Week 6, Day 2: Designing Experiments

Spend some time discussing the first journal assignment.

Was it harder than you thought?
Why?

• First get your theory.
  • What is the behavior/cognitive process that you want to examine?
  • What do you think affects that behavior/cognitive process?
• Next you need to derive predictions from the theory.
  • These should be stated as hypotheses.
  • In terms of conceptual variables (control, independent, and dependent)
  • Now you need to operationalize your variables in terms of how they will be controlled, manipulated, and measured in the experiment
    • Be aware of the underlying assumptions that link the conceptual variables to their operational variants

  Can you control your extraneous variable(s)
  • Can you keep them constant?
  • Should you make them random variables?
  • Two things to watch out for:
    • Experimenter bias - the experimenter may influence the results (intentionally and unintentionally)
      • Clever Hans - a mathematically inclined horse. It turned out that his trainer was giving (unintentional) cues to the horse that told the horse when to stop counting.
      • One solution is to keep the experimenter "blind" as to what conditions are being tested
    • Demand characteristics - cues that allow the participants to figure out what the experiment is about, influencing how they behave

  Choosing your independent variable(s)

  Method of manipulation:

  • Subject manipulation
  • Event manipulation
  • Stimulus manipulation
  • Instructional manipulation

  Choosing the right range of your independent variable
  • Review the literature
  • do a pilot experiment
  • pick a large enough range to show the effect
  • be realistic
  • consider the costs, your resources, your limitations
  • pick levels found in the "real world"