I will protect your pensions. Nothing about your pension is going to change when I am governor. - Chris Christie, "An Open Letter to the Teachers of NJ" October, 2009

Monday, February 25, 2013

"This Study PROVES..."

The fine folks at B4K have a great advantage over schlubs like me: they are unencumbered by the doubts that come from thinking about stuff...

WHAT HAS MORE IMPACT ON POOR KIDS: NEIGHBORHOODS OR SCHOOLS?

A recent Harvard study shows schools and teachers matter more.

02/12/13
This is the question asked and answered by Harvard economics professors Roland Fryer and Lawrence Katz in their recent study (here).   The short answer is: good schools and good teachers matter more. 
Oh, well, that's settled then! On the basis of one study, we now know that teachers matter more than neighborhoods! Great! Let's move on...

Uh, but before we do, could we look at page 10 of the actual study?


So when per capita income increases for a neighborhood, test scores rise in what appears to be an incredibly strong correlation. Hmm... do you think there might be more to this than the "short answer" B4K is selling?

Me too.

I was less interested in the paper than the citations, which included some really interesting studies. There does seem to be some good evidence that if you change a child's neighborhood - but don't change his family circumstances or the family income level at his school - there won't be much of an effect on his test scores.

Should that really surprise anyone? And does this at all show that teachers "matter" more than neighborhoods?

In one of the studies cited, families got vouchers to move to higher-income neighborhoods - but the schools weren't populated by students whose families had higher incomes. Well, that suggests to me that the new neighborhoods weren't that much different from the old neighborhoods. Yes, there were changes in other outcomes like adults' physical and mental health - great! But maybe larger changes are necessary to affect educational outcomes; maybe incremental changes in environment just aren't enough.

But that doesn't mean that environment isn't important. If you're going to argue that the chart above doesn't matter - and, indeed, that is exactly what B4K is saying - you need to put forward a logical counter-theory. You need to spell out the differences between high-performing and low-performing schools other than socio-economic status.

Now, B4K has a pretty simple agenda when it comes to public schools:
  • Charters, vouchers, and various other forms of "choice."
  • Merit pay.
  • Value-added teacher evaluations.
  • Gutting tenure and pretty much eliminating seniority.
  • Weakening credentialing standards.
  • Publishing teacher evaluations (no, really, they want that. Honest.)
Do we find any of these differences between high-performing and low-performing schools? Dr. Baker?

Yeah, not so much.

So how about we flip things around a bit? How about we come at this from the perspective that neighborhoods and schools are reflections of the lives of students? That the relative "failures" of neighborhoods and schools don't cause poverty, but are symptoms of poverty? 

Let's try this theory:
  • The neighborhood the child lives in is a reflection of his or her family's socio-economic status (SES). 
  • The neighborhood school is a reflection of the neighborhood families' SES.
  • Moving a family to a marginally "better" neighborhood - a neighborhood not "better" enough to have a "better" school - will have little impact on academic outcomes, especially if the SES of the family doesn't change.
To my mind, then, the answer is clear: if we raise the socio-economic status of all families in America, we have the best chance of raising the educational outcomes of all students. Certainly, B4K has presented far more evidence in this post to back up that policy prescription than "choice" and VAM and gutting tenure and publishing teacher evaluations.

This in no way means that teachers aren't important or that schools don't matter or that increasing resources to schools that serve children in poverty won't help those children. But it's time to stop dancing around the issue and be straight: the primary problem in American eduction today isn't our public school system - it's poverty.

So we can screw around with reformy, billionaire-backed polices, or we can get serious and go after poverty once and for all. We can continue to delude ourselves by twisting the research so we get the answers we want, or we can face up to what's really going on in this country. We can take one study and convince ourselves that it proves "poverty isn't destiny," or we can act like adults and have a serious conversation about the lives of America's children.

What's it going to be?

8 comments:

Anonymous said...

A few thoughts:

-When you mention determinants of low/high-performing schools, where do you consider things such as quality of instruction, well-resourced Tier 2 instruction within an RtI model, strong administration, experience level of teachers, etc? Are you suggesting that schools do not play any role in determining high or low performance?

"So how about we flip things around a bit? How about we come at this from the perspective that neighborhoods and schools are reflections of the lives of students? That the relative "failures" of neighborhoods and schools don't cause poverty, but are symptoms of poverty? "

My experience has been that there is a reciprocal relationship between family and neighborhood poverty/culture. There are generally struggles and strengths both in families and communities, and ignoring the impact of one or the other probably doesn't make too much sense.

"if we raise the socio-economic status of all families in America, we have the best chance of raising the educational outcomes of all students."

What is your plan for how teachers can do this during the school day? It may make more sense for teachers to focus on instructional strategies that have been shown to make an impact with struggling learners and kids from high-poverty backgrounds.

"the primary problem in American eduction today isn't our public school system - it's poverty."

There is a difference between cause of educational problems and potential strategies that could improve educational problems. There are many folks who have found success teaching children from high-poverty backgrounds, but they certainly didn't cause those problems to begin with. Also, none of those teachers who do find success succeed with ALL children, just as teachers from wealthy schools don't either.

When you ask, "What's it going to be," why would we chose to only focus on quality of education provided OR addressing poverty? Why would we not address both? I'm not suggesting we look to merit pay, teacher evaluation via state test, or some of the other ideas presented. However, disagreement with evaluation via test does NOT mean we believe schools are doing everything they can, or that our only way of helping kids is by directly addressing poverty.

I believe it's time we start identifying a second group of folks out there are believe we can do more as educators, but who don't look towards overly simplistic reform efforts such as evaluation via test and merit pay. All educators cannot simply be lumped into those two categories, and in fact most reasonable folks aren't in either.

Galton said...

Jazzman, I find the last sentence in this study appalling' "A vital policy question is how to generate systematic
large-scale improvements in school and teacher quality for low-income students growing up in high-poverty neighborhoods."

Duke said...

-When you mention determinants of low/high-performing schools, where do you consider things such as quality of instruction, well-resourced Tier 2 instruction within an RtI model, strong administration, experience level of teachers, etc? Are you suggesting that schools do not play any role in determining high or low performance?

"This in no way means that teachers aren't important or that schools don't matter or that increasing resources to schools that serve children in poverty won't help those children. "



My experience has been that there is a reciprocal relationship between family and neighborhood poverty/culture. There are generally struggles and strengths both in families and communities, and ignoring the impact of one or the other probably doesn't make too much sense.

Exactly my point.



What is your plan for how teachers can do this during the school day? It may make more sense for teachers to focus on instructional strategies that have been shown to make an impact with struggling learners and kids from high-poverty backgrounds.

OK. But nothing B4K suggests is an "instructional strategy."




There is a difference between cause of educational problems and potential strategies that could improve educational problems. There are many folks who have found success teaching children from high-poverty backgrounds, but they certainly didn't cause those problems to begin with. Also, none of those teachers who do find success succeed with ALL children, just as teachers from wealthy schools don't either.

"if we raise the socio-economic status of all families in America, we have the best chance of raising the educational outcomes of all students."



When you ask, "What's it going to be," why would we chose to only focus on quality of education provided OR addressing poverty? Why would we not address both? I'm not suggesting we look to merit pay, teacher evaluation via state test, or some of the other ideas presented. However, disagreement with evaluation via test does NOT mean we believe schools are doing everything they can, or that our only way of helping kids is by directly addressing poverty.

I believe it's time we start identifying a second group of folks out there are believe we can do more as educators, but who don't look towards overly simplistic reform efforts such as evaluation via test and merit pay. All educators cannot simply be lumped into those two categories, and in fact most reasonable folks aren't in either.



This blog, and many of the people at the left, are those educators. Of course schools and teachers can get better - I have said this over and over again.

But worrying about teacher and school quality when our society has so much poverty and inequality is like worrying about your tire pressure when your car is making a bad clunking noise under the hood. Tire pressure is important; you can damage your car if it isn't right. But messing with it won't help much when the transmission falls out.

Duke said...

Galton, maybe we should be asking how to generate large-scale improvements in the students' lives instead, huh?

Anonymous said...

Duke, it sounds like we're in agreement with most things. Your last paragraph is where I think we will still find disagreement. I do love car analogies, so I appreciate that, but I have 2 points of disagreement: the relative weight you assign to poverty vs. schools/teachers, and dependence of teacher quality on fixing poverty. Here's what I mean:

1) In terms of relative importance, I'm not sure I'd assign teacher/school quality to "tire pressure" and poverty to "transmission." In other words, I'm wondering if we disagree about the amount of effect a teacher/school can have. I'm wondering if you arbitrarily chose tire pressure/transmission, or if you are implying that poverty is a much bigger variable than teacher quality? My experience has been that this varies by teacher, school, neighborhood, etc. I haven't found that poverty is always more important than teacher quality. I have certainly found situations that are more difficult to address, and that are rooted in poverty or other environmental stressors. However, I have found many situations in which the teacher was a greater, not lesser, determinant than poverty. Has this been your experience?

In the graph you include, I'm not sure you can conclude that poverty affects ELA scores over teacher/school quality. The graph does not seem to control for teacher/school performance/input (i.e., quality). So, it's unclear if there is also a relationship between poverty level of school population and quality of instruction. My experience has been that, while there are some great teachers that love teaching in high-poverty schools, there are also many teachers who prefer teaching in non-high-poverty schools and attempt to leave. So, if we held "quality of instruction" constant, for example, it would be interesting to see if/how that particular graph might change. It may show a less clear relationship between poverty and ELA scores.

2) In your car analogy, no amount of fixing tire pressure will allow the car to run unless you also fix the transmission. In other words, the success of tire pressure is completely dependent on fixing the transmission. What you're saying, then, is that no amount of adjusting teacher/school quality would have an effect on education unless poverty is first fixed - that fixing poverty is a pre-condition for teachers to work effectively. This has not been my experience.

Perhaps a better car analogy would be considering the teacher to be the mechanic, and the poverty to be the quality of road conditions the car routinely drives on. Sometimes the road may be so awful, and the car so damaged, that the mechanic is simply unable to do anything. However, many times a good mechanic can improve the quality of the car even if it's in bad shape because of road conditions. A bad mechanic is much less likely to be able to do so if the problems with the car become more complex and severe.

So, I absolutely think it's worth worrying about the quality of teachers/schools. Not only do I think it's really the only thing we CAN worry about as educators (unless we get involved in social policy outside the classroom), but I think it can be highly effective even if poverty is never fixed.

giuseppe said...

Overall, teachers are doing the best they can in spite of the huge poverty problem in this country as compared to the other rich industrialized democracies which have a much lower child poverty rate. The schools in affluent school districts score just as well if not better on those international tests (PISA) than the other high scoring countries like Finland (which has a very low child poverty rate). No matter all the verbiage, deflections and general BS, poverty does matter and often determines school outcomes whether we like it or not.

Tobey said...

Hi Jerseyjazzman,

I quite appreciated this post, thank you for sharing it.

In particular, I recognize the worth of your assertion that "if we raise the socio-economic status of all families in America, we have the best chance of raising the educational outcomes of all students."

A consequence of this is that raising educational outcomes becomes inextricable from mediating inequalities. From here it I believe that it becomes imperative to advocate for policies which are capable of diminishing inequalities - not enhancing them. And that's an agenda that flies in the face of the "choice" narrative in which parents are pushed into shopping for learning experiences - much as they might shop for a new car or condo.

This point aside, another aspect that I appreciated is your attention to misinterpretations of research. For instance, some folks will pick up on this study as "proof" of X, Y, or Z, instead of /evidence/ of X, Y, or Z. This provides a context in which complex problems are rendered as simplified solutions, and misapprehensions can be mistaken as insights.

Regards,
@symphily (https://twitter.com/symphily)

Mrs. King's music students said...

The poverty isolates you and makes you feel powerless. I've worked in city schools before but not like Camden. I spend way too much time managing (or containing) bad behaviors and trying to make crumbling mouse infested instruments play. For their part, the kids are buying everything I'm selling in the bandroom and would be moving up smartly in a better situation. But here, most of them won't because their parents don't have cars and don't want to put their kids on public busses to get them to the better schools on the other side of town. Really, the lack of transportation is the single biggest setback in inner cities. If someone could solve that they'd be a hero.