Opportunity costs: How much does it matter where Jay and Wally go to high school?

Two weeks ago I speculated about what an extra $50,000 in annual income would mean for how Jay and Wally grow up.  I looked at national data and concluded that the lion’s share of that money would end up going into their educations—either as private school tuition or (as several readers pointed out) as a mortgage payment in a high-rent suburb with well-regarded public schools.

Since then I’ve been chewing on the question: To what extent will the K-12 schools Jay and Wally attend determine how much opportunity and success they have in life?  And, related to that question, to what extent will the amount of money Caroline and I make affect our children’s chances in life?

These are, I think, critical questions to come up with some kind of precise or at least well-reasoned answer to.  A desire to give your kids the best possible future drives many of the biggest decisions parents make in life—from career decisions, to where to live, to where to send one’s kids to school.  Given that, it’s worth trying to get a handle on how the success equation actually works- to make sure we’re actually putting the right amount of resources and energy into providing the right kinds of inputs.

Family Income and Children’s Outcomes

The relationship between family income and children’s outcomes is notoriously difficult to tease out.  For sure, average parental income is higher among kids who attend Ivy League colleges than it is among kids who drop-out of high school.  But just because high-parental income correlates with Ivy League matriculation doesn’t mean it causes it.  There are all sorts of other characteristics that tend to be packaged together with high parental income, including high parental educational attainment, certain genetic endowments, and a family culture that promotes educational success.  Because these characteristics are found together so often, it’s hard to disentangle them, and hard to determine which ones are really driving the train.

One of the most authoritative studies of the relationship between income and achievement is a book called What Money Can’t Buy: Family Income and Children’s Life Chances by Susan Mayer, professor of public policy at the University of Chicago.  The title gives away Mayer’s conclusion: Income doesn’t do much to boost kids’ prospects in life.

Mayer is concerned about welfare policy, so she focuses her analysis on low-income families and the extent to which cash transfers to parents can help kids climb out of poverty (she concludes they don’t do much).  But her analysis is applicable across the income spectrum.

As you might expect, if additional family income doesn’t make a big difference for poor kids, it makes even less of a difference for middle class kids who are already well-provided for.  Mayer argues that kids need access to books and exposure to educational experiences like trips to the museum, neither of which cost that much to provide, and that once a minimum standard of material provision is reached, each additional dollar of family income doesn’t do a lot to give a kid a leg up in life.  In her own words:

Some child-specific possessions and activities, such as the number of books a child has and how often a child visits a museum, do influence how well children score on cognitive assessments. But parents’ income is only weakly related to whether children have these amenities. This is probably because these items cost so little that their distribution depends more on parents’ tastes than their income.  Thus the amenities that are important to children’s outcomes are weakly related to parents’ income, whereas the amenities that are strongly related to parents’ income are not very important to children’s outcomes.

Mayer performs a variety of statistical analyses to draw this conclusion, all of which are accessible at a conceptual level and are very elegant to behold.  I’d encourage people to read the introduction to What Money Can’t Buy, which is available on Google Books.

School Quality and Children’s Outcomes

So what about school quality and children’s life outcomes?  Money certainly does buy access to more highly regarded K-12 schools, whether public or private, and it’s hard to see how that couldn’t help.  But Mayer’s analysis would suggest that such access is not hugely consequential for how kids turn out in life: Smart kids from families that value education are going to do well regardless of where they go to school just as less smart kids from less educationally-centered families are going to have a harder time of it wherever they are.

Intuitively, that doesn’t seem right.  How is it possible that you could take the same kid from the same family and place him in Philips Andover or the worst high school in Boston and he’d achieve the same in life either way?  But that’s kind of what academic research suggests—that who you are and what kind of family culture you come from matter a whole lot more to life success than where you live or what school you go to.

I read several papers over the last couple weeks on this topic.  They’re listed at the end of this post.  The most thought-provoking one among them was “Sibling, Peer, Neighbor, and Schoolmate  Correlations as Indicators of the Importance of Context for Adolescent Development.”

It’s a very accessible paper and I uploaded it to Growing Sideways so that anyone who wants to can read it.  The authors, public policy researchers at Northwestern, measured both achievement on a vocabulary test and delinquency rates (how often kids did graffiti, shoplifted, damaged property, etc.).  Then they looked for correlations between the scores kids received and different groups of people in their lives: their siblings, their neighbors, their grademates, their best friends.

Before looking at the results, I should note that the authors were looking at correlations, not causation.  My SAT score is highly correlated with my siblings’ SAT scores, for example, but that doesn’t mean that our relationships with each other caused us to get those scores.  As the authors of this paper note, what a correlation does is establish an upper-bound for causation: The actual degree of causation could be a lot lower than the degree of correlation, but it will never be higher.

Anyway, let’s get to the fun stuff.  The following chart shows the correlations between students and the people in their lives on a vocabulary test:

As the chart shows, there is a very strong correlation between identical twins—even stronger than the correlation between fraternal twins—which suggests that genetic ability plays a big role in verbal aptitude.  There’s also a very low correlation between grademates—which suggests that the kids you go to school with don’t have much of an effect on your facility with language.

One interesting piece of the chart is the correlation between best friends—at .44 it’s pretty high, nearly as high as the correlation between siblings.  The authors were worried, though, that the correlation between best friends reflected a selection effect—basically kids of similar aptitude gravitate towards each other.  For that reason the authors came up with a measure they called “Adjusted Best Friends” which tried to control for selection effects—and as you can see the correlation is much lower once those controls are put in place.

Here are the correlations between delinquency rates.  They’re considerably lower, across the board, than the correlations for verbal achievement:

In the Context of Family Life

So what does research like this mean for how a family plan’s it life?  Well, for one, the research I’ve cited is just a few papers and shouldn’t be taken as anything close to a definitive statement on what is an incredibly complicated set of interrelationships.  And for two, all families’ circumstances are different, and the types of choices that make sense for one family and one set of parents don’t necessarily make sense for another.

What I can do is write a little about how Caroline and I have talked about issues of income and school quality when thinking about the long-term decisions we’ll need to make as a family.

We both agree that we want Jay and Wally to be highly skilled and that we want them to achieve life success in the conventional American sense: good colleges, good jobs, the respect of their peers, influence, etc.

But we also think that Jay and Wally already have substantial advantages in their pursuit of those things, chief among them the fact that they come from a family with well-educated parents who possess above-average rates of cultural capital.

In concrete terms, this cultural capital means that Caroline and I have a sense of what it takes to be successful in school and beyond, and that we’re able to incorporate that knowledge into the day-to-day experience of raising our kids.  And I think it’s hard to overstate just how hugely important this is in determining how well kids do in life. (Pam Davis-Kean of the University of Michigan expresses this idea in her paper The Influence of Parent Education and Family Income on Child Achievement: “Parents’ ability to form accurate beliefs and expectations regarding their children’s performance are essential in structuring the home and educational environment so that they can excel in postschooling endeavors.”)

So, when I think about the relationship between our family’s income, the K-12 schools Jay and Wally will attend, and how well they’ll do on life, I think that they’re well-provided for, even if Caroline and I never experience a dramatic income bump or they don’t attend one of the best public or private high schools in the country.

Now, it feels a little cold to say that Jay and Wally will have “enough” opportunity in life and that as a result it doesn’t make sense to dramatically reorient our family life in terms of jobs, daily schedules, and where we live in order to provide them with even more opportunity.  And indeed, when you’re talking about providing for your kids’ futures, it can be hard to accept that there is ever such a thing as “enough.”

But the thing I remind myself is that every family decision is a matter of tradeoffs.  Earning the money to send Jay and Wally to top-shelf K-12 schools would require Caroline and I to make different career choices and to adopt a different daily lifestyle.  That  lifestyle would probably take away from other aspects of our family culture that we value- and my sense is that the marginal gains for Jay and Wally’s opportunities in life are too small to justify those decisions.

Readings that Contributed to this Post:

Can Family Socioeconomic Resources Account for Racial and Ethnic Test Score Gaps?” by Greg Duncan and Katherine Magnuson

The influence of parent education and family income on child achievement: the indirect role of parental expectations and the home environment” by Pamela Davis-Kean

Sibling, Peer, Neighbor, and Schoolmate  Correlations as Indicators of the Importance of Context for Adolescent Development by Greg Duncan, et al.

What Money Can’t Buy by Susan Mayer

Related Reading from Growing Sideways

What would an extra $50k a year mean for Jay and Wally’s childhoods?

School quality and housing costs

Advertisements

9 thoughts on “Opportunity costs: How much does it matter where Jay and Wally go to high school?

  1. Kevin, Thanks, this has been a great series of posts. My wife and I are expecting our first child soon and are preparing to move from our current, more urban neighborhood, into the suburbs where the schools are better.

    I appreciate the perspective here.

  2. I’m not sure I agree with your ultimate conclusion: “Caroline and I to make different career choices and to adopt a different daily lifestyle. That lifestyle would probably take away from other aspects of our family culture that we value- and my sense is that the marginal gains for Jay and Wally’s opportunities in life are too small to justify those decisions.”
    I’m going to make an assumption that in order to have that extra $50k, you’d both have to work more and be home less. Honestly, if you two are the ones with the cultural capital, wouldn’t being less involved in your kids’ lives mean they have fewer opportunities for your mentoring and support?
    Let me be clear: I don’t think working more is the worst thing ever–kids are resilient. But working more to provide some external benefit to your kids seems a little backwards after the evidence you presented here. If family is the ultimate influence of how successful our kids are, then providing more family experience might be the safest course of action.

    • We’re saying the same thing (though I guess I didn’t make my point very clearly)- I think that it’s *not* worth the extra time away from home to earn that $50k. As in, that money wouldn’t provide my kids much in the way of improved opportunity but the time it takes to earn it would detract from other aspects of family life that I value (like spending a lot of time with them). Are we on the same page or is there a nuance I’m missing?

      • I think it was more a nuance. I guess I would equate “other aspects of your family life” as an “improved opportunity”. For me, spending less free time with family is itself eliminating an opportunity for them that will enable them to be more successful. Any “outside opportunities” (provided by a private school, say) are pretty paltry in comparison.
        And maybe I’m getting so nuanced because it’s time for me to go to bed. Head swimming 🙂

  3. I look forward to reading these papers. It’s an interesting topic. Racial and ethnic group gaps on tests are one thing, “success” in life is another, and getting the most benefit from your educational setting is yet another “thing”. I’m trying to picture a student with all the socio/economic benefits of an affluent and educated family being placed in a terrible inner city school with a host of behavioral and achievement and cultural issues (e.g. majority of students and families not valuing or able to support educational achievement, severe behavioral disruptions, etc) and imagining that there is little deleterious effect on that student.

    A neighbor recently conducted a study of the local public high school students and local private high school students’ acceptances into various colleges and universities. This neighbor found that the public high school students were just as likely to attend ivy league colleges as the private school high school students. The study was very self-congratulatory, but what it failed to emphasize was that the students from both groups that were studeied were from virtually identical socio/economic backgrounds (with affluent and highly educated families who valued higher education.) Another factor into college acceptances was that even though the neighborhood public school students were not disadvantaged themselves, the fact that their high school had accepted many out-of-boundary inner city and disadvantaged students meant that the local public students tended to have higher class rankings and were considered to be “inner city students” by universities eager to fulfill their diversity requirements. I’m not mentioning this to be argumentative, but just as a cautionary note-to-self to be wary of some conclusions drawn from studies!

  4. What an interesting topic, so glad you are exploring it further. I agree intuitively that family culture is the biggest factor in a child’s success, more important than additional income by itself, but a question about Mayer’s study if you know – did the parents of those kids use the extra money to move to neighborhoods with better schools? It’s hard to believe a low income student in a failing school in East Boston wouldn’t enjoy more success in life if they went to Andover. It’s not just about money – it’s about finding another source of inheritance of cultural capital. I’m thinking of the success of some public schools (mostly charters) in helping kids from low income communities achieve at levels to match their high income peers (example: middle schoolers at Excel Academy in Boston, 72% low income, actually outperformed their peers in Newton on state Math and ELA exams last year). While parents of charter students in low income communities likely value education, I wonder if there are any studies of how students at successful charters do versus the hundreds and thousands of students who try to go to those schools but lose out in the lottery.

    • Well hello, Harraseeker. To your points I’d say: The impact of upshifting to a higher quality school depends a lot on the kid and the family and a lot on the quality of the school the kid would otherwise attend:

      1. As I said to Chris Huntington below, I think that for children with well-educated parents who are invested in their cultivation, there’s a minimum standard of school quality that needs to be met after which the marginal gains of upshifting schools is low.

      2. If your neighborhood school really stinks, or if your parents lack the cultural capital to effectively scaffold you to a certain kind of success in life, then I think the advantages of upshifting schools is greater.

      3. I think one of the primary advantages of charter networks like KIPP is that, as you pointed out, they provide low-income kids with an additional source of cultural capital. This includes guidance through the college preparation and application process, exposure to a wider range of experiences, and connections to the kinds of high-SES members of American society who tend to be KIPP benefactors.

      4. There are two studies of charter effectiveness that I’d recommend looking at. The first is a study by Mathematica. It is, to my knowledge, the most comprehensive and rigorous examination of the impact of charters that has been undertaken to date. The study used an ‘experimental design’ meaning that it compared the academic performance of kids who ‘won’ charter lotteries with the academic performance of kids who ‘lost’ charter lotteries. The Mathematica study found that overall, “charter middle schools that hold lotteries are neither more nor less successful than traditional public schools in improving student achievement in reading and math.” When the researchers looked at subgroups of charters and subgroups of students, however, they found some variation. They found that, “Study charter schools were more effective for lower income and lower achieving students and less effective for higher income and higher achieving students. In addition, charter schools in large urban areas had positive impacts on students’ achievement in math; those outside these large urban areas had negative impacts on achievement.” You can read the whole report here:

      http://www.mathematica-mpr.com/education/charterschools.asp

      The second is a study of of college completion rates among KIPP graduates. It found that while the first wave of KIPP graduates completed college at higher rates than average for their socioeconomic level (33% v. 8%), they completed college at lower rates than their academic achievement levels going into college would have predicted. The report attributes this gap to non-academic skills that are important to college success like self-discipline, persistence, and cultural comfort in a college environment. You can see the report here:

      http://www.kipp.org/ccr

  5. I don’t know, Kevin. The implications of these studies are very interesting –but I’m not convinced the school and peers have a minimal effect. At a certain level, a kid spends 7-8 hours a day at school with these peers but, say, an hour+ before school and then 3-4 hours in the evening, interacting with parents. I work at a private school where most of the students from multiple countries, 99% go to college, and, essentially, we have zero fights and minimal theft. My neighborhood school in Indiana was struggling with drop-outs and crime. The teachers at my current school –let me talk as one of the cooks in the kitchen, so to speak– go back onto the job market every few years and/or have an incentive to keep up/stay competitive, work at a good school. They actively select the schools they work in. When I was working in Indiana, I encountered a teachers’ union that made seniority a very hard thing to hire or fire, which had a whole complex of consequences.

    Intuitively, I think there is a level where education is like food. The family culture, eating and exercise habits have a bigger impact on a kid’s health and fitness than WHERE he eats. A kid who eats simple food is not necessarily disadvantaged compared to a kid who eats only expensive organic heirloom produce –but if you send a kid to the equivalent of McDonald’s every day, educationally speaking, it’s going to take a really smart and disciplined kid to come out of it intact.

    • I should clarify the frame of reference I’m using when I think about the impact schools have. It’s not that I think schools have no impact; and I do think that impact is going to be most noticeable when you compare two extremes, like an elite private school versus the most troubled inner city school. But the range I was thinking about is the difference between an average, decently functioning public school system and an elite suburban or private school. In the same way that my intuition says there’s not much of a difference between $100k and $150k for how Jay and Wally grow up, I don’t think that whether they go to a run-of-the-mill public high school or Exeter will make a big difference for what they achieve (or don’t achieve) in life. And this is significant, I think, because Caroline and I would have to make very different choices in life to send Jay and Wally to Exeter even though it might not pay big dividends in terms of the opportunities they receive.

      In a nutshell, my position is that with both income and school-quality, there’s a certain minimum standard that needs to be met after which the marginal gains are very small.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s