Experiments and Quasi-Experiments

1
Experiments and Quasi-Experiments

• (significance of group differences)

2
Overview

• Up to this point we have been discussing the
  relationships amongst variables where the same
  subject answers multiple questions
• e.g., what is the relationship between height and
  weight, what predicts bar passage, etc.
• We are now transitioning to a new topic of
  discussing group differences where different
  subjects are in different conditions
• e.g., who is happier, males or females?

3
Overview

• relationships amongst variables
• regression lines
• only tests CORRELATION
• group differences
• mean (average) of each group
• can prove CAUSATION
• The major advantage of group differences is
  proving causation

4
Correlation v. Causation

• Depressed Mood

Impaired Sleep
Cause?
Depressed Mood
Impaired Sleep
Cause?
Depressed Mood
Impaired Sleep
Cause?
Cause?
Family Conflict
5
Correlation v. Causation

• Finding Women who have a baby after age 40 are
  more likely to live page 100.
• Finding The greater the quantity of ice cream
  sold, the greater the number of murders.
• Finding The greater the number of Churches, the
  greater the amount of crime.
• Finding The more a person weighs, the larger
  his/her vocabulary.

6
Experiments

• (1) random assignment of Ss
• (2) to two or more conditions
• (3) which differ in terms of (only) IVs

7
(1) Random Assignment

• What is random assignment?
• every subject has an equal chance of being
  assigned to different conditions
• Why do random assignment?
• purpose is to prevent systematic and
  non-treatment differences among subjects in
  each condition

8
(2) Two or more conditions

• Two levels
• yes versus no (manipulate
  happiness versus no emotion)
• high versus low (manipulate high
  happiness versus low happiness)
• positive v. negative (manipulate happiness
  versus sadness)

9
(2) Two or more conditions

• Three levels
• Allows you to see direction of the effect
• Allows you to see shape of relationships

10
(3) Which differ in terms of (only) IVs

• In experiments, you manipulate variables.
• By only manipulating the IV, and keeping all
  other factors constant (via random assignment),
  then any change in the DV is due to the IV
• Thus, you can prove the IV CAUSED the DV

11
Comparing Correlation designs and Causation
designs

• Does watching violent TV make children
  aggressive?
• How would you conduct a correlational study
  testing this research question?

12
Comparing Correlation designs and Causation
designs

• Does watching violent TV make children
  aggressive?
• How would you conduct an experimental study
  testing this research questions?

13
(No Transcript)
14
Quasi-Experiments

• Quasi-experiments
• Contains aspects of both experiments and
  non-experiments because deficient in at least one
  of the three aspects of experimental designs
• Two most important are
• (1) Within-subjects measuring/manipulating same
  subjects at two or more times.
• (2) Mixed-designs containing both
  between-subjects and within-subjects designs

15
Quasi-Experiments

• When do I choose which type of design?
• Choose experiments!
• If practical issues prevent you from conducting
  experiment, then those same practical issues will
  dictates which quasi-experimental design you use.