Table of Contents
The Charter School Debate: What Research Actually Shows
Charter school research is more nuanced than either side admits. Stanford's CREDO study and lottery-based natural experiments reveal which charters work, which don't, and why.
The charter school debate has a problem: both sides pick the evidence they like and ignore the rest. Charter advocates cite Boston and New York City charter lottery studies showing dramatic test score gains. Charter critics cite the Stanford CREDO report showing that many charters perform worse than neighborhood public schools. Both are citing real research. Neither is giving you the full picture.
Here’s what the research actually shows — which is messier, more conditional, and ultimately more useful than either talking point.
Key Takeaways
- Stanford’s CREDO national studies (2009, 2013, 2023) show enormous variance — about a third of charters outperform comparable public schools, a third perform similarly, and a third perform worse.
- Urban, structured, “no-excuses” style charters (KIPP, Success Academy, Uncommon Schools) show the strongest evidence — lottery studies show gains of 0.1 to 0.4 standard deviations in reading and math for low-income students.
- Lottery-based studies are the most credible research design because they eliminate selection bias — students who apply to a charter and don’t get in are the ideal comparison group.
- Charter opponents are right that many charters underperform. Charter supporters are right that the best charters produce real gains for underserved kids. Both claims are simultaneously true.
- What makes a charter school work has more to do with specific instructional practices than the charter model itself.
Why Charter Research Is Hard to Do Well
Before getting to findings, methodology matters here. The fundamental problem in charter research is selection bias: families who choose charters may differ in motivation, involvement, and resources from families who don’t — regardless of what the school does. This makes simple comparisons misleading.
The best charter research design uses lottery randomization. When a popular charter school has more applicants than seats, it holds a lottery. Students who applied but weren’t admitted — the lottery losers — are an ideal comparison group for lottery winners, because both groups had motivated families who sought out the charter. Random assignment to treatment (charter) vs. control (traditional public school) is as close to a clinical trial as education research gets.
The second-best design is CREDO’s approach: matching each charter student to “virtual twins” in regular public schools using detailed demographic and achievement data. This controls for observable characteristics but can’t control for unobservable ones (like family motivation).
Keep both designs in mind as we go through the findings.
What CREDO’s National Studies Show
Stanford’s Center for Research on Education Outcomes (CREDO) has conducted the largest national charter school studies, examining millions of student-year records:
2009 National Charter School Study: Found that 17% of charter schools produced significantly better results than comparable public schools; 46% showed no significant difference; 37% produced significantly worse results. The average charter performed similarly to or below the average traditional public school.
2013 CREDO Update: Showed improvement — the average charter’s performance advantage over comparable public schools had grown, particularly in urban areas. Rural charters continued to underperform.
2023 CREDO Study: The most recent large national study found continued improvement in urban charter performance, with particularly strong results for low-income, Black, and Hispanic students in major cities. The charter-traditional school gap in high-poverty urban areas grew more favorable to charters over the 2010s decade.
What CREDO consistently shows: charter quality varies enormously, and the school’s location, structure, and student population matter more than the charter label itself.
The Lottery Studies — Most Credible Evidence
The lottery-based studies focus primarily on urban charter networks that are oversubscribed (more applicants than seats), which limits generalizability to heavily demanded charters. But within that scope:
Boston: Abdulkadiroğlu and colleagues (2011, American Economic Review) found lottery winners in Boston charter schools gained 0.17 standard deviations in math per year compared to lottery losers — a substantial effect equivalent to closing roughly one-third of the Black-white achievement gap.
New York City: Dobbie and Fryer (2011) found Harlem Children’s Zone Promise Academy lottery winners made large gains in both math and ELA, effectively eliminating the Black-white achievement gap for the children who enrolled.
National: Abdulkadiroğlu et al. (2016) examined lottery records across multiple urban charter networks and found consistent positive effects for urban, no-excuses style charters serving low-income students — and much smaller or null effects for charter schools that didn’t share those characteristics.
The key finding: lottery gains are concentrated in specific types of charters. Urban, structured, high-expectation charters with consistent instructional practices. Not all charters, not most charters — specific charters.
What Makes the Effective Ones Work
Researchers Dobbie and Fryer (2013) studied 35 charter schools in New York City to understand what distinguished high-performing from low-performing charters. The practices associated with higher achievement:
| Practice | High-performing charters | Low-performing charters |
|---|---|---|
| Extended learning time | Yes (25-30% longer school day/year) | No |
| High-dosage tutoring | Yes (small group, daily or weekly) | No |
| Data-driven instruction | Frequent diagnostic assessments | Infrequent |
| Human capital focus | Selective hiring, heavy coaching | Standard hiring |
| Student culture | Explicit behavioral expectations | More flexible |
Notice these are instructional practices, not the charter model itself. A traditional public school that implemented all five of these practices would likely see similar gains. The charter structure provides flexibility to implement them — but the practices are what drive results.
What Charter Opponents Get Right (and Wrong)
Right: Many charter schools underperform comparable public schools. The CREDO data is clear on this. Allowing a charter to open doesn’t guarantee better outcomes, and weak charter authorizing (the oversight mechanism) has allowed many low-performing charters to stay open.
Right: Charters can drain funding from traditional public schools in ways that harm the students who remain, particularly in shrinking urban districts where fixed costs (buildings, transportation infrastructure) don’t shrink when enrollment does.
Wrong: The claim that charters cream-skim the best students. Lottery studies control for this — students admitted by lottery are randomly selected from the applicant pool. When the school itself isn’t selecting, the gains persist.
Wrong: That evidence of charter benefits doesn’t exist. It does. The question is whether it generalizes beyond the specific conditions under which it was found.
What Charter Supporters Get Right (and Wrong)
Right: For low-income urban students with limited traditional public school options, the best urban charter networks provide measurable academic benefits. This is real, it’s well-documented, and it matters.
Right: Teachers’ union opposition to charters sometimes conflicts with evidence about what helps disadvantaged students. Not always, but sometimes.
Wrong: That charter school outcomes generally justify the model. The average charter does not outperform the average public school. The best charters do — and the average is dragged down by many weak charters.
Wrong: That market competition will naturally improve public schools. The competition effects of charters on neighboring public schools are small and inconsistent across studies.
What Parents Should Actually Look For
If you’re considering a charter school for your child, the research points to specific questions worth asking:
- Is it oversubscribed? Schools that are in demand for academic reasons (not convenience) are a signal.
- What does student data show over multiple years? Ask for 3-year reading and math growth data, not just proficiency rates.
- What’s the teacher retention rate? High teacher turnover is a red flag in any school.
- What does the school day look like? Extended learning time is one of the best-evidenced practices in the lottery studies.
- What’s the expulsion and attrition rate? Some charters show high outcomes partly by counseling out struggling students. Ask for cohort survival data.
Understanding what quality looks like in any school setting — not just charters — connects to broader questions about what actually makes a good teacher and how instructional practice drives outcomes.
What to Watch For Over the Next 3 Months
If your child starts at a charter school:
Month 1: Notice whether the school’s instructional approach differs from what you expected — especially around extended learning time, homework policy, and behavioral expectations. Ask how the school assesses learning progress.
Month 2: Pay attention to school culture signals. Are teachers confident in the curriculum? Is professional development evident? High-performing charters typically have systematic coaching structures that teachers talk about.
Month 3: Request and review your child’s diagnostic assessment data from the school. If the school can’t show you individualized progress data and explain what they’re doing differently for your child, the data-driven instruction that characterizes effective charters may not be present.
Frequently Asked Questions
Are charter schools better than public schools?
On average, no — they’re roughly comparable. But the distribution is wide. The best charters, especially urban charters serving low-income students, outperform comparable public schools significantly. Many other charters underperform. The question is whether a specific charter is better than the specific public school alternative, not whether charters in general are better.
Do charter schools take money away from public schools?
Yes, in most funding models — funding follows the student, so when a student enrolls in a charter, per-pupil funding moves with them. The impact on the school district depends on whether fixed costs also decrease, which they often don’t. This is a legitimate concern, especially in districts with declining enrollment.
Do charter schools cherry-pick good students?
The oversubscribed charters studied in lottery research don’t — lottery admission is random. Some charters do engage in soft selection (difficult application processes, limited transportation, no special education services), which can bias their population. This is a legitimate critique of some charters, not all.
What’s the difference between a charter school and a magnet school?
Both are public and free to attend. Magnet schools are part of the traditional public school district, often with specialized focuses (arts, STEM, language immersion), and are typically governed by the district. Charter schools operate independently under a charter contract with an authorizing body, with more flexibility in curriculum, staffing, and scheduling.
About the author Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.
Sources
- CREDO, Stanford University. (2013). National Charter School Study 2013. Center for Research on Education Outcomes. https://credo.stanford.edu/
- Abdulkadiroğlu, A., Angrist, J., Dynarski, S., Kane, T., & Pathak, P. (2011). “Accountability and Flexibility in Public Schools: Evidence from Boston’s Charters and Pilots.” Quarterly Journal of Economics, 126(2), 699–748.
- Dobbie, W., & Fryer, R. G. (2011). “Are High-Quality Schools Enough to Increase Achievement Among the Poor? Evidence from the Harlem Children’s Zone.” American Economic Journal: Applied Economics, 3(3), 158–187.
- Dobbie, W., & Fryer, R. G. (2013). “Getting Beneath the Veil of Effective Schools: Evidence from New York City.” American Economic Journal: Applied Economics, 5(4), 28–60.
- Abdulkadiroğlu, A., Angrist, J., Hull, P., & Pathak, P. (2016). “Charters Without Lotteries: Testing Takeovers in New Orleans and Boston.” American Economic Review, 106(7), 1878–1920.
- National Alliance for Public Charter Schools. (2023). A Growing Movement: America’s Largest Charter School Communities. https://www.publiccharters.org/
