L2: methodology of experiments Flashcards
What is an experiment?
A controlled data generating process
How doe we establish causal variation?
Controlled variation (ie. changing one variable at a time)
Drawback of lab data?
Cannot ensure external validity
Difference between experiments and lab?
Labs create an artificial environment to generate data, experiments attempt to impose control on naturally occurring processes
Benefit of field data?
They have automatic external validity
Experimental economics…(see)
lecture notes
Explain reasons for economic experiments?
Test theories, explore causes of a theories’ failure, compare environments, design policies etc.
Define a treatment?
A particular condition of an experiment
What is a session?
In a session a group of people take part in the experiment at a particular time and place
6 steps to an experiment?
1) Research question
2) Experiment design
3) Hypotheses
4) Preparing experiment (ie. where, how, etc.)
5) Run experiment and collect data
6) Data analysis and write up