Panel Data Help - Fixed effects vs Random effectsWatch
I am studying the link between state level military spending and state level inequality in the USA between 1994-2009. Specifically, I am interested in the effect that a shift towards a more capital intensive military (greater ratio of equipment/soldiers) has on inequality. I observe a balanced panel data set.
My problem is with methodology. I am trying to write the section where I justify the use of either fixed or random effects, and I cant seem to figure out which method I should use.
The only previous research into this topic is a 2001 paper, which examined 82 countries over 30 years, using random effects. (http://www.sciencedirect.com/science...49089X1100233X )
Their reasoning is as follows:
'Fixed effects can be interpreted as essentially "throwing away" between country variation present in the data. This is particularly relevant for the current study since the majority of variation in inequality is between nations, rather than within nations through time.'
One of the reasons I give for examining the US at the state level is that analysing different states has an advantage over analysing different countries because of the heterogeneity that exists between states - the example I give is that the differences in culture are far slimmer between California and New York than between the UK and Nigeria, which should allow us to net out cultural differences as a driver of inequality; to a certain extent.
My question therefore, could anyone make a suggestion as to which method suits my question/dataset better? If I haven't provided sufficient information please let me know!
Hope someone can help.
While you're waiting for an answer, did you know we have 300,000 study resources that could answer your question in TSR's Learn together section?
We have everything from Teacher Marked Essays to Mindmaps and Quizzes to help you with your work. Take a look around.
If you're stuck on how to get started, try creating some resources. It's free to do and can help breakdown tough topics into manageable chunks. Get creating now.
Not sure what all of this is about? Head here to find out more.