what if the successful school had better education? ...
Assessing this might require cooperation from the unions, which has been granted in some cases, notably Chattanooga's "Benwood Schools" experiment with value-added:
The "Benwood schools" results are impressive. The percentage of third graders reading at or above grade level rose from 23 percent in 2001 to 36 percent in 2003. Across all grades, the percentage of students at or above grade level in reading/language arts rose from 57 percent in 2003 to 77 percent in 2005. Math achievement increased from 54 percent to 70 percent during the same period. In addition to raising student achievement, the Benwood schools report that filling their teacher positions has been easier, turnover has been reduced, and teacher morale has improved.http://kyl.senate.gov/legis_center/rpc/102505.pdf
... if a hospital had lots of patients dying...
Hospitals work hard to keep that from happening, in part due to ethical and liability concerns. If teaching is like doctoring, should teacher ethics be like doctor ethics? Klahr at CMU suggests engineering (rather than medicine) is the better analog for knowledge management in education. In this case, is Underwriters Lab (rather than the Food and Drug Administration) the better analog for liability control?
The implication for education? Should near-term efforts for improvement be focused on "clinical trials" (e.g. randomized research in education, http://www.tegr.org/) or continuous imrovement, a la Deming? Or should improvement mechanisms be chosen based on efficacy regardless of professional discipline. Consider:
Jerry Price, Chief Operating Officer, Sibley Hospital:
We are the hospital for demanding Washington residents who have extremely high expectations. Dr. Deming taught us how to meet those expectations through the quality improvement process.http://www.managementwisdom.com/bemileinvi.html
Diane Ravitch, "What if Research Really Mattered?"
I looked appreciatively at the medical doctors around my bed, grateful to be surrounded by men and women who have a common vocabulary, a common body of knowledge, a shared set of criteria, and clear standards for recognizing and treating illnesses. They have access to reliable tests that tell them what the problem is, and they agree on treatments that have been validated over a long period of time. The thought occurred to me that educators have something to learn from physicians.http://www.edexcellence.net/institute/publication/ publication.cfm?id=126
I'd really like to see more reporting on these ideas.
The question I would ask is, "How do you know professional development for your staff is working?" This encourages accountability while avoiding teacher-bashing. The ethnographic dimension matters quite a bit, but given the strength of polarized opinions, ("viewiness") data is invaluable.
But I felt it was missing something, so I suggested to the editor an alternative hypothesis: what if the successful school had better education? Better teaching, better curriculum. What if teachers at the more successful high school knew more about content, and more importantly were better able to transmit content knowledge to students.
Don't forget about SES either. What if both schools pretty much taught the same crap with teachers of roughly the same ability level with the only significant difference being that one school had a high SES and the other a low one. The low SES school is handicapped by SES and the high SES is unfairly advantaged.
If you plot achievement scores vs. SES for most schools in a ny given state you'll see than most of them perform within 1/4 standard deviation of the regression line (i.e., an educationally significant difference) (with the exception of the schools with very low SES).
I think Chris C did something similar on his site recently and S&P's school matters has also done a somewhat similar analysis at the state level.
The take-away is that there really isn't much of an educationally significant difference between most public schools once you account for SES variations. Public schools simply do not and have never done a very good job with lower SES (lower IQ) children.
KDeRosa,
Are you suggesting that these: lower SES (lower IQ) are synonymous?
No, merely correlated at about 0.422. See for example Lubinski,
Incorporating General Intelligence Into Epidemiology and the Social Sciences, p. 171, FIG. 2.
I'm sure Tangoman could add much to this discussion.
WE NEED TO REALLY BROADEN THE IDEA OF SCHOOL REFORM.... CONTEXT MATTERS...
The Shame of The Nation: by Jonathan Kozol.
Some thoughts....
”We owe a definite homage to the reality around us and we are obliged,
at certain times, to say what things are and to give them their right names." (Thomas Merton)
My work takes me on a regular basis into our urban centers.(nj) As Kozol so eloquently put it:
"Virtually all the children of black and Hispanic people in the cities that I visited, both large and small, were now attending school in which their isolation was absolute..."
Some Statistics: 2000-2001 School Year
Chicago: 87% black/Hispanic
Washington, D.C.: 94% "
St. Louis: 82% "
Phila: 78% "
Detroit: 95% "
Baltimore: 88% "
NYC: 75% "
The story is really worse then even these dismal statistics. For example, Kozol talks about PS 65 in New York city were they have 11,000 children enrolled k-8th grade in 1997. Of the 11,000 only 26 were white....a segregation rate of 99.8 %..... "two-tenth of one percentage point now marked the difference between legally enforced apartheid in the south of 1954 and socially and economically enforced apartheid.
Certainly matches up with my look at urban new jersey....
"During the 1990's, the proportion of black students in majority white schools has decreased....to a level lower then any year since 1968...Almost three fourths of black and Latino students attend schools that are predominantly minority," and more then 2 million, including more then a quarter of black students in the Northeast and Midwest,
"attend schools which we call apartheid schools" in which 90-100% of students are non-white."
"If you want to see a really segregated school in the United States today, start by looking for a school that's named for Martin Luther King or Rosa Parks."
"Does segregation of children in public schools solely on the basis of race, even though the physical facilities and other 'tangible' factors may be equal, asked the court in 1954.... deprive the children of the minority race of equal educational opportunities? WE BELIEVE IT DOES."
Thoughts…. ... Does school reform go here at all??? Why not????
Intellect,
Are you suggesting that these: lower SES (lower IQ) are synonymous?
KDeRosa gets it right, in that student IQ is the single best predictor of academic success. Better than teaching methods, teacher qualification, class size, etc. Critics are hard pressed to find a variable that has a higher correlation to academic success.
Now SES has correlative value to IQ, and this arises from two vectors. The first is that education (proxy for IQ)is valued in the economy and the second is straight from The Bell Curve analysis on assortive mating. IQ is heritable and we're seeing stratification in the mating game. Gone are the days when it was common to see a wide gap in parental IQ.
Social mobility is declining slightly because mobility is partly predicated on education. We've seen, here in Jenny's comment sections, a number of studies which demonstrate that the children born of lower SES have a lower chance of achieving academically. We know it's not funding, for the NJ experience shows that funding schools in low income neighborhoods to a level that surpasses wealthy suburbs doesn't work magic. We know that taking stellar principals who have created success in high SES schools hardly makes a difference. We know that moving low SES kids into neighborhoods and schools of higher SES (better teachers and environment) doesn't achieve significant results. None of these "easy" solutions work.
What does work, and we only know this on a limited sample size, is the KIPP experience, where the school day is 2 hours longer, the school week is 6 days, and the school year is extended for additional month. The extra time on task for the students equalizes for the IQ disparity.
To the general topic in Jenny's post, and this is a point that Eric raised above, the methods that are thought to be crucial really need to be rigorously tested and I'll go further in stating that the majority of education research that is published today needs to implement a more professional standard in how the research is conducted and the journals need to establish higher standards for publication.
It's easy to presuppose that the "magic solution" is better methods, especially when all of the previous "magic solutions" have turned out to be worthless. It's intuitively appealing to turn to the medical profession for an anology, however are we certain that the correlation between physician skill level and patient outcome is the relationship we're modeling with teacher skill and student outcome. What if we modeled physician skill level against a patient's height? Certainly, the skills of the physician would have some relevance to not inhibiting the patient's growth, but the physician would be hard pressed to treat the patient in such a manner that they would grow beyond their genetic limitations. In the same manner, a lousy teacher can handicap some students by hindering their ability to reach their potential, but a good teacher can't make the student exceed their own limitations. And that's what is the essence of the argument about teacher practice and methods - that they can indeed equalize outcomes, irrespective of IQ or SES.
Lastly, if we really think that teachers matter so much to student outcomes, especially in light of the growing body of research which shows how little the environmental lessons taught by parents matters, then it's really incumbent on the Educaion Establishment to prove the case. Arguing by default - all the other remedies didn't work, so it must be teacher methods because that's all that's left in our environmental grab-bag of factors, isn't too persuasive. Now, one factor that I think might have some influence, not great mind you, but still useful, is the teacher themselves, rather than their methods. I've suggested to Jenny that she take her doctoral dissertation data and run a statistical analysis of student outcome against teacher IQ, for I think that that would be revealing. My hypothesis is that a sharper teacher is better able to analyze student roadblocks and then circumvent them, whereas a dimmer teacher might be stumped or not even recognize that the student is at a conceptual impasse until it's too late.
Sorry for my long windedness.
Thoughts…. ... Does school reform go here at all??? Why not????
School reform won't fix the problem. More time on task for the students has a better chance of success, but now we're dealing with a separate system of education for minorities. Which is preferrable - maintain uniform standards of instruction and see the devastation in the minority community, or acknowledge the problem and create a separate system in order to equalize the outcomes as best we can. That's the first order analysis.
The second order analysis arrives when taxpayers object to differential standards and demand that non-minority students get the same "enriched" school curriculum and time on task. Now the gap will widen again.
Take you pick - which unpleasant truth do you want to grapple with? 1.) Equal potential leads to unequal outcomes. 2.) Unequal potential leads to equal outcomes.
Regarding the Benwood Schools case that Eric notes above. If you peruse their site you can find this page which notes:
They had TVAAS (Tennessee Value Added Assessment System) three-year average scores in the top 25% of all teachers in the county.
So, what is to be done about the 75% of teachers who aren't top performers? Should they move to Lake Wobegone?
Further, on another page we find:
Other strategies include reorganizing the school day to allow concentrated study of reading and writing, after-school and summer school programs for all students, a full-time parent involvement coordinator, mentoring programs for new teachers, and special enrichment activities for students.
So, it took $7.5 million (5 year commitment) to raise the scores of 9 elementary schools from 22.6% performing at grade level to 35.9%. What I'd be interested in knowing is how much of the $167,000 annual per school allotment is overhead for the Benwood research program and how much goes into school resources. Additionally, how much of a budget increase are we looking at here and does the additional funding for added class time come from district resources or the Benwood grant?
STANDARDS AND VARIATION
Seems to me we have framed the debate in education as "either/or";
"black/white" ; this/that....
There is i believe another way...
There can be various ways to demonstrate quality with-in a standard!
I believe we need to think a bit differently in the context of standards in a country that is becoming more and more diverse...
"We are in an Age of High Standard Deviation." (Tom Peters)
Let is re-imagine what could be and get out of this low level box...
be well... mike
What does work, and we only know this on a limited sample size, is the KIPP experience, where the school day is 2 hours longer, the school week is 6 days, and the school year is extended for additional month. The extra time on task for the students equalizes for the IQ disparity.
What also works includes Direct Instruction (capital d and I) and Success for All. All three instructional programs have the same characteristics:
1. lots of direct instruction
2. heavy on formative assessment
3. providing additioanl tutorial/instructional resources for the lower performers.
For example, in Direct Instruction the lower performers are given additional instructional time so they can keep pace with the middle (average) group. The high performing group is usually slowed down by receiving less instructional time (they can also proceed at the natural pace through the curriculum butthen they'd eventually pull far ahead of the lower groups). The key, though, is that all groups must master all the material along way. In this way, the low performers are able to learn at the same pace as the higher performing groups.
See Summary of Presentation to Council of Scientific Society Presidents,
The Components
of Direct Instruction, The Journal of Direct Instruction Summer 2003, pp. 98-102, and
This thread at Kitchen table Math.
The problem with the KIPP model is that they begin with middle school students and get them after they've fallen well behind in the elementary school. As a result, they spend a lot of that extra time remiating the students to get them back up to speed -- a daunting task to say the least.
I'd say the two critical components is sufficient time on task so all students master the material (as Tangoman indicates), but also a sufficiently efficient curriculum so that the low performers are able to master the material in a reasonable period of time.
Your typical american instructional program (heavy on constructivism and discovery learning, light on practice and claity of instruction) fails both these requirements. Most curricula are so light on practice, they need to be heavily supplemented by the in-service teachers (non-experts in instructional design) to provide additional practice (and it's usually not enough since most teachers have no idea what mastery learning is). And, their consttructivist curricula are very inefficient in transmitting knowledge to the students that only the high-IQ high performers are the only ones who sufficiently learn the maerial to a level that approaches mastery. The result is a sorting machine by IQ wherein the high IQ kids go to college and the rest fall further and further behind as amply demonstrated by the decline between 4th and 12th grades on such tests as the NAEP.
The high performing group is usually slowed down by receiving less instructional time (they can also proceed at the natural pace through the curriculum butthen they'd eventually pull far ahead of the lower groups).
This, to me, speaks loudly of the intellectual bankruptcy of American education. Instead of striving to educate each student to their potential even this program purposely handicaps top students so that the fiction of all groups achieving at the same rate can be maintained.
It is more complicated than that. This data comes primarily from studies at low SES schools where the high performing group is typically tiny and mobility is typically about 25%. As a result there was rarely enough students to support their own class.
Of course, when it comes to the high performers in your typical public school they are typically given "enrichment" instead of being allowed to accelerate to their potential, but that is a separate issue.
What a GREAT analogy comparing a great school with an effective hospital. I left the high tech corporate world to become an 8th grade school teacher in Arizona and I guess I tend to look at effeciency a lot different than my fellow educators. It blows me away in the differences between the teachers that care and the teachers that collect a paycheck knowing they pretty much cannot be fired.
I know it is so taboo in education to even utter the words "merrit based pay" but I think until we find a way of compensating the great or even good teachers and try and push out the ineffective, our profssion will continue to be slammed by the critics - justifiably so.
GREAT site by the way.
Strausser
http://strausser.blogspot.com
The most "justifiable slam" would be NEA's retreat from new unionism—which apparently left when Weaver replaced Chase as president.
Has there been much progress with peer review? The NEA's article on "Enhancing Teacher Quality: Peer Assistance and Review" now appears only in thegoogle cache.
Commenting by HaloScan
