kitchen table math, the sequel: 2 + 2

Wednesday, August 15, 2012

2 + 2

3 factoids:


  • The syntactic complexity of the texts children read increases each year, eventually becoming more complex than anything children hear in conversation. This point is reached in the 4th grade. We have Jeanne Chall to thank for this insight, too.
So, starting in 4th grade children are supposed to learn grammar by reading texts written in a grammatical register they have never heard in conversation and never will hear in conversation. Writing is not talking, no matter how smart your parents are.

And, also starting in 4th grade, children's rate of progress in reading comprehension collapses.

Nobody seems to have noticed the coincidence. The National Reading Panel doesn't talk about syntax, E.D. Hirsch doesn't talk about syntax, and the NCTE is interested only in the question of whether formal instruction in grammar improves writing. Not reading.

No one seems to have asked himself whether it was all those precision diagrams of yore that brought children to the level of syntactical fluency that allowed 4th graders to read McGuffy Readers and 10th graders to read Dickens.

Instead, it's been left to speech-language pathologists to discover the fact that if you want children to read, you had better teach them how to read sentences, not just words.


reform writing (Robert Connors on the Erasure of the Sentence)

32 comments:

Jeremy Millington said...

I'm not sure if grammar education is as easy as 2+2. It's certainly valid to hypothesize grammar instruction aids reading. But where's the evidence?

Auntie Ann said...

I wish I had saved this link when I came across it, but about 5 years ago, I came across an article which said that children's books had been deliberately watered down twice during the 20th Century. That kids used to read books that were essentially adult books: Robert Lewis Stevenson, Jules Verne, etc. Then, people started writing books specifically for children and the complexity of the vocabulary and sentence structure began to tank.

The article was clear about two specific movements, one before WWII and one after, that did this deliberately. I've never been able to find that article again :(

Jean said...

How often do kids 9 and up generally hear books read aloud? I would bet that reading complex texts aloud to children would greatly help in comprehension, especially if it's a parent or teacher who will stop and explain every so often. We tend to think that kids over the age of 6 or so are too old to be read to every night--and I bet that's a bad thing.

Grammar is good too, though! I have to say, I make my kids do grammar (because I was never taught any). We also do read-alouds. They have excellent comprehension skills, even the one who doesn't like to read as much as I wish she did (she has a visual handicap that makes it harder work). A couple of weeks ago I gave my 12yo the 'Declaration of Independence test' from "The War on Grammar" and I was pleasantly surprised to find that she understood it pretty well and was able to explain it.

palisadesk said...

Well, *I* have been talking about the link between syntax, grammar and reading comprehension for YEARS, mainly to argue with people who take Willingham and Hirsch far too literally and don't realize there are component skills to reading comprehension besides vocabulary, general knowledge and IQ. "Low" kids can show spectacular improvement when taught some of this stuff explicitly.

I had the Eureka! moment about 12 years ago teaching 7th grade history, when one of my students -- who could read the text accurately -- burst out in frustration, 'Why don't they write this stuff in ENGLISH!!' Suddenly, I saw what she meant. The phrasing, the convoluted sentence structure with relative clauses and inverted order and the passive voice, all made the meaning of a simple narrative of explorers as opaque as a foreign language. If I paraphrased the text, she had no difficulty understanding it.

So these sentence structures need to be taught. IIRC, the Whimbey books address this pretty well, and in an engaging way. I had a sixth grade group two years ago who really enjoyed Mastering Reading Through Reasoning.

Learning another language in a formal way also develops grammatical sense and an awareness of the grammar of one's own language by contrast, as it were. I learned both Latin and French in elementary school. I'm sure it helped my grammar, too, though the first thorough treatment of grammar in English that I recall was in ninth grade.

TerriW said...

I mentioned earlier that I do a huge amount of reading aloud to my kids. Mostly because I think it's a good idea, but I've stepped up the amount I do because I have a 6yo non-reader (with some underlying issues that are being addressed before that it likely to change).

At his age, his sister was already chewing through books and he can't do that yet -- but he's still like a little sponge, and I'm trying to help him be able to absorb things aurally (?) instead of just visually or by screen. So, I read aloud a lot. A lot.

We also memorize. Poetry, lists, whatever. I had a friend (an unschooler, natch, but she wasn't asking unkindly) ask me why I was having him memorize so much poetry, and I said that I didn't want his only language absorption or processing to be informal speech-language, that since he's a non-reader (and likely will be for awhile yet), that in order to get formal, written language into his mind, I have to get it there by other means. Reading aloud, memorizing poetry.

And, honestly, he doesn't seem to mind -- he enjoys memorizing. (It doesn't feel like child abuse!) In fact, he's actually quite good at it -- perhaps that's an artifact of not reading. Because he's still living in a completely oral culture, he can't decide to not know something because he can just look it up, offloading some of his memory to paper or book.

Auntie Ann said...

If three year olds can memorize the names of every single Thomas train, they can memorize poetry!

SteveH said...

I don't know how "reading achievement" is defined or why its growth is expected to be linear. Scholastic points to NAEP data, but I'm not sure what data that is.

That doesn't stop them from blaming teaching to the test, even though it's the NAEP test that seems to define the problem.

"The push to “cover the standards” has crowded out time to teach science and social studies, subjects that engage kids and have the added benefit of teaching academic vocabulary."

They blame the student via engagement and motivation. Apparently, if you teach to the test, then you don't have time to do these other vague things which are somehow more effective.

Their solution is not to define the problem accurately and attack it directly. Their soution is to talk about motivation as if the solution can't possibly have anything to do with the curriculum.


"It's certainly valid to hypothesize grammar instruction aids reading. But where's the evidence?"

What, exactly, is the problem? Where's the evidence that engagement and motivation are effective tools for getting better NAEP test scores. Schools get to decide what they do based on anything they want, but they are the first to ask for evidence when challenged.

Building motivation and engagement are nice things to do, but they are not sufficient. Thematic learning and discovery can also be good tools. But how do they validate that these techniques work? When the NAEP test gives them bad news, they still ignore educational techniques that directly attack the problems. Somehow, education has to be indirect rather than direct and explicit because that would be teaching to the test - that would be rote learning.

Where's the evidence for this? What the relationship between rote learning and bad teaching? In math, educators see what they think is rote learning and come to the completely wrong conclusion.

SteveH said...

"...he enjoys memorizing."

Many educators dislike memorization. They go out of their way to let you know that it is "superficial". But what's the difference between memorizing and remembering? It is a speed and context issue? Is remembering best done thematically? They appear to think that facts need to be built around knowledge rather than the other way around.

In first grade, my son had a thematic unit on the Artic that included a Native American story, a little bit of geography, and some information about artic animals. How does that context work for memory? How do you better remember that information without a larger framework of facts? They hadn't talked about the solar system, the earth, or even the continents beforehand. How much will kids remember? I have, however, clearly seen how knowledge has fit into my son's learning of facts.

In seventh grade, he memorized the Periodic Table. Rote? Meaningless? No. He absorbed so much more. The brain doesn't automatically stop with facts. He could explain how elements would combine in different proportions. He also memorized pi to 120 digits. Rote? Yes. There are no connections. He has also memorized all of the countries of the world and their capitals. Rote? No. Those facts provide all sorts of opportunities for knowledge to fill in the blanks. Current events have a place to go. They are not lost with no place to fit into his memory.

It's one thing to argue against memorizing the list of presidents just as a list of names, but quite another to provide no factual framework for connecting new pieces of knowledge. In the earlier grades, I was constantantly astounded about how my son's schools never seemed to jump right in and get down to learning, especially with memorization. In sports and music, you are expected to work hard on basic facts and skills. For academics, they seem to think that it's some extraordinarily delicate process that will fail if you set any reasonable level of expectations.

Auntie Ann said...

This gels with the preference for low-levels of complexity in writing. I'm sure we've all been pissed off when MS Word flags a perfectly correct, albeit multi-clause, sentence for grammar "correction" simply because it is too complex. (How many have turned off the grammar correction feature completely?)

How much of the preference for "simplicity" comes not from a desire to teach students to write to their audience, and how much of it comes from the fact that teachers aren't used to complex sentences either. The greater the desire for simplicity in writing, the harder it will be for students to understand complex grammar and to be able to access older texts. In a few decades, works written in the early 20th Century might look as foreign as Shakespeare to future students.

One of the cures I have attempted at home, is to always try to get the kids reading books that are as old as I can find. The older the book, the more-complex the sentence structure tends to be. Unfortunately, our kids seem to want to read the current offerings, or read what everyone else is reading. I usually don't get my way.

Bostonian said...

As usual, I will remind people of the importance of intelligence for academic achievement. Quoting a review of Arthur Jensen's book "The g factor",

"Reading comprehension is the academic skill with the most economic importance. The significance of g for reading comprehension increases with age. Learning to decode (i.e. know which words go with which symbols) is heavily a function of memory and is usually mastered in the early grades. After the fourth grade (and setting aside those suffering from dyslexia) the correlation of reading comprehension with scores on intelligence tests is high, almost as high as the correlation of one test with another. Defects in reading comprehension in adults and older children seem to be essentially defects in g. Interestingly, there is a correlation of -.71 of reading comprehension in ninth graders with the standard deviation of reaction times (a simple measure of neuronal functioning). This suggests that reading comprehension ability is determined by the same basic feature of brain structure as determines reaction times and g. This is rather discouraging for those educators who believe that quality of teaching is the key variable in determining reading ability, but that is what the data shows."

Auntie Ann said...

SH: ". In sports and music, you are expected to work hard on basic facts and skills."

You would love the argument we had with our kid's school this summer. Our school has basically said that it would not be developmentally appropriate for our advanced math student (who has been tutored at home) to actually receive more-advanced work in school than his peers. And that he should essentially wait for everyone else to catch up to him.

So, we countered that when he goes out for football, we assume that those kids with advanced skills will be benched while those like our boy catch up. After all, it would not be developmentally appropriate for those advanced players to be set apart from their peers.

The school was not amused.

Upstart said...

Ha ha, Bostonian's anonymous reviewer doesn't realize that "data" is a plural word. So much for "g" -- obviously not g for grammar.

Obviously intelligence affects reading comprehension, but so does good teaching. Poor teaching has in fact been shown to *lower* IQ over time.

Bostonian said...

Upstart's nitpicking is silly. It is common to treat "data" as a singular noun in the U.S. As a 1-minute Google search would have discovered, the review is from

http://lrainc.com/swtaboo/stalkers/em_jensen.html
Mankind Quarterly, Vol. 39 (Spring 1999) No. 3, 337-354 Edward M. Miller
University of New Orleans

SteveH said...

Everybody knows that some kids are smarter than others. It's OK to separate kids in high school, but not in K-6. This social decision trumps an academic one. Add full inclusion to this and you make the problem worse. That some schools try to justify this non-separation with all sorts of ideas doesn't mean that they really don't know that some kids are smarter than others. They just talk about differentiated instruction. Our schools have even called it differentiated learning - the onus is on the child who is supposed to use that mantra to become a life-ling learner. Therefore, not directly teaching becomes a good thing apparently. If they did differentiated instruction well, that would just be a form of hidden tracking. They are just postponing the big filter until high school where they can't see all of the damage they've done. It will then be too easy to blame the kids, parents, peers, and society. The kids will probably believe it too.

To say that: "... it would not be developmentally appropriate ..." is just cover for their decision to place social issues over academic ones. They just don't like to be called on it. I understand that there can be severe issues due to tracking, but what do they expect? They increase the spread of ability in each grade and then set low expectations. They push kids along until it's too late to do anything about it.

For my son in fifth grade, they specifically told us that there was nothing it could do for our son mathematically. They had their hands full just trying to get the other kids back to some decent level of mathematical skills. At least they didn't try to tell us that it was for the best developmentally.

Glen said...

Bostonian, I agree with the importance of IQ as a foundation on which to build cognitive skills. But I think that if you narrow the view to just white students, group average IQ increased over most of the 20th century (Flynn Effect), while group average reading comprehension decreased. This may be true for other groups as well.

I don't have this data in front of me, and I don't have the desire to go hunting for it, so I'll have to add that this is just my impression. I can't prove it, but a comparison of the content of McGuffey Readers to the school reading materials today strongly suggests that (at least white) kids were exposed to more complex (written) language at school in those days than they are today. It's reasonable to assume that this higher level of training led to higher level of average skill, and that an increase in training level today would improve reading comprehension significantly.

Of course, raising the level of training with demanding grammar analysis, vocabulary study, more complex reading assignments, and so on, while helpful to everyone, could well be most helpful to those who are the least challenged by the current system. If so, it would widen the achievement gap, which the educational establishment would likely view as a failure and good reason not to do it.

cranberry said...

Auntie Ann, could this be the article: http://www.aft.org/pdfs/americaneducator/winter1011/Adams.pdf ?

palisadesk said...

Here's an article that supports Glen's observation (some of the research cited did indeed focus explicitly on the decline in reading skill and SAT scores among white students exclusively)
Schoolbook Simplification and Its Relation to the Decline in SAT-Verbal Scores

I suspect reading levels of books students are exposed to (or read from) are a contributing factor.

You don't even have to go as far back as the McGuffy readers. When I moved schools one time, I had to clear out a cupboard of books from an earlier era. I found some readers from the 50's and earlier. I was astounded to see what was in the fifth grade reader -- Irving's "The Legend of Sleepy Hollow," "The Ransom of Red Chief" (O Henry), "Dog of Pompeii" (Untermeyer), selections from Longfellow's "Evangeline," shorter poems like Emerson's "Concord Hymn," Lindsay's "Abraham Lincoln Walks at Midnight" ...and much more.

I have difficulty imagining most eighth graders reading those now.

ChemProf said...

A few months ago, after another similar discussion at KTM, I started reading Beatrix Potter to my three year old. She can now recite Tom Kitten and Squirrel Nutkin. At her age, memorization seems pretty natural. We will see if it helps her reading comprehension later, but I do think that reading L Frank Baum and Stevenson as a kid helped me. My undergrads definitely complain about the old fashioned language on the GRE. (plus it is darn cute hearing her say " I am affronted!")

Anonymous said...

Is this a chicken and egg problem? Before the growth of graded readers and the professionalization of juvenile writing, the only choice of literature for children was more grammatically complex than it is today.

Does reading more grammatically complex literature make the child more readily profit from the study of grammar? Or vice versa?

ChemProf said...

I think there may well be a "which came first" component. I also actually wonder if the emphasis on early reading (and therefore not reading aloud to older kids) is part of it. I see it with local parents - if a child shows signs of early reading, the tendency is to start them on the Bob books or similar easy readers, even of the child is young enough (and I am talking about three year olds) rather than developing their oral language and listening capacity.


It reminds me of the arguments for copy work, among classical homeschoolers, that kids need to learn to hold a sentence in their heads before they can learn to write well, as well as leaning what a good sentence looks like.

Anonymous said...

"I think there may well be a 'which came first' component."

The introduction of "look-say" as the primary method of reading instruction may well have played a part in this.

One of the major reading primers in the 1930s (pre look-say) was the Elson Basic Readers.

By grade, the vocabulary difficulty (larger is harder) and average sentence length (tokens/sentence) was:

    1: -49/7.9
    2: -47/10.0
    3: -51/13.5
    4: -38/15.9
    5: don't have
    6: -21/17.8

The 1960's Dick and Jane (which are the stereotypic look-say) looks like this:

    2: -59/8.2
    6: -23/11.3

The largest thing that jumps out at me is that the *grammar* has gotten substantially simpler. My hypothesis is that the readers being simplified drove the simplification of the rest of the text books (this was Dr. Hayes' view, too) rather than the lack of student skills driving the simplification of the primers.

Part of this is that the look-say primers were simplified even in the earlier grades. Elson 1st grade is roughly comparable to Dick-and-Jane second grade. What could have cause our 1st graders to fall a year behind the previous generation?

-Mark Roulo




Auntie Ann said...

We rely on the Flesch-Kincaid score feature of MS Word to at least get an idea of the reading level of books our kids are reading--and being assigned in class. We typed in a sample of each of the books our 6th grader read last year in school, only to find they scored pathetically low: 4.8, 5.2 & 5.6. The weren't even being assigned reading at the 6th grade level.

Sometimes you can use Amazon's book preview feature to type in a sample text into Word as a way of gauging a books reading level before buying it. It's time consuming and hard to do, but it helps. (FYI, the Harry Potter series gets more complex as it goes along.)

Scholastic's website also posts the reading levels that they assign to their books. However, those seem more tied to subject matter and content than to sentence structure and vocabulary.

With more people buying books for their kids online and sight-unseen, it is getting harder to promote complex reading in our kids. I would think this would be a great opportunity for a company like Amazon, which already has digital copies of many children's books. It wouldn't be hard o develop a solid reading level scoring system, or use an already-available one--and post it for parents as part of the information available on their website.

palisadesk said...

For more on the topic of text complexity, see this interesting post over at Curriculum Matters.

The links are informative and several address Auntie Ann's issue of development of a common "leveling" framework that would inform parentso r teachers about the level of the text in a comprehensive way.

A propos of which, I have heard good things about the Lexile framework (which takes into account text complexity, sentence length, vocabulary, syntax and more) but have not had an opportunity to see its application myself. It just looks promising.

momof4 said...

I was interested in ChemProf's comment about copying being used by homeschoolers. That's how my Normal-School 1-4 teachers started us; first by copying from the board. They explicitly explained the rules; All sentences begin with a capital letter and end with a period, proper names (discussed) are capitalized, all sentences have a noun/pronoun and verb (discussed) etc. We then progressed through teacher dictation to composing simple sentences. Weekly spelling words started with often-used words. We were NOT asked to write/edit stories etc. before we knew the conventions. (I don't remember ever being asked to write stories) We wrote a daily log - first by dictation: name, date, time, weather, other comment - and we wrote answers to questions; first one-sentence, then short answer. Everything had errors red-pencilled (we had to fix).I've always thought that approach made sense.

We also were expected to memorize things; short poems, counties in the state, parts of a plant, stages of metamorphosis, planets, rivers etc.

TerriW said...

Hey, I'm one of those homeschoolers who use copywork!

We use the resources from Peace Hill Press -- the First Language Lessons series and now Writing with Ease.

My daughter recently completed WWE Level 1 and is now 5 weeks into Level 2 -- we're slightly "behind" on that one, I suppose because her handwriting lagged far behind her other skills for some time.

[[As as aside, that's how I came to Peace Hill Press in the first place -- so many Language Arts programs roll along, assuming that reading and handwriting somehow track together. In my daughter, reading was there YEARS before handwriting, and in my son, handwriting has been there years before reading. First Language Lessons Level 1 is primarily oral and can be done with a non-reading or non-writing child. It was right up our alley for both of my kids, albeit for different reasons. But I digress.]]

WWE 1 worked like this:

Monday, you listen to a short excerpt from a classic novel or somesuch, then you ask your child specific (scripted) questions about it to ensure that they caught the meaning. (They are to answer in complete sentences.) Then you have them tell you one thing they remember about the passage, which they then watch you write down.

Next day, they have a copywork sentence from the passage. (Two choices, one longer than the other, based on handwriting ability.) There's generally something about the sentence that you are trying to highlight, for example, that months of the year are capitalized.

I seem to recall in WWE1 that the next two days of the week repeat that rotation. Eventually you work up to them using your copy of their narration as copywork.

It is extremely incremental. Then you get to WWE2, where you now start to have your narration from the read excerpt no longer be "just something you remember" but a summarization -- but there are scripted questions that you can ask about the passage that you can use to get them thinking about how to summarize the information.

Then, on Day 2, you do your copywork from the passage, with the twist that on Day 3 now, you use that same sentence as dictation -- they have to write out the sentence after just hearing it from you, not actually seeing it in front of them.

This is getting long. More to come.


TerriW said...

So, the general philosophy is that there are a lot of plates that you need to keep spinning in order to write -- there's having the thought to write about, there's putting that thought into words, there's holding that thought in your head while you get it written down, there's the actual physical, mechanical act of writing -- and, for many kids, all of those skills just aren't there, ready to go.

And forcing writing at a young age when there are deficiencies still in a few of those areas can produce a lot of "I hate to write"rs.

So, in WWE, those skills are practiced in isolation so you only have to worry about honing one at a time, and then you eventually start interleaving them.

I'm sure this approach horrifies some folks, but ... it's working very, very well for us.

TerriW said...

And this just came through my email this morning, it's from a newsletter from Andrew Pudewa, the guy behind IEW (Institute for Excellence in Writing, another popular writing program):
---------------------


If anyone out there is struggling with "journaling" and boys (or girls),
this correspondence might be of interest to you.
Andrew.
--------------------
Dear Mr. Pudewa: I attended your two day seminar at last summer and have
been borrowing the videos from a friend. I think you have some great ideas
and my 10 year old son is getting better each week at the key words. We
have made it to Level IV. We have been reading about the 7 Wonders of the
Ancient World and he has done a paragraph about each of the Wonders. He is
doing all the dress ups and two of the sentence openers, so I feel like
we've made good progress through this school year.

My concern: I also have him journal 15-20 minutes a day about anything he
wants to write about. I thought the more he wrote the better he would get.
I do not critique any of this writing. He usually writes something to the
effect - I HATE WRITING! I am frustrated. He is excelling in every subject
but this one. He keeps saying - I'm not going to be a writer when I grow
up, why do I have to do so much writing? When I hear you on the videos, you
seem to have a special sympathy for boys. Could you please give me some
wise counsel? Thank you in advance.
--------------------
Why are you forcing him to do journaling? This would be torture to me! As
far as I can tell, the practical value of doing it, especially when you say
"anything he wants to write about" is virtually nonexistent, and the torture

factor is high. Furthermore, it actually creates bad writing habits, as the
"stream of consciousness" nature of the activity precludes any real
thinking, organizing or structuring sentences or ideas. So, unless you are
forced by some authority greater than yourself to require that he do this,
I'd stop it immediately, and focus on content-based writing assignments.
That is what is most appropriate and useful for a 10-yr. old boy.

Now, if you are determined to keep this journaling idea going, then you
should do it this way:

1. Choose a subject. (Trees, Weapons, Food, Sports, etc.)
2. Ask him questions about these things.
3. Put his answers into a key word outline for him on a white/black board.
4. Limit the number of ideas he uses to the time available (which for 20
minutes would be about 5-6).
5. Let him write in his "journal" from a key word outline about things you
have helped him think of and organize beforehand. (This is similar to our
Unit VII - Creative Writing).
6. Require dress-ups.
7. Don't do a new topic every day; instead use a second (or even a third
day) to rewrite or type a final version. Remember: NO ERASING ALLOWED on
the first draft, and there's no such thing as a first draft of anything,
even "journal" entries.

That's my advice. Personally, I don't like writing either, and even now at
43 years old, if you gave me the assignment to write 15 minutes a day about
"anything," I'd whine, procrastinate, and if I did actually get beyond the
blank paper, I'd probably mouth off about how stupid the idea of
"journaling" is. Sorry to be so blunt.

God bless you in your home schooling. Always do more of what works well,
and less of what frustrates children. Consciously and intentionally smile
at least once every ten minutes, and start each lesson with a joke.

Auntie Ann said...

It wasn't the Adams article that I saw--it was written too recently--but the "Schoolbook Simplification" one might have been it.

ChemProf said...

momof4 - yes, as TerriW says, lots of homeschoolers use copywork. It is pretty common with several types (Charlotte Mason and classical being the most likely) to break apart narration (or telling the parent about what was read) and writing (which starts with copywork). And in The Well-Trained Mind, they specifically recommend against having students write stories as assignments -- it is great if kids want to, but they argue that it shouldn't be required.

Anonymous said...

Auntie Ann, you don't have to retype what is in Amazon.

See http://gasstationwithoutpumps.wordpress.com/2012/08/03/getting-text-from-amazons-look-inside/

momof4 said...

Terri: What is a "dress up"?

TerriW said...

Momof4: I don't actually use IEW, so I can only give you a half answer -- but my understanding of their method is that when you build your paragraphs, it has some check-off boxes of types of things that you can use to add color or interest or complexity to your writing (stronger verbs, adverb clauses, whatever it is they happen to be focusing on that week) -- they call those "dress-ups."

If someone here is more familiar with IEW than I am, please correct me if I have that wrong!