a thoughtful web.
Good ideas and conversation. No ads, no tracking.   Login or Take a Tour!
comment by b_b
b_b  ·  3355 days ago  ·  link  ·    ·  parent  ·  post: Quiz solution: Income Mobility

Not sure how I missed the first post. Must have been a busy day that day.

All of these analyses are difficult, because there are so many variables to control for in any one comparison (e.g. race, parents' education, mother's income). As I pointed out in the original thread, the controls in the news article were bad enough that the data aren't incredibly useful.

We need to remember also that the very idea of a decile itself isn't very useful, either, if we aren't comparing cross-era data sets. That is, the third decile might be a much lower wage in 2009 in real dollars than the third decile in 1979 (about the time when real wages for low skill jobs started contracting).

Any way you look at it, however, I'm encouraged by the numbers. It's not what I expected.

Did you happen to see this? I think that reforming the way taxes are collected might further encourage even more mobility.





wasoxygen  ·  3354 days ago  ·  link  ·  

    As I pointed out in the original thread, the controls in the news article were bad enough that the data aren't incredibly useful.

Your disappointment in the article's rigor was not very evident at first. In the this whole "land of opportunity" thing is bunk dialog you seemed to endorse the premise of "white privilege," giving yourself as an example of someone who "did technically graduate high school, but barely" and yet ended up affluent.

The big idea of the article is that good life choices, especially relating to education, do not matter. What matters is whether you are born to a rich or poor family.

I did not question the data provided. In my view the data provide evidence that education is the most important factor in having a better future, and birth status is not a good predictor of future outcomes, completely contrary to the article.

After I criticized the article's conclusions, you criticized the data. Your objection seems to be that only two of four subgroups are considered:

  poor kids who graduate
  rich kids who drop out of school
and two subgroups are ignored:

  poor kids who drop out of school
  rich kids who graduate
If we want to know which factor is more important, birth status or education, we won't learn anything by looking at the latter two groups. I think we can speculate that poor dropouts have bad outcomes, and rich graduates have good outcomes, without stretching credulity nor contradicting my conclusion about education being the best single predictor of outcomes, based on the data given. The two subgroups considered are the ones we must examine to determine if birth status or education make more of a difference.

It is particularly odd that you cite your personal story, since you (born to a "middle class white family" and having completed an advanced degree that you neglected to mention) belong to one of the subgroups that leaves the cause and effect relationship unclear. Why do you believe it was "white privilege" and not your doctorate that enabled you to become an art collector?

(Note: it's harder for poor kids to get a good education, obviously. For best results, start rich and stay in school. But if you want to end up in the top 20%, the data show that it's better to start poor and stay in school than to start rich and drop out. This is what meritocracy looks like.)

b_b  ·  3354 days ago  ·  link  ·  

    Why do you believe it was "white privilege" and not your doctorate that enabled you to become an art collector?

I think my current circumstance is a combination of hard work, good choices, and luck. DIsentangling them is not so easy. All are contributing factors.

As to your former points, I do believe that whiteness and my middle class upbringing have helped me, and I think that the data are not very useful. These aren't contradictory points. Were I not middle class, for example, I may not have had a support network while I took a couple extra years to get on the right path for me.

I'll certainly admit that I shared the article without fully digesting it, because I thought it was interesting, and I thought it would be a conversation starter (obviously it was, since we're still talking about it). After your critique, I went back and took a closer look, and it seemed obvious to me that they were comparing data sets in which two variables had been manipulated--an automatic DQ in that type of analysis. Perhaps the source material did a better job (I assume so, because no journal would publish such quality data), but I didn't go back and read it.

    I think we can speculate that poor dropouts have bad outcomes, and rich graduates have good outcomes, without stretching credulity...

Sure, but to what extent? That's the point. There are certainly many contributing factors, and without knowing them in quantitative terms speculation is all we're left with.

In the end, we're arguing about the small space between, because we're in complete agreement that education is the most important thing. No matter how shitty I was as a kid, I'd be nowhere without an advanced degree.

wasoxygen  ·  3354 days ago  ·  link  ·  

    I think my current circumstance is a combination of hard work, good choices, and luck.

You did not mention good looks.

We have a question: Is education associated with improved life outcomes? Somebody gathers data and finds that people with more education tend to have better outcomes than people with less education. That is evidence that helps us answer the question.

It does not tell the whole story. Perhaps being tall helps too. (Seriously.) We assume that there are both tall and short people in both the dropout and graduate groups, so any height effect will cancel out. This might be a bad assumption, and we would have to ask another question and gather more evidence to find out.

We will never determine all the ingredients that contribute to each success; that would require a biography of every individual. But that doesn't mean we can't make justified generalizable conclusions with some level of confidence.