Are the uninsured more likely to end up in poverty?
Health insurance has a number of benefits, but perhaps one of the largest is its ability to shield individuals (and families) from catastrophic financial ruin. In the United States, health expenses are one of the leading causes of personal bankrupcy. Also, given that health insurance in the United States is primarily determined by employment, there are inequities in who is covered by health insurance.
We will do a brief investigation on the association between health insurance and poverty.
ACS_citizen_min_wage_shp4)health_insurance)Link the two datasets above together. You should begin with ACS_citizen_min_wage_shp4 and then merge it to health_insurance.
-Hint 1: load the following libraries: tidyverse, haven, sf and if you are able thatssorandom
-Hint 2: To load the ACS_citizen_min_wage_shp4.rds dataset, you’ll need to use readRDS
Generate two variables:
Poverty Rate (Hint: Haven’t we already done this? See Lab 8)
An indicator if a county has above median level of health insurance coverage. We will define these counties as “High Health Insurance Coverage”. (Hint: Lab 2)
Create a map of the United States that shows health insurance coverage on a county level. Do you notice any interesting spatial patterns?
Bonus: See if you can do a map that focuses on the West Coast (California, Oregon, Washington, etc . . )
Compare poverty rates between counties with high health insurance coverage vs. low. Do a t-test and present the difference in poverty rates visually. Report the relevant results from the t-test below the figure.
Using a bivariate linear regression, estimate the association between poverty rates and high health insurance coverage. You should then think about a possible confounder/omitted variable. Include that as a control and compare your results from this multivariate regression to the bivariate one.