By Jay Mathews
Washington Post Staff Writer
Tuesday, April 24, 2007 1:22 PM

Some critics decry the way the Knowledge Is Power Program presents itself as the savior of inner city education. My answer: KIPP doesn't do that. We sloppy journalists do.

Let me present Exhibit A: The latest annual report card from the KIPP Foundation in San Francisco. It has 93 pages of remarkable data. (See, there I go again, making KIPP the miracle cure. Let me change that to "interesting" data.) The report card tells how well each of the KIPP schools is doing, but it does not claim to be saving our cities.

I understand why we education reporters try to make KIPP sound like more than it is. We are starved for good news about low-income schools. KIPP is an encouraging story, so we are tempted to gush rather than report. We don't ask all the questions we should. We don't quote critics as often as we ought to. We don't emphasize how new and incomplete the KIPP data is. But none of that is KIPP's fault. Data costs money, and KIPP tries to use most of its funds to educate kids.

One of the best things about KIPP, a network of 52 independent public schools in 16 states and the District, is that it tries very hard to make the statistics it has available to everyone. Focusing on results is one of the organization's basic principles. Anyone can order a free copy of the new report card by going to www.kipp.org. And on page 57 you will find numbers that help explain why KIPP is firing its middle school in Buffalo, N.Y., the sixth time a KIPP school has left the network.

The KIPP people put this more gently. In an April 20 letter to the New York Charter School Association, KIPP chief executive officer Richard Barth said the KIPP Foundation "will end its partnership" with the KIPP Sankofa Charter School in Buffalo "and remove 'KIPP' from the school's name." KIPP spokesman Steve Mancini wished the school well in its plans to continue without the KIPP label. "It is not meeting KIPP standards," he said, "but we think it is providing a viable option for that community."

What does that mean? Let's start with the KIPP standards, as described by KIPP, not awestruck reporters. Most KIPP schools are grade 5 to 8 middle schools, although a few elementary and high schools have opened. Here is KIPP's view of KIPP on page 1 of the report card:

"KIPP schools are free, open-enrollment, college-preparatory public schools where under-served students develop the knowledge, skills, and character traits needed to succeed in top quality high schools, colleges, and the competitive world beyond. . . . Over 90 percent of KIPP students are African American or Hispanic/Latino, and more than 80 percent of KIPP students are eligible for the federal free and reduced-price meals program. KIPP students are in school learning for 60 percent more time than average public school students, typically from 7:30 a.m. until 5:00 p.m. on weekdays, every other Saturday, and for three weeks during the summer. Rigorous college-preparatory instruction is balanced with extracurricular activities, experiential field lessons, and character development. In spite of the long hours, average daily attendance at KIPP schools is 97 percent."

Impatient reporters tend to skip that part so we can get to the test scores, such as last year's jump in math for fifth graders at the KIPP Adelante Preparatory Academy in San Diego, led by Kelly Wright, from the 29th to the 84th percentile, or the jump in reading scores at KIPP Indianapolis College Preparatory, led by Omotayo Ola-Niyi, from the 20th to the 44th percentile. Almost all the schools in the book showed strong improvement in neighborhoods in which students usually have few gains on national percentile scales from year to year.

Then we get to page 57. It says last year's fifth-graders in Buffalo, led by Uchenna Cissoko, made small gains, but the sixth grade news was bad. Those students finished the year with math scores that were eight percentile points lower than those same students achieved when they arrived at Sankofa's fifth grade two years before. They also had a nine percentile point drop in language arts. The only good news in that cohort of sixth graders was a three percentile point increase in reading. Unlike most KIPP schools, Sankofa had not beaten the average proficiency rates for its school district on state tests.

Cissoko told me her results "had not been up to the KIPP standards," but she had not lost faith in KIPP and would continue to use KIPP methods even though she would no longer have the KIPP name. She said she had difficulty recruiting enough good teachers, in part because the Buffalo public schools do not have a Teach for America program. Teach for America, a non-profit organization, recruits top graduates of selective colleges for two-year stints in low-income schools. Many KIPP teachers are selected from among the most successful veterans of those two-year Teach for America enlistments. KIPP founders Dave Levin and Mike Feinberg began in Teach for America. Barth, the KIPP Foundation CEO, helped start Teach for America and is married to its founder, Wendy Kopp.

Cissoko said KIPP helped her advertise for Teach for America veterans to come to Buffalo, but she did not get enough. Cissoko, a math teacher, also said she wished she had gotten more help from KIPP in building her reading program. "It would be prudent for the foundation to keep close contact with the school educators and make sure that people who are working for the foundation are providing the support to the schools when they request that it be provided," she said. She said she hopes to improve the school with help from its new board chairman, Samuel Savarino, a construction company owner and award-winning community leader.

In his letter, Barth said KIPP decided to strip Sankofa of the name only after "nearly a year of intensive support by the KIPP Foundation, including: assistance with the re-design and implementation of a school-wide professional development plan for teachers; coordination of the development of a program to incorporate writing and literacy standards across school curricula; and providing (at the KIPP Foundation's expense) experienced KIPP principals to mentor the school's instructional leaders, including a four-month residency by a KIPP employee with 25 years of school leadership experience."

Cissoko got help and encouragement from Tracy McDaniel, 50, principal of KIPP Reach College Preparatory school in Oklahoma City, whose own sixth graders last year were up 33 percentile points in math, 22 percentile points in reading and 26 percentile points in language arts since starting at KIPP. McDaniel first recruited Cissoko for KIPP leadership training in 2002, when she was getting rave reviews as a 24-year-old classroom teacher in Buffalo. They grew so close that he officiated at her wedding.

McDaniel said Cissoko had a great start, but lost good teachers and had trouble replacing them. She tried to regain her momentum, he said, but could not get back up to the KIPP standard. KIPP leaders, he said, "make a promise to kids" to raise their level of achievement and prepare them for good high schools and college. The standards, and the possibility of losing the KIPP name, are important motivators, he said. The foundation took the action it did, he said, because Cissoko "did not meet the benchmarks that KIPP put in place last summer."

McDaniel and Cissoko said they are still close, and will continue to work on her school's improvement. The most important part of the KIPP model is the independence of principals, who run their schools as they like as long as they show good results. Many KIPP principals say that leads to creative solutions to the many problems that plague their schools despite their glowing press clippings. One such difficulty is families pulling their kids out of some KIPP schools after just a year.

At the KIPP Bridge College Preparatory school in Oakland, Calif., for instance, of the 87 students who enrolled in fifth grade in 2003, 32 later moved out of the area and 30 had parents who decided to remove them from KIPP for other reasons. Twenty-two went back to their regular public schools -- nine left because they did not like the long KIPP day and 13 because KIPP wanted their children to spend another year in fifth grade.

KIPP Bridge principal David Ling said when he told parents repeating the grade would help get their children up to grade level, they often said they thought their children were already excellent students, and would be stars back at their regular schools. I call this the American Idol syndrome, similar to the insistence of untrained singers and their parents on that show that they are great because everyone has been telling them that for years.

The retention issue has been a hot topic in KIPP conferences and email traffic. How can they help these students reach national standards if they quit because of wounded pride? The KIPP schools in Baltimore are now in their third year of a solution they call the Rapid Readers program. It serves all fifth-graders who test below the second grade level in reading. Their families are told from the beginning that it may take them five years to get to eighth grade level. There are no surprises. If they don't like that idea, they are free to withdraw, but most don't. During their first fifth grade year, they spend three hours a day on reading. By the end of their second year in fifth grade, they are ready for sixth grade. At most, only one or two of these dozen families each year, according to KIPP Baltimore executive director Jason Botel, have transferred their children back to the regular school system.

The KIPP report card, and recent KIPP history, show several KIPP schools have struggled. At the KIPP DIAMOND Academy in Memphis, last year's fifth-graders dropped from the 30th to the 29th percentile in reading. Mancini said the board of the Memphis school has moved to correct that situation by persuading Jamal McCall, the highly regarded co-founder and assistant principal at KIPP WAYS Academy in Atlanta, to become principal of KIPP DIAMOND.

So far, two schools have left the KIPP network voluntarily -- one in Dekalb County, Ga., because of philosophical differences one in Sacramento, Calif., because of a desire for more local control. KIPP removed its name from a school in Atlanta because the school's board failed to create what KIPP considered a viable financial plan. Schools in Chicago and Asheville, N.C., that had trouble recruiting enough students gave up the KIPP label and merged with local public schools. KIPP this year dropped plans to turn a Denver public school, Cole Middle, into a KIPP school because none of the people who had successfully completed the KIPP principal training program wanted to go there. Nonetheless, more than 90 percent of schools started by KIPP since 1995 are still operating.

Trying to follow my own advice, I asked some well-informed critics about the latest KIPP statistics. Gerald W. Bracey, an educational psychologist, author and columnist, said "I would give 'sloppy' journalists only half the blame for the KIPP-as-magic-bullet phenomenon. I'd put at least half on Heritage, Walton, Fisher and other conservative think tanks who have used KIPP to say that if it can happen there it can happen anywhere."

Caroline Grannan, a public school advocate and blogger who follows charter school issues, said "KIPP schools succeed for some students--but it's a select subset of students. KIPP is evidently not the solution to the challenges facing urban public education. It would be wonderful to see the vast private funding that's poured into the KIPP schools, which serve just that limited subset, benefiting a larger segment of high-need students."

KIPP schools are supported mostly by tax dollars, as public charter or contract schools. Per pupil funding varies widely, from $4,800 in Oklahoma City to $13,000 in D.C. The report card said its schools spend on average another $1,100 to $1,500 per pupil to cover the costs of the extra learning time, annual field trips and facilities.

The report card said the average student who has been with KIPP three years started at the 44th percentile in math and the 34th percentile in reading at the beginning of fifth grade. By the end of seventh grade that student was at the 83rd percentile in math and the 58th percentile in reading. It is important to note that many students who start KIPP in fifth grade move away or decide to return to the regular school system, so the number who stay three years is relatively small. Also, these results are from tests given not by the school district but by KIPP officials so they can place students properly, diagnose weak areas and keep track of their progress. Such internal assessments lack external monitors and thus, at least in theory, can be subject to cheating. But at least three independent studies have decided the KIPP gains are real, and more such research is under way.

I continue to look for programs that have done better than KIPP in raising the achievement level of low-income children, the central problem in American education today. I have not found any yet, and will keep looking. But I will also continue to watch KIPP, with caution and with some hope.