Experiments Flashcards
What are the 3 basic steps of an experiment?
- Start with hypothesis
- Introduce stimulus
- Compare outcome with and without stimulus
Independent variable?
treatment or experimental stimulus (you manipulate)
Dependent variable?
Physical conditions, behaviour that change in response to the treatment (you measure)
Pre-test?
Measure of the dependent variable prior to the treatment
Post-test?
Measure after exposure
9 steps in conducting an experiment? Neuman
- Formulate hypothesis
- Decide on design
- Decide how to introduce the stimulus
- Develop valid and reliable measure of the dependent variable
- Set up experimental setting and run pilot test
- Locate subjects
- Conduct experiment
- Debrief subject
- Examine data
Classical experimental design?
- Randomly assign experimental and control group
- Post-test
- Stimulus for experimental group
- Post-test
Types of experimental group designs?
- One- shot case study design
- One group pre-test post-test design
- Static group comparison
- Two group post-test only design
One- shot case study design?
One group (no control group), only post-test, no random assignment
One group pre-test post-test design?
One group (no control group), Pre and post-test, no random assignment
Static group comparison?
Two groups, post-test only, no random assignment
Two group post-test only design?
Same as static group comparison
What are the 9 threats to internal validity?
- History
- Maturation
- Testing
- Demoralisation or compensation rivalry
- Diffusion of treatment (contamination)
- Experimental mortality
- Instrumentation
- Statistical regression
- Experimenter expectancy or compensatory treatment
Internal validity: history?
When an event occurs during the experiment and influences the dependent variable
Internal validity: maturation?
Change in subject’s psychological, emotional,etc state during the experiment
Internal validity: testing?
Hawthorne effect
Internal validity: Demoralisation or compensation rivalry?
Change in responses of control group due tot he deprivation of stimulus (get competitive)
Internal validity: Diffusion of treatment (contamination)?
When participants communicate with each other and thus prepare themselves/ change their attitudes/expectations
Internal validity: Experimental mortality?
When subjects withdraw from the experiment
Internal validity: instrumentation?
When measure of variable pre-test and post-test not consistent
Internal validity: Statistical regression?
Problem of extreme values - regression to the mean
Internal validity: Experimenter expectancy or compensatory treatment? + solution
Researcher’s behaviour threatens validity. Solution: double blind: subjects + researchers giving stimulus don’t know which are control of experimental group
What are the positives aspects of assessment?
- Cause and effect relationship are easy to see
- Isolation of dependant/ experimental variable
- Replication is easier
- Good for hypothesis testing
What are the negatives aspects of assessment?
- Complexity of real life not captured
- Limited to certain types of questions