Methodology

Psy 5054 ]


Experimental Methodology

  • Theories
  • Empirical Thinking/The Role of Experiments
  • Critical Thinking/Evaluating Experiments
  • Statistical Tests
  • Should you read Stanovich?
  • Experimental Reports

Theories

  • The Role of Theories
  • General Criteria for Evaluating Theories

The Role of Theories

  • "A theory is a set of principles that can be used to account for, predict and even control certain phenomena."
  • Most theories make causal claims.
  • Science can be viewed as a continuous process of formulating, evaluating, reformulating, and abandoning theories.
  • All theories are wrong!
  • Theories can lead us astray, so it is important to be critical of theories.
  • Gravity
  • Disease

General Criteria for Evaluating Theories

  • Is it consistent with observable facts?
  • Sufficiency (Not Uniqueness)
  • Does it explain everything that needs explaining?
  • The fact that one theory is sufficient doesn’t mean that other theories aren’t sufficient as well.
  • Parsimony
  • The fewer assumptions a theory makes, the better!
  • Generality and Completeness
  • Does it explain one domain or many?
  • Does it explain all of the domain, or just some of it?
  • Power/Falsifiability
  • Could it explain anything, including the impossible?
  • Can it be disproven?

Empirical Thinking/The Role of Experiments

  • Let’s Make a Deal!
  • What should you do?
  • Ask Marilyn

Let’s Make a Deal!

  • Montie Hall gives you a choice of 3 doors.
  • Behind one is a car, behind the other 2 are goats.
  • You choose door A.
  • Montie shows you that there is a goat behind door B, and offers to let you switch to door C.

What should you do?

  • No Time to Prepare
  • Keep door A?
  • Switch to door C?
  • Time to Prepare
  • Reason it out?
  • Ask an expert?
  • Something else?

Ask Marilyn

  • Who is Marilyn Vos Savant?
  • The "Three Doors" Column in PARADE Magazine
  • The Experts’ Response
  • Who’s right?

Always Switch

Critical Thinking/Evaluating Experiments

  • Five Questions
  • Psychokinesis (PK) Example
  • Does this constitute evidence for PK?
  • Some Things that Go Wrong with Experiments

Five Questions

  • What exactly is the claim am I being asked to believe?
  • What evidence supports the claim?
  • Are there any alternative interpretations of the evidence?
  • What additional evidence do I need in order to evaluate the alternatives?
  • What is the most reasonable conclusion?

Psychokinesis (PK) Example

  • "PK is the ability of some people to bypass their motor mechanism and, by volition, reach directly into space and time outside themselves to exert a force."
  • Evidence for PK?
  • A California televangelist gave a sermon in which he advised viewers on methods for identifying Satanists. One was to watch for a hand signal they use to identify one another.
  • During a subsequent telecast, an "experimenter" and 4 "trained witches" tried to use PK to force the televangelist to make the sign himself.
  • During the 30 minute live broadcast, he made the sign 23 times!

Does this constitute evidence for PK?

  • What exactly is the claim am I being asked to believe?
  • What evidence supports the claim?
  • Are there any alternative interpretations of the evidence?
  • What additional evidence do I need in order to evaluate the alternatives?
  • What is the most reasonable conclusion?

Some Things that Go Wrong with Experiments

  • Operational Definitions of Cause (IV) and Effect (DV)
  • Reliability
  • Validity
  • Baseline Measures/Control Groups
  • Correlation versus Causality
  • The Directionality Problem
  • The Third Variable Problem (Confound)
  • Randomization and Matching
  • Procedural Artifacts
  • Replication and Converging Operations
  • Generalizability

Statistical Tests

  • You don’t need to know statistics to make it through this course.
  • It is helpful to understand the following at a conceptual level:
  • Correlation Coefficient (r)
  • t-test (t)
  • ANOVA (F)

Should you read Stanovich?

  • Yes, if you don’t feel comfortable with the following concepts:
  • Independent Variable (IV)
  • Dependent Variable (DV)
  • Operational Definition
  • Reliability and Validity
  • The Directionality Problem
  • The Third Variable Problem (Confound)
  • Randomization and Matching
  • Replication and Converging Operations
  • Generalizability
  • Otherwise, you needn’t bother!

Experimental Reports

  • Abstract
  • Introduction
  • Method
  •         Participants
  •         Materials
  •         (Design and) Procedure
  • Results
  • Discussion
  • References

The End!

 


Psy 5054 ]

The views and opinions expressed in this page are strictly those of the page author. The contents of this page have not been reviewed or approved by the University of Minnesota

This page was last updated on 02/01/00.