July 8, 2008

Relying on the Coleman Report

I'm embroiled in a bit of a debate over at Russo's with frequent commenter (though not here) John Thompson over the conclusions I drew for the Baltimore first grade reading achievement study posted below.

The argument revolves around the popular premise that the effects of low socio-economic status cause low student achievement. I argue that large-scale empirical research like the Baltimore study and Project Follow Through refute such causal interpretations which are merely based on the correlational data we have on SES and student achievement, like that found in the Coleman report.

John, however, raises the following point regarding the Coleman Report:

You should realize you wouldn't be digging yourself into such a deep hole if you would back off from gratuitous attacks on others, like the people who issued the Bolder Broader challenge, and didn't try to refute the Coleman Report with sweeping comments.


I am not trying to refute the Coleman report. I am refuting the causal implications John and his Broader, Bolder allies draw from the Coleman Report.

The Coleman Report of 1966 was based on achievement data on over a half million students. The Report noted the disparity that existed between at-risk students and those not at-risk. The report compared schools of equal physical characteristics serving both groups. Coleman's finding was that money spent on smaller classes, laboratories, counseling, higher teacher salaries, and higher teacher qualifications were not correlated with academic achievement.

If the physical characteristics of the schools and all the other factors made no difference in student performance, there still remains the possibility that the instruction at-risk students received was inadequate. This possibility, which has been shown to exist in such large scale controlled research studies as Project Follow through and the Baltimore study serves as clear indictment of the educational system and a clear premise that the instruction provided to at-risk kids needed to improve.

The premise goes like this: if the instruction is inadequate, school factors won't matter and SES effects will predominate. Conversely, if the instruction is adequate, school factors will predominate rather than SES factors. This premise is consistent with the Coleman Report findings which examined an educational system in which inadequate instruction predominated, before we discovered that adequate instruction did, in fact, exist.

This data has not been refuted by the Broader, Bolder proponents. They simply ignore it and hopes that no one will notice. This is precisely the reason I always bring it up--to watch them go through rhetorical gymnastics trying to explain it away.

17 comments:

Parry Graham said...

I want to make sure I understand your argument. Is it your contention that, at the time of the Coleman report, inadequate instruction was pervasive throughout the entire public K-12 education system nationwide? In other words, inadequate instruction was the norm across the country, and the learning that occurred happened in spite of, rather than as a result of, efforts by public school teachers?

KDeRosa said...

Parry, my argument is that typical instruction was and continues to be largely inadequate for teaching at-risk populations and that such instruction was and continues to be the norm throughout the entire K-12 education system nationwide, but for one-off higher-performing schools.

Yes, some learning did and does occur, but the learning is usually insufficient for at-risk students to maintain a grade-level pace.

In the 70s and 80s we learned how to improve instruction and how to replicate those instructional programs. Those prpgrams were tested in large-scale controlled experiments which found that at-risk students could achieve natinal norms at least in elementary schools which was the level these instructional programs were developed for. these programs have not been adopted by many schools ans there penetration remains low.

For these reasons the SES/student achievement correlation persists.

RMD said...

does anyone know where to get a copy of the coleman report?

KDeRosa said...

i thinks it's oop

Anonymous said...

The Coleman Report is available for download at the following URL:

http://webapp.icpsr.umich.edu/cocoon/ICPSR-STUDY/06389.xml

Anonymous said...

The key findings of the Coleman Report relevant to Ken's debate with John Thompson at Russo's are the following:

"The quality of teachers shows a stronger relationship to pupil achievement. Furthermore, it is progressively greater at higher grades, indicating a cumulative impact of the qualities of teachers in a school on the pupils’ achievement."

"It should be noted that many characteristics of teachers were not measured in this survey; therefore, the results are not at all conclusive regarding the specific characteristics of teachers that are most important. Among those measured in the survey, however, those that bear the highest relationship to pupil achievement are first, the teacher’s score on the verbal skills test, and then his educational background-both his own level of education and that of his parents. On both of these measures, the level of teachers of minority students, especially Negroes, is lower."

IMHO the Coleman Report has been widely misread and misinterpreted. Judge for yourselves.

Anonymous said...

Are people really arguing about whether or not the conclusions of a 40 year old report need to be revisited and revised in light of recent data.

Do Bigger, Bolder and its proponents realize that Pat Moynihan, one of the authors of the report (see "On Equality of Educational Opportunity", Random House, 1972) used the findings of the report to argue that additional resources in schools would not overcome the effect of poverty? That Jencks had a chapter in the same book (nuff said), as did Hanushek, who is regularly marched out by the right in opposition to increased education spending?

Anonymous said...

I have made two basic points. You present a simplistic diagnosis of educational failure, as you continue to blame everything on "bad" instruction. Secondly, you present an even more simplistic solution, "good" instruction. Actually my point you cited was nor regarding the Coleman Report. My point was your intemperate rhetoric. I guess I got too immodest because I figured that freshmen can't be that much harder to deal with than an adult who is always sure that he is right.

Numbers are numbers and reality is reality. Numbers can be a wonderful tool for understanding reality. Numbers can even be better tools when used in controlled experiments. Until you learn some modesty and respect for scholarly discourse, I question whether you will master the art of teasing real understanding from social science.

I'll give one simple example. You study the numbers in regard to the pitches thrown by Little League pitchers, and don't worry about whether the coaches are damaging the kids' arms by throwing too many curves and total pitches. At the end of sixth grade, the win/loss numbers may look good.

What measures have you taken to study the damage caused by Direct Instruction, scripting, and excessive testing? How much damage would you accept for one of your children in order to raise scores for another? Don't just come back with your standard reply saying that bad DI, bad scriping, and bad testing is bad, while you support good instruction which does good things. Tell me how you have contemplated the issue. How do you approach the trade-offs that are inherent in any approach, but which are particularly worrisome with your approach. By the way, for me it is personal. Our 100seniors were the 8th grade class of nearly 300 that earned banner headlines for increasing test scores dramatically. We can debate whether the damage was worth it,or how it could have been minimized, but you can not honestly deny that it exists.

You, like many accountability hawks, are gambling everything on "the Head" while neglecting "the Heart." Then you ridicule the practical wisdom of teachers (or in my case a teacher with an extensive academic background) who stress emotional, peer pressure, and physical dynamics. You may think that you are on the side of the angels, helping poor children of color. I am offended, however, by the way you seem to diminish the complexity and the humanity of my kids. Without developing some modesty and some appreciation of the full glories of the learning process, you could never master the art of teaching.

Finally, I love to learn from research. But it is not even close to being as useful as dead reckoning in wrestling with either classroom instruction, or the diagnosis and the solution of educational problems.

Jo Anne C said...

John Thompson,

How does teaching students to read diminishes their humanity?

Can you please explain the damage that you think DI is causing students?

Anonymous said...

"What measures have you taken to study the damage caused by Direct Instruction, scripting, and excessive testing? "

A billion dollar study that involved over 200,000 students that shows that DI students have higher self esteem and stronger cognitive skills than child-centered programs that emphasized self esteem and cognitive skills.

ari-free

RMD said...

Mr. Thompson,

If by "damage" due to scripting you mean totally psyched about school and learning, and very capable to face the world, then DI kids are truly damaged.

jh

P.S. Please go visit a well-run DI school and see the "damage" for yourself. Ask the kids what they think. And if you think their being brainwashed to answer that they like school, then you give DI too much credit. No one can put words in kid's mouths.

Tracy W said...

You present a simplistic diagnosis of educational failure, as you continue to blame everything on "bad" instruction. Secondly, you present an even more simplistic solution, "good" instruction.

Okay, please explain in words of small syllables, so our little minds can understand it, how come, if the solution of "good" instruction is simplistic, schools have seen drastic increases in educational performance by improving instruction, not by changing their student intake?

What measures have you taken to study the damage caused by Direct Instruction, scripting, and excessive testing?
This was studied by Project Followthrough. Kids taught by Direct Instruction had the highest results on basic skills, cognitive skills, and affective skills (the emotional stuff).
http://www.projectpro.com/ICR/Research/DI/Summary.htm

As for excessive testing, for someone who is so scathing about Ken's mental ability, you show a remarkable lack of understanding yourself. Excessive *anything* causes harm. That's the definition of the word excessive. Drinking too much water causes death. If we are to say anything useful, the right question is "Does level x cause harm?"

Anonymous said...

of course I'll check out your cites.

But if I hear you guys correctly, then I don't even have to. "Good" is DI. And the DI web cite proves that it is good.

Personally, I don't have a dog in this fight, or I didn't until I got into the unfortunate habbit of debating this.

Frankly, when I bring my layperson's curiosity to a debate over Success for All, I don't get this trepidation. They make good points, and they don't sound like a cult.

I would have read your stuff and filed it away as FYI if it had not been for the nonstop attacks on everyone else.

For all I know, your silver bulet approach might be a once-in-a-lifetime deal, and you've got the one true gospel. But when you get out of elementary school and you attack the solutions that make sense for my postage stamp of education, I don't see your philosophy as credible.

But I will check your cites.

KDeRosa said...

John, you're not bringing your best game to this debate. But let me see if I can respond to the points you have made that others haven't already addressed.

Complexity. Simple explanations are preferred over complex explanations having the same predictive capacity. As such, your complex explanation, which I'll note that you have failed to set forth so far, would only be better than my simple explanantion if it more accuratetely reflects/predicts the actual data. Further, the nature of my simple explanation implies the nature of the solution.

Ad hominems: I have a high tolerance for ad hominems, but you should realize that they don't prove your point. How was my rhetoric "intemperate"? My rhetoric is either wrong or it is right. If it was wrong, then explain how it was wrong.

Numbers and reality. If the numbers don't accurately reflect reality you need to explain why and provide support. If you have some superior understanding, then you need to provide some support and explanation for that.

Your baseball analogy: Your pitches thrown question has an emperical answer which implies a controlled experiment. If "damage" were being done, then that damage should show up in subsequent statistics. In any event, the difference between your baseball example and the research on elementary education programs are not completely analogous. In education the instruction received in middle school is a confounding variable that doesn't exist in your baseball example.

DI and damage: You have not yet established that DI causes damage. Until you can establish that fact it's a non-issue. The fact that some test prep methods supposedly cause some unidentifiable damage does not mean that DI does.

Your practical wisdom: This is an appeal to authority, another logical fallacy. Again, if you posess some practical wisdom or superior knowledge that I don't you should be able establish that wisdom/knowledge and present a coherent argument based thereon. So far you have failed to do this.

"Good" is DI. And the DI web cite proves that it is good. Some research on DI has been conducted by DI advocates. Much of it, like Project Follow Through and this Baltimore study, was not, but rather was conducted by independent reseachers.

DI as a cult. I use DI as an example because it has the largest research base and that research is readily available on the net. It also has the best evidence of success. I could just as easily use Success for All or other instructional programs as my example, but if figure why bother.

"But when you get out of elementary school and you attack the solutions that make sense for my postage stamp of education, I don't see your philosophy as credible": What solutions are these and what evidence, besides your own subjective belief, do you have that they are or will be successful. Why isn't my philosophy credible? Why doesn't what happens in elementary school relevant to what happens later? Why do your solutions "make sense," where is your evidence?

I look forward to your answers, John. The other commenters and I are sticklers about backing up your assertions with real evidence. You need to support your arguments. if you think you have some special authority that makes you more knowledgable than the rest of us, you need to prove it.

Anonymous said...

Mr. Barone's assignment is to read his response, and report on how many and which logical fallacies he employs.

And to the world at large, just because it makes me throw things, "cite" is a verb; "citation" is the noun ("quote" and "quotation" are the same).

Anonymous said...

yeah, maybe we are a "cult" because we get people commenting on DI and how scripts ruin kids, etcetera, etcetera, and . .

They have no idea what they're talking about!

so yes, if you get a strong reaction here, it's because we've seen DI work, and yet people continue to be threatened by it and criticize it, and yet they're not willing to do their homework on it

when you see kids charged up by DI and happy and psyched because they're learning a ton, then come back again and make your points

Anonymous said...

I'd want to Know Coleman's Analyses (not the databases)are available on the Internet.

Thanks