Null hypothesis watch

Several readers spotted a story on the results of a Bill Gates initiative to improve teaching. The null hypothesis won.

https://www.washingtonpost.com/news/answer-sheet/wp/2018/06/29/bill-gates-spent-hundreds-of-millions-of-dollars-to-improve-teaching-new-report-says-it-was-a-bust/?noredirect=on&utm_term=.e24c304f9896

24 thoughts on “Null hypothesis watch

  1. What’s truly annoying is that the Gates initiative tried all of the education reform shibboleths. None of them worked. Moreover, despite all the talk about charters, there’s vanishingly little evidence charters improve outcomes independent of skimming and creaming.

    Happily, education reform policy objectives were largely wiped out in ESSA. But it’s incredibly frustrating that independent researchers are still reflexively pro-reform. No one will stand up and acknowledge reality: we’re doing a fairly good job, and the reform objectives have only increased teacher scarcity, drove up salaries, and give some people the idea they have the right to have private school selection at public prices.

    • Are you still up in arms about people wanting a decent education for their talented kids if they can’t afford private school or sky high rents. Are you pro destroying Stuyvesant because it’s too Asian?

      • People can want whatever the hell they want. They just have to pay for it. The taxpayers shouldn’t have to fund some people’s desire for private school.

        And I have no objection to Stuy being “too Asian”, as it would be no matter what. But they need to change the test, and make it much harder to game.

        • Bringing out the talents of our best and brightest through proper educational environments is in societies interest including the less fortunante. Destroying Stuyvesant out of some act of egalitarian rage is pure nihilism.

        • I had assumed that the kids at Stuy and similar schools were driven, diligent, and “objectively smart” in the sense that see seem to have high IQ as typically measured. Are you saying that they are mostly in such schools because they are gaming the tests and somehow “cheating” or proceeding in a way that is unethical?

          An issue I see: Many people improve on tests if they take them over and over. Therefore, taking tests over and over to improve one’s performance is not unethical–yet who is motivated to do such things, especially between the ages of 10 and 13?

          It’s somehow unfair but not unethical that high stakes testing exists and some youngsters are practicing while others don’t know they exist. Reeves touches on this issue in _The dream hoarders_–there is a literature on this topic that I don’t know but that he mentions, using a different metaphor, that involves fair tournaments in which the healthy and well-nourished (imagine early modern Europe) battle it out in physical combat against the scrawny and malnourished.

          • That is one of the most predictable dynamics in human history; winners change the system to ensure that their children succeed. Who is so noble as to let their children fall into poverty so that someone else’s kids can have a fair chance? Nobody.

  2. Was anyone else bothered by Strauss’ pooh-poohing of Gates’ initiative for being devoid of research when implemented (“America’s wealthiest citizens choose pet projects and fund them so generously that public institutions, policy and money follow — even if those projects are not grounded in sound research.”)? The whole rest of the article was about how this was part of a research project. She set up a nice Catch 22 which essentially enforces the status quo because it demands we only do things that are “grounded in sound research” while not considering how sound research is conducted.

  3. Paging through the Rand review on teacher effectiveness discussed in the article (https://www.rand.org/pubs/research_reports/RR2242.html ), it seems to me that the Gates grants were highly productive: we now know that teachers are not the centerpiece of education and efforts to improve education need to be focused elsewhere in areas like curriculum, self-directed learning, personal matching with alternative educational approaches, etc. The US has long been dominated by the view that schools are places whose purpose are to provide teachers self-fulfillment opportunities. Nowhere more so than in higher education, where the primary purpose of colleges and universities is seen as little more than providing sinecures for people of the right political persuasion and lots of degrees.

    However, if you want to stoop to Rand studies (I’m not a big fan of their work which is generally Center for American Progress-lite), maybe you should also be aware of their recent education reports on interventions which may, in time, reject the null:

    – Are Content-Specific Curricula Differentially Effective in Head Start or State Prekindergarten Classrooms?(https://www.rand.org/pubs/external_publications/EP67635.html )
    “Findings from this study demonstrate that content-specific curricula are effective in different preschool settings and should be leveraged to improve children’s educational outcomes.”

    -High-Quality Early Childhood Programs Can Change Lives
    The most comprehensive look to date at the benefits of early childhood education found that 102 of 115 programs improved at least one outcome for children beyond a statistical doubt. And the economic and social benefits continue to pay dividends, sometimes well into adulthood. Jan 3, 2018 ( ttps://www.rand.org/topics/education-and-literacy.html?page=4 )

    -How Does Personalized Learning Affect Student Achievement?
    “…students in the PL schools started the year significantly below national norms in both mathematics and reading, but moved closer to the norms during the school year. In mathematics, students gained about 2 percentile points but remained significantly below national norms; in reading, students also gained about 2 percentile points and were performing approximately at national norms by spring.”
    (https://www.rand.org/pubs/research_briefs/RB9994.html )

    -Dual-Language Immersion Programs Raise Student Achievement in English
    Portland Public Schools students randomly assigned to dual-language immersion programs outperformed their peers on state reading tests by 13 percent in grade 5 and by 22 percent in grade 8. Students gained proficiency in a second language with no drop in their performance in math or science. Nov 3, 2017 ( https://www.rand.org/topics/elementary-education.html )

    Do these papers demonstrate interventions that make a “long-term, scalable, replicable difference?” No. But they do show interventions that have made differences. And maybe micro-level successes are what is important and not questing for macro-level magical one-size-fits-all panaceas. Different schools for different people.

  4. While it may be “early days” yet much, testable by research, and over multiple, individual variations, is being shown and learned (about learning) from the pilot programs of the Kahn Academy, to which the Gates Foundation has been a major benefactor.

    Check out your “Null hyposthesis” with the real measurements of what the support given Kahn has enabled them to introduce and achieve.

      • Hopefully, you are really interested.

        While there is extensive (and detailed) information on YouTube, as well as on the Kahn Academy website (recommended), a very simplified description of a major achievement has been (and is) the cooperative effort with specific (diverse) **public ** schools in CA., that have replaced collective *objective* “teaching systems” of public education systems with individualized subjective “learning systems” over a wide variety of topics (math, biology, sciences, etc.) at the several levels (and age levels). This methodology also provides instructors with real-time information on individual student learning; pinpointing learning difficulties (or “slacking”) [rather than waiting for “group” averages to assume results].

        These steps are possibly only the beginning of a new basis for classroom and **personal** learning.

        Do take a look at what they are doing.

        • I have provided an answer (with details) whixh has been “awaiting moderation” since 10:55 a m 7/5.

          Not complaining; explaining.

          • My comments here get stuck in moderation all the time. Not complaining either, and it’s a small price to pay if it gets rid of spam and trolls.

        • I suspect that the use of the term “Kahn Academy” in my answer may have caused a “security” check block.

  5. Good school associated with high parental involvement: PTA, after school activities, pancake fundraisers, soccer, field trips,… Dunno cause and effect.

  6. If the null hypothesis is correct, why not just shut down the school system? After all, the educational intervention – school – achieves nothing.

    • That’s NOT what the null hypothesis says!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

      It says that in America, any non-boutique change to the EXISTING SYSTEM will have little or no lasing effect. That’s not because American schools are turning most of their students into well-educated adults. Such an outcome is impossible in a compulsory system with a school-leaving age of 16 or more. It’s because American schools are doing about as well as they can. For their own reasons, neither the right nor the left is willing to believe that.

      Of course, there are many school systems around the world where the null hypothesis is false. See e.g., the awful stories and statistics in Lant Pritchett’s The Rebirth of Education: Schooling Ain’t Learning.

      • That was a pretty good working definition of the null hypothesis in K-12 education, to my mind. Maybe not perfect, but aysmptotically (sp?) approaching a great definition, and in one sentence.

        “In America, any non-boutique change to the EXISTING SYSTEM will have little or no lasting effect.”

        I will suggest a corollary: low hanging fruit is still around, somewhere, but the existing system precludes grasping for them.

        Possibilities… (besides boutique solutions)…I started listing my pet hobbyhorses and then caught myself.

        A good analytical treatment of the issue, methinks, is still Myron Lieberman’s _Public education: An autopsy_ which is available full text online through ERIC or a similar clearinghouse.

        • I’d be really interested to know what you think the low hanging fruits are. (Like cliches, hobby horses sometimes exist because they are true.)

          • 1. More emphasis on literacy instruction at an early age. Methinks the worst students fall behind in reading early on and never catch up. This may not actually “be their fault,” and the teaching system can improve outcomes. Somehow. I like Diane McGuiness’ work on this, but I’m not an expert and have no training in the field.

            2. Kick out the trouble-makers and program wreckers, out of the mainstream classrooms and schools, and perhaps into the labor market as early as possible. My guess would be that in many schools there is a 1% of students with serious conduct disorders who make it harder to teach.

            3. Allowing children back into the labor market, based on a “the door swings both ways” policy–easy to drop out, easy to drop back in. Can’t kids go to work camps and do crafts, rather than sit in a classroom? Or just make them run around the sports field for three hours till they drop? Something, anything, to provide less emphasis on holding kids “in custody” in classrooms.

            4. Generally, more emphasis on “the door swings both ways” so students can assign themselves to things they like. Oddly, last month I heard a lecture on Winston Churchill where the speaker asserted that young Winston was hopeless at classical languages and was instead permitted to triple on English.

            I could make a longer list. Anyone could. As Linus Pauling said, “The way to have lots of good ideas is to have lots of ideas and then throw the bad ones away.”

            allowing adults to teach without certification is a an obvious one. Real obvious. Make a list of the smartest, most effective, most successful, most thoughtful, most imaginative, most fun, most captivating, most *something* people you can think of. Now, ask yourself, how many of them could teach in a public school in the USA next year based on the current system?

  7. To my mind, the superior performance of the Dutch educational system, which is dedicated to family choice, and in which dozens of different pedagogical strategies are readily available to choose from, soundly rejects the null hypothesis. In study after study, union-free charter schools in the US do as well but are so few and far between that the true potential of meaninful choice has not been realized. The US will remain the educational laughingstock of the world as long as we continue to pursue a one-size-fits all panacea. To the extent that Dr Kling’s “null hypothesis” insists that only a panacea is legitimate reform, it is counterproductive. “Boutique” schools are the answer and reform and pedagogical reforms should be accomplished at the micro-level, not at some macro-level of agglomeration. Rejecting the “null hypothesis” is a fool’s errand. Children are not fungible. Repeat, Children are not fungible. For god’s sake Dr. Kling, please open your heart to the idea of specialization in education.

    • I tend to agree with this–lots of boutiques, with free assignment of students and staff based on decentralized management. There would be a lot of cycling through, and a lot of mistakes and wasted effort, but also a lot of “good fit” between students and programs.

      A big challenge is that centralized management by school boards would be intimidated and flummoxed by trying to manage or administer such a system–or even to monitor it.

      Another big challenge is it might be a magnet for crooks and scammers to float semi- bogus programs.

      A third big challenge is there are going to kids who are so problematic that almost no program wants to accept them. I’m not sure what to do with that issue. Some troublesome youngsters benefit from a military or paramilitary ethos–what to do with the rest is beyond me. In the old days they would be in the wild west, Cossacks on the frontier, mining gold in the Amazon, fur traders assaulting native women in the northern forests, etc. At sea. in the galleys. An empirical question, and one size doesn’t fit all.

Comments are closed.