Observational validity

• How can we improve validity in an observation

  -covert may show more genuine and real behaviour than overt
• make sure behaviour categories

  -are not too broad or overlapping
• ensure variables are

  -operationalised
• ensure the sample used is

  -representative