Key Issues in Studying Charter Schools and Achievement: A Review and Suggestions for National Guidelines
The Charter School Achievement Consensus Panel; Julian Betts and Paul T. Hill, Principal DraftersCenter on Reinventing Public EducationMay 2006
Key Issues in Studying Charter Schools and Achievement: A Review and Suggestions for National Guidelines
The Charter School Achievement Consensus Panel; Julian Betts and Paul T. Hill, Principal Drafters
Center on Reinventing Public Education
May 2006
For charter supporters, August 17, 2004, is a day that will live in infamy. That day, the New York Times unleashed its AFT-spun appraisal, based on the National Assessment of Educational Progress, that charter schools were "lagging behind" traditional public schools. (See here.) All hell broke lose. Full page ads denouncing the article were taken. Rival studies were rushed to print. It was a fight to the death. Reason, facts, and rigorous analysis, it seemed, couldn't make much difference. And that was a big problem. So professor Paul Hill and the leaders of several foundations set out to create a space where research would triumph over rhetoric. The result is the National Charter School Research Project (of which Fordham is a minor funder), housed at his Center, and this white paper is one of its first major contributions. It can best be described as a peace offering, the conditions of a possible truce in the charter research wars. It focuses (probably too narrowly) on how to best answer one key question: "whether students in charter schools are learning more or less than they would have learned in conventional public schools." The paper walks the reader through seven major types of studies (such as experimental and "fixed effects") and considers their strengths and weaknesses in addressing this question. No method is perfect, argues the paper; even well-regarded experimental studies have limitations (namely, that their results might not be applicable to the entire charter universe). But some methods are definitely better than others, and studies that look at average student achievement at one point in time-as did the AFT's-are shown to be the least reliable. But even as the field matures, the research is not necessarily getting any better. The panel ranked a majority of the charter studies published between 2001 and 2005 as "fair" or "poor." (Unfortunately, the reader is not told which studies received which rating.) If researchers take this paper's recommendations to heart, the track record for the next few years should be much stronger. Before you launch your next charter school study, read the white paper here.
Teaching Inequality: How Poor and Minority Students Are Shortchanged on Teacher Quality
Heather G. Peske and Kati Haycock
The Education Trust
June 2006
Education Trust collaborated with teams in three Midwest states (Illinois, Ohio, and Wisconsin), and in the largest school districts in each of those states (Chicago, Cleveland, and Milwaukee), to examine the distribution of high-quality teachers across schools. The teams will put out full reports later this summer, so this small report just highlights selected findings, which are depressingly familiar: students in predominantly poor and minority schools are less likely to have a high-quality teacher in their classrooms. Sometimes these teachers are novices (teaching less than three years) or relatively inexperienced (less than five years experience), and sometimes they have no background in the subjects they teach. In Wisconsin, for example, 25 percent of teachers in the highest-minority schools were novices, while the state's lowest-minority schools had 10 percent novice teachers. Low-performing schools in the Badger State "had approximately twice the percentage of novice teachers as high-performing schools." Why is this so? Partly because federal law contains a loophole that "allows districts to ignore disparities in teacher qualifications across different schools," and because teachers in urban schools are given no salary bonus for serving in more difficult environments. What's more, teachers with seniority oftentimes have first dibs on open jobs in cushier schools. Veteran teachers, therefore, abandon tough urban schools, leaving them mostly to rookie teachers. The goal of educational equity is thwarted. Is there any hope? Next month, every state is required to submit to the federal government a plan to make 100 percent of their teachers highly-qualified, with a special focus on high-poverty and high-minority schools. We already know that states have, in the past, used games to drastically inflate their highly-qualified numbers. This report hammers home exactly which students are most hurt by those actions. Read it here.
Why American Students Do Not Learn to Read Very Well: The Unintended Consequences of Title II and Teacher Testing
Sandra Stotsky
Third Education Group Review
Vol. 2, No. 2, 2006
At the end of the day, our children's ability to read depends upon whether ed school instructors know how to teach the science of reading. A growing body of evidence suggests they are failing (see here.) So how, then, do their graduates continue to receive licenses to teach elementary school or to serve as reading instructors? This new paper by Sandra Stotsky provides a simple but disturbing answer-the licensing exams prospective teachers must pass don't test their knowledge of research-based reading. Stotsky analyzed exams offered by Educational Testing Service (both national and state exams), National Evaluation Systems, and the American Board for Certification of Teacher Excellence, to see how much of their tests covered instruction in phonemic awareness, phonics, and vocabulary building. Parents whose children are struggling to read will find little solace in Stotsky's results. She reviewed eleven tests used for certifying elementary teachers and found that most failed to cover the basics of scientific-based reading instruction. Three of the four PRAXIS II exams reviewed, required by 46 states, only dedicate between 1 and 7 percent of their content to this critical material (though the fourth PRAXIS fared substantially better). "These aspects of reading instruction receive such minimal attention," Stotsky writes, "that test-takers could fail every question...and still pass the test no matter where the passing score is set." The record's no better for tests administered to reading teachers, reading specialists, early childhood teachers, or special ed teachers. Gladly, some of the customized state tests (California's and Massachusetts's, for example) are much stronger, as is the ABCTE exam. Alas, that's the only test studied that is not required for certification, but is merely an alternative route to teaching. Pity. Read the report here.
Fat American thighs
This week, Mike and Rick chat about how to walk, chew gum, and teach science at the same time; the D.C. teachers' union's sea change; and the language of love. Paul Peterson (not the butterfly collector) grants us an interview, and Education News of the Weird is a distant hum. All in under 20 minutes-it's a celebration!
Spicy Chile
Chile's country-wide education protests are now over. The fallout from the three-week crucible that saw nearly 800,000 students take to Chile's streets (sometimes violently), however, will not be soon forgotten. The walkouts began innocently enough, with students asking the government to provide free bus passes and to waive university entrance exam fees. But protests grew angrier as high school students insisted that the government centralize school management and pour profits from the country's copper trade into education. The financial inequity between wealthy private schools and public schools has only worsened over the past decade, and those disparities provided ample fodder to keep the youthful anger stoked. Though non-student groups (unions, anti-government groups, etc.) certainly pushed up the level of violence, the students' voices ring loud and clear-they want better schooling and a better life. We wonder how many American students are as passionate about their own educations.
"How Chile's growth skipped its schools," by Jen Ross, Christian Science Monitor, June 14, 2006
"Will Chile's President Flunk the Test?," by Mary Anastasia O'Grady, Wall Street Journal, June 9, 2006 (subscription required)
"Chile students clash with police," BBC News, June 6, 2006
"Schools out," The Economist, June 1, 2006 (subscription required)
Group think
In last Sunday's New York Times, Jeffrey Rosen, a well-known legal scholar, wrote a longish article related to the Supreme Court's decision to hear two cases that challenge racially influenced admissions policies in public schools.
Rosen didn't pick sides in the upcoming battles. Rather, he explained why the growing nuance of racial issues, especially for conservatives, has begun sowing political divisions where once only unity existed.
Let's examine Rosen's point from the perspective not of conservatives or liberals, but of education reformers. In the education community's debate over racial issues, there also exists much nuance and substantial division. Especially since the advent of No Child Left Behind and its goal of closing racial achievement gaps, those divisions have grown starker and exposed some seeming contradictions.
For example: How can one support racial classifications in NCLB, but eschew them as, say, part of K-12 public school admissions?
For an answer, we must start with NCLB, which derives its name from the idea that all children are capable of faring well academically. To that end, the law demands that schools, districts, and states disaggregate their student test data, breaking them down by subgroups. Racial subgroups are a key part of the system.
NCLB's system of disaggregation makes good sense. A chasm exists between the academic performance of America's racial groups, and closing that gap is essential for the social harmony and economic competitiveness of the nation. Thus, it seems foolish to not record data on the gap and closely monitor the progress of racial subgroups which are having trouble meeting academic marks.
Yet, when it comes to the school admissions policies which will soon be examined by the Supreme Court, race should play no part. In last week's Gadfly, we wrote: "Why should districts obsess over the racial make-up of their schools when they should be stressing about what's actually going on (or not) inside their classrooms? Especially when most urban systems are ‘majority minority,' isn't it time for them to focus on achievement first?"
A contradiction? Not at all.
To be sure, the disaggregation of racial group data mandated by NCLB is not unproblematic. The system lumps all individuals into a handful of broad racial groups (always a dicey proposition), and it unquestionably shows that, as a group, black and Hispanic students do significantly worse academically than their white and Asian counterparts.
That can be a hard pill to swallow. But what is the alternative? Ignoring a blatant problem simply because it seems safer or less unsettling?
No. This alternative is feckless. However, there will come a day when educators can track the achievement of individual students over time, rather than groups, allowing a laser-like focus on the needs of all children who are trailing behind-many of whom are likely to be black or Hispanic. An accountability system that ensures strong achievement gains for all such students will mitigate the need for racial classifications. For now, we rely on an imperfect system, but one that has done much good.
On the other hand, diversifying schools by race has outworn its necessity. Jonathan Kozol will disagree, but today's schools suffer far less from a lack of racial diversity than from a lack of educational competence. Minority youngsters will have their future prospects brightened far more by learning a strong, core curriculum than by eating lunch with a planned, multi-racial community of students.
When possible, racial classifications ought to be avoided, mostly because they tell us so little about individuals, and because they can so easily drift into unhealthy territory. They should be used only when necessary, and then with the utmost caution and respect.
Confronting our nation's achievement gap and creating a unified and strong academic background for all American students warrants the careful use of racial data groups. Engineering diverse schools is far less compelling.
Lose a mile, give an inch
Give them credit for progress-even if it's painfully incremental. In a decidedly uncharacteristic move, the Washington D.C. Teachers' Union approved a new contract that will introduce bonus incentives for teachers and give principals more autonomy at a handful of pilot schools. The union's president, George Parker, says that the changes are necessary if the city's public schools are to remain competitive with D.C. charters, whose attendance has jumped by 70 percent over the past five years, giving them approximately 25 percent market share of the public school population. D.C. Superintendent Clifford B. Janey said, "You can't move a reform agenda unless you have a genuine relationship with the union." What he couldn't say is that such a relationship is much more likely when the union is feeling the heat. Parker is more honest: "The landscape has changed. Our parents are voting with their feet. As kids continue leaving the system, we will lose teachers. Our very survival depends on having kids in D.C. schools so we'll have teachers to represent." Eureka! But Mr. Parker, we're doubtful that allowing reform at less than dozen schools is likely to stem the tide. Still, you get an "A" for attitude.
"District Briefing: Teachers Approve Contract," by V. Dion Haynes, Washington Post, June 8, 2006
"Bonuses, Relaxed Rules Proposed," by V. Dion Haynes, Washington Post, June 6, 2006
Buzz kill
A new front has opened in the cell phone wars. In a ring tone realm once ruled by Fur Elise and the latest jam from 50 Cent (aka Fiddy), a new craze has taken center stage-"Teen Buzz," the almost-silent cell phone alert. Almost-silent because youngsters can hear it while older folks (who typically lose their ability to discern higher-pitched sounds) cannot. The original sound was developed by a Welsh security company which sold it, not as a ring tone, but as a way for British shopkeepers to keep loitering youths away from their stores. The tone is emitted at high volumes from the storefronts, allowing adults to buy groceries in peace while young hooligans and their dirt bikes are quickly disbursed. (Reports make no mention of Moms and Dads who shop with baby Johnny in tow.) Now the onetime weapon has been usurped by the youthful masses. Kids are using the high pitch to receive text message alerts in class, and their teachers are none-the-wiser. Losing the tech-savvy battle to China and India? Not if these youngsters have anything to say about it.
"Students Find Ring Tone Adults Can't Heart," Associated Press, June 12, 2006
Condition of education 2006
The National Center for Education Statistics doesn't always do right by its annual "Condition of Education" report (COE), which has sometimes been humdrum and sometimes dizzy from pro-administration spin. But this year Education Statistics commissioner Mark Schneider and his team have produced an uncommonly interesting and, at almost 400 pages, sizable report. It differs from the group's facts-only "Digest of Education Statistics" in that COE points to trends, patterns, and notable changes, thus making it informative, not just informational.
Yes, its value is diminished by the nation's archaic ed statistics apparatus. Even the latest numbers, for example, are nearly always a year or two old (the finance data are typically 3 years old). Moreover, important data is simply unavailable (e.g., the cost and value of teacher benefits); the dropout and graduation definitions remain murky; and Congressional constraints on what can be asked of kids and parents means some information cannot be gathered.
Still, there's much here of value, both for K-12 and higher education, beginning with an excellent (if depressing) summary of comparative international data on academic achievement, which generally show U.S. students reading about as well as their counterparts in other countries but faltering by age 15 in both math and science. (In 4th grade, young Americans do OK across the board.)
These items in the 2006 COE caught my eye:
--Despite much recent ferment about early childhood education, participation in pre-K programs seems to have plateaued over the past 15 years at 50 to 60 percent of 3- through 5-year-olds and remains slightly lower for poor children.
--Total pre-K-12 enrollments are rising slowly, due both to increasing birthrates and to immigration, but it's lumpy across grades and regions. Over the next five years, for example, high school rolls will shrink a bit while the pre-K-8 ranks will grow. And regional differences are significant, with the Northeast and Midwest continuing to lose pupils while the South and West add them. (Surely that helps to explain why battles over things such as charter school caps are so much bloodier in New York, Ohio, and Michigan than in Colorado, Florida, and California, where conventional schools can lose students to charters and still not face declining enrollment overall.)
--Private schools cling to about a 10 percent "market share" as they have for several decades, but the Catholic portion of that continues to shrink (from 55 percent in 1990 to 46 percent in 2004), offset by growth in other religious and secular schools.
--One in five American school children is now Hispanic, but this, too, is really lumpy: less than 7 percent in the Midwest versus 39 percent in the West. Meanwhile, black enrollments are essentially stable at 16 percent. The Hispanic share of K-12 enrollments first exceeded the black share in 2001.
--Seventy percent of black 4th graders are poor (eligible for free/reduced lunch)-and 72 percent of black kids are enrolled in schools where most pupils are poor. For Hispanic 4th graders, both numbers are 73 percent.
--The share of 5- through 17-year-olds who speak a language other than English at home rose steadily from 1979 to 2000, but over the past few years the number seems to have stabilized at 18 to19 percent.
--The number of women enrolled in undergraduate colleges surpassed the number of men for the first time in 1978, and that gap keeps widening. In 2004, colleges had four women students for every three men. In graduate school, it's nearly 3 to 2, and 2005 was the first year that women outnumbered men in professional schools.
--Young Americans continue to have overly optimistic notions of how far they will go in formal education. In 2004, more than two-thirds of 12th graders expected to earn a bachelor's degree or more. (Indeed, more expected to complete graduate/professional school than to stop after undergraduate training.) This included 67 percent of black and 57 percent of Hispanic 12th graders. While immediate college matriculation rates are almost that high (69 percent for white high school grads, 63 percent for blacks, 62 percent for Hispanics), college completion rates are far lower: in 2005, only 29 percent of the 25- through 29-year-old population had bachelor's degrees or higher. Only 18 percent of blacks and 11 percent of Hispanics in that age group held such diplomas. (Note, though, that the latter figures are percentages of the entire age cohort, not of high school graduates.)
--A huge fraction of U.S. school children now attend "schools of choice": more than half of K-12 parents reported in 2003 that they had the "opportunity" to send their kids to a "chosen public school." It appears that 15 percent actually sent them to a "chosen" public school (including charter schools), to which must be added the 10 to 11 percent in private schools, the 1 to 2 percent who are home schooled, and what seems to be 24 percent who moved into their current neighborhood because of the schools. Though there is some duplication in those numbers, it looks to me like a third to a half of U.S. schoolchildren's families are exercising school choice of some sort.
--Class-size data are elusive but it's easy to calculate the student/teacher ratio in U.S. public schools, which has been below 17 to 1 since 1998. Even allowing for special ed, AP physics, and 4th year language classes with 5 kids in them, one may fairly ask why a country with fewer than 17 kids per public-school teacher remains obsessed with class-size reduction. (When I was in fifth grade, the national ratio was about 27:1.)
--Total expenditures per pupil in U.S. public schools reached $9,630 in 2003-up 23 percent in constant dollars over the previous 7 years. At 17 kids per teacher, that translates to almost $164,000 per teacher. Why, then, are teachers not terribly well paid? Because (using the NCES categories) the U.S. spends barely half of its school dollars on "instruction."
And that's just the tip of the COE iceberg. You really should peruse it for yourself. You can find it here.
Teaching Inequality: How Poor and Minority Students Are Shortchanged on Teacher Quality
Heather G. Peske and Kati Haycock
The Education Trust
June 2006
Education Trust collaborated with teams in three Midwest states (Illinois, Ohio, and Wisconsin), and in the largest school districts in each of those states (Chicago, Cleveland, and Milwaukee), to examine the distribution of high-quality teachers across schools. The teams will put out full reports later this summer, so this small report just highlights selected findings, which are depressingly familiar: students in predominantly poor and minority schools are less likely to have a high-quality teacher in their classrooms. Sometimes these teachers are novices (teaching less than three years) or relatively inexperienced (less than five years experience), and sometimes they have no background in the subjects they teach. In Wisconsin, for example, 25 percent of teachers in the highest-minority schools were novices, while the state's lowest-minority schools had 10 percent novice teachers. Low-performing schools in the Badger State "had approximately twice the percentage of novice teachers as high-performing schools." Why is this so? Partly because federal law contains a loophole that "allows districts to ignore disparities in teacher qualifications across different schools," and because teachers in urban schools are given no salary bonus for serving in more difficult environments. What's more, teachers with seniority oftentimes have first dibs on open jobs in cushier schools. Veteran teachers, therefore, abandon tough urban schools, leaving them mostly to rookie teachers. The goal of educational equity is thwarted. Is there any hope? Next month, every state is required to submit to the federal government a plan to make 100 percent of their teachers highly-qualified, with a special focus on high-poverty and high-minority schools. We already know that states have, in the past, used games to drastically inflate their highly-qualified numbers. This report hammers home exactly which students are most hurt by those actions. Read it here.
Why American Students Do Not Learn to Read Very Well: The Unintended Consequences of Title II and Teacher Testing
Sandra Stotsky
Third Education Group Review
Vol. 2, No. 2, 2006
At the end of the day, our children's ability to read depends upon whether ed school instructors know how to teach the science of reading. A growing body of evidence suggests they are failing (see here.) So how, then, do their graduates continue to receive licenses to teach elementary school or to serve as reading instructors? This new paper by Sandra Stotsky provides a simple but disturbing answer-the licensing exams prospective teachers must pass don't test their knowledge of research-based reading. Stotsky analyzed exams offered by Educational Testing Service (both national and state exams), National Evaluation Systems, and the American Board for Certification of Teacher Excellence, to see how much of their tests covered instruction in phonemic awareness, phonics, and vocabulary building. Parents whose children are struggling to read will find little solace in Stotsky's results. She reviewed eleven tests used for certifying elementary teachers and found that most failed to cover the basics of scientific-based reading instruction. Three of the four PRAXIS II exams reviewed, required by 46 states, only dedicate between 1 and 7 percent of their content to this critical material (though the fourth PRAXIS fared substantially better). "These aspects of reading instruction receive such minimal attention," Stotsky writes, "that test-takers could fail every question...and still pass the test no matter where the passing score is set." The record's no better for tests administered to reading teachers, reading specialists, early childhood teachers, or special ed teachers. Gladly, some of the customized state tests (California's and Massachusetts's, for example) are much stronger, as is the ABCTE exam. Alas, that's the only test studied that is not required for certification, but is merely an alternative route to teaching. Pity. Read the report here.
Key Issues in Studying Charter Schools and Achievement: A Review and Suggestions for National Guidelines
The Charter School Achievement Consensus Panel; Julian Betts and Paul T. Hill, Principal Drafters
Center on Reinventing Public Education
May 2006
For charter supporters, August 17, 2004, is a day that will live in infamy. That day, the New York Times unleashed its AFT-spun appraisal, based on the National Assessment of Educational Progress, that charter schools were "lagging behind" traditional public schools. (See here.) All hell broke lose. Full page ads denouncing the article were taken. Rival studies were rushed to print. It was a fight to the death. Reason, facts, and rigorous analysis, it seemed, couldn't make much difference. And that was a big problem. So professor Paul Hill and the leaders of several foundations set out to create a space where research would triumph over rhetoric. The result is the National Charter School Research Project (of which Fordham is a minor funder), housed at his Center, and this white paper is one of its first major contributions. It can best be described as a peace offering, the conditions of a possible truce in the charter research wars. It focuses (probably too narrowly) on how to best answer one key question: "whether students in charter schools are learning more or less than they would have learned in conventional public schools." The paper walks the reader through seven major types of studies (such as experimental and "fixed effects") and considers their strengths and weaknesses in addressing this question. No method is perfect, argues the paper; even well-regarded experimental studies have limitations (namely, that their results might not be applicable to the entire charter universe). But some methods are definitely better than others, and studies that look at average student achievement at one point in time-as did the AFT's-are shown to be the least reliable. But even as the field matures, the research is not necessarily getting any better. The panel ranked a majority of the charter studies published between 2001 and 2005 as "fair" or "poor." (Unfortunately, the reader is not told which studies received which rating.) If researchers take this paper's recommendations to heart, the track record for the next few years should be much stronger. Before you launch your next charter school study, read the white paper here.