The final report from a three-year study of a batch of KIPP charter schools in the San Francisco Bay Area probes key issues that have sparked debate about the national network of independently run public schools, including student achievement and attrition.
, issued Sept. 16, comes amid wide acclaim for the KIPP network, along with charges that the schools 鈥渃ream鈥� the strongest students from low-income communities.
The study concludes that the middle schools run by KIPP, which stands for the Knowledge Is Power Program, have posted 鈥渟trong achievement gains,鈥� especially in the 5th and 6th grades, and points to signs that the schools are not simply drawing better students.
鈥淏ay Area KIPP schools do not appear to have attracted higher-scoring students over time, and the three schools for which we have comparison data have attracted lower-scoring and more minority students relative to the neighborhood population,鈥� the study says.
At the same time, the report, by SRI International, a Menlo Park, Calif.-based research institute, finds high student attrition at the KIPP schools, and notes that lower-performing students leave most often. Of the cohort of entering 5th graders at four Bay Area campuses in 2003-04, a total of 60 percent had left before the end of 8th grade, the report says.
The KIPP network includes 66 mostly charter schools in 19 states and the District of Columbia serving more than 16,000 students. Kipp leaders have recently been paying closer attention to attrition, which is seen at rates that vary widely across schools. Critics have suggested high attrition at some campuses may distort achievement gains over time. (鈥淜IPP Student-Attrition Patterns Eyed,鈥� June 13, 2007.)
Achievement Analyzed
The new study, which examined a total of five Bay Area KIPP campuses, follows a preliminary report issued in 2006. (鈥淣ew KIPP Schools Seen as Faithful to Model, Despite Variations,鈥� April 12, 2006.)
The SRI report examines student achievement through multiple lenses, including an analysis that paired up KIPP students from three Bay Area schools with a similar group of students who attended schools in their host districts.
鈥淎 critical question,鈥� the study says, 鈥渋s whether or not student-achievement gains can be attributed to KIPP.鈥�
The researchers concluded that the mathematics gains for 5th graders over one academic year were positive and statistically significant in all cases, and in the majority of cases when it came to English language arts.
The study described the size of the effect on student achievement as 鈥渕odest to substantial.鈥� It also found positive and statistically significant effects for students who were new to the KIPP schools in the 6th grade, when matched with a comparison group.
The matching analysis was limited to one year of data because student attrition and in-grade retention substantially changed 鈥攁nd reduced鈥攖he KIPP sample.
KIPP attrition rates were higher than for the host districts, though the district figures did not count students who may have switched schools within systems, the study finds. Still, the study notes that 鈥渆ven after accounting for attrition, these three schools were not attracting high-performing students relative to their host districts.鈥�
Steve Mancini, a spokesman for the San Francisco-based KIPP Foundation, pointed to the study鈥檚 data on incoming students.
鈥淭hat is pretty significant,鈥� he said, 鈥渂ecause it does counter a misconception about KIPP that we鈥檙e attracting the highest-achieving students.鈥�
Jeffrey R. Henig, a professor at Teachers College, Columbia University, who is working on a review of KIPP studies, said the level of student attrition reported in the SRI study was surprising, and an important caveat to discussions of achievement at KIPP schools.
鈥淚 think people鈥檚 jaws will drop a little bit at that,鈥� he said.
But, he added, this is not to say KIPP isn鈥檛 helping students.
Mr. Henig said his reading of the existing research on kipp suggests that the schools鈥� effects on students are positive, though not as dramatic as some have claimed, when factors such as attrition are taken into account.
Mr. Henig said he was also struck by data on teacher turnover. Annual turnover rates for the Bay Area kipp schools ranged from 18 percent to 49 percent from 2003-04 to 2007-08, the study says.
鈥淭eacher turnover, a result of both ambitious young teachers鈥� moving on and the demanding nature of the job, poses challenges for Bay Area school leaders and may have implications for the sustainability of the model,鈥� it says.
The study also examined the practices in the KIPP schools. It concludes that core practices that likely contribute to student achievement include the emphasis on high expectations for academic performance and behavior, devoting more time to learning, and a focus among KIPP school leaders on 鈥渃ontinuous improvement.鈥�
