Fully agree. Where schools have failed worse, there are obvious issues upstream causing other problems too. Schools and teacher quality did not just randomly degrade.
FWIW, I vibecoded a little survey tool at votto.app to measure sentiment on takes like this across different traits and media community affiliations because I'm genuinely fascinated by how different groups land on these debates and what other opinions are correlated with them. EG, respond to "Schools are being blamed for educational decline that has much more to do with smartphones and broader social conditions than classroom policy" and filter to see how other groups answer. Dedicated page for Freddie readers here: votto.app/freddiedeboer
Explore if for no other reason than liking cool graphs... other communities have more robust data, but would love to start accumulating data for this crowd which will inevitably split in fascinating ways...
I continue to be bemused that Singapore is regarded as a model for other countries to attempt to emulate. The entire area of Singapore is only slightly more than half of the Brisbane City Local Government Area, which is itself only a fraction of the area of metropolitan Brisbane. As well as geographic compactness, Singapore has the kind of political economy and political ecology that is only possible in an entirely urban polity. It does not face the complications faced by a country with a countryside, a rural population, and land-based and resource-based industries. This is not to mention the other aspects of Singapore that simply can't be replicated in other countries.
It's been taken down now, but I remember reading Adrian D'Souza's post "Fat Shaming Kids In Singapore" and thinking...my God, at what cost are they getting these results? Profoundly different society that violates so many of our norms around individualism and use of force. You can still learn something from reducto ad absurdums, as a sort of proof of concept, but...if any given lessons can't actually be practically implemented at scale, there's limited probative value to be had. And it's not like GLP-1s were invented there anyway, there's often more than one way to skin a cat.
Sure - I don't have zero inclinations towards paternalism or strong state intervention, sometimes the juice is worth such risky squeezes. Like from our mutual other blog follow, taking capital-C Crime and also public order really seriously is, in fact, a huge deal! Don't need to take it all the way to caning people for chewing gum or whatever, to be clear, but...like our current host argues, with some cases like the severely mentally ill, we're clearly making wrong and dumb choices that leave everyone worse off without even protecting the motivating sacred principles. That's a really wide action space to find *some* sort of happy American medium. Cede the commons -> degrade trust in governance/outsource to unaccountable NGO-industrial complex -> further reduce ability to maintain commons -> rinse and repeat. We still have time to get off that track before derailment.
(The pachyderm on the premises of "demographic differences" is of course hard to discuss candidly. One wants to avoid pulling a David A rhetorically there, or such proposals are DOA in the places that most need them.)
It actually bothers me that I can't come up off the top of my head with a non-speciesist equivalent phrase? Apologies for the fighting words. No felines were harmed in the making of this comment.
...The trouble I have with the "it's the phones" argument is that it presumes the American *social* adoption of smartphones—more specifically, the widespread gifting of smartphones to kids—was somehow universal, and applies to western Europe. While you can definitely make the *numerical* case for adoption, that's merely a correlation and does not suggest causation that western Europe adopted smartphones in the same way Americans did. Consider that, since the iPhone was announced, Apple/iOS has been the primary (if not majority) leader of smartphone OS market share in the US. But in practically every other country (except I *think* Japan?), Google/Android has been the leader by a country mile. That on its own suggests a different cultural dynamic in play, to say nothing of the fact that Euros used different chat programs than Americans (Skype was very popular in Europe in the early 2010s while being relatively niche in the States, as was/is Telegram).
I am highly skeptical to the idea that Euro parents and Euro schools reacted to the smartphone in the exact same or similar way Americans did. And if they reacted differently, and thus the devices' influence is different, then it's plausible that smartphones could not on its own feed into educational declines on a broader level, which weakens the American case.
Further, the broader declines between 2018 and 2022 we have another factor involved: COVID. We know through research that even limited lockdowns greatly impacted educational performance. So to not mention that (which many in the accountability camp, most of whom you could say were pro-lockdown, would desperately ignore) seems questionable.
A. The initial chart you showed before you edited was Europe, which proved my country mile thesis correct on the continental level (and using the UK is dithering, since you'll get a different answer on whether the UK is European depending on location in the UK, type of people you talk to, what they voted for in the 2016 referendum, what party they voted for in recent elections, their feelings on the Royal Family, and most importantly the time of day).
B. The cost and app factors simply reinforces the point I made that about the social adoption of smartphones. Americans adopted smartphones in a very distinct way that we cannot say with certainty could be reproduced anywhere else except maybe Canada. In the States, the iPhone *isn't* a luxury item, but an item that possesses social symbolism and status, for example.
When you focus entirely on the device and its abilities without considering how it is used in the social context, you make a lot of massive assumptions when you proclaim that it is a universal problem. Freddie's theory is based on how the smartphone was adopted and used in the *United States,* which becomes assumption when applied elsewhere. The only thing that can be proven is that social media (which is a different thing) has had widespread adoption and impact on children, given the recent drive to ban access in many countries. Smartphones are a different matter, and Freddie is resting too much weight on it.
Hell, part of me thinks the American case is not strong because of that. While smartphone use certainly plays a role in the decline, I suspect it's more accelerant than cause.
The UK might not be part of the European Union, but it is definitely considered to be part of Europe (just curious as to where else you would classify it). I think the point is that the introduction of the smart phones has distracted and sucked up student attention, and it does not really matter if it’s an iPhone or an Android, or if people are using WhatsApp vs Telegram, or what social media platform is preferred, the point is people have a distraction device they are carrying around all day long. I can’t think of a significant way in which Europeans use smart phones differently, and I would say that the adoption of widespread smart phone use in Western Europe was only about 2-4 years behind America. As for COVID, there was a steep decline, but the declines in general are prior to COVID.
"I can’t think of a significant way in which Europeans use smart phones differently, and I would say that the adoption of widespread smart phone use in Western Europe was only about 2-4 years behind America."
What matters is why the children were affected. And I can see how Europeans could have handled that differently: Parenting is a lot different—culturally, socially—in Europe than it is in America. Even if they adopted smartphones at similar rates, that doesn't mean that parents acted the same way, i.e. many many American parents treating smartphones as a means to render the kid docile when they need to travel/are stressed out/need to cook dinner/etc. In fact, I wouldn't be surprised if Euro parents *didn't* go that route.
Now that a lot of Europe is going anti-phones in schools, we'll soon have a lot of data. Tentative results seem to suggest pupils are a lot better behaved in a phone-free classroom.
This all makes sense to anyone who’s talked at all to teachers who have been in the classroom over the last ten or fifteen years. No, the schools didn’t somehow do away with “accountability.” Teachers have always been expected to do the impossible by making sure every child is above average and the bottom quarter of all-but-unteachable kids are nevertheless regarded as future potential college graduates. The challenge has always been to teach kids who don’t want to learn or who don’t come from a home environment that supports learning. That’s it. That’s the whole game. More tests / assessments / curricula / benchmarks were never going to change that.
I always feel like there's a distinct lack of defining-terms which makes School Accountability(tm) debates a lot of talking past each other, similar to how there's still extensive rules-lawyering around "we didn't *really* close schools during covid" and such. It's a complex topic with lots of moving parts, often in contradiction with each other, and you can paint most any picture you want just by choice of which to emphasize (and over what timeframe).
But it's still fun to read both sides anyways. School being a long time ago and not having any kids or relatives still in those trenches means I don't really learn what goes on in modern education except via takes like these, or sometimes anecdata from coworkers still plodding towards matriculation. (It sounds a lot easier than I remember college being, but then, I wasn't trying to juggle a job too!)
Education is one area where theory triumphs over evidence. It has always been this way. There is all manner of theory about how to teach and every year there is a new "latest thing" that promises to give better results. Teachers are taken out of the classrooms multiple days each year for "professional development" (I am not making this up) to learn the latest thing, but where are the results?
I don't think the best teachers follow the theoretical B.s. It would seem to me studying the most successful school districts would give us a clue to what works. Compare what they do to the ones who fail. Send teachers and administrators from the worst performing schools to the most successful schools.
My wife is a teacher. Nobody is giving her "weeks" off every year for professional development. Admittedly, she gets 4 days a year where she's forced to attend meetings where all the teachers roll their eyes at admin's latest bullshit....but the idea that teachers are getting weeks off...I've never seen it. And she's taught in two deep blue states
ABSTRACT: I use mapping and age trajectories of advanced cognitive skills to better understand why these skills are more prevalent in some local areas than in others. The study begins by explaining what advanced cognitive skills are. It offers a nonspecialist’s review of recent brain science that indicates that adolescence is the key period for the development of advanced cognitive skills. The paper considers three main explanations for why the prevalence of advanced cognitive skills varies substantially across US counties. Is it early childhood factors which could generate endogenous responses that are important later when advanced cognitive skills are developing? Is it factors whose influence is greatest during adolescence—the period when brain science argues that experience would most directly affect advanced cognitive skills? If so, adolescence is indeed the age of opportunity but also risk. Is the variation among counties explained by migration of individuals toward areas where other people have advanced cognitive skills similar to their own? Evidence based on cognitive skill trajectories, maps at different ages, and longitudinal regressions suggests that all three of these explanations play a role in generating areas where advanced cognitive skills are prevalent and areas where they are not—advanced cognitive skill deserts.
[Concerning the disadvantage index and 2020 Presidential election voting patterns, the strongest correlation (for all disadvantage study variables) was 0.51, between percent of county population with a bachelor's degree or higher and percentage who voted for Biden. (The correlation with percentage who voted for Trump was -0.52.) The correlation between
standardized value social mobility measure from Chetty and Hendren (mean household income rank for children whose parents were at the 25th percentile of the national income distribution) was -0.45 for percentage who voted for Biden and 0.43 for percentage who voted for Trump. (The differences in D/R correlations occur because in some counties, %D + %R < 100% -the result of a small percentage of votes being cast for candidates of other political parties.)]
*Kathryn J. Edin et al, Injustice of Place: Uncovering the Legacy of Poverty in America (Mariner Books, 2023).
Excellent analysis as always, Freddie. Quick question about this line: "The PISA declines visible in American math and reading scores over the 2003–2022 period aren’t remotely anomalous..." Don't the charts preceding this line show an increased reading score from 2003-2022? Not sure if that line was a typo/mistake, or if there's something I'm missing.
My read is that, similar to what you've said here and elsewhere, learning outcomes are as volatile and variable as people themselves (we are, after all, assessing human beings), and likely a consequence of larger forces related to social environments and technological changes. It's obviously worth investigating trends and trying to identify what helps and what harms -- and as a teacher there are clearly better pedagogical practices than others -- but the fact that the US mirrored global trends on some measures and (at least temporarily) bucked them on others feels like stronger evidence that we can't neatly draw broad conclusions about what causes these fluctuations.
As a career teacher two years into retirement it’s numbing and dreary to be confronted yet again with how distant public discourse about education is from the realities of teaching and schooling. Big Data ed stats, even when filtered, massaged, crafted and re-packaged for local-level consumption, are broad-brush indicators of the sort that entrance wonks, pols and armchair QBs but are best seen as trailing evidence; that they’re so prone to manipulation as to be virtually useless is part of the point here. The demand for results on that front also walks in discordant harmony with our political churn, as noted. There have been some attempts to make it mean something that were given play, but the froth of Big Data actually tends to obscure that, at least for purposes of facile commentary.
For several years California required students to pass a high school exit exam to qualify for graduation. It was not a particularly demanding test and most passed it in 10th grade, yet it went down in flames on the basis of a court ruling in the early 20teens. In terms of establishing a baseline for student performance it had a significant positive effect and was difficult to game. Cali was well ahead of the game in addressing cultural bias and, though more softball than high-stress for many, it was a major student attention-getter. I don’t follow ed research closely but the CAHSEE deserves a close look as a historical test case, complete with substantial before/during/ after data sets.
High school exit exams have become a popular policy tool in the current movement in U.S.
public schooling toward more explicit standards of instruction and accountability. In this paper, we investigate the effects of a high school exit exam requirement on students’ achievement, persistence in high school, and graduation rates, using data from three cohorts of students—one of whom was not subject to the exit exam requirement and two of which were—from four large school districts in California. We find that the exit exam requirement had no positive effect on student achievement, small negative or zero effects on students’ persistence in high school, and large negative effects on graduation rates."
I appreciate that you bothered to respond to my comment - thanks. I am uncertain what you presume about my beliefs regarding ed. research or assessment, but you do seem to be making assumptions on thin grounds. In any case, thanks for passing along the paper. I read the first 22+ pages. FWIW, I did originally assume that there was a good chance someone had taken a strong look at the impact of the CAHSEE (given the fact that Cali is a huge state with a large public higher ed. domain and grants numerous EdDs every year) and the study does fit the bill in that regard.
Though I am not well versed in statistics and unqualified to critique the finer methodological points, there are some evident weaknesses, and some of the conclusions based on the data are suspect on the authors’ own terms. Also, having been produced several years before the CAHSEE was discontinued, it does not, in fact, attempt to weigh the utility and efficacy of the test in any truly long-term context in the way that I described. That would be difficult in any event, and, as noted in the paper, almost impossible for the math element, since students’ math class enrollment diverges widely early in high school. But, as noted, the state has administered a separate assessment (the CST) and continues to do so, therefore it is clearly conceivable that data from the CST’s ELA component could be comparable across the entire time span, from before, during, and after the CAHSEE era, and brought into comparison. It would be interesting to see the results, though whether or not they would lead to solid, actionable conclusions is uncertain.
What is not uncertain, though, is the potentially disruptive effect of external factors. To their credit, the authors, apparently writing in 2008-9, made note of the possible effect of the labor market on high school enrollment. The kind of kid - particularly among boys - who struggled with the CAHSEE would’ve been exactly the sort to take an available construction job during the early years of CAHSEE implementation. The comprehensive demolition of mortgage lending standards accomplished in the late 90s by the repeal of the Glass-Stegal Act helped lay the groundwork for the real estate bubble that blew up in early 2000s, and the boom produced a substantial amount of low level construction employment among the very cohort the authors focus their conclusions on.
Ultimately the study is unpersuasive for another reason. Did the test affect graduation rates? Yes - and that was, in a very real sense, part of the point. Given the numbers cited the effect was notable but far from devastating. What the authors failed to do was establish the value of a high school diploma that does not signify learning above the 8th grade level, and therefore any utility or purpose it might have beyond that of a GED for students who would rather work than stay enrolled in high school, serving their time to collect a worthless piece of paper.
The test also had an actionable assessment aspect. Students who didn’t pass were enrolled in remedial classes. Given the low bar for passing that was appropriate, and all were aware of the stakes. This formalized tracking that would’ve occurred anyway and established tangible stakes difficult even for distracted jocks to ignore. I remember it all well because I taught the classes.
True - I only taught an elective (art) for my last 8 years and paid little attention to the testing. Given the CAASP testing regime direct comparison is stickier. Having looked at the score reporting website, though, it looks like ELA results approached 60% proficiency for 11th graders but dipped a bit post-Covid and may now be rebounding.
Old data about the CAHSEE in the Wikipedia article about the test showed a significant rise in pass rates over the years it was administered. That squares with my experience teaching ELA for a couple years (2005-2007) after it was fully implemented. It did force students to pay attention to their performance. Test administration also prohibited use of phones; no one had a smartphone yet, but the possibility of texting test Qs/As during a bathroom break was headed-off, and the writing section was by hand in a ‘bluebook.’ I didn’t have special affection for it all but it did seem to establish a legit baseline of expectations.
The CAASP system does incorporate an improved accommodation for EL students, which was, I think, the only real weakness of the CAHSEE. Given the evolving demographics that was needed. But I think it was otherwise a legitimate requirement and shouldn’t have been entirely scrapped. By the time it was ended pass rates were strong and had improved a lot in both math and ELA. Forcing students - particularly on the lower end of the achievement spectrum - to pay attention to their preparation and performance on a task with meaningful stakes did produce positive motivation.
I think the perspective of most of those you reference like Yglesias is shaped by where they live. Like me they live in deep blue urban environments where we have seen the attacks on standardized testing, on academic tracking and on graduation requirements materialize and create a mess. Happy to provide specific details (from Boston) but suffice it to say that my neighborhood was eliminated from entry to an exam school here because of demographic characteristics (akin to blocking well off suburban neighborhoods from entry to AP classes at their public high school).
If you cannot measure performance well enough to manage it, cannot attribute causation well enough to reward or punish staff, and cannot claim schools have enough leverage to move outcomes much, then your framework is education nihilism dressed up in humane rhetoric.
We know Freddie rejects relative metrics. But what about absolute gains? Setting aside children with severe disabilities or profound cognitive impairment, why is it unreasonable to expect schools to ensure that all children reach basic reading and math proficiency? If they fail to do so, why should that be blamed on genes or bad parents rather than on the school?
And if it is in fact unreasonable to expect schools to teach basic literacy and math, then why have public schools at all? Why not give the money directly to parents for homeschooling, daycare, or private school?
So I keep coming back to the same questions.
Do bad teachers exist? If so, what makes them bad?
Can bad teachers be identified?
Can bad teachers be turned into good teachers? If so, how?
Is it sometimes infeasible to turn bad teachers into good teachers? If so, should they be fired? If not, then what should happen to them?
Do bad teachers have a significant effect on children’s absolute learning gains? Note that I said absolute, not relative.
Freddie’s position seems to boil down to this: we cannot really know whether teachers are bad, and even if they are, it does not matter much because teachers have little influence over whether children actually learn. But if that is true, then why keep them? If teachers cannot be expected to teach children to read, write, and do math, what exactly are they for?
" why is it unreasonable to expect schools to ensure that all children reach basic reading and math proficiency?"
Because "proficient" in the tests used doesn't mean "can read and add" but "above grade level". In fact, schools do in fact get almost all children to reading and mathing and the ones that don't reach it have severe cognitive limits.
And all your questions that you think are so profound have, in fact, been answered. Very few teachers are actually bad, and no, they have no real input on student outcomes.
"If teachers cannot be expected to teach children to read, write, and do math, what exactly are they for?"
You don't understand the research. Teachers do, in fact, teach children to read and write and do math. What they can't do is *improve* children's ability to some desired level, because that's a factor of cognitive ability.
i am really confused. In the same sentence you said “most teachers aren’t bad” and teachers have no control over student outcomes. so what exactly is a bad teacher and if a teacher is good how are they good if they can’t affect student outcomes.
And I never said my questions are profound. They are however earnest and I’ve yet to hear any good answers if any from anti education reformers. either you folks tie yourself into knots avoiding the obvious conclusion from your premises is that teachers are mostly useless or evade questions of accountability if in fact teachers matter a lot.
"that all children reach basic reading and math proficiency"
As far as I know, the way the accountability movement defines and measures proficiency is not relative to some absolute standard, but against standards specific to age groups based on something like the average performance of said age group.
Put another way, if a new pupil comes into 3rd grade with 1st-grade reading skills, and by 6th grade can read at 4th grade level, in absolute terms they're advancing just fine but they're utterly failing the end-of-6th-grade benchmark.
I have four children, the oldest graduated in 2017, the next in 2019, then 2023 and my youngest is graduating this spring(2026), all from the same school district. There was a marked difference in their education. COVID was absolutely part of the problem but the total reliance on smartphones and chromebooks, first because everything was online and then even when in person school started again was and continues to be the biggest problem. It has made me a firm believer that we have to go back to books, pen and paper for at least K-8 and get smartphones out of school.
Is there any unequivocal evidence that educational successes which are reported here and there are scalable? I have always thought that holding teachers responsible for student achievement would invariably open the educational market up to more manipulation and there is evidence enough that has happened. Or, put another way, do schools truly educate the masses or do they more correctly provide a sort of template that some are able to take advantage of and some are not? Part of being able to do so depends on students' intellectual capacity and motivation which is not stressed in this article.
I like what you say, but wonder if it is tautological. When comparing schools and educational methods, what are we really comparing? Are the data broken out by cognitive ability in the studies we read, or is most of this obscured by "averages?"
I was definitely being hyperbolic - but it's certainly a profession in which you get a lot of non-teachers very confidently making judgemental claims about the job.
Fully agree. Where schools have failed worse, there are obvious issues upstream causing other problems too. Schools and teacher quality did not just randomly degrade.
FWIW, I vibecoded a little survey tool at votto.app to measure sentiment on takes like this across different traits and media community affiliations because I'm genuinely fascinated by how different groups land on these debates and what other opinions are correlated with them. EG, respond to "Schools are being blamed for educational decline that has much more to do with smartphones and broader social conditions than classroom policy" and filter to see how other groups answer. Dedicated page for Freddie readers here: votto.app/freddiedeboer
Explore if for no other reason than liking cool graphs... other communities have more robust data, but would love to start accumulating data for this crowd which will inevitably split in fascinating ways...
I continue to be bemused that Singapore is regarded as a model for other countries to attempt to emulate. The entire area of Singapore is only slightly more than half of the Brisbane City Local Government Area, which is itself only a fraction of the area of metropolitan Brisbane. As well as geographic compactness, Singapore has the kind of political economy and political ecology that is only possible in an entirely urban polity. It does not face the complications faced by a country with a countryside, a rural population, and land-based and resource-based industries. This is not to mention the other aspects of Singapore that simply can't be replicated in other countries.
It's been taken down now, but I remember reading Adrian D'Souza's post "Fat Shaming Kids In Singapore" and thinking...my God, at what cost are they getting these results? Profoundly different society that violates so many of our norms around individualism and use of force. You can still learn something from reducto ad absurdums, as a sort of proof of concept, but...if any given lessons can't actually be practically implemented at scale, there's limited probative value to be had. And it's not like GLP-1s were invented there anyway, there's often more than one way to skin a cat.
Idk, we’d never get to where they are, but I feel like Singapore is a great model for basically everything.
Sure - I don't have zero inclinations towards paternalism or strong state intervention, sometimes the juice is worth such risky squeezes. Like from our mutual other blog follow, taking capital-C Crime and also public order really seriously is, in fact, a huge deal! Don't need to take it all the way to caning people for chewing gum or whatever, to be clear, but...like our current host argues, with some cases like the severely mentally ill, we're clearly making wrong and dumb choices that leave everyone worse off without even protecting the motivating sacred principles. That's a really wide action space to find *some* sort of happy American medium. Cede the commons -> degrade trust in governance/outsource to unaccountable NGO-industrial complex -> further reduce ability to maintain commons -> rinse and repeat. We still have time to get off that track before derailment.
(The pachyderm on the premises of "demographic differences" is of course hard to discuss candidly. One wants to avoid pulling a David A rhetorically there, or such proposals are DOA in the places that most need them.)
most of what people like in Singapore is also achieved in Switzerland, without any need for authoritarianism. worth considering!
Can we replicate by changing our country name to start with S?
Has anyone measured for negative changes when Swaziland became Eswatini?
Now I'm interested in the Swedish data.
EXCUSE ME!
At this rate I'll be reduced to wearing Nutria.
It actually bothers me that I can't come up off the top of my head with a non-speciesist equivalent phrase? Apologies for the fighting words. No felines were harmed in the making of this comment.
...The trouble I have with the "it's the phones" argument is that it presumes the American *social* adoption of smartphones—more specifically, the widespread gifting of smartphones to kids—was somehow universal, and applies to western Europe. While you can definitely make the *numerical* case for adoption, that's merely a correlation and does not suggest causation that western Europe adopted smartphones in the same way Americans did. Consider that, since the iPhone was announced, Apple/iOS has been the primary (if not majority) leader of smartphone OS market share in the US. But in practically every other country (except I *think* Japan?), Google/Android has been the leader by a country mile. That on its own suggests a different cultural dynamic in play, to say nothing of the fact that Euros used different chat programs than Americans (Skype was very popular in Europe in the early 2010s while being relatively niche in the States, as was/is Telegram).
I am highly skeptical to the idea that Euro parents and Euro schools reacted to the smartphone in the exact same or similar way Americans did. And if they reacted differently, and thus the devices' influence is different, then it's plausible that smartphones could not on its own feed into educational declines on a broader level, which weakens the American case.
Further, the broader declines between 2018 and 2022 we have another factor involved: COVID. We know through research that even limited lockdowns greatly impacted educational performance. So to not mention that (which many in the accountability camp, most of whom you could say were pro-lockdown, would desperately ignore) seems questionable.
Agree on COVID, but the apps are the same whether on iPhone or Android. Your "country mile" assertion is incorrect. https://gs.statcounter.com/os-market-share/mobile/united-kingdom. iPhone is a luxury phone. Cost drives the vast majority in every country.
A. The initial chart you showed before you edited was Europe, which proved my country mile thesis correct on the continental level (and using the UK is dithering, since you'll get a different answer on whether the UK is European depending on location in the UK, type of people you talk to, what they voted for in the 2016 referendum, what party they voted for in recent elections, their feelings on the Royal Family, and most importantly the time of day).
B. The cost and app factors simply reinforces the point I made that about the social adoption of smartphones. Americans adopted smartphones in a very distinct way that we cannot say with certainty could be reproduced anywhere else except maybe Canada. In the States, the iPhone *isn't* a luxury item, but an item that possesses social symbolism and status, for example.
When you focus entirely on the device and its abilities without considering how it is used in the social context, you make a lot of massive assumptions when you proclaim that it is a universal problem. Freddie's theory is based on how the smartphone was adopted and used in the *United States,* which becomes assumption when applied elsewhere. The only thing that can be proven is that social media (which is a different thing) has had widespread adoption and impact on children, given the recent drive to ban access in many countries. Smartphones are a different matter, and Freddie is resting too much weight on it.
Hell, part of me thinks the American case is not strong because of that. While smartphone use certainly plays a role in the decline, I suspect it's more accelerant than cause.
The UK might not be part of the European Union, but it is definitely considered to be part of Europe (just curious as to where else you would classify it). I think the point is that the introduction of the smart phones has distracted and sucked up student attention, and it does not really matter if it’s an iPhone or an Android, or if people are using WhatsApp vs Telegram, or what social media platform is preferred, the point is people have a distraction device they are carrying around all day long. I can’t think of a significant way in which Europeans use smart phones differently, and I would say that the adoption of widespread smart phone use in Western Europe was only about 2-4 years behind America. As for COVID, there was a steep decline, but the declines in general are prior to COVID.
"I can’t think of a significant way in which Europeans use smart phones differently, and I would say that the adoption of widespread smart phone use in Western Europe was only about 2-4 years behind America."
What matters is why the children were affected. And I can see how Europeans could have handled that differently: Parenting is a lot different—culturally, socially—in Europe than it is in America. Even if they adopted smartphones at similar rates, that doesn't mean that parents acted the same way, i.e. many many American parents treating smartphones as a means to render the kid docile when they need to travel/are stressed out/need to cook dinner/etc. In fact, I wouldn't be surprised if Euro parents *didn't* go that route.
Now that a lot of Europe is going anti-phones in schools, we'll soon have a lot of data. Tentative results seem to suggest pupils are a lot better behaved in a phone-free classroom.
This all makes sense to anyone who’s talked at all to teachers who have been in the classroom over the last ten or fifteen years. No, the schools didn’t somehow do away with “accountability.” Teachers have always been expected to do the impossible by making sure every child is above average and the bottom quarter of all-but-unteachable kids are nevertheless regarded as future potential college graduates. The challenge has always been to teach kids who don’t want to learn or who don’t come from a home environment that supports learning. That’s it. That’s the whole game. More tests / assessments / curricula / benchmarks were never going to change that.
I always feel like there's a distinct lack of defining-terms which makes School Accountability(tm) debates a lot of talking past each other, similar to how there's still extensive rules-lawyering around "we didn't *really* close schools during covid" and such. It's a complex topic with lots of moving parts, often in contradiction with each other, and you can paint most any picture you want just by choice of which to emphasize (and over what timeframe).
But it's still fun to read both sides anyways. School being a long time ago and not having any kids or relatives still in those trenches means I don't really learn what goes on in modern education except via takes like these, or sometimes anecdata from coworkers still plodding towards matriculation. (It sounds a lot easier than I remember college being, but then, I wasn't trying to juggle a job too!)
Education is one area where theory triumphs over evidence. It has always been this way. There is all manner of theory about how to teach and every year there is a new "latest thing" that promises to give better results. Teachers are taken out of the classrooms multiple days each year for "professional development" (I am not making this up) to learn the latest thing, but where are the results?
I don't think the best teachers follow the theoretical B.s. It would seem to me studying the most successful school districts would give us a clue to what works. Compare what they do to the ones who fail. Send teachers and administrators from the worst performing schools to the most successful schools.
My wife is a teacher. Nobody is giving her "weeks" off every year for professional development. Admittedly, she gets 4 days a year where she's forced to attend meetings where all the teachers roll their eyes at admin's latest bullshit....but the idea that teachers are getting weeks off...I've never seen it. And she's taught in two deep blue states
Have you seen?:
Caroline M. Hoxby, “Advanced Cognitive Skill Deserts in the United States: Their Likely Causes and Implications,” Brookings Papers on Economic Activity, Spring 2021, pp. 317-351 DOI: https://doi.org/10.1353/eca.2021.0006https://www.brookings.edu/articles/advanced-cognitive-skill-deserts-in-the-us-their-likely-causes-and-implications/
ABSTRACT: I use mapping and age trajectories of advanced cognitive skills to better understand why these skills are more prevalent in some local areas than in others. The study begins by explaining what advanced cognitive skills are. It offers a nonspecialist’s review of recent brain science that indicates that adolescence is the key period for the development of advanced cognitive skills. The paper considers three main explanations for why the prevalence of advanced cognitive skills varies substantially across US counties. Is it early childhood factors which could generate endogenous responses that are important later when advanced cognitive skills are developing? Is it factors whose influence is greatest during adolescence—the period when brain science argues that experience would most directly affect advanced cognitive skills? If so, adolescence is indeed the age of opportunity but also risk. Is the variation among counties explained by migration of individuals toward areas where other people have advanced cognitive skills similar to their own? Evidence based on cognitive skill trajectories, maps at different ages, and longitudinal regressions suggests that all three of these explanations play a role in generating areas where advanced cognitive skills are prevalent and areas where they are not—advanced cognitive skill deserts.
(Data are also available via the URL above and in the Stanford Education Data Archive (SEDA) within https://edopportunity.org/ at https://edopportunity.org/opportunity/data/downloads/.)
I’d like to see if there are any noteworthy correlations between
•Hoxby’s data,
•voting patterns,
•Kathryn Edin et al disadvantage index* https://poverty.umich.edu/research-funding-opportunities/data-tools/understanding-communities-of-deep-disadvantage/ and
•Woodard’s Nations classifications https://www.nationhoodlab.org/.
[Concerning the disadvantage index and 2020 Presidential election voting patterns, the strongest correlation (for all disadvantage study variables) was 0.51, between percent of county population with a bachelor's degree or higher and percentage who voted for Biden. (The correlation with percentage who voted for Trump was -0.52.) The correlation between
standardized value social mobility measure from Chetty and Hendren (mean household income rank for children whose parents were at the 25th percentile of the national income distribution) was -0.45 for percentage who voted for Biden and 0.43 for percentage who voted for Trump. (The differences in D/R correlations occur because in some counties, %D + %R < 100% -the result of a small percentage of votes being cast for candidates of other political parties.)]
*Kathryn J. Edin et al, Injustice of Place: Uncovering the Legacy of Poverty in America (Mariner Books, 2023).
Excellent analysis as always, Freddie. Quick question about this line: "The PISA declines visible in American math and reading scores over the 2003–2022 period aren’t remotely anomalous..." Don't the charts preceding this line show an increased reading score from 2003-2022? Not sure if that line was a typo/mistake, or if there's something I'm missing.
My read is that, similar to what you've said here and elsewhere, learning outcomes are as volatile and variable as people themselves (we are, after all, assessing human beings), and likely a consequence of larger forces related to social environments and technological changes. It's obviously worth investigating trends and trying to identify what helps and what harms -- and as a teacher there are clearly better pedagogical practices than others -- but the fact that the US mirrored global trends on some measures and (at least temporarily) bucked them on others feels like stronger evidence that we can't neatly draw broad conclusions about what causes these fluctuations.
Yes, that's a fair point. There is an awful lot of controversy about PISA's reading test, but technically we have improved over time.
As a career teacher two years into retirement it’s numbing and dreary to be confronted yet again with how distant public discourse about education is from the realities of teaching and schooling. Big Data ed stats, even when filtered, massaged, crafted and re-packaged for local-level consumption, are broad-brush indicators of the sort that entrance wonks, pols and armchair QBs but are best seen as trailing evidence; that they’re so prone to manipulation as to be virtually useless is part of the point here. The demand for results on that front also walks in discordant harmony with our political churn, as noted. There have been some attempts to make it mean something that were given play, but the froth of Big Data actually tends to obscure that, at least for purposes of facile commentary.
For several years California required students to pass a high school exit exam to qualify for graduation. It was not a particularly demanding test and most passed it in 10th grade, yet it went down in flames on the basis of a court ruling in the early 20teens. In terms of establishing a baseline for student performance it had a significant positive effect and was difficult to game. Cali was well ahead of the game in addressing cultural bias and, though more softball than high-stress for many, it was a major student attention-getter. I don’t follow ed research closely but the CAHSEE deserves a close look as a historical test case, complete with substantial before/during/ after data sets.
There's was lots of research on the California Exit Exam, and it showed exactly the opposite of your beliefs here:
https://search.issuelab.org/resources/9116/9116.pdf
High school exit exams have become a popular policy tool in the current movement in U.S.
public schooling toward more explicit standards of instruction and accountability. In this paper, we investigate the effects of a high school exit exam requirement on students’ achievement, persistence in high school, and graduation rates, using data from three cohorts of students—one of whom was not subject to the exit exam requirement and two of which were—from four large school districts in California. We find that the exit exam requirement had no positive effect on student achievement, small negative or zero effects on students’ persistence in high school, and large negative effects on graduation rates."
I appreciate that you bothered to respond to my comment - thanks. I am uncertain what you presume about my beliefs regarding ed. research or assessment, but you do seem to be making assumptions on thin grounds. In any case, thanks for passing along the paper. I read the first 22+ pages. FWIW, I did originally assume that there was a good chance someone had taken a strong look at the impact of the CAHSEE (given the fact that Cali is a huge state with a large public higher ed. domain and grants numerous EdDs every year) and the study does fit the bill in that regard.
Though I am not well versed in statistics and unqualified to critique the finer methodological points, there are some evident weaknesses, and some of the conclusions based on the data are suspect on the authors’ own terms. Also, having been produced several years before the CAHSEE was discontinued, it does not, in fact, attempt to weigh the utility and efficacy of the test in any truly long-term context in the way that I described. That would be difficult in any event, and, as noted in the paper, almost impossible for the math element, since students’ math class enrollment diverges widely early in high school. But, as noted, the state has administered a separate assessment (the CST) and continues to do so, therefore it is clearly conceivable that data from the CST’s ELA component could be comparable across the entire time span, from before, during, and after the CAHSEE era, and brought into comparison. It would be interesting to see the results, though whether or not they would lead to solid, actionable conclusions is uncertain.
What is not uncertain, though, is the potentially disruptive effect of external factors. To their credit, the authors, apparently writing in 2008-9, made note of the possible effect of the labor market on high school enrollment. The kind of kid - particularly among boys - who struggled with the CAHSEE would’ve been exactly the sort to take an available construction job during the early years of CAHSEE implementation. The comprehensive demolition of mortgage lending standards accomplished in the late 90s by the repeal of the Glass-Stegal Act helped lay the groundwork for the real estate bubble that blew up in early 2000s, and the boom produced a substantial amount of low level construction employment among the very cohort the authors focus their conclusions on.
Ultimately the study is unpersuasive for another reason. Did the test affect graduation rates? Yes - and that was, in a very real sense, part of the point. Given the numbers cited the effect was notable but far from devastating. What the authors failed to do was establish the value of a high school diploma that does not signify learning above the 8th grade level, and therefore any utility or purpose it might have beyond that of a GED for students who would rather work than stay enrolled in high school, serving their time to collect a worthless piece of paper.
The test also had an actionable assessment aspect. Students who didn’t pass were enrolled in remedial classes. Given the low bar for passing that was appropriate, and all were aware of the stakes. This formalized tracking that would’ve occurred anyway and established tangible stakes difficult even for distracted jocks to ignore. I remember it all well because I taught the classes.
The state hasn't done the c.S.T's since the CAHSEE was discontinued. Only 11th grade test.
True - I only taught an elective (art) for my last 8 years and paid little attention to the testing. Given the CAASP testing regime direct comparison is stickier. Having looked at the score reporting website, though, it looks like ELA results approached 60% proficiency for 11th graders but dipped a bit post-Covid and may now be rebounding.
Old data about the CAHSEE in the Wikipedia article about the test showed a significant rise in pass rates over the years it was administered. That squares with my experience teaching ELA for a couple years (2005-2007) after it was fully implemented. It did force students to pay attention to their performance. Test administration also prohibited use of phones; no one had a smartphone yet, but the possibility of texting test Qs/As during a bathroom break was headed-off, and the writing section was by hand in a ‘bluebook.’ I didn’t have special affection for it all but it did seem to establish a legit baseline of expectations.
The CAASP system does incorporate an improved accommodation for EL students, which was, I think, the only real weakness of the CAHSEE. Given the evolving demographics that was needed. But I think it was otherwise a legitimate requirement and shouldn’t have been entirely scrapped. By the time it was ended pass rates were strong and had improved a lot in both math and ELA. Forcing students - particularly on the lower end of the achievement spectrum - to pay attention to their preparation and performance on a task with meaningful stakes did produce positive motivation.
I think the perspective of most of those you reference like Yglesias is shaped by where they live. Like me they live in deep blue urban environments where we have seen the attacks on standardized testing, on academic tracking and on graduation requirements materialize and create a mess. Happy to provide specific details (from Boston) but suffice it to say that my neighborhood was eliminated from entry to an exam school here because of demographic characteristics (akin to blocking well off suburban neighborhoods from entry to AP classes at their public high school).
Would Yglesias et al really disagree with "phones are heavily to blame", though?
A lot of the real reason for "accountability" was also just straight up union busting.
If you cannot measure performance well enough to manage it, cannot attribute causation well enough to reward or punish staff, and cannot claim schools have enough leverage to move outcomes much, then your framework is education nihilism dressed up in humane rhetoric.
We know Freddie rejects relative metrics. But what about absolute gains? Setting aside children with severe disabilities or profound cognitive impairment, why is it unreasonable to expect schools to ensure that all children reach basic reading and math proficiency? If they fail to do so, why should that be blamed on genes or bad parents rather than on the school?
And if it is in fact unreasonable to expect schools to teach basic literacy and math, then why have public schools at all? Why not give the money directly to parents for homeschooling, daycare, or private school?
So I keep coming back to the same questions.
Do bad teachers exist? If so, what makes them bad?
Can bad teachers be identified?
Can bad teachers be turned into good teachers? If so, how?
Is it sometimes infeasible to turn bad teachers into good teachers? If so, should they be fired? If not, then what should happen to them?
Do bad teachers have a significant effect on children’s absolute learning gains? Note that I said absolute, not relative.
Freddie’s position seems to boil down to this: we cannot really know whether teachers are bad, and even if they are, it does not matter much because teachers have little influence over whether children actually learn. But if that is true, then why keep them? If teachers cannot be expected to teach children to read, write, and do math, what exactly are they for?
" why is it unreasonable to expect schools to ensure that all children reach basic reading and math proficiency?"
Because "proficient" in the tests used doesn't mean "can read and add" but "above grade level". In fact, schools do in fact get almost all children to reading and mathing and the ones that don't reach it have severe cognitive limits.
And all your questions that you think are so profound have, in fact, been answered. Very few teachers are actually bad, and no, they have no real input on student outcomes.
"If teachers cannot be expected to teach children to read, write, and do math, what exactly are they for?"
You don't understand the research. Teachers do, in fact, teach children to read and write and do math. What they can't do is *improve* children's ability to some desired level, because that's a factor of cognitive ability.
i am really confused. In the same sentence you said “most teachers aren’t bad” and teachers have no control over student outcomes. so what exactly is a bad teacher and if a teacher is good how are they good if they can’t affect student outcomes.
And I never said my questions are profound. They are however earnest and I’ve yet to hear any good answers if any from anti education reformers. either you folks tie yourself into knots avoiding the obvious conclusion from your premises is that teachers are mostly useless or evade questions of accountability if in fact teachers matter a lot.
"that all children reach basic reading and math proficiency"
As far as I know, the way the accountability movement defines and measures proficiency is not relative to some absolute standard, but against standards specific to age groups based on something like the average performance of said age group.
Put another way, if a new pupil comes into 3rd grade with 1st-grade reading skills, and by 6th grade can read at 4th grade level, in absolute terms they're advancing just fine but they're utterly failing the end-of-6th-grade benchmark.
I am explicitly disavowing that kind of analysis and talking about absolute gains.
I have four children, the oldest graduated in 2017, the next in 2019, then 2023 and my youngest is graduating this spring(2026), all from the same school district. There was a marked difference in their education. COVID was absolutely part of the problem but the total reliance on smartphones and chromebooks, first because everything was online and then even when in person school started again was and continues to be the biggest problem. It has made me a firm believer that we have to go back to books, pen and paper for at least K-8 and get smartphones out of school.
Is there any unequivocal evidence that educational successes which are reported here and there are scalable? I have always thought that holding teachers responsible for student achievement would invariably open the educational market up to more manipulation and there is evidence enough that has happened. Or, put another way, do schools truly educate the masses or do they more correctly provide a sort of template that some are able to take advantage of and some are not? Part of being able to do so depends on students' intellectual capacity and motivation which is not stressed in this article.
Schools actually do educate the masses. What they don't do is move people outside of the range of their cognitive ability.
I like what you say, but wonder if it is tautological. When comparing schools and educational methods, what are we really comparing? Are the data broken out by cognitive ability in the studies we read, or is most of this obscured by "averages?"
The latter.
Folks just hate teachers.
That's probably why teachers are always in the top five most highly respected occupations.All that hate.
I was definitely being hyperbolic - but it's certainly a profession in which you get a lot of non-teachers very confidently making judgemental claims about the job.