Experimental and Quasi-Experimental Designs: Slide 1

Chapter 11
Experimental and Quasi-experimental

Designs
Educational Research by John W. Creswell. Copyright © 2002 by Pearson Education. All rights reserved. Slide 1
Key Ideas
 History of experimental resign
 Key characteristics of experimental design
 Types of experimental design
 Steps in conducting experimental research
 Criteria for evaluating experimental
research
Educational Research by John W. Creswell. Copyright ©2002 by Pearson Education. All rights reserved. Slide 2
A Brief History of Experimental
Designs
 Schuyler 1903: control groups
 McCall 1916: randomly selected groups
 McCall 1925: book on experiments
 Fisher 1936: statistical methods book
A Brief History of Experimental
Designs
 Campbell and Stanley 1963: 15 types of
experimental designs evaluated in terms of
threats to validity
 Cook and Campbell 1979: four types of
validity
 1980 and beyond: computer enhancements
to experimental design2
Key Characteristics of
Experimental Designs
 Participants selected and assigned to groups
– control
– experimental
 An intervention is applied to one or more
groups
 Outcomes are measured at the end of the
experiment
Key Characteristics of
Experimental Designs
 Procedures are designed that address
potential threats to validity
– Internal
– External
– Construct
– Statistical Conclusion
 Statistical comparisons of different groups
are conducted
Selecting Participants and
Assigning Them to Treatments
 Decide on the experimental unit of analysis
to be treated
– individual
– group or groups
– organization
 Randomly assign individuals to groups
control for extraneous characteristics that
might influence the outcome
Selecting Participants and
Assigning Them to Treatments
 Control for extraneous factors
– random assignment (equating groups)
– pretest/posttest
– covariates
– matching participants
– selecting homogenous samples
– using blocking variables
Controlling for Covariates
No Covariates
Independent Dependent
Variable Variable
Covariate Introduced
Independent Dependent
Covariate:
Variable: Variable:
Parents Who
Type Rates of
Smoke
of Instruction Smoking
Variance
Removed
Variance
Matching Process Based on
Gender
John
Jim Experimental
James Group
Josh
Jackson
Jane
Johanna
Julie
Control
Jean
Group
Jeb
Applying an Intervention or
Treatment
 Identify a treatment variable
 identify the conditions or levels of the
variable
 Manipulate the treatment conditions
The Experimental Manipulation of a
Treatment Group
Independent Variables Dependent Variable
1. Age (can’t manipulate)
Frequency of Smoking
2. Gender (can’t manipulate)
3. Types of Instruction (can
manipulate)
a. Lecture (control)
b. Lecture + Hazard Instruction
(Comparison)
c. Lecture + Hazard Instruction +
slides of damaged lungs
(experiment)
Threats to Statistical Conclusion
Validity
 Low statistical power due to low sample
size
 Violation of assumptions of statistical tests
 Use of unreliable measures
Threats to Internal Validity
 History
 Maturation
 Regression
 Selection
 Mortality
 Interactions with selection
Threats to Internal Validity
 Diffusion of treatments
 Compensatory equalization
 Compensation rivalry
 Resentful demoralization
 Testing
 Instrumentation
Threats to Construct Validity
 Lack of good operational definitions
 Apprehensiveness by participants
 Participants “guessing” what the researcher
hopes to find
Threats to External Validity
 Interaction of selection and treatment
 Interaction of setting and treatment
 interaction of history and treatment
Treatment Comparisons in an
Experiment
Phase 1: Relationship Picture
+
Error Correction Treatment Spelling Accuracy
Phase 2: Timeline Picture Test 1 Test 2 Test 3
Class A: Regular Spelling Practice (Control) 6 Weeks 6 Weeks 6 Weeks
Class B: Reduced word list (Comparison) 6 Weeks 6 Weeks 6 Weeks
Class C: Error Correction (Experimental) Error Correction Treatment

6 Weeks 6 Weeks 6 Weeks
Phase 3: Statistical Comparisons
Class A Class B Class C F-value
Test 1 10.3 10.8 9.9 0.27
(3.6) (4.3) (3.9)
Test 2 10.7 10.6 13.9 4.90*

(3.3) (3.8) (4.2)
* p<.05
11.1 10.3 13.1
Test 3 3.31*
(3.3) (3.6) (3.8)
Types of Experimental Designs:
Between Groups
 True Experiments
– Pre- and posttest
– Posttest Only
 Quasi Experiments
– Pre- and posttest
– Posttest Only
 Factorial Designs
Types of Experimental Designs:
Within-Group or Individual
 Time series experiments
– interrupted
– uninterrupted
 Repeated measures experiments
 Single-subject experiments
– A/B design
– Multiple baseline design
– Alternating treatments
Means and Main Effects of Eight
groups in Factorial Design
Depression
Low Medium High

Mean rate Mean rate Mean rate
Health lecture of of of
smoking smoking smoking
Main Effects
Type of Instruction of Type of
Mean rate Mean rate Mean rate Instruction
Standard lecture of of of
smoking smoking smoking
Main Effects of
Depression
Graphs Showing Main and
Interaction Effects Standard lecture
(a) No interaction Effects High
(Parallel) Health lecture
Extent of Smoking
(b) Interaction Effects
(a) Low
(Crossed) Low Medium High
High Standard lecture
Extent of Smoking
Health lecture
Low
Low Medium High
(c) Interaction Effects High
(Not Parallel) Standard lecture
Extent of Smoking
Health lecture
Low
Low Medium High
Steps in Conducting
Experimental Research
 Decide if an experiment addresses the
research problem
 Form hypotheses to test cause-effect
relationships
 Select an experimental treatment and
introduce it
 Identify study participants
Steps in Conducting
 Choose a type of experimental design
 Conduct the experiment
 Organize and analyze the data
 Develop an experimental research report
Criteria for Evaluating
 Does the experiment have a powerful
intervention?
 Does it employ few treatment groups (e.g.
only two)?
 Will participant profit from the intervention?
 Is there a systematic way the researcher
derived the number of participants per group?
Criteria for Evaluating
 Were there an adequate number of
participants used in the study?
 Were valid, reliable, and sensitive measures
or observations used?
 Did the study control for extraneous factors?
 Did the researcher control for threats to
internal validity?
Applying What you Have
Learned: An Experimental Study
Review the article and look for the following:
 The research problem and use of quantitative research
 Use of the literature
 The purpose statement and research hypothesis
 Types and procedures of data collection
 Types and procedures of data analysis and
interpretation
 The overall report structure

Experimental and Quasi-Experimental Designs: Slide 1

Uploaded by

Copyright:

Available Formats

Experimental and Quasi-Experimental Designs: Slide 1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Experimental and Quasi-Experimental Designs: Slide 1

Uploaded by

Copyright:

Available Formats

Chapter 11

Experimental and Quasi-experimental

Class C: Error Correction (Experimental) Error Correction Treatment

Test 2 10.7 10.6 13.9 4.90*

Low Medium High

You might also like