PS11A Psychology Laboratory

1
PS11A Psychology Laboratory Methods Abnormal,
Social Developmental

• Charlene Coore
• 2006-2007

2
Introduction to Experiment

• Experiment vs. Correlational Study
• Internal Validity

3
Experiment vs. Correlational Studies

• An experiment is an investigation in which two
  (or more) equivalent groups of subjects are
  treated exactly the same way, except that the
  independent variable is different. Differences
  in measurements of the dependent variable can be
  attributed to the difference in the independent
  variable.
• Correlational study An investigation in which
  relationships between or among variables can be
  identified, but you cannot establish cause and
  effect because of the influence of uncontrolled
  variables.

4
Experiment vs. Correlational Studies

• An example of correlational study vs. experiment
• Placebo an inert substance or treatment that
  had no physical effect on the participants.
• Between groups research design a research
  design in which the performance between 2 or more
  groups is compared.

5
Experiment vs. Correlational Studies

• Equivalent Groups
• Random Assignment equal chance
• Selection Bias the confound arising when there
  are differences between the comparison groups
  within a study.
• Matching identifying pairs (triplets etc...) of
  participants who measure similarly on a
  characteristic that is related to the dependent
  variable and then randomly assigning each of the
  participants to separate experimental conditions.
  

6
Controlling Extraneous Variables

• Extraneous variables variables other than the
  independent variable that can affect the
  dependent variable.
• Confounded results results of an investigation
  that can be explained in various ways because of
  the presence of one or more extraneous variables
• An example
• A note on Pretesting.

7
Internal validity Confounds

• A Reminder on common confounds
• Experimenter Bias
• Demand Characteristics

8
Internal validity Confounds

• How do we protect internal validity?
• Single-blind procedure a research procedure in
  which either the participants or the experimenter
  do not know to which condition the participants
  have been assigned. E.g. either the participants
  know that they are getting alcohol but the
  researchers do not know who got the alcohol or
  the researchers know who got the alcohol but the
  participants dont know.

9
Internal validity Confounds

• Double-blind procedure a research procedure in
  which neither the experimenter nor the
  participants know to which condition the
  participants have been assigned. (Both
  experimenter and demand characteristics can be
  eliminated)

10
Internal validity Confounds

• Instrumentation effects - the confound arising
  when a measuring device fails to measure in the
  same manner across observations.
• Subject Mortality/Attrition the loss of data
  when participants withdraw from a study or their
  data cannot be used.
• Non-systematic subject mortality the loss of
  data when participants withdraw from a study or
  their data cannot be used for reasons unrelated
  to the experiment itself.
• Systematic subject mortality the loss of data
  occurring when more participants in one
  experimental condition than in another withdraw
  from the study.