QuasiExperimental and SingleCase Designs

1
Chapter 12

• Quasi-Experimental and Single-Case Designs

2
Definition

• The quasi-experimental approach is used when it
  is impossible to randomly assign participants to
  comparison groups and when a researcher is faced
  with a situation where only one or two
  participants can participate in the research
  study (single case designs).

3
Reliability is weaker

• Causal explanations can be made when using
  quasi-experimental designs but only when you
  collect data that demonstrate that plausible
  rival explanations are unlikely, and the evidence
  will still not be as strong as with experimental

4
Three quasi-experimental designs

• The nonequivalent comparison-group design
• The interrupted time-series design
• The regression discontinuity design

5
Nonequivalent Comparison-Group Design

• Contains a treatment group and a nonequivalent
  untreated comparison group which are administered
  pretest and posttest measures.
• The groups are nonequivalent because you lack
  random assignment
• Only when you can rule out the effects of
  confounding variables can you confidently
  attribute the observed group difference at the
  posttest to the independent variable.

6
Example?
7
Interrupted Time-Series Design

• Compares pretest and posttest responses obtained
  from a single group of participants.
• Participants are pretested and posttested a
  number of times
• The pretesting phase is called the baseline
• A treatment effect is demonstrated only if
  post-treatment responses differs from the
  pretreatment responses.
• The main potentially confounding variable that
  cannot be ruled out is a history effect. The
  history threat is a plausible rival explanation
  if some event other than the treatment co-occurs
  with the onset of the treatment.  

8
Example?
9
Regression Discontinuity Design

• Looks for a discontinuity in regression lines
  between individuals who score lower and higher
  than some predetermined cutoff score on an
  assignment variable.
• It is a design used to determine if individuals
  meeting some predetermined criteria profit from
  receiving some special treatment.

10
For example

• Use a standardized test with a cutoff at 50
• Administer treatment to those scoring in pretest
  above 50
• Administer posttest to both groups, and see if
  there is a discontinuity in the regression line.
  If there is none, the treatment didnt work.
• This is actually quite a strong design, and
  methodologists have, for a number of years, been
  trying to get researcher to use this design more
  frequently.
• Figure 12.9 on pg 340 clarifies

11
Control
Treatment
No discontinuity in scores
Treatment Didnt work
50
12
Control
Treatment
Regression Discontinuity
Treatment worked
50
13
4 Single-Case Experimental Designs

• Treats single participants, one at a time. Or one
  single group of individuals.
• A-B-A design,
• A-B-A-B design,
• Multiple-baseline design
• Changing-criterion design

14
A-B-A and A-B-A-B Designs

• The A-B-A design - participant is repeatedly
  pretested (the first A phase), then the
  experimental treatment condition is administered
  and the participant is repeatedly post-tested
  (the B phase)
• Finally, the treatment is stopped and participant
  is tested again (second A phase)
• The effect of the treatment is demonstrated if
  the pattern differs from A to B then reverts back
  to A

15
For example,

• Student talks out of turn frequently (A)
• Teacher administers treatment (B)
• Teacher stops administering treatment (A)
• If student returns to talking out of turn
  frequently, treatment was effective.
• One limitation of the A-B-A design is that it
  ends with baseline condition at the end of the
  experiment.
• This limitation can be overcome by including a
  fourth phase which adds a second administration
  of the treatment condition so the design becomes
  an A-B-A-B design.

16
 Multiple-Baseline Design

• This design focuses on two or more different
  behaviors in the same individual
• Or on the same behavior exhibited by two or more
  individuals
• Or on the same behavior exhibited by one
  individual but in different settings

17
Example Same behavior in four people

• Step 1 Get baseline pretest on all four people
• Step 2 Treat first person only, dont treat the
  other three
• Step 3 Treat second person only
• Step 4 Treat third person only
• Step 5 Treat fourth person only
• If results show change during treatment,
  treatment is effective

18
A polluting variable

• The other three may change behavior because one
  of the four is changing.
• For example, if all four talk out of turn, and
  one is silenced through treatment, the other
  three may grow quiet because one is silent, not
  because of the treatment

19
Changing-Criterion Design

• This is a single-case design that is used when a
  behavior needs to be shaped over time or when it
  is necessary to gradually change a behavior
  through successive treatment periods to reach a
  desired criterion.
• Step 1 Get baseline
• Step 2 Administer treatment
• Step 3 Administer stronger treatment
• Step 4 Administer even stronger treatment

20
end