Part 1 - Tutorial Flashcards
What could be experimental aims?
1) speaking to theorists = testing theoretical predictions
2) searching for facts = repeat/modify experiments for more robust findings
3) searching for meanings =
4) “whispering in the ears of the princess” = dialogue with practitioners e.g. policy makers
start with theory, then make lab experiment -> inductive or deductive reasoning?
deductive
example for inductive reasoning in experiment?
repeat experiment over and over and observe something, then make new theory from that
What is decent reward structure?
- incentive compatibility,
- reasonable payment,
- monotonicity
- salience
What is monotonicity?
= more is better, prefer more to less
What is salience in experiments?
= instructions about payoffs are clear
What is dominance of rewards in experiments?
reward should be dominant reason for decision in experiment, other reasons for choice should be able to be neglected
What is the “parallelism precept”?
about critique on economic experiments being very artificial - that you can generalize if general conditions are similar
What does Asian disease example illustrate?
positive vs negative framing
What do we mean by “treatment” in an experiment?
particular condition of the experiment -> often “main” treatment and “control” treatment
What do we mean by “session” in an experiment?
experiment usually in several sessions -> in one session, one group of people take experiment
What do we mean by “rounds” in an experiment?
often a number of rounds/periods played by participants
What do we mean by “subjects” in an experiment?
participants in the experiment
How does deductive reasoning work?
top - down approach:
- > from general to specific
- > if x and y (premises) then z (lead to conclusions)
- > logical conclusion from premises
How does inductive reasoning work?
bottom - up - approach:
- > from specific to general
- > detect regularities from specific situations
- > as it depends on the strength of the argument and cannot be verified, some researchers do not accept this
What are the four types of validity?
– Conclusion Validity
– Internal Validity
– Construct Validity
– External Validity
When do inferences possess internal validity?
When we already assume there is a relationship, is a causal relation between variables properly demonstrated?
What is conclusion validity?
whether ther is an actual effect between variables (cause + effect on observational arena)
What is construct validity?
Connecting measurement + theory -> did we implement the constructs correctly, are we measuring the right effect?
What is external validity?
generalizability in terms of time, place, setting, people
How can history affect internal validity?
History: events between first and second measurement
How can maturation affect internal validity?
Maturation: processes within individuals like growing older/hungrier
How can testing affect internal validity?
Testing: effects of taking a test upon second testing
How can instrumentation affect internal validity?
Instrumentation: changes in calibration or observers change findings
How can statistical regression affect internal validity?
Statistical Regression: operating when groups are selected on basis of extreme
scores
How can biases affect internal validity?
Biases resulting in differential selection of respondents from the comparison groups
How can experimental mortality affect internal validity?
Experimental Mortality: differential loss of respondents
What is a Reactive or interaction effect of testing for external validity?
pretest might increase or decrease the respondent’s sensitivity or responsiveness to experimental variable — results unrepresentative for un-pretested universe
What are Reactive effects of experimental arrangements for external validity?
variables have a different effect in an experiment rather than in a non-experimental setting
What is Multiple-treatment interference for external validity?
likely to occur whenever multiple treatments are applied to the same respondents, because effects of prior treatments usually not erasable