approach is
   appropriate
3. Specifying the independent variable(s)
4. Specifying all potential dependent variables
5. Stating the tentative hypotheses
6. Determining availability of measures for
   potential dependent variables
7. Pausing to consider success potential of the
   research
8. Identifying full potential of intervening
   variables
9. Making a formal statement of the research
   hypotheses
10. Designing the experiment
11. Making final estimate of success potential of the
    study
12. Conducting the study as planned in steps 1
    through 11
13. Analyzing the data
14. Preparing a research report

3
Internal and External Validity

• validity How valid are the
  findings within the study?
• validity How much can the
  findings be inferred or generalized to other
  populations, settings, or treatments?
• Must have good internal validity to have good
  external validity, but doesnt guarantee good
  external validity.

4
Threats to Internal Validity

• History
• Maturation
• Testing
• Instrumentation
• Statistical Regression
• Selection
• Experimental Mortality
• Interaction of Selection and Maturation or History

5
Threats to External Validity

• Interaction Effect of Testing
• Interaction Effects of Selection Bias and
  Experimental Treatment
• Reactive Effects of Experimental Setting
• Multiple-Treatment Interference

6
Types of Designs

• designs variety of ways a research study may be
  structured or conducted
• Three main types of designs
• experimental designs
• experimental designs
• designs

7
True experimental design

• random sampling of participants, random
  assignment to groups, all threats to internal
  validity controlled
• Not always possible

8
Quasi-experimental design

• Lack either random sampling or random assignment
• Controls threats to validity of history,
  maturation, testing, instrumentation, selection,
  and experimental mortality

9
Preexperimental designs

• no random sampling of participants, limited
  groups, control few threats to validity

10
Methods of Control

• physical manipulation researcher controls all
  aspects of participants environment and
  experience
• selective manipulation
• statistical techniques

11
Common Sources of Error

• Hawthorne Effect participants in an experiment
  may perform differently because they know theyre
  in a study
• Placebo Effect participants receiving treatment
  believe the treatment will have an effect (use
  single-blind approach)
• John Henry Effect Control group might try
  harder in attempt to outperform experimental
  group
• Rating Effects halo effect, overrater or
  underrater error, central tendency error
• Experimental Bias Effect try to control using
  double-blind approach
• Participant-Researcher Interaction Effect Gender
  issues, age, etc.
• Post Hoc Error Assuming cause-and-effect
  relationship where one does not exist