Do Graphs Really Show the Whole Picture?

Ideally, graphs are supposed to be tools to aid in our understanding of a topic. But do they clarify or confuse? That depends. When the topic is about something as controversial as healthcare, misleading graphics can become a problem as well.

Rationing

David Leonhardt, writer for the New York Times, sheds light on the health care controversy as he discusses the reality of rationing in the health care system throughout his article. At the end of his article he discusses the effects of rationing and includes a graphic to illustrate.

Visualizing the Consequences of Rationing

The graphic that is included in the article isn’t included in the text. That is the biggest problem. There is a small image of it in the left column next to the article, which you can click on. Doing so opens a new window with the enlarged graphic. The graphic shows a few things
· Spending on health care as share of GOP 2006
· 5 year survival rates of people (in percentages) with diseases including breast cancer, kidney transplant, childhood leukemia, Non-Hodgkin’s lymphoma, and colorectal cancer in England, Canada, Australia, New Zealand and the U.S.


Graph Pros
One things that I did like about the statistical graph was that it did include one example of where the U.S. had high survival rates, a few in the middle, and one where the U.S. had low survival rates. So it did appear somewhat balanced.

Also, Leonhardt does tie the graphic in nicely with his article.

Graph Cons
One of the first things that stuck out to me, that is somewhat trivial, was that the percentages spent on health care as share of GOP were from 2006. The article was written in 2009, which leads to me to question if that was the most recent statistics they had or if those statistics were the ones that fit best with the point they were trying to get across.

Another thing that stuck out was the examples chosen to include in the graphic. The writer could have purposely chosen diseases that the U.S. didn’t have high rates of survival in. Put another way, the author could have included a country or several countries that had terrible health care and compared the U.S. to them. There is a reason the author chose the countries he did to compare with the U.S. and the diseases he did.

Another question that arose when looking at the graphic was the sample size. One would hope the sample size would be the same for each country and that the sample size would be large.

Response to Leonhardt’s article

Leonhardt’s article received some response. One of which was from Ezra Klein wrote a short response agreeing with Leonhardt. Klein
includes the same graphic found in Leonhardt’s and comments on the graphic but also includes information about countries not on the graph such as Japan and France.

What to take with you

The most important thing to remember when reading a graph or statistics of any kind, is why did the author include this particular one? Could it have been said or presented in a different way?