Do KIPP Elementary Schools Improve Student Outcomes? Evidence From Admissions Lotteries

• In Event: Understanding the Effects of KIPP as It Scales Up

Abstract

Objective: This paper measures the impact of KIPP elementary schools on students’ math and reading achievement and other outcomes in early grades.

Background: KIPP schools seek to engage students and parents in the educational process, expand the time and effort students devote to their studies, reinforce students’ social competencies and positive behaviors, and improve academic achievement. As public charter schools, KIPP schools have greater autonomy to set their own policies than do most traditional public schools, but are accountable to their authorizers for achieving satisfactory performance.

Prior research has found that attending a KIPP middle school leads to positive effects on student achievement (Gleason et al. 2014; Tuttle et al. 2013; Ferguson et al. 2012; Angrist et al. 2012; Woodworth et al. 2008). For example, using quasi-experimental methods, Gleason et al. (2014) found that entering a KIPP middle school led to increases in math and reading achievement of 0.36 and 0.21 standard deviations after three years, respectively. These are large effects, equivalent to about 90 percent of a year of extra learning in math and about two-thirds of a year of extra learning in reading (Bloom et al. 2008). However, existing research does not tell us anything about impacts of KIPP schools at levels other than middle school. This paper addresses this gap by presenting evidence on the KIPP model’s effectiveness in elementary schools.

Methods: We used kindergarten and pre-kindergarten admissions lotteries for the 2011–2012 school year as randomized experiments to measure the impacts of eight KIPP elementary schools. Students offered admission via the lottery are included in the treatment group; those not offered admission through the lottery are included in the control group. This design ensures that treatment and control group students are similar at baseline (pre-KIPP) in terms of demographics as well as other key factors such as motivation and parental support.

Data Sources: To measure students’ academic achievement in reading and math, we administered the Woodcock-Johnson III (WJ-III) assessment in the third follow-up year after the lottery, when most pre-kindergarten applicants were in kindergarten, and most kindergarten applicants were in grade 2. To measure other outcomes including behavior, we administered a survey to parents in the second follow-up year.

Results: The results, forthcoming in late 2015, will measure the impact of eight KIPP elementary schools on students’ achievement on four WJ-III subtests (two math and two reading) three years after the admissions lottery.

Significance: Given existing evidence of the KIPP model’s success at the middle school level, it is of interest to education researchers whether the model’s key features also lead to improvements in student achievement at the elementary school level. Given that the KIPP charter school network is the largest in the U.S., new information on these schools’ effectiveness is practically significant for policymakers aiming to improve student achievement.

Authors

• Gregory Chojnacki, Mathematica Policy Research, Inc

• Christina Clark Tuttle, Mathematica Policy Research, Inc

• Thomas Coen, Mathematica Policy Research, Inc

• Philip Gleason, Mathematica Policy Research, Inc.

• Virginia Knechtel, Mathematica Policy Research, Inc