America's Best Schools?

"While the average fifth-grader enters KIPP in the bottom third of test-takers nationwide (28th percentile), the average KIPP eighth-grader outperforms nearly three out of four of test-takers nationwide (74th percentile) on norm-referenced reading and math assessments," the report card says. "In the fifth-grade year, approximately 40 percent of KIPP schools outperform their respective districts on state reading exams, and just over 60 percent do so in math. By the eighth grade, 100 percent of KIPP schools outperform their districts in both subjects."

"These results show that longer hours, great teachers, and a structured learning environment are producing significant academic gains and putting students on the road to college," the report says.

The report gives a two-page summary of each of the 37 schools. There are several places where I would raise caution flags, as the KIPP people do themselves, since judging schools is a very tricky business and all experts who have tried it have been burned at one time or another.

KIPP has commissioned two independent reports that have given its schools good reviews. The most recent, published in August 2005 by the Educational Policy Institute, said KIPP schools in 2003-2004 experienced "larger and significant gains" as compared to urban public schools. But KIPP is a moving target, with many more schools and many more children every year, and such assessments tend to lose their relevance quickly.

Among the most important things to say, for instance, about that astounding jump in math and reading achievement from the 28th to the 74th percentile is that it is based not on an independent study but on standardized tests that KIPP teachers gave to their own students. KIPP schools are run by their mostly young principals, not by the KIPP Foundation, and the principals decide which tests to use and conduct the testing. The schools in the report gave the Stanford 9 or Stanford 10 tests or the Iowa Test of Basic Skills.

These are popular tests used by many school districts to see how their students compare to a national sample of students, and are as good as you are going to get in the world of relatively inexpensive off-the-shelf tests. But this is still KIPP testing itself, and that should be kept in mind. There are several cases of public school principals changing standardized test papers to enhance their results when they have had an opportunity to do so.

What makes me confident that this is unlikely to happen at KIPP is that I have had several chances to see how KIPP people handle information and the press, and have yet to catch them being anything but honest and open about what was happening at their schools. The KIPP system for selecting and training principals, a year-long program, has become a national model. And the fact that a few KIPP schools do not look good on some tests in this report reinforces my view that they are acting in good faith.

I also think that on balance it is better that they are choosing and administering the tests themselves, compared to other choices available to them. This way they can continue to give the same tests for the next several years and give those of us watching them a reliable means of seeing how they are doing over the long term. Most regular school systems change the tests they are using every few years, making it impossible to make long-term comparisons. Finding an independent group willing to test KIPP students annually, and indefinitely, without spending a great deal of money would be next to impossible, so the self-testing is probably the best we can hope for.

KIPP is doing something else that no other large collection of low-income student schools I know of is doing: It is reporting these test results by cohort. That is, we are seeing how the same students improved from grade to grade, rather than doing what most schools do, compare the scores of this year's third-graders to last year's third-graders, which does not tell you if your child and his classmates have improved or not.

The 2005 report card also shows how KIPP schools compare to district and state averages on the state tests required under the federal No Child Left Behind Act. Those results too are mostly favorable to KIPP, but the different state tests have such widely varying standards of proficiency that they make it difficult in many cases to see just how well or poorly KIPP schools are doing.

As the KIPP people also acknowledge, that impressive fifth- to eighth-grade gain in math and reading achievement is based on just the five KIPP schools that have been open long enough to graduate an eighth-grade class. It will take a few more years to see if the other schools can match that impressive record, since KIPP middle schools start with just a fifth grade and add just one grade a year, taking four years to have a full school.

Most of the schools are showing healthy gains in nearly all grades, but some are not. KIPP LA Prep in Los Angeles reported a drop in reading from the 40th to the 39 percentile for sixth-graders in spring 2005. Sixth-graders that same year at the KIPP South Fulton Academy in Atlanta dropped from the 44th to the 38th percentile in reading, and sixth-graders at the KIPP Ascend Charter School in Chicago dropped from the 35th to the 34th percentile in reading.

The KIPP Ujima Village Academy in Baltimore was significantly above the average scores for that city in reading, but its seventh-graders showed a drop from the 38th to the 33rd percentile on the Stanford 9 reading test in spring 2005. The KIPP Reach College Preparatory school looked impressive when compared to the average for other public schools in Oklahoma City, but its seventh-grade's reading score dropped from the 63rd to the 43rd percentile in spring 2005 compared to what those same students did the previous spring.

There are many more details in the report, most of them impressive. But time has a way of wearing down the most successful educational enterprises, so we shall have to see what happens next.

It is encouraging to me that in several instances KIPP principals and teachers whose students were not improving have been shown better ways to do their jobs, and if that hasn't worked, have been fired or allowed to resign. Very few students leave KIPP. It prides itself on helping those students struggling the most. But it is far less patient with educators who ignore suggestions and let things slide.

Two schools, the KIPP Chicago Youth Village Academy and Atlanta's KIPP Achieve Preparatory Academy, have had the right to use the KIPP name revoked effective at the end of this school year. Mancini said there were many efforts to help them, but the Chicago school still "struggled with low enrollment and low reading scores relative to the district average" and the Atlanta school "struggled with financial reporting and viability and did not properly administer voluntary tests that would demonstrate growth over time."

The usual reaction to mediocre results in most public schools is to pray to all the available deities for a change, but not do very much to make that happen. KIPP Nation has gone beyond even the federal requirements by reporting both its internal and external test results. That is a sign that it is willing to risk revealing its failures so that everyone knows it is serious about fixing them. In American education that is rare and worth rooting for.