So the Obama campaign posted this yesterday. Discuss.
Wow… A great example how not to use colors…
I’m guessing the colors were decided on even before they picked the style of presentation. :P
I don’t mind blue and red, but the “shades of red” in the legend are just plain stupid^Wmisleading: 51-75% in grey (and not on the map)?
Haha, yeah the subtle partisanship is clever :-P
Agreed. How hard would it have been to make the 26-50, 51-75 and 76-100 buckets gradually darker shades of red?
The blue makes sense, it is a “soothing” color, imparting a sense of well-being.
PSYOPS in action.
Perhaps someone will make a chart “How Much More Men Pay for Car Insurance”…
Nice little use of the “standard” political colors to make it feel like the other party is the reason for the difference.
Basically health insurance is not allowed to take any standard risk factors into account. I wonder if they will start taking regional differences into account?
I agree with you that it’s unfair for men to have to pay more for auto insurance too, but the two situations are a little different, IMO. Men are more likely to be involved in auto accidents, and hence the higher insurance rates. Women are more likely to use health care services, in large part due to maternity care needs and screening for cancer, etc, which is why their health insurance rates are higher. To me, there’s a huge difference in volition or inevitability between the two.
How can it be that having an abortion is a choice, but having a baby is inevitable? And the justification for men paying higher for car insurance is the same as for women paying more for health insurance – statistically, that guarantees the optimal spread of risk and profit for the insurance company, which is their raison d’etre.
Perhaps they should have made a map showing in which states the insurance premiums will go up for private insurance (the same as the second chart – all states) and showing in which state the employers will be less likely to offer health insurance to employees (the same – all states).
I *hope* there will be a *change* …
In general ‘the other party’ probably are responsible for the situation.
It’s hard out there for a man, AM I RIGHT FELLAS?
Someone should make that chart, because from what I’ve googled, the greatest difference in auto insurance is in Wyoming, where men pay 20% more on average than women. Compare to health insurance, where in Wyoming, women pay 75% more on average than women.
Coupled with the fact that health insurance is an order of magnitude more expensive than auto insurance (and, of course, the fact that you don’t have to own a car but you do have to own a body), the net is still enormous discrimination against women.
But you know, them mens. They got it rough.
All of those who are replying by saying this is comparable to men having to pay higher auto insurance. PLEASE read: Only young men pay higher auto insurance, and that is because they statistically get in MANY more accidents than young women. Furthermore, healthcare is much different than driving. Driving is a privilege, and young men statistically abuse that privilege. Health care is not more costly for women because they abuse it, it is because they are more likely to actually be seen by their doctor, especially because they get pregnant…. BY MEN! Anyways, just to diffuse the comparison :) Thanks. (I tried to post it here the first time ;)
And we still don’t know if women would pay less than they are currently. How about some actual dollar amounts..
It doesn’t matter if both men and women pay more, as long as men and women pay the same.
Aren’t Leftist values wonderful?
This is a horrible case of data presentation because of all that ‘data not available’ – seriously? The President can reform health care nationwide and can’t get health insurance rates in a dozen or more states? That’s ludicrous. Then there is the source of the data. He can’t get his data from an industry standard publication or a government agency? He needs to use a partisan advocacy group? That hints of corruption in the data.
As for the issues of policy themselves; already the states are empowered to make decisions about this and other health care policy matters. The voters in California have spoken, as have those in Texas.
BenK – Most of the gray is the darker shade representing “0-25% more,” not “data not available.” The only states with no data that I see are in north New England. But your comment underscores other comments already posted here – that this has some color issues. Why use two shades of gray to represent no data, 0%, and 1-25%? Can be misleading and easily misunderstood.
Actually, it looks like only three states are shown as “data not available”: Maine, Vermont, and New York. The color choices aren’t the best, though, so it’s easy to misread which light-gray is which.
I guess that is another comment on the color. I see 3 states where the data isn’t available. I think the dozen or more you reference are in the 0-25% category…
hmmm… those prior comments weren’t there. I didn’t really think this required a third comment saying the same thing.
I stand corrected on how much data is missing. Although, missing the data for NY? That’s no small oversight. Also, this does highlight how the actual color scale is absolutely unacceptable.
Also, one can read quite a bit into the eradication of state lines on the right side. This strong statement of political philosophy does not, frankly, respect the Constitution.
Also, what about the fact the women pay for more haircuts? When are we going to see a Haircut Affordability Act?
Men don’t value their health nearly as much as women, thus they are less likely to go as often.
I agree though, who made the color choices on this? I couldn’t immediately find the data no available states.
“When are we going to see a Haircut Affordability Act?”
I’ll give this a try…
When no other viable and affordable haircut options or markets exist.
AND when haircuts become critical to health and wellness of our society.
AND when haircuts are denied because of pre-existing hair conditions.
AND when 50 million citizens are un-haircutted.
AND when haircuts are provided through a system of pooled resources where the user is different then the payer.
AND when the haircut market includes significant adverse selection and information asymmetry.
AND when that adverse selection negatively affects the haircut market for others in the haircut market (your lack of haircut drives up the cost of my haircut).
AND when raising haircut costs significantly outpace inflation.
Just a few, I’m sure we can come up with more.
AND when Emergency Salons are required to give you a haircut without regard to your ability to pay if you haircut situation is deemed life threatening.
@KErickson, Men also don’t have a babies grow inside of them.
Sorry, didn’t mean flippant with a serious issue. Yes brnpttmn there are massive differences between the two markets. All your points are valid.
If it were another market (say non-life dependent), and if there is support in the data to charge one set of people more than another, would it be wrong to do so in a market?
Some states permit insurance companies to offer maternity and women’s health as a separate policy rider. This is the case in NC, where the cheapest publicly available insurance for a young woman adding maternity coverage comes close to doubling the total premium. I am surprised they did not take this into account on their data, since it would lead to a lot more ‘red states’ – of some strange shade of red or gray (I think that’s been covered well at this point).
Feels like one of those ‘know your data’ issues.
I am trying to leave the politics out of it, but for the ‘men pay more in car insurance’ folks out there, the last time I checked there is only one driver of a car. Maternity coverage implies the involvement of a guy at some point in time or another, but the health burden is entirely on the woman.
There is unfortunately no way to leave politics out of this situation because the relationships among regions, religions, families, individuals, genders and generations are all intimately involved. Now, maternity does not necessarily require a guy – taking banked donations as the only contribution. Part of choice and control involves responsibility, including financial responsibility, even liability.
It’s easy to over-emphasise the maternity component of healthcare costs for women; there are actually a heck of a lot of other illnesses that women are more susceptible to, simply by virtue of having a lot more ‘going on’ internally than men do (obvious puns…).
Here in New Zealand, males and females pay pretty much the same premium for life and critical illness insurance – what women save on life insurance is offset by critical illness expenses.
Is is true that there is no state in which women (according to the data from National Women’s Law Center) pay less than men?
Also, I am not sure that NWLC is trustworthy on this issue.
If the state allows gender rating, then women most likely pay less than men if they’re older than 55. It’s just that there is more “weight” in the younger ages when you calculate the average differential. I think the numbers in the exhibit are correct, but we all know averages can be deceiving sometimes.
You can verify that, in general, 60 year old women get charged less than 60 year old men by going to ehealthinsurance.
For the record, I support the ACA.
Just noticed the footnote at the bottom. This applies to 25 year olds only, so they aren’t calculating an average.
It’s politics as usual.
Scare tactics to ensure the female vote in November.
If this is your idea of “scare politics,” what would this be? http://bit.ly/HpRfny
Identify or interest politics, sure, but on the scale of scare politics this doesn’t really register.
Could this not be sorted out by splitting the cost of maturity care with the biological father (or anyone who wants to take the cost but isn’t the biological father)? Or as another incentive stop teenage/unwanted pregnancy, having people declare the desire to get pregnant and increase their insurance to cover it. Wait thought all this through and its a huge mess with many hole for people to take advantage and to screw innocent victims. However, its the thought that counts.
I like the “poetical” approach of the maps, without having to use visual lies. The “before” map shows a complicated situation of inequality, and the “after” map shows how equal would be the situation, and also reinforces the concept of a equal country, without state borders. It’s a great way to transmit a message, and they didn’t had to cheat on data (at least not visually).
What I’m not sure if it’s good is that the first map shows some states that already have this situation, and we can make an interpretation like, these states are “examples” to be followed. I should know more of the USA politics to have an opinion.
I don’t like the illogical color scale on the first map, though I like the stripes to show another kind of information. The “red is bad” association of color works great for a Democrat, lucky him.
They should have included one more map. The third would look just like the second but the caption would read, “States Where Women’s Health Care Costs will Decrease with the Affordable Care Act.”
I think this is a wonderful piece of propaganda. The map on the left is so difficult to relate to it dissuades you from reading it. It actually pushes you over to the clean map to the right. The odd color choices on the left go against the norm to drive home this awkwardness. I think the average reader knows that there should be a logical progression in the choice of colors. Meanwhile, the texture of the fill is wonderful for it’s lack of flatness.
As an infographic this is awful because it communicates a forced agreement as opposed to an understanding of the data. Agree with the right because it is less painful to your eyes than the other. This is purely political, but that is to be expected from a political campaign.
So, let’s make the map for how much less men pay for heath insurance.
Why should men pay more, if women cost more?
Now that women make up the majority at college, and are well positioned for success in the white collar marketplace, they can afford to pay their way.
Women pay more because they cost to insure …. It is absurd to think that payment should not be based upon costs … Why should anyone be required to pay for someone else … There is nothing special about healthcare that requires the healthy to pay for the sick
This is a *highly* misleading visualization, regardless of whether you support ACA or not.
Why do I say this?
1) The perceived visual change is proportionally much greater than the change in the underlying data.
In most heat maps, there is a gradation of color that demonstrates the progression in small increments. However, in this case, the data “steps” are fairly large. For example, if a state moved from 25% to 26%, they would change from gray to red! It would have been much more accurate to move from light yellow to yellow to orange to red etc. for every 5%, which would have been easy for the campaign staff to do.
Is it possible that a choice was made to represent a continuous data set in a very discrete way in order to heighten the contrast between the groups of states?
Also, in looking at any particular state in the way depicted here, we have no idea whether a state is at the top or the bottom of these ranges. For the skeptical, this introduces the possibility that these distribution ranges were chosen for a reason. For example, what if most of the states in the 25-50% block were in the 25-30% range? Without seeing the state-level data, we have no way to know if the ranges presented to us are a fair representation of the true data.
2) The color scheme and sequence is irrational, confusing, and could easily give a casual reader a false impression.
Assuming we’d made the same color choices (insert chuckle here), many of us would sequence them in order of color intensity – medium gray, dark gray, red, dark red (excluding light gray and blue, which convey no comparative data). However, notice that the actual sequence is medium gray, red, dark gray, dark red!
This is so irrational that it is hard to imagine that this was not a conscious choice.
Why would someone do this? I’m glad you asked!
What do you first notice when you look at this graph? If you’re like me, you noticed red versus blue.
However, this is a false comparison – these colors represent states where ACA has absolutely no effect (blue) and states that have a medium effect (red, 25-50%). However, a casual reader is likely to interpret the bright red and blue as states with relatively low and high differences in cost, respectively.
Ok, so what? My main issue here is this: why was the color red used for the second-lowest tier of states?
Notice that bottom two tiers of states, 0-25% and 25-50%, make up the large majority of states. Only a few states have a difference of 50% or more. However, if the gray colors had been used for the 2 lower tiers, and the the red colors had been used for the two higher tiers, only a few states would have been colored red.
That’s not very dramatic! Is it possible that red was assigned to the 25-50% range so that there would be more “red” on the left chart? This would heighten the appearance of the “problem” which is being solved by ACA.
Look! There’s so much less “red” on the chart on the right!
This color sequence could also easily give a casual reader the impression that a large number of states have a large difference between rates for men and women, when in fact these states are in the middle in terms of this difference.
One other note is that the highest level of difference in rates (dark red) is less vibrant than the red used for 0-25%. Again, this would be confusing to a casual reader, who might think that the bright red states represent the largest differences.
3) As others have mentioned, using red and blue in a political context could imply political responsibility for the “before” and “after” charts.
4) This chart is a representation of the country by the geographical size of the state, rather than population!
This introduces all sorts of problems, where a reader could get the wrong impression of the true impact of the law because, for example, Texas is 4-5 times large geographically than Florida. What really matters here, and what ought to be shown, is the number of women in each state that will be affected.
5) The segment of the population being represented here is highly focused, and a casual reader could easily generalize this to differences between women and men of all ages. The choice to use 25 year-olds (when women are at a child-bearing age) also serves to display a much more dramatic difference in rates. Also, because men and women are much less expensive to insure when they are young, the differences in rates are more pronounced. While the campaign staff did point out the focused nature of the comparison, these types of footnotes are often missed by a casual reader.
6) The chart on the left compares what women pay to what men pay, pre-ACA. However, it *should* compare what women pay pre-ACA to what they will pay post-ACA, which illustrates the true difference. However, this would result in a smaller apparent difference between the pre- and post-ACA environments, and might lessen the impact of the comparison.
7) Unless this comparison is based only on prices for single adults (and that does not appear to be the case), the reality is that many of these women are part of a family where both a man and a woman are insured. So in the post-ACA world, the price for the man will go up to subsidize the woman’s expected costs, but the total price for the family will not actually change very much! How many of these women are single or in a family environment where their total premium costs will actually go down? That’s what I’d like to know.
I don’t think it is too much of a stretch to say that displays like this are designed for visual appeal and impact, as well as political persuasion, rather than to provide a cold look at the data.
Wouldn’t it be fun for me to create a similar visual based on the same data but with a different “worldview”, and see how different it might look? Objectivity is in the eye of the beholder, friends!
All of those who are replying by saying this is comparable to men having to pay higher auto insurance. PLEASE read: Only young men pay higher auto insurance, and that is because they statistically get in MANY more accidents than young women. Furthermore, healthcare is much different than driving. Driving is a privilege, and young men statistically abuse that privilege. Health care is not more costly for women because they abuse it, it is because they are more likely to actually be seen by their doctor, especially because they get pregnant…. BY MEN! Anyways, just to diffuse the comparison :) Thanks.
sarah are you some kind of idiot? whats your point. there are different insurances with different risk groups in those insurances. so the comparison is not “diffused” by you.
Ralph, you are so sweet;) no. That’s not my point. I mistakenly posted this comment here, but meant to post it in response to a different string of comments. My point might be better understood by you, well if you read that string first. My point is responding to people who were saying, “Why is it ok for young men to pay higher rates for auto insurance, but not OK when women pay higher health insurance.” Driving=privilege; Healthcare=necessity. Some advice, Ralph: If you want to have real conversations, your opening line should be more graceful than, “Are you some kind of idiot.” Don’t let the anonymity of the Internet allow the conversation to be so sour, friend. Take care.
Become a member.
Learn to visualize your data.
From beginner to advanced.
What you get
The individual data points of life are much less predictable than the average. Here’s a simulation that shows you how much time is left on the clock.
For various occupations, the difference between the person who makes the most and the one who makes the least can be significant.
We’ve seen that we can learn from what people search for, through the eyes of Google suggestions: state stereotypes, national …
I call myself a statistician, because, well, I’m a statistics graduate student. However, the most important things I’ve learned are less formal, but have proven extremely useful when working/playing with data.