Harry Brighouse provides a summary of "Sold a Story," a documentary podcast from Emily Hanford. The topic is reading instruction:
Very brief synopsis: the methods routinely used to teach children to read in the US don’t work well for large numbers of children, and the science of reading has been clear about this for decades. Three academics in particular — Lucy Calkins of Teachers College, and Irene Fountas and Gay Su Pinnell of the Ohio State University — are responsible for promoting these bad practices (which are pervasive), and persisted in doing so long after the research was clear.
Harry ponders what this means for the integrity of university research in general:
The way universities work mean that a substantial amount of Fountas and Pinnell’s work will have been looked at, at least formally, by colleagues in other, but related, disciplines, deans, etc, over many many years. It’s hard to believe — or at least its hard for me to believe — that nobody thought to themselves “the research designs here look very weak”/”the methods by which these results have been found look very dodgy”/”what the hell kind of research is supporting these recommendations which are, on their face, obviously bad?”. It’s easy to believe — or at least it's easy for me to believe — that people would have thought those things and not taken action.
Why did so few people fight back against this stuff? Harry hypothesizes that it's a sort of academic omertà that prevents both experts and non-experts from going after their colleagues. That may have its place, but in this case it caused a lot of damage for a very long time—and continues to do so. So he asks: "What obligations do academics have to scrutinize the actions of colleagues who are operating outside of their immediate areas of expertise?"
Good question. Usually this sort of thing doesn't end well, since experts are, among other things, expert at dismissing challenges from outside their field. But should they be?
Would it have killed you or anyone else that I had to click through, to find out that central to the "reading instruction" issue was phonics, or the ignorance of its importance to boosting reading skills?
That summary leaves out one pressure involved: the academics making the bad recommendations were making bank from them.
Academics are over-represented in my circle of friends. I have no trouble believing that their peers wouldn't attack them - there is probably no upside, mostly only downsides, in picking fights like that. Your research probably doesn't intersect, so you have no academic reason. You almost certainly have no authority over them, so unless you're personally close and can present it as help, you can only really attack their cash cow. I think we can predict how that will go.
Administration has no reason to step in - if one of theirs is doing well and getting attention, you're not going to stop the fun.
Entrepreneurial academics are odd cases. The incentives that drive them come from outside academia, so the normal feedback mechanisms, imperfect as they are, don't have much power. I'm not sure how you fix that without a fairly radical rethink of university power relations.
Posted Without Comment
https://jan.ucc.nau.edu/~jar/Reading_Wars.html
Thanks, good ref. Many complicated interests intersecting here.
A few months ago one of the NPR public affairs shows did an episode on whole language that might have been based on this documentary, or related to it in some way. It was a pretty complete evisceration of whole language. But unless I missed something, it didn't bring in No Child Left Behind and all the pressures around that, which seems like a big omission. Whatever, it sure seems like the knives are now out for whole language in a big way. Doubtless wielded by some of these interests.
This sounds like the kind of story I'd expect you to refute. We have a system in place to have experts in a field review each other's work: peer review. And experts in a field are motivated to correct each other. That's called Science. Krugman is certainly not worried about pointing out details that Friedman might have overlooked or misinterpreted.
Asking the dean of a college to carefully read and think through each paper published by a researcher at the college seems like a high bar.
"And experts in a field are motivated to correct each other."
Hopefully, but it is not guaranteed. It depends on these experts doing better when the field of research really progresses and and doing worse otherwise. This is not always guaranteed, and there is no obvious mechanism that guarantee it.
"...And experts in a field are motivated to correct each other. ..."
Not if the entire field is nonsense. Or even partially nonsense as it contains an universally accepted (within the field) premise that happens to be false.
I was not familiar with the name Lucy Calkins, so I looked her up on Wikipedia. To see a wonderful example of your characterization of “if the entire field is nonsense,” you can read this Wiki article, which sounds very much like the sort of cotton-candy bureaucratic b.s. that drove me out of government work 50 years ago:
https://en.wikipedia.org/wiki/Teachers_College_Reading_and_Writing_Project
In the immortal words of Clara Peller, “Where’s the beef.”
K-12 education is not an academic discipline, and K-12 education professors are more of a club designed to further their careers, IMHO, than a community devoted to seeking truth. Research into reading instruction is best conducted by cognitive psychologists and neurology specialists.
GBS did his best to make phonics work: https://en.wikipedia.org/wiki/English-language_spelling_reform
Went through the links to find out that the (unhelpful) reading approach when encountering an unfamiliar word is:
See if there's a picture nearby to provide context - and then guess
See if the sentence/syntax can provide context - and then guess
Skip the word
The approach had passing acknowledgment of how words are spelled (to be sure it's not some other word: e.g. house / horse), but pronunciation was peripheral.
In other words, the phonics debate once again.
The brain has lots of things stored. Images, words already encountered (as images of letter sequences), and vocabulary. From that, one can read, up to a point. But learning new words is not as fast as it could be.
What does a reader do when a new word appears? If it can't be determined with image/vocabulary, what other resource is there that could help?
Spoken words you have heard. You may never have encountered the word "platypus" but you might have *heard* it said in a sentence (e.g. "A platypus lays eggs and is duck-billed"), and if the reader sounds out p-l-a-t-y-p-u-s there is a good chance that a sound look-up will bring back a memory of how it was used, thus providing the context that is missing in a sound-free approach.
Abandoning phonics is basically depriving the student of a huge, helpful database. Why exclude it?
As a chemistry PhD, it’s very hard for me to make comparisons with other academic disciplines.
But it’s also hard to overlook the general differences in the behavior of academics n fields where the subject matter lends itself to less interpretation than the fields where interpretation occupies a significant role.
Education is a great example of a field of study that is by no means, “fake”, but there are soooo many factors in even a simple study, simply getting consensus on the criteria for its evaluation is an enormous challenge.
My mother used to teach second grade, so she read a bunch of the literature, including a metastudy covering the research on learning to read. The strongest result: children in the test group do better than children in the control group. If you've read about why medical research uses double blind studies, this may be unsurprising. Studies of methods of teaching reading are not, and cannot be, double blind because the teachers know which methods they are using.
This means that an assertion like, “the science of reading has been clear about this for decades,” may be true, but would have to be based on something other than direct observation of which methods of teaching reading work best.
Much more efficient that in political science, where the standing joke about survey research is the PI having to ask the sponsors what result they want.
(To be clear, I'm not saying the education projects are deliberately rigged, but that the double-blind rationale really does hold water even where it can't be implemented.)
It's not up to university administrators to "police" researchers, unless it's to address actual scientific misconduct (e.g., fudging/manipulating data). The job of "policing" is up to others in the field to provide new evidence that challenges, or supports, existing conclusions. The problem is that public policy is always lagging behind new research findings, and new research that overturns old paradigms can take a generation as the new guard replaces the old guard. Even well-intentioned policy-makers are going to pay more regard to someone who has risen to the top of the academic ranks, published more papers, is a leader in professional societies, and influences what research gets funded. Oh, and then there are politicians who don't give an actual crap, and are just looking for research to confirm their biases or advance an agenda.
I wonder how kids in China are taught to read. Obviously phonics can't apply to an ideographic writing system. Yet China is an economic powerhouse so they must be doing something right.