SERforum

 

Join Andrew Olshan, for SERforum Live!

“Career and Professional Development”

JOIN SERforum – Live
December 5th, 12pm EST

 

Do you ever find yourself struggling to figure out a question about epidemiologic methods, or other topics in epidemiology, and don’t know who to ask? The SERforum allows for individuals to answer questions that come up in our daily work around substantive and methodological topics in epidemiology.
 
All topics may be viewed, but to read and post comments, SER membership is required. If you are member, login! Not a member, join us!

You need to log in to create posts and topics.

Impact of state-level policies on individual-level risk behaviors

I am conducting a study on the impact of state-level policies on individual-level risk behaviors. Can you help me understand the advantages of choosing a random effects versus a fixed effects differences-in-differences model?

I am assuming you have units grouped in states. I am unsure of the D-I-D aspect in this setting, but the key benefit of multilevel models is that they control for within and also between group variability. A non-random effects model naively does not and increases risk of a type I error (rejecting the null when it is true).

Side note, depending on the context - sometimes people living on a border may be able to defy the policy or have a different policy impact (e.g., drive across border to buy fireworks, so a ban may not impact them the same as rest of state, which this may need to be accounted for in geographical groupings.