politics | Geographical Perspectives

The Political Geography of Cancer in the US

Recently I posted data graphics illustrating the Political Geography of Heart Disease in the US. This post will take the same approach to examine Cancer mortality in the US.

I won’t rehash my approach here. If you’re interested please go back and read the initial post.

Here’s the scatter plot with Cancer Death Rate along the horizontal axis and % Voting for Mitt Romney along the vertical axis.

And here’s the breakdown by quadrant.

Quadrant 1 (upper left). These are politically conservative States (over 50% voted for Romney) with lower death rates due to cancer. States in Quadrant 1 include Alaska, Arizona, Georgia, Idaho, Kansas, Montana, Nebraska, North Dakota, South Dakota, Texas, Utah, and Wyoming.

Quadrant 2 (upper right). These are politically conservative States with higher death rates due to cancer. States in Quadrant 2 include Alabama, Indiana, Kentucky, Louisiana, Mississippi, Missouri, Oklahoma, North Carolina, South Carolina, Tennessee, and West Virginia.

Quadrant 3 (lower right). These are politically progressive States with higher death rates due to cancer. States include Delaware, Florida, Iowa, Maine, Michigan, Ohio, Oregon, New Hampshire, Pennsylvania, Rhode Island, Vermont, and Wisconsin.

Quadrant 4 (lower left). These are politically progressive States with lower death rates due to cancer. States include California, Colorado, Hawaii, Illinois, Maryland, Massachusetts, Minnesota, Nevada, New Mexico, New York, Virginia, and Washington.

Looking back at the same graphics for Heart Disease I see quite a bit of overlap. I plan to add posts for the other common causes of death and then I will do some sort of comparison/consistency analysis.

The Political Geography of Heart Disease in the US

As reported by several news outlets, the CDC recently published a report entitled, Potentially Preventable Deaths from the Five Leading Causes of Death – United States, 2008-2010.

The report includes tables with number of deaths for the 5 leading causes of death. They have some sort of formula for estimating number of “Deaths Expected” and then the authors compare those estimates to actual deaths to arrive at a number of “Potentially Preventable Deaths”. As I read about where preventable deaths were highest I couldn’t help but notice a familiar geographic pattern.

In order to see whether America’s cultural regions might express themselves through these health outcome data I decided to dig a bit deeper. Recently, I looked at Presidential Election voting percentages and compared them to Gender Wage Gap data. The results were intriguing so I’m taking a similar approach here.

Note: typically I don’t like to analyze State level data but in this case each State is both a health insurance regulator and a political actor. So here it makes sense.

First, I pulled in population counts and calculated death rates due to Heart Disease for each State. I decided simple mortality rates would suffice rather than using a more complex age-adjusted calculation. I realize this may produce some bias for States with older than average or younger than average populations. Since I’m not a slave to the silly world of academic/scientific publication I can do whatever I want. Na-nana-na-na. I don’t think it will make much difference.

Then, I compared these death rates (deaths per 1 million people) with the percent voting for Mitt Romney in the 2012 Presidential Election by producing a scatter plot. I love scatter plots – don’t you? Here’s the resulting graphic with Heart Disease Death Rate along the horizontal axis and % Voting for Mitt Romney along the vertical axis.

Now, I wish I had time to build a fancy cartographic brushing display so you could interact with the scatter plot and a map of the lower 48. Sadly, I don’t have so much time available so I’m providing a set of small multiple graphics to help you see what I’m seeing in this scatter plot. I could examine some of the closely clustered points to illustrate cultural homogeneity among several smaller groups of States but instead I’ve decided to simply divide the graphic into 4 quadrants.

Quadrant 1 (upper left). These are politically conservative States (over 50% voted for Romney) with lower death rates due to heart disease. States in Quadrant 1 include (in alphabetical order) Alaska, Idaho, Kansas, Montana, Nebraska, North Dakota, South Dakota, Texas, and Wyoming.

Quadrant 2 (upper right). These are politically conservative States with higher death rates due to heart disease. States in Quadrant 2 include Alabama, Georgia, Indiana, Kentucky, Mississippi, Missouri, Oklahoma, North Carolina, South Carolina, Tennessee, and West Virginia.

Quadrant 3 (lower right). These are politically progressive States with higher death rates due to heart disease. States include Michigan, Nevada, Ohio, and Pennsylvania.

Quadrant 4 (lower left). These are politically progressive States with lower death rates due to heart disease. States include California, Colorado, Florida, Hawaii, Illinois, Iowa, Maine, Maryland, Massachusetts, Minnesota, New Hampshire, New Mexico, New York, Oregon, Rhode Island, Vermont, Virginia, Washington, and Wisconsin.

Maybe because I’m a geography nerd I see something you may not when I stare at this graphic. To me it illustrates how regional culture is a key factor in so many behaviors, from voting preferences to economic disparities to health outcomes. Perhaps these data will be made available at the County level so I might explore in the context of my Cultural Ecoregions. It may also be interesting to see how cohesive these groupings turn out to be for the other leading causes of death.

Presidential Politics and the Gender Wage Gap in America

The Paycheck Fairness Act of 2014 was recently defeated in the Senate with Republicans voting unanimously against the bill. Conservatives at the Heritage Foundation argued the bill is based on faulty statistics, would hurt women’s employment prospects and invites too much government interference in the economy. On the other side of the political aisle, Democratic politicians and Women’s rights advocates argued it was a simple issue of fairness. I’m not familiar enough with the language in the bill to do any fact checking or know how it might improve upon or detract from the Equal Pay Act of 1963. In any case, I’m fairly confident both sides are misrepresenting the issue, making it difficult to have an open discussion. After all, it’s politics, right?

Out of curiosity I downloaded the State Gender Wage Gap data along with voting results from the 2012 Presidential Election in an effort to see how voting patterns correlated with wage gap data. As I tell my students, correlation does not imply causation so these data prove nothing. That said, there is a fairly strong positive correlation (Pearson Correlation Coefficient = 0.76) between % Obama votes and Female Wages as a % of Male Wages by State.

My wife tells me that I should restate the preceding sentence in plain English. Translation: States with a low percentage of votes going to Obama in 2012 are more likely to pay women less; likewise, States with a high percentage of votes going to Obama in 2012 are more likely to pay women more. Like I said, this relationship does not prove that political orientation causes wage discrimination. But, there is a correlation.

I’ve produced a scatter plot to help visualize the data relationship. Each circle below represents one State.

I plan to circle back on this topic to examine the geographical dimension. In the meantime, here is the same graph with a few labels.

The Geography of Voter ROI

Terrific map from Nate Silver’s Five Thirty Eight blog at the New York Times. Apparently my move from Michigan to Colorado increased the value of my vote significantly. The map is interactive so if you visit the Five Thirty Eight blog and find the map you can mouse over your state and see how important your vote might be in November.