2d. There will always be reviewers that hate case studies. You can't win them. You can win reviewers who are borderline if you can show them that you can estimate understudied populations/ new mechanisms with the data. Do that.
2e. With the dissertation > Ask a question you genuinely don't know the answer to-- it'll make the process more interesting to you and you won't be disappointed if you get a result you didn't expect.
If you’re interested in the nitty gritty: In the US, we have no federal sick leave --12 states DC have sick leave laws and no policy is alike! This variation means that we can compare different design features to see which features are most effective at improving health.
I looked at three elements of policy design: policy generosity (accrual rate), inclusion (firm size threshold exempt worker types) and autonomy (prescribed types of uses) to see if these design features had differential impacts on flu rates
1. Income volatility (changes in income) are large and have consequences for health! More volatility > less food security, lower self-reported health, tough choices between paying bills or getting care and more stress 2/