Download The Experimental Method: Understanding Causation through Controlled Experiments and more Study Guides, Projects, Research Design in PDF only on Docsity!
The Experimental Method
What is an experiment? How is it different from other
methods? < Purpose: to demonstrate causation, that A ---> B
- What are the requirements to demonstrate causality?
- Correlation
- Order. A must precede B.
- Control over other variables
- Extraneous variables and alternative explanations
The Experimental Method
< How are experiments different from other
types of research?
- Manipulated independent variable
- Control of organismic variables either by
- Random assignment of units of analysis to conditions of the independent variable, or by
- Assignment of each unit to all conditions, with controls on order of presentation
- Control of other variables by holding them constant
- What are extraneous variables?
- Can explain the findings of a study without resorting to the hypothesis.
- Lead to an alternative explanation of the findings from the one you had.
- “In an airtight experiment, there is only one rival hypothesis: chance.”
The Experimental Method
< Within vs. Between subject experiments
- Within has
- complete control of all organismic variables
- Possibility of one condition to influence others, possible “on stage’ and “demand” effects
- Between has
- Less chance of subjects “catching on”
- Higher error rates < Comparing the correlational and experimental methods. example: testing the hypothesis that people with low self-esteem will be more attracted to an accepting other
- Correlational: measure self-esteem, observe response when other expresses interest
- Experimental: manipulated self-esteem, then do the same
The Experiment Exercise
How do you do a between-subjects experiment? Doob &
Gross as example < Devise a “set-up”, including a cover story if needed < Construct independent variable and way to assign subjects to categories < Figure out how to measure the dependent variable
A student example: the waitress and the mints
< Set-up < Independent variable and random assignment < Dependent variable
Walk through instructions for experiment exercise
Further examples of past student experiments
Internal Validity
Campbell and Stanley: types of pre-experiments,
experiments, and quasi-experiments
X O Or O X O
- True experiments: random assignment to conditions
X O Or O X O
— O O – O
- Quasi-experiments – same as experiments, except with no random assignment to conditions
Internal Validity
< An example of a quasi-experiment: jobs for at risk youth in Milwaukee
- Describe study
- Purpose of study
- Initial design
- How it turned out
- What is the problem?
- Possible extraneous variables?
Internal Validity
Using these terms, what are the problems with pre-
experiments?
X O Or O X O
What are the problems with quasi-experiments?
X O Or O X O
— O O – O
(Without random assignment to conditions)
Why are experiments OK?
The Darley and Batson Experiment
Design
< Sample < Is this a between- or a within-subjects experiment? < “Set-up” < Measurement
- Independent variables
- Dependent variable < Findings
- What is an interaction effect? < Threats to internal validity
The Goldstein and Arms Study
Design
< Sample < Is this a between- or a within-subjects experiment? < “Set-up” < Measurement
- Independent variables
- Dependent variable < Findings < Threats to internal validity < Is this an experiment? Why or why not?
