PowerPoint PresentationCopyright ©2014 by Pearson Education, Inc.
All rights reserved.
Research Methods, Design, and Analysis, Twelfth Edition Christensen
• Johnson • Turner
CHAPTER
Introduction
Quasi-experimental design
a research design in which an experimental procedure is applied,
but all extraneous variables are not controlled
typically lacking random assignment
Introduction
control by design
making a complex prediction that few rival hypotheses can
explain
Copyright ©2014 by Pearson Education, Inc. All rights
reserved.
Research Methods, Design, and Analysis, Twelfth Edition Christensen
• Johnson • Turner
Nonequivalent Comparison Group Design
Includes both an experimental and control group
Participants not randomly assigned to groups
Pre-test important to determine equivalence of groups
large difference between groups on pre-test may indicate selection
bias
Threats frequently reveal themselves in the outcome
Copyright ©2014 by Pearson Education, Inc. All rights
reserved.
Research Methods, Design, and Analysis, Twelfth Edition Christensen
• Johnson • Turner
Copyright ©2014 by Pearson Education, Inc. All rights
reserved.
Research Methods, Design, and Analysis, Twelfth Edition Christensen
• Johnson • Turner
Outcomes with Rival Hypotheses
could be caused by a number of rival hypotheses
selection-maturation
selection-history
Outcomes with Rival Hypotheses
no change in control group
treatment condition starts higher and increases
rival hypotheses
Outcomes with Rival Hypotheses
no change in control group
treatment condition starts lower and increases
rival hypotheses
Outcomes with Rival Hypotheses
Crossover effect
treatment group scores significantly lower than the control group
at pretest
significantly higher at posttest
rival hypotheses unlikely with this type of result
Copyright ©2014 by Pearson Education, Inc. All rights
reserved.
Research Methods, Design, and Analysis, Twelfth Edition Christensen
• Johnson • Turner
Ruling out Threats to the Nonequivalent Comparison Group
Matching
example
Head Start programs can be matched on income, intelligence, and
parental involvement
matching not a perfect replacement for random assignment, but can
be used when random assignment is not possible
selection-regression effects may occur when using extreme
groups
Statistical control techniques
Causal Inference from Nonequivalent Comparison Group Design
To increase internal validity
self-selection increases bias
matching
Interrupted Time-Series Design
Design in which a treatment effect is assessed by comparing the
pattern of pre- and posttest scores for a single group of research
participants
Looking for discontinuity in the series of dependent measures
Example
Interrupted Time-Series Design
Use of multiple pretest and posttest measurements demonstrates
reliability of effect
Visual inspection of pre and posttest measure very important to
determine treatment effect
Improvement over one-group pretest-posttest design
Primary weakness
Copyright ©2014 by Pearson Education, Inc. All rights
reserved.
Research Methods, Design, and Analysis, Twelfth Edition Christensen
• Johnson • Turner
Regression Discontinuity Design
Design that assigns participants to groups based on their scores on
an assignment variable and assesses the effect of a treatment by
looking for a discontinuity in the groups regression lines
Copyright ©2014 by Pearson Education, Inc. All rights
reserved.
Research Methods, Design, and Analysis, Twelfth Edition Christensen
• Johnson • Turner
Regression Discontinuity Design
individuals who score above some cutoff score receive the
treatment
all individuals are posttested
Regression Discontinuity Design
LOAD MORE