A Statistical Analysis and Graphic Data of the Affordable Care Act in America

The political support for the Affordable Care Act is something that is bound to vary throughout the country. Each state has an opinion on the new healthcare initiative enacted by Barack Obama, and it can easily be seen which states supported it. Looking at the information from the dataset health.csv, one can easily analyze and construct plots and tables that help to view correlations and lack thereof. One major correlation between two factors is that of percent_favorable_aca and Obama_share_12, which in plain English is much simpler.

Looking at the percent of people in favor of the Affordable Care Act and the percentage of those who voted for Barack Obama in 2012, one can see a very clear correlation. Putting the data of percent_favorable_aca and Obama_share_12, the graph in image 1 shows the correlation. It is quite clear that there is a strong correlation, which isn’t so unexpected, but is very strong. The fact that the states that supported Obama’s reelection are in favor of the Affordable Care Act is something that is expected but not directly linked. Elaborating, it is clear that those states and people who voted for Obama in 2012 wanted the ACA but it may not have been the soul reason for his reelection. There of course must be more variables to Barack Obama’s victory in 2012, but it is evident people wanted to have a new healthcare system.

Looking at the data, there is an almost perfect linear relationship between the support in states for the Affordable Care Act and the reelection of Barack Obama, making the data surprising. Again, the results aren’t something unexpected, but their near perfect structure is something that is uncanny. To look deeper into the data, one can use the information provided and run separate tests and analyze more information. Putting the data in the spotlight, this observation is one that has great significance, as it shows that Obama supporters wanted healthcare reform quite linearly and predictively. This correlation is not coincidental, clearly deriving from the promises Obama had made in his campaign and his eventual creation of the Affordable Care Act. One more variable that seems to be correlated is of course ideology score, which defines states’ political leniency. Looking at the variables and correlations discussed, it is clear that the best thing to do next is to calculate linear regression. Linear regression in statistics is basically a way to predict the data’s behavior beyond the information given and observed. In a way, linear regression is predicting the data based on the existing information given. Linear regression is useful as it allows anyone who desires the ability to speculate more information, for example, how many votes Obama would get if there were more democrats in each state, or how much support a bill would get if there were a certain party majority in the senate.

Even though the information is available for the real life numbers, the reality is that predicting is very important for statisticians and politicians alike. When running a linear regression test for the data of percent_favorable_aca and Obama_share_12, some interesting yet expected numbers come up, which could be seen clearly in the graph in this specific case. When run, the linear model test gives a positive number that is quite high and an intercept that is alike. A 0.793 as the coefficient shows that there is not only strong correlation, but generally a high chance of predictability in this scenario. Furthermore, this number shows that there is a clear connection between those who supported Obama and those who support his initiative on healthcare reform. Looking at the intercept, one can see a 5.440 which means that when the percent in favor of the ACA is made to be 0, the support for Obama is 5. With this information, one can predict the statistics of one variable or the other with just one number, and may extrapolate the data. This is important for real world situations where one can use this knowledge to predict bill votes, turnout, and overall favor for any regulation that may be related. In contrast, running a linear regression on ideology score and people in favor of the Affordable Care Act gives a negative number which if high enough could mean a negative correlation. The reality though, is that a -41.19 means a weaker correlation than before, and also negative.

Looking at this number, it is clearer to understand that predictability for ideology score and people in favor of the Affordable Care Act is less reliable but still doable. Furthermore, the intercept lies at 45.89, and as stated above, that helps to understand the data and create an equation. Linearly, a y=mx+b equation to calculate missing variables works, and in this situation, m would be the slope and b would represent the intercept. In conclusion, looking at statistical analysis and graphing the data helps to visualize the information and draw conclusions. By using graphs in R to determine correlations and compare two variables, one can understand how one thing affects the other. The way in which support for Obama and his Affordable Care Act correlated showed that there is clear demand for healthcare reform in the United States, as voters chose Obama for his promises. Moreover, using regression models to predict the data that is not shown can help with extrapolation and real world issues. In this case, and many others, looking beyond the given data is important and sometimes the goal. In all, the use of statistics in politics is important for information gathering, analyzing, and concluding.

Did you like this example?

363

19