Friday, January 25, 2008

Answers to Criticisms of the Four Subject Curriculum: An apologia for teaching dead languages

Well I profess to being surprised that my minimalist education proposal has caused such a fuss. If only those who fussed at me knew how much I liked being fussed at. What a casus belli I seem to have innocently introduced into the education discussion! I guess that's what happens when you challenge the status quo.

I attribute most of the dissent to my proposal damnant quod non intelligunt: that the critics of this proposal are not necessarily understanding what I have proposed. Now I take some of the responsibility for that, since my original post was meant as a sort discussion starter or abstract, although, apparently, some have mistaken it for some sort of ultima ratio.

I want to begin answering some of the objections to my proposal, and I want to start out with one aspect of this one, since I think it is the most general of the criticisms of my Four Subject Curriculum:
So, your four critical subjects are: a language that no one speaks anymore, a formal system, a method of entertaining yourself and others, and a skill that involves inputting and outputting information. There a few very large gaps, which at least include critical thinking, learning where you fit in the world, and interactive learning.
I want to address just the first implicit assertion in this one: that studying a dead language is somehow inferior to studying living languages.

Yes, my plan does include a language no one speaks anymore--if you mean that ad litteram. Yes, no one speaks Latin per se, but that does not ipso facto mean it is not useful to teach. And, in fact, it is not a priori true that a language that is no longer spoken conversationally should not be a major part of a curriculum. Certainly it is not prima facie true that a dead language is better than one that is not, but neither is it true that just because something is ad patres that it shouldn't be considered as a valid curricular subject.

There have been a good many post mortems pronounced over the Latin language, but I think they are a little premature. Is a language useless just because it's dead? G. K. Chesterton once said that a language must die in order to become immortal--or, to put my own Latin spin on it: morte sola lingua immortalis fit.

So why, then, in the year 2008, Anno Domini, would someone consider Latin important enough to be one of the only four subjects in a curriculum? Well for one thing, I will make the claim right here that there is a direct correlation between the lifelessness of a language and its academic worth.

That's right. I'm saying that, academically speaking, a dead language is better than a living one. Now I know as soon as this is posted I'll be pronounced non compos mentis, and the object of challenges to prove it.

I'll save them the trouble.

You don't have to rely on my ipse dixit. Here are the performance results of students with various language backgrounds on the SAT test via the College Board's last National Profile Report:
Notice that the top three languages--Latin, Hebrew, and Greek--are all dead languages. None of them are spoken in the modern world. Yes, Hebrew is spoken in modern Israel and Greek is spoken in Greece. But the vast majority of people with backgrounds in these languages taking the SAT have not studied modern Hebrew or modern Greek, but the ancient versions of these languages which haven't been spoken for centuries.

The languages that do the best job of preparing students for college entrance exams have long ago assumed room temperature.

Quod est demonstrandum.

28 comments:

Anonymous said...

What students are likely to study dead languages? Maybe smart ones?

Post hoc ergo propter hoc?

I looked at http://professionals.collegeboard.
com/portal/site/Professionals/ and discovered that students who took AP classes also did better. Does this imply all students should take AP classes and average test scores would rise?

[I can't find the data supporting your graph; the breakdown by language I found is not the same as yours. Could you please provide a reference?]

Can you also please provide some better evidence, which accounts for other variables affecting the test scores, to substantiate your claim?

Some other observations:

1) What's the number on the left of the graph? Combined SAT score or just score on one part? If not combined score, why not?

2) That's an absolutely horrible plot. The 3-D aspect and the non zero origin give a seriously distorted picture of the results. I thought your advice was "simplify". What's wrong with a straightforward 2-D plot?

3) Is the purpose of education to "teach students how to think, and give them a basic familiarity with Western civilization" or to do well on standardized tests?

4)"If only those who fussed at me knew how much I liked being fussed at."
It is clear from your earlier writings that often the primary purpose of your posts (e.g., on ID) is not to convey ideas. You seem to deliberately try to mislead people so you can then complain about their comments. But could you at least make it a bit more challenging?

PS Are you advocating four subjects or more, with the rest subsumed under Reading?

PPS I've tried looking up your books but they are not available at either my public or school library. From an online excerpt I see that you have used a chair as an example. Would that explain your apparent animosity towards science?



jah

Anonymous said...

If only those who fussed at me knew how much I liked being fussed at.

This is the standard self-justification used by attention-hogging mediocrities after they've been proven wrong: take all the debunkings as some sort of vindication -- "If I'm so wrong, why are you giving me so much attention?" Or, in the case of creationists: "Look at all the controversy I've caused! We have to teadch the controversy!" Holocust-deniers and creationists do it A LOT, and it's really a bit childish.

I attribute most of the dissent to my proposal damnant quod non intelligunt: that the critics of this proposal are not necessarily understanding what I have proposed.

If we haven't understood your arguments, it's probably because you stated them poorly -- which is evidenced by the fact that: a) you haven't actually stated how or why our counter-arguments are wrong; and b) in the previous thread, you had to go back and explain that you really didn't mean what your choice of words indicated you meant. This, again, is a standard dodge used by people with bad arguments: after the arguments are thoroughly refuted and discredited, simply pretend that nobody really understood what you really meant.

I notice you don't actually defend your previously-stated "simplified" home-schooling curriculum here. I also notice that none of the respondents who took your side actually practiced your recommendations -- they all added extra subjects to their kids' learning.

Notice that the top three languages--Latin, Hebrew, and Greek--are all dead languages...

Your needlessly (and possibly deceptive) 3-D graph (which doesn't include a source, as all graphs should) does not account for any other possible factors in SAT performance. Did learning a dead language improve the students' performance? Or did the students get a better education that happened to include dead languages? Consider the resource angle: a school that has the resources to hire a Latin teacher would also be able to hire better teachers in all other subjects, and provide better facilities all around; while a school that couldn't afford a Latin teacher, would also be less able to compete with wealthier schools for the best teachers.

Furthermore, you have simply ignored the point about the merits of teaching living languages.

All in all, your bad argumetns and evasions, combined with your smug (and gratuitous) smattering of Latin phrases, only drive home the point that Latin is more a pretension than a benefit: a means of sounding more educated and erudite without actually engaging with the real world. Even if you actually know what all those phrases mean, it's certainly obvious that learning Latin didn't train your brain very well.

..... said...

I know Pig Latin. Does that count?? My mom tried to teach me Carney once, but I'm a slow learner.

Heck yeah, you CAN learn science through reading. You can learn History through reading. You can learn science by playing in the mud. You can watch The Magic School Bus, too! Thinking out of the box rox!

It's time to learn humility. Come on, let's all take a deep breath and laugh at ourselves!

..... said...

Music is WAY better than painting!! It rocks all of your senses! ( even smell, if your neighbor has B.O.!) Dictation is fun, exciting and challenging. You must try dappling in composition, too! You learn teamwork beyond compare when you're in an ensemble attempting to keep good intonation, stay in step with one another rhythmically, play effective dynamically and expressively-all while keeping your eye on the conductor and focusing on your own technical accuracy and tone quality and anticipating the nervous jitters that sometimes accompany upcoming solos. Rock on!

Anonymous said...

Mostly very good points Motheral, but give the guy a little slack. Would you put your name on a plot like that?


jah

Martin Cothran said...

Well, I see some have caught the spirit of this post, which was slightly tongue-in-cheek. For those who haven't caught the spirit, Frigate! (which is Latin for "take a chill pill).

Anonymous, you have a valid point here on the post hoc ergo propter hoc point (and I see you picked up on the Latin play in the post). I was being intentionally hyperbolic with my "QED" remark at the end of the post. I have always wondered whether, for example, the argument that music students do better academically is a result of the fact that smarter students take music rather than that music makes smarter students.

I actually think this does have something to do with the results like those I posted. But, first, I don't think that it completely explains it, and, second, it doesn't address the question why smart students choose to take these languages.

The questions we haven't adequately addressed (and I do not even pretend to have done so in these short posts--this a blog people!) are: Does Latin give you an academic advantage and does it do this more than most other languages? And, second, if it is true that part of the reason for the high test scores of those who study it is the fact that they were smart to begin with, then why is it that people who are smart to begin with study it?

I think discussing these two questions may very well be enlightening.

Having taught Latin for 15 years to hundreds of students (and having studied Greek, Spanish, and German at various times) I am pretty confident in my "Yes" answer to the first question. My reasons for believing this come from my own experience, which I don't expect anyone else to necessarily take as authoritative. But there are other, more objective arguments for it that I think are authoritative which I will discuss in my next post.

In regard to why it is that smart students take dead languages, I don't know the answer to that one, but my instincts tell me that the answer may still reflect well on these languages. After all, there are a lot better arguments against the virtues of something than that intelligent people choose to take advantage of it.

I just took the graph from an e-mail newsletter I had done, which had been done with more complicated aesthetics in mind, so I understand the complaints about the format. If I have time, I'll take anonymous's suggestion to do it 2D. In any case, here the link to the report: http://www.collegeboard.com/prod_
downloads/about/news_info/cbsenior
/yr2007/national-report.pdf

Anonymous said...

I also notice that none of the respondents who took your side actually practiced your recommendations -- they all added extra subjects to their kids' learning.

I'm not sure if you're including me in that group -- I'm the one that answered the question about how many homeschoolers limit what they do to mom or dad and a few learning aids.

We didn't add extra subjects to our kids' learning. In fact, we didn't assign them any subjects to study in the first place. They were free to learn what they wanted, when they wanted, and how they wanted. We were there to help and offer the benefit of our education and experience along the way.

So, I guess I'd have to say that we're even more minimalist than the proposed four-subject curriculum, since we scrapped the curriculum completely.

One Brow said...

One of the better ironies was including a quote about critical thinking, and then using an argument that displayed exactly why critical thinking is so important.

Well played, sir!

Martin Cothran said...

One Brow,

Maybe you could be a little more explicit about the argument you think lacked evidence of critical thinking. Maybe you could even present an argument against it.

I gave one reason why I think Latin is a useful educational subject: students who take it perform better on the SAT. And there's more where that came from.

Maybe, as I'm presenting these arguments for Latin, those who disagree could state exactly what evidence they would accept--and then we could see whether they're really interested in getting to the right answer, or just making snippy remarks from behind pseudonyms.

Anonymous said...

Martin: given that, according to your graph, Hebrew and Greek improved performance (or at least went along with improved performance) at least as much as Latin (Hebrew possibly more, depending on how we read that crappy graph), your case for Latin as a "brain training" aid looks rather weak. Also, other languages appear not to do as well, but the margin really isn't that great.

Furthermore, the vertical axis has a range from "440" to "580," whatever that means; if it started at "0", the differences would look a lot less significant.

And now Martin is again getting snippy about "making snippy remarks from behind pseudonyms;" which never seems to be a problem until after his arguments get refuted. So I guess that means this particular debate is coming to an end. (BTW, Martin, if snippiness is a problem for you, perhaps you should be less snippy yourself: you weren't very polite (or factual, or logical) when complaining about "health nazis" and non-smokers earlier.)

Anonymous said...

Martin,

It goes without saying that correlation isn't the same as causation, however, I think your limited curriculum idea for K-8 is interesting. I don't know if anyone else has mentioned this yet, but I would guess the biggest problem with your proposal to teach only Latin K-8 and not other languages for public schools would be that there aren't enough good latin teachers. I expect students would benefit from becoming fluent in any other foreign language.

Anonymous said...

I noticed that Latin, Greek and Hebrew were the top three languages on this chart (for what that's worth); which leads me to a wild guess that may or may not be true, and, if true, may or may not be relevant: Roman Catholic schools are more likely to teach Latin, Eastern Orthodox schools are more likely to teach Greek, and Jewish schools are more likely to teach Hebrew. So maybe the superior SAT performance is due to something in the religious schools' curriculum, rather than the specific language itself.

Anonymous said...

MC: Maybe, as I'm presenting these arguments for Latin, those who disagree could state exactly what evidence they would accept

I am more than willing to make a few more tries at answering Mr Cothran's claims, but he has to hold still for a bit.

The crux of his original post was that only four subjects were necessary for a good education, except his definition of one subject included a number of what had previously been separate subjects.

But never mind that, now he wants to discuss why the study of Latin produces such excellent scholars. I suspect this argument may also illustrate why a grounding in science can be beneficial.


Here's a brief [not all that carefully thought out] example of what would constitute good evidence for his claim. [Unfortunately only approximations are possible in the real world.]

The first statement to be tested is (in Mr Cothran's words) "Does Latin give you an academic advantage" (academic advantage being estimated by the SAT score).

The SAT scores (in the link provided by Mr Cothran earlier) can be examined to see if those students who take Latin score statistically significantly higher than those who take one of the other twelve subjects listed under 'Course Work or Experience'. (I'll link to the data later.)

The real problem with this approach (as pointed out by several people earlier) is that just because a student who studied Latin scored higher than a student who didn't study Latin, the difference in scores may not be due to the study of Latin itself but instead to other parameters. For example Latin may only be offered in good private or public schools which do a better job of teaching in general than public or private schools which do not offer Latin. Students who are home schooled and study Latin may have more interested and better educated parents with more resources than other students who have less educated parents who both work and do not have the option of home schooling.

So here, as Mr Cothran requested, is a draft of a protocol:

1) Acquire a list of all students of kindergarten age (some may have had preschool but that shouldn't matter).

2) Divide them into three groups based on whether their guardians plan on home school, public school, or private school.

3) Randomly select a goodly number (say thirteen thousand) from each pool. This gives three groups of 13,000. Randomly subdivide each of the groups of thirteen thousand into 13 groups of 1000. Within each of the 3 major groups, randomly assign each subgroup of 1000 one of the 13 SAT recognized 'Course Work or Experience' categories.

4) Send the kids to school for 12 years or so and test at the end. Compare test scores to see the effect of each of the thirteen categories.

[There are a number of issues not addressed but the basic plan should be clear.]

jah

Anonymous said...

These excerpts from "The Scholastic Achievement of Home School Students." at http://www.ericdigests.org/2000-3/home.htm
further illustrate some of the other variables which may affect test performance:

MAJOR FINDINGS - DEMOGRAPHICS
Home school parents in the study had more formal education than parents in the general population; 88% continued their education beyond high school compared to 50% for the nation as a whole.

Many home school parents were formally trained as teachers. Almost one-fourth of home school students (24%) have at least one parent who is a certified teacher.

The median income for home school families ($52,000) was significantly higher than that of all families with children ($36,000) in the United States.

Almost all home school students (98%) were in married couple families. Most home school mothers (77%)did not participate in the labor force; almost all home school fathers (98%) did work.

Home school students watched much less television than students nationwide; 65% of home school students watch one hour or less per day compared to 25% nationally.

LIMITATIONS

In spite of the large size of the student sample, there are notable limitations to the study. Foremost, home school students and their families are not a cross-section of the United States population. The act of home schooling distinguishes this group in terms of their exceptionally strong commitment to education and children. As highlighted above, there were major demographic differences between home school families in this study and the general United States population.

This was not a controlled experiment. Students were not randomly assigned public, private or home schools. As a result, the reported achievement differences between groups do not control for background differences in the home school and general United States population and, more importantly, cannot be attributed to the type of school a child attends. Thus, the study was not designed to compare home schools with public or private schools. Such comparisons would be fraught with problems. Home schooling is typically one-on-one. Public schools typically have classes with 25 to 30 students and an extremely wide range of abilities and backgrounds. Home school parents are, by definition, heavily involved in their children's education; the same, unfortunately, is not true of all public or private school parents. Home schools can easily pace and adapt their curriculum; public and private schools typically have a mandated scope and sequence. The list of differences could continue.

-----------------
Mr Cothran has to address these issues if he wants to convince anyone who doesn't already agree with him.


jah

Anonymous said...

Here are links to an objective presentation of the data from the SAT report referenced by Mr Cothran [the data were typed into Excel and checked once - any reported typos will be corrected]:

1) Here is Mr Cothran's chart. It is a plot of the SAT 'Critical Reading' score vs 'Course Work or Experience'. http://www.4freeimagehost.com/uploads
/639968c8b985.gif

2) Here is a plot of the total SAT score vs 'Course Work or Experience'. http://www.4freeimagehost.com/uploads/
c8dbbd012a29.gif

3) Here is a plot of the 'Mathematics' score vs 'Course Work or Experience'. http://www.4freeimagehost.com/uploads/
7515848aa0e7.gif

I have not made a chart of the percentage of students listed under each of the thirteen categories. It is worth pointing out that 70% of the students are listed under Spanish. Presumably for many of them, Spanish is the primary language ('Experience' rather than 'Course Work'). It might be expected that these students would not perform as well as the others.

Mr Cothran wrote "Latin is a useful educational subject: students who take it perform better on the SAT".

From the charts it is clear that

1) The students with the highest mean scores in both the total SAT and the 'Critical Reading' section are those who took AP/Honors Courses. This group comprises 21% of the total. The mean scores of those taking Latin, Hebrew, or Greek (8% altogether) aren't even in the top fifth.

2) It's even worse in 'Mathematics' (one of Mr Cothran's four subjects); the students who fared best studied (in descending order) Chinese, Korean, AP/Honors Courses, Hebrew, and Japanese. Latin students are in 6th place (still in the top third though, due to the small number of students listed under the other languages).

The 'Writing' results are similar to those for 'Critical Reading'.

jah

Anonymous said...

If the above links don't work, try these:

http://i249.photobucket.com/albums
/gg215/jahigginbotham/mathSAT.gif

http://i249.photobucket.com/albums
/gg215/jahigginbotham/mathSAT.gif

http://i249.photobucket.com/albums
/gg215/jahigginbotham/mathSAT.gif


jah

Anonymous said...

http://i249.photobucket.com/albums/
gg215/jahigginbotham/criticalreadingCothran.gif


http://i249.photobucket.com/albums/gg215/
jahigginbotham/combinedSAT.gif

http://i249.photobucket.com/albums/gg215/
jahigginbotham/mathSAT.gif

Sorry 'bout that. Obviously I am learning.

jah

Anonymous said...

Correction: Since students may fall in more than a single 'Course Work or Experience' category, the total is 133%. Please adjust percentiles accordingly.

Also I imagine many students may choose Spanish since it is thought to be the easiest to learn (as well as most useful in day-to-day encounters).

One more check down the hall and I am out of here.

jah

Martin Cothran said...

Wow. Jah, go for it. That's good stuff.

While I'm making my way through Jah's stats, I wanted to issue a mea culpa on the citation. As Jah pointed out, these were the critical reading scores, not total SAT scores. That was plain from the graph, but I noticed I didn't mention it in the text of this post.

I did mention them on the earlier post from which I snipped the graph: http://vereloqui.blogspot.com/2007/08
/only-good-language-is-dead-language.html

But I appreciate you putting time into looking at this. These are helpful.

Martin Cothran said...

aagcobb,

I actually agree with you on the matter of any language being academically advantageous. My somewhat tongue in cheek assertion about dead languages being better than living languages is attributable, I think, to the fact that most ancient languages seem to be inflected. I'm working on another post to take account of some of Jah's points, and to posit another hypothesis about this that maybe Jah could test for me.

But I agree with your general point: ceteris parabis, studying any foreign language is better than not studying any at all.

Anonymous said...

Martin,

I've enjoyed your suggestion because its worthwhile to consider innovative ways in which schools can be improved. I think studying a foreign language in K-8, even if its not Latin, is a great idea, because thats when chldren's brains are most amenable to picking up language, and I believe studying a foreign language will help stimulate brain activity. That made me consider making the modest suggestion that an innovative school district could establish one foreign language to be studied in each school, be it latin, french, german or whatever, and every course in the school would be taught in that language. Total immersion in a foreign language is the best way to teach it, and in nine years by moving around the schools, children could learn multiple foreign languages! I like your suggestion of a limited number of courses to master in K-8, but I would add physical education because kids need physical activity, especially in this day and age when boys are inclined to just sit around playing video games.

Anonymous said...

jah, while it does provide some interesting data to look at, the big Rudner study on homeschooling that you cited has some fatal flaws. Here's what some of his research peers had to say about it:

Although the results of Rudner's analyses are likely valid for the particular population he studied, his insufficient attention to the data's bias has led to an erroneous picture of homeschooling.

The study sample was a large group of homeschoolers who used standardized testing services provided through Bob Jones University. Most homeschoolers are not required to administer standardized tests to their children, and many who do test would not choose BJU as their provider. The researchers noted that

The unfortunate result is an inaccurate portrayal of homeschoolers as a white, Christian, monolithic population.

Here's a link to their response: http://epaa.asu.edu/epaa/v7n13.html

If you want to see more current demographic information that is far less problematic, take a look at the most recent National Center for Education Statistics report that is done based on a random sample of American households:

http://nces.ed.gov/pubs2006/homeschool/TableDisplay.asp?TablePath=TablesHTML/table_3.asp

They found that homeschool families are slightly LESS well off than the general population. They found only about 7 percentage points difference in whether parents had continued their education beyond high school, not 38. 80% lived in two parent households, not 98%. Only 54% were of the two parents with only one working variety, not the 77% or so in Rudner's sample. In fact, 25% had two parents both working full time, and 16% were single parent families with the parent working full time. They didn't ask about television viewing or whether parents had teaching certificates.

So, while the data Rudner gathered is an interesting look at that particular sample, it isn't useful for generalizing to the whole homeschooling population because it is far from representative of it.

One Brow said...

Maybe you could be a little more explicit about the argument you think lacked evidence of critical thinking. Maybe you could even present an argument against it.
Why, the entire idea that showing a comparsion of the scores, without using a designed experiment or any controls, gives any indication at all the is a causitive effect between one and the other. A very excellent satire, sir. Your continuing to engage in it is sublime.

I gave one reason why I think Latin is a useful educational subject: students who take it perform better on the SAT. And there's more where that came from.
Is there as proof as I can present between the high correlations rates of soft drink consumption and new cases of polio arising in the 1930s?

Maybe, as I'm presenting these arguments for Latin, those who disagree could state exactly what evidence they would accept-
You now pretend to the impression that someone in here thinks Latin is not worth studying at all. Your satire is subtle, indeed. One might think you were serious.

-and then we could see whether they're really interested in getting to the right answer, or just making snippy remarks from behind pseudonyms.
Well, I don't use any psuedonyms personally, but they don't particularly bother me. I would think this is part of the satire if you had not mentioned it before.

Of course, I am rather fond of a nick name I have had for more than half my lif, so much that I consider it to be as real a name as any other. Still, anyone who was intereted in my legal name, could find it easily enough from clicking a couple of links or a Google search, were they motivated enough.

Anonymous said...

Hi sar,

The sole point of quoting parts of the Rudner study was to show that there are other factors than the mere study of Latin which can explain the higher test scores of students who have taken Latin. These factors affect home, public, and private schools.

I'm sorry, but I can't figure out how your comments have any bearing on this whatsoever.

I'll strive to be clearer in the future; that is one of the main reasons I write these comments.

jah

Anonymous said...

Aagcobb said.
... studying a foreign language in K-8 ... is a great idea, because thats when chldren's brains are most amenable to picking up language, ... one foreign language to be studied in each school, be it latin, french, german or whatever, and every course in the school would be taught in that language.
------
This is a side issue, but does anyone have any statistics or educated guesses of
1) what percentage of students of dead languages actually learn to converse in them?
2) what differences there are between learning to translate (and to do some composing) in a dead language and learning to be fluent in a living one?

jah

Anonymous said...

jah, if your purpose in quoting the Rudner study was, as you said,

there are other factors than the mere study of Latin which can explain the higher test scores of students who have taken Latin. These factors affect home, public, and private schools.

I can't figure out why you quoted a homeschool study, and a fatally flawed one, at that, rather than something relevant - say a study of the demographics of students who study latin, versus those who don't. In any case, I'm sorry to have confused you.

MY purpose was to correct the demographic information, lest people reading who are unaware of the self-selection bias in Rudner's sample think those are the real demographics of the homeschooling population. That kind of misinformation only perpetuates stereotypes, and doesn't serve to help anyone understand anything.

Rudner didn't examine SAT scores, and I can't help you with anecdotal information from my family either, since all three of my children entered college without taking SATs or ACTs (nor did they study Latin, though I should note that all have large latin vocabularies, as they were fascinated with word origins and also have taken interest in other subjects that use many latin words and phrases).

Anonymous said...

sar:I can't figure out why you quoted a homeschool study, and a fatally flawed one, at that, rather than something relevant - say a study of the demographics of students who study latin, versus those who don't.

Thanks for pointing out the error. I quoted the home school study since it was handy. [Presumably most researchers consider the same factors.] I didn't quote a relevant study because I didn't find any. I am not the one making the claim that studying Latin is a better way of improving SAT scores. I have reached no conclusion precisely because I have seen no data.

sar:In any case, I'm sorry to have confused you.

Yep, I need no assistance in that area :-)



BTW, this link is not active:
http://nces.ed.gov/pubs2006/
homeschool/TableDisplay

jah

Anonymous said...

BTW, this link is not active:
http://nces.ed.gov/pubs2006/
homeschool/TableDisplay


You just didn't catch the whole link. (It wrapped onto the next line.) Here it is again:

http://nces.ed.gov/pubs2006/homeschool/TableDisplay.asp?TablePath=TablesHTML/table_3.asp