Live data from Hacker News

Sold a Story: How Teaching Kids to Read Went So Wrong

features.apmreports.org

181–190 of 524 posts

Re: Sold a Story: How Teaching Kids to Read Went So Wrong

#181
Again from the podcast:

Hanford: So much of this research isn’t new. And this idea that readers use context, multiple sources of information to solve words, identify words as they’re reading, that was really taken on by researchers back in the 70s and 80s, as an interesting question. Like, is that what we do? And they showed quite definitively that that wasn’t the case. I mean, were you sort of aware of that research and how clear that was already by the 90s?

Calkins: Um, again, you’re asking me to go back and figure out what was in my mind at one point or another. Um, but I would say that, that you have to remember that that research was not – I don't think that there were classrooms that were doing classroom-based methods that were exciting, and poignant and beautiful, and, you know, getting kids on fire as readers and writers, that were using that that train of thinking. You know, it was part of an entire gestalt that was different than ours.

These people belong in gaol. Introducing teaching methods that were known to be wrong because they don't like the style of the effective training. Therefore introducing new training which is less effective. This has damaged countless lives. Reading is a fundamental requirement of modern society, and failing to teach people because you don't like an effective approach is criminal.

Again from the podcast:

"Good reading instruction isn’t boring for children. Maybe adults find parts of it boring. But this shouldn’t be about what adults want. It should be about what kids need.

And there’s no reason that reading instruction aligned with scientific evidence can’t be exciting and beautiful. I think Lucy Calkins sees it that way now too. Because instruction aligned with the science of reading is what she says she’s now selling."

Re: Sold a Story: How Teaching Kids to Read Went So Wrong

#182

Earlier quoted context omitted.

>English has a phonetic alphabet after all, right? No, it doesn't - if it used IPA, it maybe would, but English has very inconsistent pronunciation for otherwise similar words written in Latin alphabet, a matter that many English native speakers seem to overlook: https://www.youtube.com/watch?v=1edPxKqiptw

It's an alphabetic writing system where the letters largely correspond to sounds, in contrast to logographic writing system where the symbol corresponds to the entire word (like Chinese or Japan). Yes, there are special rules and outliers that you need to learn in English, but it seems absurd to not classify it as phonetic because it's not purely phonetic. This is doubly so when discussing phonetic vs whole-word lear…

> Yes, there are special rules and outliers that you need to learn in English (...)

I think you're greatly understating it. It's most of English, it's present everywhere as you try to learn the language. It's present from the very beginning, when you need to figure out why "are" and "area" are not pronounced the same, until the very end, when you have mostly mastered the language but now need to be able to understand everyone else's pronunciation while also accounting for them most likely pronouncing some words incorrectly.

Re: Sold a Story: How Teaching Kids to Read Went So Wrong

#184

No one is talking about the impact of kids seeing parents read on phones instead kids seeing parents read on books. When kids see a parent reading a book there's no doubt as to what they're doing. They are reading. OTOH, even if a parent is reading the most wholesome text on the most wholesome app, the fact that they are doing it on a phone admits some possibility of confusion in the eyes of the child observing. Ther…

I'm not sure how you are reading to your kid but mine usually sits on my lap. He knows what is on the screen whether it's words or something else.

There's a "nonzero chance" your dead tree book is a game too ala Sudoku.

Re: Sold a Story: How Teaching Kids to Read Went So Wrong

#185

No one is talking about the impact of kids seeing parents read on phones instead kids seeing parents read on books. When kids see a parent reading a book there's no doubt as to what they're doing. They are reading. OTOH, even if a parent is reading the most wholesome text on the most wholesome app, the fact that they are doing it on a phone admits some possibility of confusion in the eyes of the child observing. Ther…

Great point. Children also have a powerful desire to know their parents are paying attention to them. Phones disrupt this as well.

Re: Sold a Story: How Teaching Kids to Read Went So Wrong

#186

Earlier quoted context omitted.

I am shocked this comment exists, when the counter-example is literally written into it. English uses the latin alphabet which is a phonetic alphabet. It's why english readers can also 'read' Spanish without understanding any of the words. Anyone who says otherwise should be promptly ignored.

To be fair, English pronunciation is not as regular as Spanish. Spanish is a purely phonetic language, while English is phonetic but irregular.

People keep saying this, and I agree english is more complicated, but it is otherwise regular. The vast majority of English words are regular, and for those that are not regular, sounding them out incorrectly in context with the other words quickly disabuses you of any incorrect reading.

My mom was a public school teacher in a poor district, and the main problem with her kids learning to read is that they did not actually know English. This would make it much more difficult to figure out how to read properly, since they'd never be able to correct 'wrong' pronunciations. I believe a lot of focus on reading is mis-placed when the children cannot speak English properly. Better to focus on diction and vocabulary.

Re: Sold a Story: How Teaching Kids to Read Went So Wrong

#188

Earlier quoted context omitted.

> if you just teach phonics you aren't actually helping kids to read (they've tried) because reading is more than just sounds Of course, decoding isn't sufficient on its own (I can 'read' Portuguese without understanding it). But for a child who has oral fluency in a phonetic language, decoding is the main thing they need to learn, to allow them to bootstrap additional reading skills.

I have heard many teachers make this same criticism and it's so odd to me. Little children learn to read first by literally reading out loud. It's only later that 'silent' reading develops, and for most of us (I'd venture), we silently read by internally 'hearing' the audio version. At least I do (I'm listening to myself narrate this comment as I type it out silently). How else can you learn to read if not sounding o…

> Little children learn to read first by literally reading out loud.

IME, little children learn to read first by recognizing whole words that other people read out loud, and the best learn-to-read curricula recognize and leverage that heavily, and use it to progress to add in phonics skills via using words that vary in one phoneme to reinforce what the common part sounds like. Phonics is useful as one of the skills used to identify unfamiliar words when children progress to reading independently, but even there its not the only thing going on (context is important).

> How else can you learn to read if not sounding out?

Whole word recognition. That written languages have existed that are logographic rather than phonetic makes it clear that it is possible to read, and learn to read, by means other than phonics, which is only potentially useful as part of the process in phonetic languages. But even where it can be part of the process, its not the whole story.

Re: Sold a Story: How Teaching Kids to Read Went So Wrong

#189
post #2

As someone who's both worked as a teacher and as a software engineer, my feeling is that what happened in reading education is roughly what engineers' jobs would be forced by their CTOs to use software methodologies and languages that came from their university professors who've never really spend much time (at least not in well over a decade if at all) doing actual salaried software development where code had to be…

You are right it is `required methodologies`, but wrong on the source. The source is corporations which sell curriculums which are mandated by regulation. These curriculums are designed to be easy to sell to governments, which means they tick all the regulatory boxes, but don't actually teach. Our public education system is not designed to educate, it is designed to return profit to a few corporations which supply it…

The US Gov really needs to start in-housing critical tools if only to build internal competency. There also probably needs to be more incentives for individuals inside the USG to excel and produce value. There are too many perverse incentives for unscrupulous contractors to milk tax payers (see the national parks situation [0]).

There is a huge resource imbalance with the amount of money spent generating hype/misinformation targeting USGov acquisitions vs how much is spent on the other end doing research and fostering expertise.

Maybe the way contracts are written/awarded could also just be overhauled for a similar effect.

[0]: https://mattstoller.substack.com/p/why-is-booz-allen-renting...

Re: Sold a Story: How Teaching Kids to Read Went So Wrong

#190
post #164
post #80

I just listened to this podcast a few weeks ago. What really struck me was just how obviously wrong the Guided Reading curriculum and three cueing theory was even just on its face, and how many people went along with replacing phonics with Guided Reading despite that. For example, the podcast recounts a Guided Reading lesson where the teacher covers up the words the kids are supposed to be reading with a piece of tap…

It worked well enough for BERT: https://en.m.wikipedia.org/wiki/BERT_(language_model)

> It worked well enough for BERT: https://en.m.wikipedia.org/wiki/BERT_(language_model)

Great, the next education fad will be trying to teach kids like some software engineer trained his algorithm.

"Hey kids, start your integration over this new 1TB data set. Remember to adjust your weights after each iteration!"

Post reply on HN