Prove My Friend Wrong: Come Up with Ideas for an Economic Study

[clowntable]

Didn't mean to hijack.

1) Does graduating in the top X% of a non Ivy League school or graduating in the top Y% of an Ivy League school lead to a higher income (X<Y). Could be solved to find out at which X and Y the average income are the same. Pretty sure it's been done before.

2) Is there a correlation between interior problems and exterior policies (does unemployment lead to war). Needs to be done worldwide and even then the sample size might be to small.

3) Does an increase in the number of children in a family lead to a higher rate of divorces?

p.s.: econ: play more WW.

[econophile]

[ QUOTE ]
Didn't mean to hijack.

1) Does graduating in the top X% of a non Ivy League school or graduating in the top Y% of an Ivy League school lead to a higher income (X<Y). Could be solved to find out at which X and Y the average income are the same. Pretty sure it's been done before.

[/ QUOTE ]
The paper asks a similar question: Estimating the Payoff to Attending a More Selective College

[ QUOTE ]
2) Is there a correlation between interior problems and exterior policies (does unemployment lead to war). Needs to be done worldwide and even then the sample size might be to small.

[/ QUOTE ]
The most relevant things I can think of offhand are Andrei Schleifer's series of papers with La Porta, Lopez-de-Silanes, and Vishny that examine the relationship between legal institutions and economic outcomes and Daron Acemoglu's work on the relationship between political institutions and economic growth.

[ QUOTE ]
3) Does an increase in the number of children in a family lead to a higher rate of divorces?

[/ QUOTE ]
This paper finds that couples who have female children are more likely to get divorced: The Demand for Sons: Evidence from Divorce, Fertility, and Shotgun Marriage. One way of looking at the effects of number of children on divorce is to compare families that have twins to those that don't.

[ QUOTE ]
p.s.: econ: play more WW.

[/ QUOTE ]
I expect my WW partcipation to increase after I finish my thesis.

[NoahSD]

clown,
np. Nice suggestions.

Econ,
Nice. Are you just an econophile or an econ student/professor/omist?

[gumpzilla]

[ QUOTE ]
Are you just an econophile or an econ student/professor/omist?

[/ QUOTE ]

[ QUOTE ]
I expect my WW partcipation to increase after I finish my thesis.

[/ QUOTE ]

Usually indicates Ph.D student.

[NoahSD]

Right.. I suck. At least I got to make a nerdy word joke.

[Borodog]

[ QUOTE ]
[ QUOTE ]
Are you just an econophile or an econ student/professor/omist?

[/ QUOTE ]

[ QUOTE ]
I expect my WW partcipation to increase after I finish my thesis.

[/ QUOTE ]

Usually indicates Master's student.

[/ QUOTE ]

[clowntable]

I'm currently working on my thesis and am neither [img]/images/graemlins/wink.gif[/img]

[RicoTubbs]

The trouble with this type of thing is that you need both:
a) an unanswered, but interesting, economic question
b) a dataset that is likely to provide evidence on that question (for example, it will not only provide evidence that the answer is X, but that X is unlikely to be caused by some other correlated omitted variable).

Here's something that I think is interesting:
Behavioral finance (see Richard Thaler or Kahneman/Tversky, for example) has documented that investors' behavior seem to depart from the classic view of unemotional, rational, home economus. So, for example, studies have found that investors are loss averse and are likely to hold onto losing positions longer than winning positions because selling a losing position would guarantee a loss, while holding it (even though it might be negative EV) would provide the possibility of reversing the loss, which brings great utility.

Anyway, I think it would be interesting to study whether these behavioral patterns apply for poker players as well. There's a rich dataset, if you could collect multiple players' online databases (ideally, get it from the site), so you could do something like:
- Do players tend to play longer when they are losing vs. when they are winning (i.e., are they exhibiting loss aversion)

- Do players tend to increase limits when they are winning vs. losing? If winning, it would indicate kind of a rational "gambling a constant % of bankroll", but if losing it would indicate more of a behavioral bias.

I thought Levitt would do this, especially with the data collection through pokernomics, but I haven't seen anything.

Hope this makes sense - I'm a little drunk right now.

[LetItBe]

I was an econ major, did a paper on the effects of economic growth on the environment. (Turned out to be an upside-down U graph...found it interesting.)

[MrMon]

Is there is correlation between teacher pay and student success in public schools? You can choose to measure student success however you wish. Controlling for cost of living in a given area should be used somehow, perhaps pay of teachers as a percentage of average local wage or something like that, to level out the difference between say NYC and North Dakota. Or, you could just compare urban districts and their pay and success rates, so you have a more uniform student body.

You could also evaluate supervisor salaries and student success, but that obviously has no correlation by anecdote.