Validity

1
Validity

• Review

2
Definition

• Does it (method/measure) measure what it is
  intended to measure?
• Internal Validity (was there a genuine effect?)
• External Validity (can we generalise the
  findings?)

3
Studies

• Poor Validity
• Milgram (?)
• Zimbardo
• Tajfel
• Loftus
• Gould
• Gardner Gardner

• Good Validity
• Piliavin
• Rosenhan (?)
• Samual Bryant
• Hodges Tizard
• Raine
• Deregowski

4
Threats to Internal Validity

• Type 1 and 2 Errors.
• Reliability to measures and procedure.
• Random errors.
• Order effects/testing.
• Drop-out.
• Sample bias.

5
Threats to External Validity

• Construct validity
• Mono-method bias.
• Demand Characteristics.
• Researcher bias.
• Generalisation to other populations.
• Ecological validity.

6
Methods of Testing Validity

• Content/Face Validity.
• Predictive Validity.
• Concurrent Validity.
• Construct Validity. (Relates well to
  theory/expectations).

7
Issue

• It is easy to note that methods that are usually
  reliable are not always valid and that methods
  low in reliability are often valid.
• It may be that much of human experience can not
  be studied in a restrictive scientific way to be
  valid.