As somebody who's spent some time studying the impact of various levers (income, educational achievement, maternal education etc) on university/college enrolment I'd like to throw in an economist's perspective that the NY Times did not include in the chart.
You may draw the conclusion from the chart that income prevents the poorest from attending university because they can't afford the fees. This is highly misleading. There's substantial evidence [at least in rich countries, see US example below] that when you control for educational achievement (i.e. how well students achieve in pre-uni schooling), income level is a much weaker predictor of whether someone attends university. The primary driver of low university enrolment is educational achievement. i.e. the main chain of causation is this: low income -> low education achievement -> low college attendance. The key to raising university enrolment among poor students is to provide better education/development opportunities and aspiration.
This is important because the policy implications are vastly different according to how well you understand the data. The UK is an interesting case in point where tuition fees (£9k per year) have been levied in England for almost a decade; whereas this has been abolished in Scotland. Despite this, Scotland has the worst university enrolment among its poorer students within the union. [1]
Edit: Ok I found the paper with US data [2]. The Evidence on Credit Constraints in Post Secondary Education by Pedro Carneiro and James Heckman (FYI, a nobel laureate).
Note the graphs on pages 17 and 20 which illustrate the point most clearly. When you compare them, you can see that once you control for a range of important variables - peers within the same AFQT test scores have very similar enrolment rates across all income quartiles.
Another perspective on it is (one I think Americans are specifically averse to) is that these are all simply correlated to "class," for some definition for class. This is really another way of saying people follow in their parents footsteps.
If you are born into a reference group (parents, uncles, etc.) where most people have children in their early 20s, you are more likely to as well^. It's expected and seems normal. If you are born into reference group where educational achievement, college attendance, profession etc are within some band then you are more likely to find yourself in that band too.
If your reference group is a family and friends circle where college is a rarity, then attending college becomes overachieving. Most people are no overachievers. For others anything less than a top college, higher degree, top-of-your class is underachievement. People do what they feel is expected of them.
This is one of those things that people take for granted in individual cases (like their kids) but not at group levels. IE if you and your wife are highly educated, you expect that your kid will go to college. You don't think his chances are represented by national averages adjusted for income, race and intelligence.
^The parenthood point is a bit of a strange one because cultural shifts have been at work over the last few generations.
A lot of this relates to where people grow up. If nobody in your high school expects to go to college then that's a huge influence. Even something as simple as signing up for the SAT's is simply off the radar. Not to mention things like how much education your parent’s friends and coworkers have.
Income strongly correlates with zip code so it's hard to figure out the 'root' cause such things.
> when you control for educational achievement (i.e. how well students achieve in pre-uni schooling), income level is a much weaker predictor of whether someone attends university. The primary driver of low university enrolment is educational achievement.
The Times reported the exact opposite finding a year ago:
> whether a student graduates or not seems to depend today almost entirely on just one factor — how much money his or her parents make. ... About a quarter of college freshmen born into the bottom half of the income distribution will manage to collect a bachelor’s degree by age 24, while almost 90 percent of freshmen born into families in the top income quartile will go on to finish their degree.
> ... ability turns out to be a relatively minor factor behind this divide. If you compare college students with the same standardized-test scores who come from different family backgrounds, you find that their educational outcomes reflect their parents’ income, not their test scores. Take students like Vanessa, who do moderately well on standardized tests — scoring between 1,000 and 1,200 out of 1,600 on the SAT. If those students come from families in the top-income quartile, they have a 2 in 3 chance of graduating with a four-year degree. If they come from families in the bottom quartile, they have just a 1 in 6 chance of making it to graduation.
> The key to raising university enrolment among poor students is to provide better education/development opportunities and aspiration.
I'm going to risk calling you out on this, because I think what goes unstated here is that public resources cannot compete with private ones when it comes to education. It's sad to look at Johnny Trust Fund, and see that he is born with a true educational leg up over poor children, and there's really no way around it. If Johnny's parents have the will and motivation to spend hundreds of thousands of dollars a year educating Johnny, they can. I'm sorry, but no country can compete with that. Even increasing per-pupil spending in public schools to that of the Phillip's Academies is arguably impossible in all but a handful of the top GDP/capita nations, let alone all the private tutoring and traveling and internships that Johnny has access to.
So, yeah, if you manage to educate poorer children as well as wealthy children, they would achieve similar university performance, but getting their educational background to parity with one-another is so far out of the realm of possibility for most nations that I don't think it should even be mentioned.
It's not clear that money per student is the metric that matters.
Parental involvement, for example, is crucial for motivating students, imparting various skills (ie time management, studying) and setting appropriate expectations[1]. This is more a matter of cultural and structural factors. Parents that have the time and motivation to care for their children's education could have a significant effect without worrying about per-student outlays.
Perhaps there's a way to replicate this by restructuring school to provide positive parent-like influences? I don't know, but it's not a matter of throwing money at the problem, one way or another.
I don't think I claim that's the only metric that matters. If I did, that was a mistake. Money is a big influencing factor in-terms of saying what kinds of resources a child has access to for their education, because it's simple to turn money into tutoring or other life experiences. But obviously, for example, having parents who themselves are well-educated and understand how to educate children is a resource that can be more valuable than just having parents with money. There are, of course, many many other examples of non-monetary resources that are accessible by some children and not other, though.
They solution I can see to this would be moving away from the conveyer-belt institution-and-teacher centric model of education which makes it impossible to distribute the same educational opportunities to all, solely because the money will attract the resources and teaching talen which will then be trapped within one institution.
If instead we had a model that allowed students to participate in an institituion independant network that created classess and assigned teachers based on smaller resource loads, week long classess as opposed to 2 year long courses, we could have a system that much more easily lends itself to individual needs and gives more freedom to all students regardless of income level. This same problem could also be tackled by bridging the teacher-student polarity. Students are very capable of teaching themselves in small groups and acheiving goals at the same rate as if they were with teachers. By removing the teacher from the model you reduce resource costs hugely.
Here students are giving by the system what they put in. By participating and working with multiple small groups of students we build social skills and open the door for more specialized and specific learning at every level, not just the private school. If the digital revolution needs to shake things up anywhere in education its in the model and class structure.
tl;dr:A system that gets rid of large classess and focuses on connecting students to teach themselves can provide the same(and highly effective) educational opportunities to all (at a very reasonable cost).
We don't have to do as well at educating low income children as some theoretical money-is-no-object family... you;ve got a strawman here:
"Even increasing per-pupil spending in public schools to that of the Phillip's Academies is arguably impossible in all but a handful of the top GDP/capita nations, let alone all the private tutoring and traveling and internships that Johnny has access to."
It is reasonable to expect that we can educate kids at a standard where income doesn't affect outcomes for most folks.
That is an interesting point. I wonder what it would be like if any student could potentially attend any school based on their academic merit simply by making all private and public institutions free to attend if accepted. Of course, this is not realistic because private institutions would probably not exist and the wealthy have other upper-legs that are not related to tuition, such as professional-tutoring, geographic location, and so on. I'm still curious how this would change the data though.
This is why I'm much more in favor of vouchers. From the papers I've read, its a less costly option, encourages/rewards parental involvement, and retains the freedom to choose (and the competition/movement that comes with it).
With vouchers applicable to private schools, you don't have to burden yourself with the entire cost or keep him in a school that he/she is not compatible with. At the very least, vouchers amongst public schools seems like a no brainer but I've met a lot of resistance from some friends that are teachers.
Vouchers don't work as well as you'd expect. The reason is that parents are not rational agents on behalf of their children. Especially parents of lower socioeconomic status tend to value convenience (how close the school is) to the quality of the school.
The effect of vouchers in the real world is that the best schools get a little bit better while the worst ones get a lot worse.
So is there no genetic component to this at all? I see parental income and educational achievement mentioned, but not parental genes.
Kids who got got good genes from their parents are more likely to go to college. And the parents who gave the kids good genes are likely to have high income.
If that were the case you would expect to find that adoption did very weird things to educational attainment in families. And i can't find any evidence to assert that that's the case.
"Genetics is the most important ingredient here." is a pretty extraordinary claim, and extraordinary claims require extraordinary evidence.
See: http://www.nyu.edu/econ/user/bisina/Chapter5_Sacerdote.pdf Excerpt: "Surely it would be difficult to deny that genetic effects matter. Just look at how much more biological siblings resemble each other on education and income than do adoptive siblings."
So maybe genetics is not the most important ingredient, but saying genetics is not relevant to this discussion is astounding. Smarter parents beget smarter kids -- that's not a controversial statement. IQ is heritable to a large extend. Smarter parents tend to be wealthier. (This is not true everywhere, but it is mostly true in the USA, the land of opportunity.) Smarter kids do better in college. That's not a controversial statement either.
I don't know about _most_ important, but genetics is definitely a large factor. Quoting from Heritability of Lifetime Income[1]
> we find that 24% (54%) of the variance of women’s (men’s) lifetime income is due to genetic factors and that the contribution of the shared environment is negligible
The Promises and Pitfalls of Genoeconomics[2] sees similar results, coming up with 46% and 58% heritability in women and men, respectively.
There's plenty of research here. "Heritability of income" is a good search term.
""Humans are descended from monkeys" / "The Earth goes round the Sun" / … is a pretty extraordinary claim, and extraordinary claims require extraordinary evidence."
Dismissing as "extraordinary" any claim to which you are politically opposed is not good science. (Frankly, the Sagan quote is in itself not good science until you can get everyone to agree on what is "extraordinary".)
Meanwhile, an entirely appropriate riposte to GP would have been:
""Genetics is the most important ingredient here." is a claim, and claims require evidence."
Very well-written. I'm a UK citizen, and although I'm fairly left-wing, the tuition fees really don't bother me. So long as we ensure there's sufficient credit available on good terms, they just aren't a problem.
Your Telegraph article makes the argument that the fees paid by all students allow the universities in England to offer more money in bursaries to the truly disadvantaged. Which contradicts your argument that money isn't important.
A lot of great comments here on the content and presentation.
I'm really interested, also, in the technical aspects of how this was put together.
For instance, the app was able to recognize the shape of the line I drew. On my first attempt, it correctly identified that my line was an S shape; then when I started fiddling around, it was able to identify that my line was (roughly) straight. I wonder what method they used to identify those shapes.
Yeah, I was really interested too. From what I can get from playing around in the console, they (probably jashkenas and his cronies) they're using d3 to do the rendering. Since d3 has convenient map / filter / reduce chains, I assume (though have no proof), they can figure out the shape of your line by doing ye olde' rise / run calculation for each point. If your slopes gets significantly bigger at some point, you must have a S shape. They also force you draw only 1 point per x-increment, so they don't have to handle bad users drawing vertical lines.
Glad to see so much discussion of this here! Always cool to see our work on HN.
To answer this question, yes, we (Jeremy's cronies) are indeed using D3 to do most of the drawing and interactions. He also forced us to use Underscore.
The heat map showing everyone's responses is done in Canvas.
I just a took really quick look at the javascript they used and it looks like they matched the lines to a limited set of line shapes based on various heights. So, if the low income part of the line is between these certain values, the medium low income between these, etc, then they show this result.
Here's a list of the possible results:
[{
key: "kindaSShaped",
sentence: "Your line looked a little bit like an S, with a gradual increase for the poorest families and a leveling off for the rich. That’s what we guessed the first time, too. But we were wrong. It’s a straight line."
}, {
key: "underEstimatePoor",
sentence: "You underestimated the chances of college enrollment for the very poorest children. In reality, about one in four children in America’s poorest families go to college. (You guessed around " + e.round(t[0]) + " percent.)"
}, {
key: "overEstimatePoor",
sentence: "You overestimated the chances of college enrollment for the very poorest children. In reality, about one in four children in America’s poorest families go to college. (You guessed around " + e.round(t[0]) + " percent.)"
}, {
key: "justRightOnPoor",
sentence: "You correctly guessed that children from the very poorest families face tough odds in going to college – only about one in four do."
}, {
key: "underEstimateRich",
sentence: "You underestimated the chances of college enrollment for the very richest children. In reality, about 94 percent of children from America’s richest families go to college. (You guessed around " + e.round(t[20]) + " percent.)"
}, {
key: "overEstimateRich",
sentence: "You overestimated the chances of college enrollment for the very richest children. In reality, about 94 percent of children from America’s richest families go to college. (You guessed around " + e.round(t[20]) + " percent.)"
}, {
key: "justRightOnRich",
sentence: "You correctly guessed the exceptionally high rates of college enrollment for children from the very richest families – about 94 percent."
}, {
key: "veryStraight",
sentence: "Your line was very straight, which means you correctly guessed one of the more striking findings of this research: The relationship between college enrollment and parental-income rank is linear."
}, {
key: "kindOfStraight",
sentence: "Your line was relatively straight, reflecting one of the more striking findings of this research: The relationship between college enrollment and parental-income rank is linear."
}, {
key: "lineButNotSteepAddition",
sentence: "Your line was not steep enough: There’s more inequality than you guessed."
}, {
key: "lineButTooSteepAddition",
sentence: "Your line was too steep: There’s more equality than you guessed."
}, {
key: "perfectAddition",
sentence: "Your guess was extremely accurate. Is that you, <a target='_blank' href = 'http://www.rajchetty.com/'>Raj Chetty</a>?"
}, {
key: "youAreWeird",
sentence: "We’re going to be honest here: Your line was pretty odd. Did you even try?"
}, {
key: "youAreCanada",
sentence: "Your line, which was relatively horizontal, was admirable: It suggests that everyone has the same chance of going to college regardless of parental income. <a target = '_blank' href = 'http://www.nber.org/papers/w17218'>But not even Canada</a> is that egalitarian."
}, {
key: "trustFundDip",
sentence: "You guessed that there would be a small decrease in college attendance for the richest families – something we’ve called the trust-fund dip. It makes some intuitive sense – if you’re born into an extremely wealthy family, it’s reasonable to think that there’s less incentive to attend college. But the trust-fund dip does not exist."
}, {
key: "youIgnoredOurNote",
sentence: "It looks as if you didn’t use the free point we gave you, which may be one of the reasons you performed worse than other people."
}, {
key: "downwardSlope",
sentence: "Your line sloped downward. Did you believe that rich children were less likely to go to college than poor children? Or did you misunderstand the axis?"
So do we keep bikeshedding and tweaking little levers or do we actually try and fix the fundamental problem?
I can't be the only one who is tired of seeing all of these snippets about small parts of a large system.
We need to deal with wealth inequality. We live in a world... hell, we live in countries, in cities with 10 orders of magnitude wealth difference.
A western person earns 10e6 - 10e7 over a lifetime, revenue.
A western person stashes away 10e0 - 10e7 over a lifetime, balance sheet.
Why? It is predominantly not about decision making and choices to save. Yes, people do make suboptimal decisions. But mostly, people are simply unable to save because the baseline of living sucks up all of their work.
In the UK a poor person rents instead of owning because we allow the group of capitalists to own all housing and restrict new building.
Until we deal with that we are just pissing around.
Savings rates: US 17% India 32% China 51%. Apparently the Chinese manage to save without having that "baseline of living" sucking up all their work. Yet Chinese people earn far less than Americans .
How is that possible? It's almost as if your "baseline of living" is actually "living high on the hog as if tomorrow doesn't exist".
The average person in the UK cannot save 51% of net income. It is totally impossible. Difficult even for someone earning above the median wage.
Is paying the rent 'living high on the hog'? I don't get it. Rent alone on a studio flat in loads of places is 10K. If you move away you earn less than you save on the rent.
What's the suggestion? Live 2 to a room? Live in a car?
Theoretically possible if you're determined I suppose.
The 5'th decile in the UK earns a little under double what the 1st decile does. Since the people in the 1st decile aren't dying, that suggests that yes - someone in the 5'th decile could live like a person in the 1'st decile and save 50%.
Pretty much every story you could tell to explain this away (e.g., folks in the City earn 2x as much but cost of living is 2x higher) kind of undermine any real inequality story.
If that's the case, then you need to adjust your income/wealth/etc inequality numbers by that same factor. I.e., the inequality for a person living in London and earning 2x as much as someone in Preston might need to be adjusted down to perfect equality (assuming the cost of living in London is also 2x Preston). So why didn't you adjust your numbers in this way?
Also, if inequality is merely a matter of where one lives, why do you care?
As I said, this sort of observation seriously undermines basically every complaint about inequality.
Well, my argument is that most cannot save. Relevant amounts, anyway (a grand a year isn't going to really get you anywhere).
Your rebuttal was that the 5th decile earning twice as much surely can, because the 1st decile survive by spending less.
My argument is that the 5th decile are predominantly in higher cost of living areas and end up in roughly the same situation.
But let's just concede and say that they can. No doubt some can - there are people earning median UK wage in the north.
The bottom guys still can't save at all. Do they just not matter? Should 10/20/30/40% of the population be stuck renting for their entire lives? Is that reasonable?
I am not making the claim that a London worker earning double a Preston worker is a problem.
My claim is that people earning far more than that being able to build masses of capital whilst others cannot build capital at all is a problem.
It's not really about the numbers at all; it's more the idea that me "owning" 20 homes and forcing others to labour if they want to live in them is OK.
I dunno. I kind of don't have enough interest in continuing this, I gain nothing from convincing you. I know people who cannot save, many of them, they exist. They aren't intelligent enough to get out of that hole unless they won a lottery or similar. I don't need to convince you that they exist, because I can like, go and visit them. I want to fix that, you seemingly don't, best of luck.
Depends on how much disposable money you have not the absolute value plus cultural issues are in play eg putting savings into gold as you don't trust the banks
Reading my post again I can see why you would think that.
I just feel frustrated, that's all. It feels like the premise is just so obviously silly.... like these articles are just fluff. Am I insane?
Of course people with wealth do better at everything. It's definitional. That is what wealth is... better schools, better food, better holidays... bigger better more. I don't see the value in continually studying how bad the poor have it and how well the rich have it.
It feels to me like a textbook example of a fruitless endeavour. Like trying to make a human powered bicycle that's faster than a fighter jet, or something. It's just... daft! Even if you think of a way to do it it's bloody silly. Stick both people in the same vehicle.
That is what wealth is... better schools, better food, better holidays...
No, that's consumption. Wealth is ownership of stocks, bonds, REITs, and other investment vehicles. It provides no value whatsoever to the holder until it is consumed.
If consumption is what you are worried about, you should focus on consumption inequality, which is vastly lower than either wealth or income inequality and has not risen anywhere nearly as sharply.
You can't consume if you don't have any wealth to spend. Conversely, the wealthier one is the lower one's marginal propensity to consume. Very wealthy peoplewho literally have more money than they know what to do with end up investing the rest, but thanks to financialization (among other things) that investment doesn't generate as much demand for labor as it used to, making it harder for people without surplus capital to accumulate wealth.
I know you love to characterize poor people as living high on the hog but it's a bullshit characterization. The least educated are the most vulnerable to the constant imperative to consume with which people are confronted in western society.
Conversely, the wealthier one is the lower one's marginal propensity to consume.
I'm glad you agree that stegosaurus is wrong, and that increased wealth inequality isn't a strong driver of consumption inequality or educational inequality.
I know you love to characterize poor people as living high on the hog...
Since you are using the phrase "high on the hog", I think you actually meant to respond to my other post. But that post characterizes rich people (namely Americans) this way, not poor people (Chinese, Indians).
I'm confused about the 'high on the hog thing' from your other post (I just didn't feel like replying to both), because stegosaurus was talking about inability to save during the baseline cost of living, which you went on to suggest was treated very differently in the US.
I think you're equating the US being a rich society compared to others with all the people in the US being rich in absolute terms, disregarding their relative poverty within the US. But you know very well that absolute measures of wealth are pretty meaningless given the fact of geographic dispersion. If you're talking to someone who doesn't know how to get ahead in the US and feels pinched by a lack of disposable income, observing that they're far better off than a peasant in some other country isn't responsive to their problem, notwithstanding the truth of the matter. It's not like this person struggling with poverty in the US can just relocate to the peasant area and start leveraging his capital advantage.
I think that what yummyfajita is trying to get at is the idea that as societies the US, UK etc spend up to their limits rather than saving.
The problem is that this doesn't really generalise into individuals being able to opt out of the spending.
If everyone refused to pay 50% of their income on rent then rents would fall (unlike, say, food production, the homes won't just cease to exist).
But an individual can't affect that change. If they want to work in a city, they rent in that city.
There are a few oddball solutions that can work in unique cases. Living with parents works if they are willing and your hometown has a decent employment market. Living in a car might work in some places. House sharing might work if you can find reasonable housemates and you are well paid (if you have no hope of ever saving much, then it's pointless to inflict a crap life upon yourself for a future that will never come).
The idea that people spend loads on crap just doesn't add up to me though. When rent, energy, taxes and minimum food (e.g. 20pw rice and beans diet) add up to most of your income, does it really matter what you do with the small amount left over?
edit: Oh; and the capital advantage on moving doesn't exist. Most people have zero net worth.
I hate being the typical negative commenter who points out that correlation doesn't equal causation, but in the case of this article I think it's warranted.
Just because high-income families send kids to college more does not mean that high income causes it. A recent study of Swedish lottery winners showed that the kids of lottery winners did not do better on measures of health or education, suggesting that wealth does not have a direct or short-term impact on child outcomes. [http://www.ifn.se/eng/publications/wp/2015/1060]
I was surprised that nearly 95% of the children of the highest-income earners went to college, especially with everyone talking nowadays about all the good career options that don't require any college education at all. (Trade school, entrepreneurship, etc.) Then I thought about (what I believe to be) the typical person at very top of the income scale. Someone who has not only a lot of wealth in terms of money, but also reputation, connections, favors owed, etc. For a kid with parents like that, the easiest start for many of them may be to get a degree from a prestigious school and then have their parents call in some favors or sign on with the family business.
Lottery winners are in a totally different life situation from the rest of the wealthy, so I don't think lottery winner results are generally applicable like that.
That may be true in Sweden, but perhaps there are other reasons that health and education aren't strongly correlated to wealth... for Sweden.
Sweden is famous for its public services (funded through high taxes), so their citizens receive free/low-cost health and education anyway. Thus, wealth isn't as big a factor for receiving those services as it is in other countries.
Pointing out correlation doensn't equal causation is fine, but drawing conclusion from another country and translating back to the U.S. while ignoring significant differences isn't correct either.
The causation is probably the reverse. Parents who send their kids to college make more money. And that's possibly indirect through the fact that people who go to college earn more money, so we're probably seeing the effect of having well-educated parents.
It would be really interesting to see if income plays a part once you control for parental education levels. Segment by both parents have college degrees, one parent has a degree and neither parent has a degree. We'd probably see a much weaker correlation with income at that point.
That's a fair question. When many people make the correlation!=causation argument, they do so to express skepticism, often without any evidence or positive contribution to the discussion. However, in this case, I think the argument is warranted because there is a large body of evidence that suggests wealth does not directly affect whether kids go to college. I linked to one interesting study, but I can link to more if you'd like.
In any case, I hoped my comment and link added value to the discussion. If not, I'd be happy to remove them.
I do not understand why poor people have kids. There should be a guideline for married couples, which tells them that if family income is less then 90K in NYC, you can not afford to have kids, if it is more than 120K then they can have one kid etc.
"Having kids" not for everyone. Kids degrade your lifestyle, drain you financially and limit your chances if you are poor. It's a SCAM.
Yeah, like having a vacation home in Florida is not for me. If I want i can have it, i can apply for mortgage and buy one tomorrow, but i can not afford it. I should not do it.
Why does it matter? (for reference I went to a decent university having had a poor upbringing).
Are we substantially better off in a world with 1000 arbitrary people going to e.g. Cambridge vs one in which 1000 upper class children go? (expand the concept to university in general if you wish).
You still have unbalanced outcomes based on what eventually comes down to a lottery of some sort. Even if you choose the absolute best 1000, are the next 10000, 20000, 30000 actually materially worse?
Going beyond that, surely the concept of prestige in and of itself is problematic? Should you get a better life because you have a higher IQ? Why? Do we actually want that? Would we design the world that way if we actively chose?
I certainly wouldn't... why should my friend have a shit life due to quirk of birth? Why should my other friend have a dramatically better life due to the same?
We seem to strive to normalise things that are/were seen as handicaps, like race, disability, sex, etc. etc. But we are seemingly unwilling to accept that the idea of ranking humans is broken. It only makes sense as a hack to encourage productive behaviours - it might not even be needed any more soon.
I suspect it matters because we assume those with attributes such as greater IQ would utilise the education better then a random group of lottery winners. We then assume that that would result in greater communal good.
We also live under the illusion of attempts at fairness rather than perpetuating privilege.
I think it's great exercise, but why don't they just use an absolute scale for parent's income? "About $2,400 in annual income separates the bottom two dots, while nearly $1 million separates the top two", I would never have guessed this either.
I think this way scales better - it's easier to visualize than the massive gap would be. And where we are in the news cycle is to focus on problems common to all classes, not the problems that divide classes.
Perhaps the visualization is just better this way, but your question is at least an interesting one to think about. How would the shape of the curve change, and might it suggest slightly different conclusions?
For example, it seems to me the absolute income version would be steeper at low incomes, perhaps amplifying the perceived impact of things like a living wage or basic income on educational outcomes. (I'm talking mainly about the psychological impact of the curve's shape, not raw statistical truths.)
I completely agree. Percentiles are a very unintuitive way of thinking for this sort of data. It's really easy to know how much one made last year, but it takes a while to figure out what percentile one's income was in. Differences in Percentile vs differences in Income is even harder.
I felt strongly about this the moment I read the article on NYTimes - I reproduced the data set to the best extent I could and actually went about plotting against absolute income:
That's a far more intuitive chart, that a lot of us will understand better. I also graphed how much these children actually go on to earn against absolute income. I compared these results in a blog post - http://unside.t4you.in/data/intuitive-axes/
Agreed. I suppose it makes the aggregate result a bit more tidy, which makes it easier to show that most people had relatively the same idea about the relationship, but for me at least having that dot to strive for in the middle skewed my projection from my initial imagining.
That would be very interesting. I would expect the higher income side to remain practically unchanged while the lower income side would move up to 50% or so, but still be far below the higher income side.
I think a more interesting line to see would be graduation rates...At least in the US, the average graduation rates from college are surprisingly (at least to me!) low at around 55%.
If your family makes just above $150K per year in the US (not uncommon, especially in high cost of living areas like NYC, SF, DC, etc), you fall in the category where you have no "need" and according to some outdated government formula can afford the whole tuition, usually 20-60K per year.
The reality, however, is that 150K is the minimum living wage for a family with two kids in a place like SF, and usually not enough.
The sad part is that someone who makes $500K a year or has millions in the bank falls in that same category.
The people who make $500k / year feel the same way you do: http://www.nytimes.com/2009/02/08/fashion/08halfmill.html. Virtually no one considers themselves "rich" because they tend to put their "minimum acceptable" level at whatever lifestyle they're currently living, so obviously they can just barely afford it. Lack of perspective (historical, global, or even just from outside of your own social circles) is a real problem.
We can't all live exactly where we want to live. There are plenty of places in the SF bay area where 150k is fine. There are plenty of places in the SF bay area where even $2mil/year isn't fine.
If you choose to sacrifice your childrens' college education so you can live in SF, that's your prerogative. Most families don't even have this choice. Complaining that you're minimum living wage at these amounts is hugely insulting to the vast majority of the population.
Living wage is different than the poverty line. I wouldn't be surprised if $150k was really tight for a family of 4 in SF proper.
Running the calculations, a payroll calculator indicates that $150k/yr turns into $3925 take home every 2 weeks or just over $100k/yr. The housing alone for a three bedroom (assuming everyone shares a room) would run at least $60k/yr. That leaves $40k/yr to support 6 people. That may be possible, but it's not easy and it's damn near impossible to be saving for college on that amount of money.
(2) $40k/year is enough money to support four people (or even six), if you've already paid taxes and rent. That's more than $500-800 per month for food/clothes/transportation/entertainment, etc.
Of course, many people would like families to get more than $150,000 of income, but I hope we all agree that it's common to get by on less than $150,000 of family income, as most families do.
If by "get by" you mean survive, then, certainly. But are they living a good comfortable life and are able to afford good food, a second car, good health insurance, good dental care, music lessons for the children, a decent family vacation and a well maintained house? Doubt it. Every hard working educated parent deserves this and being able to put their kids though college in a civilized society that the USA claims to be.
This is the case for a lot of things. Pretty much all "middle-class" tax credits phase out by the time you hit $80k, regardless of cost of living, because the IRS doesn't scale anything by zip code and so assumes that a household making 6 figures must be loaded.
The only government subsidies that are worth anything are Pell grants and they phase out well below 150k. Also, they are limited to like 6k a year, way less than your 150k family can kick in.
The elite private schools have a similar formula that sort of screws over these people. But you have no right to an elite education. Just go to Berkeley instead of Stanford or UVA instead of Duke.
Having had a kid accepted to UVA, UChicago and UC Berkely this year and being a resident of VA, I can tell you that board and tuition at UVA are still over 30K even for in-state. UChicago (a private shool) was the only one that gave a (quite generous, actually) merit grant.
But congrats on your kid getting a merit scholarship at U of C. That's great and I'm sure you are very proud. I went of school and U of Illinois and Northwestern so I'm supposed to hate U of C, but it's a great school.
This is nonsense. If you correct for Socio-economic status, you'll find the best predictor is IQ. You can give a kid an IQ test at age 10 and know whether or not he'll graduate college.
Since IQ is heritable, kids from more successful homes will be more successful.
You may draw the conclusion from the chart that income prevents the poorest from attending university because they can't afford the fees. This is highly misleading. There's substantial evidence [at least in rich countries, see US example below] that when you control for educational achievement (i.e. how well students achieve in pre-uni schooling), income level is a much weaker predictor of whether someone attends university. The primary driver of low university enrolment is educational achievement. i.e. the main chain of causation is this: low income -> low education achievement -> low college attendance. The key to raising university enrolment among poor students is to provide better education/development opportunities and aspiration.
This is important because the policy implications are vastly different according to how well you understand the data. The UK is an interesting case in point where tuition fees (£9k per year) have been levied in England for almost a decade; whereas this has been abolished in Scotland. Despite this, Scotland has the worst university enrolment among its poorer students within the union. [1]
Edit: Ok I found the paper with US data [2]. The Evidence on Credit Constraints in Post Secondary Education by Pedro Carneiro and James Heckman (FYI, a nobel laureate).
Note the graphs on pages 17 and 20 which illustrate the point most clearly. When you compare them, you can see that once you control for a range of important variables - peers within the same AFQT test scores have very similar enrolment rates across all income quartiles.
[1] http://www.telegraph.co.uk/news/uknews/scotland/10354393/Fre...
[2] http://www.ucl.ac.uk/~uctppca/credit.pdf