March 22, 2010
The White House called it "the most significant health reform legislation this country has seen in decades." While this may be true, the long-term results will likely be determined at the polls the next few elections. We'll see. How has the healthcare debate impacted President Obama's approval ratings? The Guardian Datablog provides us the data.
First, I looked at the relationship between all four data points: For/Against the legislation and Approve/Disapprove of the President.
The upper-left and lower-right quadrants tell the story. Those that approve of the President are in favor of reform. Those that are against the President are against reform. I bet the majority those on the upper-left are Democrats (and watch MSNBC) and those in the lower-right are Republicans (and watch Fox News). Such a polarizing society we have become...sigh!
From there I looked at the trends in the polling. The poll results start June 27, 2009 and continue through March 20, 2010.
Do these trends support the scatter plots? Well, not really. Again, on the upper-left of the dashboard there appears to be a strong correlation between those that are for reform and those that approve of the President. However, it's clear that support for both the President and reform have been waning, even for his most ardent supports.
However, the revelations in the scatter plot are supported by the polls for those that are against legislation and disapprove of the President. See the bottom-right of the dashboard.
What does this all tell us? It tells me that politicians only look out for themselves; they make the decisions that they think will get them re-elected.
One other note to consider. As Michael Tomasky writes today:
- A record $3.47 billion was spent lobbying Congress last year. Not all of that was about health care, but a hefty chunk of it was, and the vast majority of it by corporations and associations that wanted to kill the bill outright or shape it to reflect their financial priorities.
Maybe this is the real impact on the results we're seeing.
March 3, 2010
Some of the issues I see:
- Of course, the 3D perspective. BARF!
- The data labels on the bars...Are they really necessary to communicate the message?
- The background color choice gives me the impression I'm somewhere deep in space with all of that black.
- Stop-light colors. We all understand why people use stop light colors, but this doesn't work for those that are color blind.
- The cylinders. A flat bar is always a better choice.
- The angle of the chart leads you to believe the height of the bars is different, when in fact, they're all the same height.
Here are some options:
However, given the small set of data, would a simple table be best? Really all I care about are the problems. None of the charts above give that "pop." What would I do? I would create two tables.
First, the raw values.
Second, the percent of the total for each claim type.
Does anything stand out to you in these tables?