Thread Closed 
Week6: Whether to Switch Obamacare Plans or Stay the Course
09-29-2014, 02:41 AM (This post was last modified: 09-29-2014 02:45 AM by MukteshKhole.)
Post: #1
Week6: Whether to Switch Obamacare Plans or Stay the Course
Data Source: McKinsey Center for U.S. Health System Reform

Image and Article: link

[Image: Image1.png]

[Image: Image2.png]

The visualization shows percentage change in the premium rates for Health Insurance silver plans. The author is trying to convey the information that it might be cheaper to shop for new Health Insurance plans rather than to renew cheaper plans bought in 2014.
The data from McKinsey Center for U.S. Health System Reform shows that people who renew their plan from 2014 will have to digest an 8.4% increase in premium rate on an average. While if they switch to a new cheaper plan in 2015 they will have to bear an increase of just 1% on an average.

The graph is interactive, providing options to select between 'Renew cheapest silver plan from 2014' and 'Switch to cheapest silver plan in 2015'. Each of these options shows a typical (average) percentage change in the premium rates. Data Set type is Geometric, attribute type is ordinal (color saturation).
The visualization is using color saturation channel to show percentage change in premium rate. Which is a good decision as the colors used are quite intuitive and saturation clearly highlights the areas where the change is maximum. A legend at top shows what color saturation level maps to what percentage change. It also shows a typical (average) percentage change for a given choice (renew plan from 2014 or switch to a new plan in 2015).
Another aspect which I liked about this visualization is if we hover over a particular area it shows the amount that one needs to pay for the cheapest plan available also the percentage change in premium rate on an overlay.

Let us analyze this visualization using Tufte's principles:
Labeling using color coding is sufficient to understand the main message of the plot. Data variation is shown as color saturation. Using color coding, option to select action (renew or buy a new plan) and overlays (layering) user is presented with a large number of ideas, in a small space. Though the amount of ink used is more (because of color coding) it makes it easier to read the plot. Thus we see that most of Tufte's design principles are followed
.

Let us analyze this graph using goals of information visualization:
The visualization is making large amounts of data coherent without any clutter. Information is presented from two different viewpoints - renew old policy and switch to a new one. Information is presented at various levels of details - color coding to give an overview and overlays to give more detailed information. This color coding and ability to select different options (switch to new plan or continue old one) allows visual comparison. It does tell a story about the data which is the message the author is trying to convey - that is might be cheaper to switch to a new Health Care plan. So in all this visualization conforms to all goals of information visualization.

A downside of this visualization is that the color used for states for which no data is available closely matches the color showing 0 to -5% change in premium rate. This was a bit confusing. Also side-by-side comparison of districts/states on the lines of whether to renew the plan or to buy a new one is not possible here, so we need to either go back and forth between the two options to see which state/district is at loss for a given action (A separate table summarizes this but it is not part of this particular visualization).

Overall, this a good visualization which clearly tells the story about increase in premium rates and that it might be better to buy a new silver plan in 2015 rather than to renew old one.
Find all posts by this user
09-29-2014, 08:25 AM
Post: #2
RE: Week6: Whether to Switch Obamacare Plans or Stay the Course
I think the visualization was very good. The color palette seems to work very well (except for the "no-data" color as was mentioned) and the rollover text allows an easy comparison of switching vs renewing within one area. My biggest complaint with this visualization was the use of the word "Typical". Reading the article indicates that the "Typical increase" is the average increase. In every day English we use "average" to talk about the "mean", so we can reasonably guess that the "Average increase" is the "Mean increase". The word "Typical" is more flexible. Just reading the visualization I didn't know if the slider showed the "Mean increase", the "Median increase", or the "Mode increase". I had to guess that it was the mean.
Find all posts by this user
09-29-2014, 12:28 PM
Post: #3
RE: Week6: Whether to Switch Obamacare Plans or Stay the Course
I would agree with your critique. I really liked how you pointed out the colors are a good color scheme. I think that this is really good. I would agree that it does clearly show the data and helps the consumer to read the information about the plot. I would also say that the hue of the color does help to show what the data is trying to read. I would also agree that it might be a bit confusing if the color of the states is similar to the color of the lowest value. If you don't know the state geographical area then it might be confusing. But I am not so sure that it is that big of deal because you as the consumer will be driven to the darker color of red.
Find all posts by this user
09-29-2014, 05:51 PM
Post: #4
RE: Week6: Whether to Switch Obamacare Plans or Stay the Course
This is a graph that conveys the idea which it is designed for - to show that switch to a new plan generally is cheaper than to renew the original plan. However, I wouldn't agree it is clear enough. First, it took me a while to realize what the grey color and orange color stand for. Second, it is tedious to switch the categories back and forth between "Switch to cheapest silver plan in 2015" and "Renew cheapest silver plan from 2014" in order to see the difference in each area. I would rather to see a simple table listing all the areas and rates for the two categories. Third, I think it needs more categories for grey color to show the difference of the rates below 0. For example, the area in the middle has a rate as low as -29.3%, yet the saturation is the same as the one in the upper left - only -6.2%. I think it needs at least one more grayer color to show the difference between these two.
Find all posts by this user
Thread Closed 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)