QuasiExperimental Designs - PowerPoint PPT Presentation

1 / 16
About This Presentation
Title:

QuasiExperimental Designs

Description:

McBurney any research or variable where the experimenter lacks ... Lousy design. Possible confounds. Time series analysis. Economics. Rhythmical patterns ... – PowerPoint PPT presentation

Number of Views:66
Avg rating:3.0/5.0
Slides: 17
Provided by: alanm1
Category:

less

Transcript and Presenter's Notes

Title: QuasiExperimental Designs


1
Quasi-Experimental Designs
  • Quasi-experiment
  • Conditions created by researcher or others
  • Nonrandom assignment to conditions
  • Usually intact groups
  • McBurney any research or variable where the
    experimenter lacks complete control
  • Evaluation research

2
Nonequivalent control group designs
  • Nonequivalent posttest only
  • Selection, maturation confounds

3
Nonequivalent control group pretest posttest
design
  • Nonequivalent control group
  • Control group is not randomly selected from the
    same population as the experimental group
  • Nonequivalent groups assessed before and after
    intervention
  • Interpreting results
  • Clear

4
Nonequivalent control group pretest posttest
design
  • Interpreting results
  • Not clear
  • Regression

5
Nonequivalent control group pretest posttest
design
  • Interpreting results
  • Not clear
  • Proportional

6
Nonequivalent control group pretest posttest
design
  • Interpreting results
  • Clear

7
Nonequivalent control group pretest posttest
design
  • Attempt to eliminate all other explanations
  • Identify variable producing any initial
    differences
  • Focus on change in dependent variable

8
Mixed factorial designs with non manipulated
variable
  • Independent variables
  • Customer race (manipulated within subject
    variable)
  • Clerk race (subject variable)

9
(No Transcript)
10
Designs without control groupstime series
  • The one-group pretest-posttest design
  • Lousy design
  • Possible confounds
  • Time series analysis
  • Economics
  • Rhythmical patterns
  • Seasonal adjustments

11
Interrupted Time Series Design
  • Multiple measures before and after interruption
  • Interruption independent variable

12
Patterns
13
Patterns
14
Patterns
15
Interrupted Time Series Design
  • Threats to internal validity
  • History
  • Increase validity by use of multiple time series
  • Non-equivalent control group time series

16
Developmental and Evaluation Studies
  • Developmental
  • Cross-sectional
  • Cohort effects
  • Longitudinal
  • Cross-sequential
  • Different cohorts tested longitudinally
  • Evaluation
  • Formative
  • Summative
  • Four basic questions
  • Program achieve goals
  • What aspects contributed to success or failure
  • Can the program be improved
  • How valid is program theory
  • Resistance to Program Evaluations
Write a Comment
User Comments (0)
About PowerShow.com