Quasi-Experimental Strategy: Nonequivalent Group, Time-Series, and Developmental Designs

1
Quasi-Experimental Strategy Nonequivalent Group,
Time-Series, and Developmental Designs

• Chapter 9
• George S. Robinson, Jr., Ph.D.
• Department of Psychology
• North Carolina AT State University

2
Quasi-Experimental Research Strategy

• Involves a comparison of groups or conditions
• Uses a nonmanipulated variable to define the
  groups or conditions
• The variable is usually a participant variable or
  time variable

3
Nonequivalent Group Design

• Researcher can not control the assignment of the
  groups
• Thus, the groups of participants are nonequivalent

4
Nonequivalent Group Design cont.

• Threats to internal validity for nonequivalent
  group design
• Assignment bias
• Participants in one group necessarily have
  different characteristics from the participants
  in the other group

5
Nonequivalent Group Design cont.

• Differential research design (non-experimental
  design)
• No manipulation
• Uses participant characteristics to assign groups
• Attempts to compare pre-existing groups
• (e.g., IQ scores from Blacks and Whites, or
  Leadership scores from men and women)
• (ex post facto after the fact)

6
Nonequivalent Group Design cont

• Posttest-only Nonequivalent Control Group (static
  group comparison)
• Uses pre-existing groups
• One group serves in the treatment condition
• The other group serves as a control group (does
  not receive the treatment)
• There is no random assignment to groups
• X treatment O measurement
• X O (treatment group)
• O (nonequivalent control group)

7
Nonequivalent Group Design cont

• Pretest-Posttest Nonequivalent Control Group
  Design
• Compares two nonequivalent groups
• One group is measured twice, once before
  treatment, and once after treatment
• The other group is measured at the same two
  times, but does not receive treatment
• O X O
• O O

8
Nonequivalent Group Design cont

• Threats to a Pretest-Posttest Nonequivalent
  Control Group Design
• History effects
• Instrumentation
• Practice effects and fatigue
• Maturation
• Regression toward the mean
• Threats from differential effects (differential
  history, differential instrumentation,
  differential practice/fatigue, differential
  maturation, differential regression)

9
Time-Series Design

• Time-Series Design
• A series of observations is made over time
• One-group posttest design (pretest-posttest
  design)
• One measurement before treatment, and one
  measurement after treatment for a single group
• O X O

10
Time-Series Design cont.

• Time-Series Design
• Series of observations before treatment, and a
  series of observations after the treatment
• O O O X O O O
• Interrupted Time-Series Design
• A series of observations before an event occurs,
  and a series of observations after an event the
  event is not a treatment or manipulated by the
  experimenter

11
Time-Series Design cont.

• Equivalent Time-Samples Design
• A long series of observations in which a
  treatment is alternatively presented and
  withdrawn
• O observation, X treatment, N no treatment
• O O O X O O O N O O O X O O O N O O O X

12
Developmental Research Strategy

• Developmental research strategy
• Used to examine changes in behavior related to
  age
• Cross-Sectional Research Design
• Uses different groups of individuals
• Each group represents a different age
• The groups are measured at one point in time and
  then compared
• Longitudinal Research Design
• Measuring a variable (commonly) in the same
  individuals over time

13
Terminology in Quasi-Experimental and
Developmental Designs

• Quasi-independent variable
• Variable that is used to differentiate the groups
  of participants or the groups of scores
• Dependent variable
• Variable that is measured to obtain a score for
  each individual