Some School Districts Do a Better Job Educating Poor Kids than Others


I’m playing around with Datawrapper, which provides cool ways to display data– don’t quite have the hang of it, but making progress. Anyway, my inaugural effort shows the considerable variability between school districts in pass rates for English Standards of Learning (SOL) tests.

We all know that the socio-economic status of a student is a major predictor of their academic achievement. Because school districts draw their student bodies from very different socioeconomic backgrounds, it is not fair to compare the academic achievement of Virginia school districts without adjusting for demographics. Therefore, for this map I compare the English SOL pass rate for disadvantaged kids, kids who are poor enough to qualify for free school lunch.

Virginia school districts range from an 87.5% pass rate for disadvantaged kids in West Point, a mill town on the edge of Hampton Roads, to 49.3% for Danville, a mill town in Southside; and from 85.96% in Highland County, the locality with the smallest population in Virginia, to 51.68% in the City of Richmond, the state capital.

If disadvantaged kids in Danville, Petersburg, and the City of Richmond have dismal standardized test results, local educators can’t blame the outcomes on poverty alone. Other localities have poor kids, too, but they have significantly better outcomes. What could explain the variability between school districts?

One possibility is that some districts spend more money per student than others. Perhaps West Point and Highland County spend more per student than Danville and Richmond. The “more money” hypothesis seems less than plausible from a superficial look at the map above, which shows that the pass rates for disadvantaged kids tend to be lower in the affluent Northern Virginia localities. But maybe there’s an explanation that transcends spending per student. Maybe Northern Virginia school districts have more hard-to-educate English-as-a- Second-Language students. The issue warrants closer examination.

Another explanation of the variability seen in the map might be that poverty is worse in some localities than others — not more widespread, but more intense and socially destructive. In cities like Richmond, Petersburg and Danville perhaps the poverty is more concentrated in a few neighborhoods, or poor kids are more concentrated in a few schools, or the degree of social breakdown and dysfunction is greater.

Yet another potential explanation is that school districts have different racial/ethnic mixes and that different ethnic groups put a greater premium on succeeding academically than others. For example, Asians might study harder than their socioeconomic peers in other racial/ethnic groups. Or Hispanics might encourage their kids to drop out of school, become wage earners and contribute to their families.

Yet another option: Maybe some school districts do a better job with the resources and student populations they have.

Finally, a related possibility: Perhaps the move from traditional disciplinary practices to restorative justice disciplinary practices (my pet theory) has eroded discipline and promoted classroom disorder with deleterious consequences for kids who want to learn.

Clearly, the data in this map tells us only so much. But one limited conclusion does seem inescapable. Blaming poor educational results on the prevalence of “poor kids” in the school district goes only so far.

Assuming I can figure out how to create fully functional maps, I’ll be exploring these competing theories in the future.

There are currently no comments highlighted.

14 responses to “Some School Districts Do a Better Job Educating Poor Kids than Others

  1. This could be good – as well as dangerous!

    I applaud the effort to look more comprehensively at the data.. Kudos!!!

    but we also know a couple of things we need to take into account.

    1. Counties and their borders do not really define the pockets of the disadvantaged kids. For instance, we know that Richmond and probably Portsmouth probably have district-wide problems but how about a county like Henrico compared to Lancaster or even Albermarle or Fairfax?

    2. – Clearly from looking at the map you produced – the rural counties have a different experience than the more urban counties. One important difference is that many rural counties don’t have a lot of schools and don’t have schools aligned according to neighborhoods… the schools are much more homogenous with respect to the mix of kids from diverse economic circumstances. I bet you could show that on a map also if you’d map schools rather than counties… Henrico, for instance would likely show huge differences between schools within Henrico but the rural counties would not since they have far fewer schools not really aligned according to the economic demographics of the neighborhoods.

  2. “Other localities have poor kids, too, but they have significantly better outcomes. What could explain the variability between school districts?”

    I don’t have a solid answer to this very good question. When I think about K12, I often think about one statistic — that the average child spends about 13% of his/her waking hours in school from birth to age 18. With that in mind, roughly 87% of the contextual influences on human capital development come from outside-of-school environments (home, neighborhood, etc.).

    Even looking just at students with free school lunch, I think there are meaningful differences in their families, out-of-school environments, and schools across localities in Virginia, as you have noted. I don’t know what those differences are, but I think the binary aspect of free school lunch (yes/no) means some of the disadvantages and contextual differences are not adequately captured in that one control variable.

    • I totally agree. There is a lot of variation even within that “disadvantaged/free lunch group.” You hear anecdotal stories about amazing moms fighting for their kids, and kids who just defy the odds.

    • An interesting chart, but I suggest there are different ways to look at the problem. And what is the problem exactly? Does not the problem depend on the facts surrounding it? And if we have the wrong facts, can we define the problem at all? Much less the solution, if there be any solution at all?

      One central question is whether on not the problem is written in the stars, or in ourselves? And is that a distinction without a difference? For example:

      Is one demographic grouping less academically talented than another? If so, what is that the case. And if that be so, what can we do about that “fact”? Can we change that reality. And if so, can we do it within only limited parameters? Or can we change that reality altogether? Turn it upside down? Or can we instead equalize results within demographic groups? And if that be so, what then are the consequences, intended and unintended, seen or unforeseen, positive or negative? And how do we gauge and weigh those consequences, before we act to maximize our positive results and minimize our negative results, or create the most beneficial mix of possible results?

      Or is all of the above a fools errand? Nothing more than tilting at windmills? Or the useless babbles of a hard core racist, or of hard core realist? Indeed is our throwing out of our heads these demographic studies of groups the essential beginnings of throwing our own deeply built in biases and limitations out the window, so we can get for the first time seriously onto the tasks of really solving real problems in realistic and practical ways?

      One thing seems for sure. Mostly we been going backwards in education for the great majority of our kids now for some 60 years. All these self congratulating schools of education, the teaching and researching of how to educate, not to mention the Federal Department of Education, what have all these institutions done for us? Anything at all? Where and what are their results? Positive or negative? And how much have they cost us, in how many ways, if we’ve gone backward so far over the past 60 years?

      • Jim here tries to isolate factors in Virginia’s system of k-12 education that thwarts the ability of students to learn in Virginia’s classroom. I share Jim’s concern, but I would enlarges his concern, specifically:

        I suggest that today’s k-12 classrooms often not only thwart student learning therein, but also often destroys whatever learning their students have achieved outside the classroom before their arrival inside the classroom. And that our society and its educational system carries on this destructive behavior inside and outside the classroom on a daily basis. And that the bulk of the blame for these horrible results lie not in the k-12 classrooms, but within Virginia’s system of higher education, and within our our society generally, including within the state.

        Thus, I asked series of questions posted immediately above, including this typo corrected version of part of those questions, namely:

        “One central question is whether on not the problem is written in the stars, or in ourselves? And, is that a distinction without a difference? For example:

        Is one demographic grouping less academically talented than another? If so, WHY is that the case. And if that be so, what can we do about that “fact”? Can we change that reality. And if so, can we do it within only limited parameters? Or can we change that reality altogether? Turn it upside down? Or can we instead equalize results within demographic groups? And if that be so, what then are the consequences, intended and unintended, seen or unforeseen, positive or negative? And how do we gauge and weigh those consequences, before we act to maximize our positive results and minimize our negative results, or create the most beneficial mix of possible results?

        Or is all this a fools errand? Nothing more than tilting at windmills? Or the useless babbles of a hard core racist, or of hard core realist? Indeed is our throwing out of our heads these demographic studies of groups the essential beginnings of throwing our own deeply built in biases and limitations out the window, so we can get for the first time seriously onto the tasks of really solving real problems in realistic and practical ways?

        One thing seems for sure. Mostly we been going backwards in education for the great majority of our kids now for some 60 years.”

        These are big complex questions. A modest start to finding answers might be found is this comment posed by Nathan Cofnas who’s reading for a DPhil in philosophy at Oxford. Mr Cofnas here is commenting on a year long project undertaken by scholars James Lindsay, Helen Pluckrose, and Peter Boghossian who sent fake papers to academic journals said to specialize in activism or “grievance studies.” Their mission was to expose how easy it is to get “absurdities and morally fashionable political ideas published as legitimate academic research.”

        “To date, their project has been successful: seven papers have passed peer review and been published, including a 3000 word excerpt of Adolf Hitler’s Mein Kampf, rewritten in the language of Intersectionality theory and published in the Gender Studies journal Affilia,” according to the editors of “the Grievance Studies Scandal: Five Academic’s Respond, published on October 1, 2018 in Quillette Magazine. Mr. Cofnas, the first of the five scholars to respond in the Quillette article, says:

        “Twenty years ago, Alan Sokal called postmodernism “fashionable nonsense.” Today, postmodernism isn’t a fashion—it’s our culture. A large proportion of the students at elite universities are now inducted into this cult of hate, ignorance, and pseudo-philosophy. Postmodernism is the unquestioned dogma of the literary intellectual class and the art establishment. It has taken over most of the humanities and some of the social sciences, and is even making inroads in STEM fields. It threatens to melt all of our intellectual traditions into the same oozing mush of political slogans and empty verbiage.

        Postmodernists pretend to be experts in what they call “theory.” They claim that, although their scholarship may seem incomprehensible, this is because they are like mathematicians or physicists: they express profound truths in a way that cannot be understood without training. Lindsay, Boghossian, and Pluckrose expose this for the lie that it is. “Theory” is not real. Postmodernists have no expertise and no profound understanding.

        Critics of Sokal point out that his paper was never subjected to peer review, but this time there are no excuses. LBP’s papers were fully peer reviewed by leading journals. The postmodernist experts showed that they had no ability to distinguish scholarship grounded in “theory” from deliberate nonsense and faulty reasoning mixed in with hate directed at the disfavored race (white) and sex (“cis” male).

        King Solomon said of the fool: “His talk begins as foolishness and ends as evil madness” (Ecclesiastes 10:13). Can a disregard for evidence, logic, and open inquiry combined with a burning hatred for large classes of people perceived as political opponents (“racists,” “sexists,” “homophobes,” “transphobes,” etc.) possibly lead to a good result? The editors and peer reviewers who handled LBP’s papers have revealed their true, vicious attitudes.

        The flagship feminist philosophy journal, Hypatia, accepted a paper (not yet published online) arguing that social justice advocates should be allowed to make fun of others, but no one should be permitted to make fun of them. The same journal invited resubmission of a paper arguing that “privileged students shouldn’t be allowed to speak in class at all and should just listen and learn in silence,” and that they would benefit from “experiential reparations” that include “sitting on the floor, wearing chains, or intentionally being spoken over.” The reviewers complained that this hoax paper took an overly compassionate stance toward the “privileged” students who would be subjected to this humiliation, and recommended that they be subjected to harsher treatment. Is asking people of a certain race to sit on the floor in chains better than asking them to wear a yellow star? What exactly is this leading to?” For more of this fine article please see Quillette.com END OF QUOTE

        I find that this absurdity, namely, the ongoing efforts of many university Departments of Arts and Sciences to destroy education and undermine western civilization as they pit different groups of people within America one against the other, is directly responsible for the efforts behind the wife of Senator Kaine, and her ilk, to push restorative justice disciplinary practices in Richmond’s k-12 schools that are eroding discipline and promoting classroom disorder with terrible consequences for kids who want to learn. And, I suggest these awful practices are tearing our country apart, threatening race and class wars in our nation. These wars all around us now are spreading like wildfire.

        But why should we be surprised, when rules governing classroom behavior now require teachers to punish an proportionate number of white kids in order punish black or brown kids creating disorder. Only the hate of ideologues such as fascists could built and promote such a perverted system of injustice in America’s classrooms.

  3. I have analyzed school performance as a function of many variables, including race. My report on the DC-area schools (MD, DC, VA) can be found at http://www.fcta.org/Pubs/Reports/2014-04b-fac.pdf. A similar study for all schools throughout Virginia comes to the same conclusion(http://www.fcta.org/Pubs/Reports/2014-08a-fac.pdf). Fairfax County schools perform better than others because they have such a high Asian population. The correlation between Hispanic and Free-and-Reduced-Meals students is too strong to separate the effects. When corrected for demographics, all Virginia schools have similar SAT scores, with Fairfax slightly underperforming.

  4. The problem with the geography.

    Some counties have few schools so their free and reduced is truly county-wide.

    Other counties with higher populations have many more schools but they are often aligned according to the economic demographics of the neighborhoods. In other words – there will be schools in the same county that have high percentages of free & reduced and other schools in the same county that do not but we won’t see this if the data of both the low and high schools is aggregated on a county basis. The data is effectively submerged.

    So the data for the counties with many schools gets distorted. The county shows one percentage of free/reduced but the reality is that the good schools have very low free/reduced percentages and the low-income schools have high numbers but what we see is an “average”.

    To be sure – there will be culture and other differences but by and large the most significant commonality is the economic status of the family the kids come from because more often than not – their parents are not well educated and as a direct result – are low income. But if you look only at county level data – you’ll not see this.. and instead you’ll be thinking about free/reduced on a county basis rather than on a school basis.

    Once one actually recognizes this – it takes us away from the narrative that the govt or public schools are monolithic “failures” and instead to the view that for kids of educated parents -the schools are “good” and for kids of parents who are not well-educated – SOME , not all , of the schools are legitimate “failures”.

    And of course, the burning question is – could non-public schools do a better job with THESE economically-disadvantaged kids and if so, would those schools voluntarily report the same success/failure metrics than the public schools must report?

  5. On the counties with many schools idea.

    How about we take a county like Henrico or Fairfax and do a map that shows each school along with it’s free & reduced percentage – and SOL reading scores?

    Then show the top numbers for the average free & reduced county-wide and the aggregate county level SOL reading scores?

    I think if you do this -we’ll all develop a better appreciation of what the data is telling us – as well as what it is not.

  6. This is a good post that shows some opportunity for further analysis. Perhaps Cranky might take a swing at this data. The variations in performance are sizeable, even between jurisdictions that seemingly have similar demographics. Perhaps there is something that can be identified as a driver of different outcomes. Perhaps it is something Fred has already covered. I haven’t had d chance to review that yet.

  7. There are some schools in Virginia that do well at teaching kids from low-income families. And there are some that are not. It takes a different kind of teaching to successfully teach these at-risk kids and the truth is that teachers that have those skills do not want to teach at schools where despite their best efforts – they get dinged for not totally “fixing” the problems.

    In other words, they’re human… and they don’t want to have their careers harmed by having more thrown at them than they can handle.

    The most friendly situations are schools where there is a mix of kids – some average, some high and some low.. and special help is given to those who are behind.

    The most unfriendly are neighborhood schools serving predominately low income neighborhoods and the problems are systemic and even the skilled can overcome by themselves unless the school system itself supports the principal and in turn the skilled folks. Otherwise – the skilled teachers will bail.. if they think it is a sinking ship.

    I’d wager the very same thing would happen no matter if the school if public or private. No teacher wants to get thrown to the wolves and if they have good credentials – they can choose better places to have a career.

    The more folks like Cranky hammer the problem schools – the more the teachers in those schools will leave… and the harder it will be for the administration to recruit GOOD teachers.

    This is why VDOE is revamping it’s ratings so that schools that are behind – but making progress – get recognized for their efforts – which motivates the folks at those schools to keep at it.

  8. The more folks like Cranky hammer the problem schools – the more the teachers in those schools will leave

    Yeah, let’s blame Cranky for the horrendous churn in failing schools! What a howler!

    I can guarantee you, if you interviewed the dozens of teachers who departed from the hard-case schools, not one will attribute their decision to people like Cranky dissing the schools. In almost every case the instance will be related to disciplinary issues or personality conflicts with administrators.

    The young teachers who sign up to teach at the hard-case schools are idealistic. They want to help poor kids. They’re just not prepared for the conditions they encounter. Many burn out quickly.

    • People like Cranky are using the very transparency that the schools actually do have to hammer them for things they cannot control and that in turn makes the schools not so keen about releasing more info … it’s a lose-lose…for them.

      Cranky never seems to get into the difficult areas that make teaching low income demographics so hard. He is full of condemnation and little else including what he thinks should be done to fix.

      Young teachers sign up for whatever job they can get -and get thrown to the hardest schools that veteran teachers won’t touch with a 10 foot pole because they KNOW that they cannot quickly or easily fix the problem and people like Cranky will hammer them if they “fail”.

  9. I love it when Richmond area people examine problems. I guess immigration just isn’t an issue in Richmond. I say that because I only had to read two sentences until I had a key question … How many of those failing the English SOLs have English as a second language? Arlington County is 15.4% Hispanic or Latino. In Henrico, it’s 6%. 28% of Arlington residents were foreign born as of 2000. 94% of people living in Henrico County were citizens in 2016. The corresponding number for Arlington was 87%.

    44,118 of Henrico County, VA citizens are speakers of a non-English language, which is lower than the national average of 21.6%.

    61,349 of Arlington, VA citizens are speakers of a non-English language, which is higher than the national average of 21.1%.

    Given that Henrico has a population that is 39% bigger than Arlington the difference in speakers of a non-English language is startling.

  10. The thing about Richmond schools – is that Henrico has some individual schools just like Richmond and you never hear it from Cranky because he is focused on Richmond alone – NOT the issues of which Henrico – as well as Fairfax do also share.

    How can “Good” school systems like Henrico and Fairfax have some schools that are more like Richmond schools but are not called “failures” like Richmond is?

    English as a second language is also an issue – true… and again – the schools have no choice – the law requires them to figure out how to educate kids who do not speak English – as well as educate kids whose own parents have crappy educations and low incomes because they don’t have good educations.

    WHERE did these parents with bad educations come from to start with?

    Did they just suddenly spring up like mushrooms after a rain?

    Does anyone think that kids who have parents with bad educations and low incomes are part of the issue – that exacerbates the “failed” schools?

    Does Cranky publish data for the low scoring kids in terms of the education and income levels of their parents?

    I do NOT give Richmond or any school system INCLUDING HENRICO a “bye” on this issue but anyone who asserts that it proves that the schools are a “failure” is either ignorant or just not interested in understanding the actual problem. And the proof of this is that almost never do the critics talk about how to address the issue except the ridiculous idea that non-public/charter schools can fix it – but of course those critics are NOT in favor of the same level of transparency to judge those schools efforts.

    So what this boils down to is essentially a jeremiad with no intention of talking about how to remedy.

    That’s why I ask what the real purpose is of the relentless hammering …

    If Cranky and the critics were actually pointing to other schools that were having success and advocating that Richmond change… yes… that would show some good intent on their part – but that’s not happening.

Leave a Reply