1. Intro Flashcards
What is the standard economic assumption about honesty?
People will lie if it’s net beneficial to them
Describe die in a cup task
- Throw a dice twice
- Report the first roll
- Get paid according to reported roll
What treatments can be used in die in a cup experiment and how do they impact results?
Increase payouts, this appears to have no affect on honesty
Motives for being honest
- Lying aversion- people balance feelings of guilt with benefit from lying
- Lies in disguise- people refrain from lying to maintain a positive self image
Results of die in a cup experiment
People are partially dishonest but stakes have no impact.
What did the experiments selling newspapers in Austria find?
That a moral appeal (“thank you for being honest”) increases payments
How does honesty vary across societies that differ in corruption and rule of law?
Strong correlation between quality of institutions and honesty
Lost wallet findings
People are more likely to return the wallet if it has money in it
Natural experiment
Exogenous events like Covid 19
Natural field experiments
The experimenter conducts an experiment in a naturally occurring environments where the participants don’t know about the experiment
External validity
If lab conditions are also present in the real world
Why is field happenstance data, field experiments, and survey data so important?
Because there is uncertainty around the validity of lab experiments
Experiment
A controlled data generating process
What does control refer to?
- Rules of the game
- Experimenter knows the theoretical predictions
- Treatments allow causal inference
- Evidence is replicable
Induced value theory
- Basic idea: use money as a reward medium and make earnings in an experiment depending on the decision
- Goal: to map any incentive structure assumed in an economic model into an experiment that shares these incentives
- Assumption: people want more money as opposed to less
Assumptions of Induced Value Theory
- Monotonicity: subjects must prefer more of reward and not become satiated
- Salience: the reward depends on the subjects actions
- Dominance: changes in subject’s utility come from the reward not other motives
How to run an experiment
- Explain rules of experiment in neutral, non- associative wording
- Don’t use deception
- Participants do experiments and get paid
Where can you run experiments?
Classroom, computer lab, internet, household surveys, in the field
Pros and cons of classroom experiments?
+easy to administrate
- anonymity not ensured
- communication
- messy logistics
Pros and cons of computer lab experiments
+high control
-artificial environment
Pros and cons of internet experiments
+large numbers
-less controlled than lab
Pros and cons of household surveys
+representative
-control is unclear, only simple experiments possible
Pros and cons of field experiments
+natural environment
- hard to replicate
- poor control
