Week 2 : Elements of the Research Process Flashcards
What are models?
Abstractions capturing relevant features.
How are models judged?
By how useful they are, not by how realistic they are.
How do we calculate the chance of a rational voter?
R = p*B - C
What do p, B & C stand for?
p: Probability that a voter’s vote is decisive, that the vote will yield the preferred outcome.
B: How much more a voter prefers his/her party over the opposite, i.e. utility gain from getting the preferred outcome.
C: Costs of voting
What does it mean if R is positive?
The benefits outweigh the costs and an individual is more likely to vote.
What is a causal theory based on X and Y?
If X - the independent variable - is causing Y - the dependent variable.
How does the independent variable p have a causal effect on R?
As the probability of casting the decisive vote increases, we should see higher turnout.
How does the independent variable B have a causal effect on R?
For individuals that feel stronger about the consequences of the vote outcome, we should see higher turnout.
How does the independent variable C have a causal effect on R?
In areas where voting is more costly the turnout should be lower.
What data do we need to measure R?
- Turnout data from all states.
- Two different time periods.
- To measure the difference in level of education, average age of voters and if the registration has been made easier etc.
Give two examples of how we can we use hypotheses to formulate policy recommendations.
- Educate people about how elections make a difference (affecting B).
- Making participation easy (affecting C) or incentivising it (voter party).
How many steps are there to assess causal statements?
Four
What is the first step of assessing causal statements?
Is there a credible causal mechanism that connects X to Y ?
What is the second step of assessing causal statements?
Can we rule out the possibility that Y could cause X?
What is the third step of assessing causal statements?
Is there covariation between X and Y ?
