LAUREN COVELLI, Vanderbilt University Peabody College Poster.pdf · • Senate Bill 2 of 2013allows the commissioner to accelerate expansion of effective charter schools and close

• As the charter sector grows, the percentage of students in TPSDs passing their annual state tests increases. This is true for all subjects and all students, all students in math, all students in reading, Hispanic students in all subjects, and students qualifying for free and reduced-price lunch.

• However, these positive impacts are not observed for black students or white students as separate subgroups.

• These results are consistent with previous research on the Texas charter sector finding test score gains for TPSD students after charter market penetration (Booker et al., 2008).

• As the charter sector grows, TPSDs are spending more on instruction, both in terms of the percent of their expenditures that are allocated towards instruction as well as thereal dollars spent per pupil.

• As the charter sector grows, the proportion of TPSD revenue coming from federal sources increases while the percent of revenue coming from state sources decreases.

• These results are consistent with the theory that charter competition may spur districts to reallocate resources towards academicachievement, though previous research testing this theory finds mixed results (Arsen & Ni, 2012).

• As the charter sector grows, the percent of black students and ESL/bilingual students enrolled in TPSDs decreases while the percent of white students enrolled in TPSDs increases.

• There appear to be no observable effects ofcharter sector growth on the percent ofHispanic, FRPL, or SPED students in TPSDs.

• These results are consistent with previous research on the Texas charter sector, which finds that charters serve a higher proportion of English Language Learners and black students than TPSDs, while TPSDs serve a higher proportion of white students (CREDO, 2017).

Source: Texas Education

Agency

• 1,038 districts• 1994-95 to 2016-17• Demographics,

finance, and performance

Source: NCES Common Core

of Data

• Charter school enrollment by year merged onto geographic districts

• 1998-99 to 2016-17

Final Analytic Sample

• 203 unique districts with charter schools

• 3859 unique district-by-year observations

• 1998-99 to 2016-17

𝑌!" = 𝛽# + 𝛽$𝑃𝐶𝑇𝐶𝐻𝐴𝑅𝑇𝐸𝑅!" + 𝛿! + 𝜃" + 𝛿! ∗ 𝑡 + 𝜀!"𝑌!": District by year outcomes include demographics, performance, and various

measures of revenues and expenditures.𝛽#: The expected change in Y when the charter market share increases by one

percentage point.𝛿!: District fixed effects control for time invariant unobserved district characteristics.𝜃": Year fixed effects control for annual shocks experienced by all districts.𝛿! ∗ 𝑡: A district-by-year interaction allows for district specific time-trends.𝜀!": Robust standard errors are clustered by district.

• TX charters were first authorized in 1995 by the state legislature.

• The first charter schools opened in fall of 1996 enrolling approximately 2,500students on 17 campuses.

• The charter sector expanded by about 13% annually.

• Senate Bill 2 of 2013 allows the commissioner to accelerate expansion of effective charter schools and close underperforming charter schools.

• By 2016, TX charters enrolled 250,000 students across 629 campuses.

• 35% of charter schools are in the Dallas Fort Worth area.

• 25% of charter schools are in Houston.

TEXAS CHARTERS

LAUREN COVELLI, Vanderbilt University Peabody College

EMPIRICAL STRATEGYDESCRIPTIVE STATISTICSTo what extent does charter sector growth change TPSD demographic composition,

finances, and performance?

QUESTION

DATA & SAMPLE

DEMOGRAPHICS

The Impact of Charter School Expansion on Traditional Public School Districts:Evidence from Texas

Leaky (31%)

Galveston (30%)

Austin (12%)

Houston (17%)

Dallas(19%)

Fort Worth(5%)

Figure 1. Percent of Students Enrolled in Charter Schools by Geographic District (2015)

Figure 2. Percent of District Enrollment in Charters by Year

Table 1. Effects of Charter Expansion on TPSD Demographic Composition

As the charter market share increases by 1 percentage-point…

Table 2. Effects of Charter Expansion on TPSD Revenues and Expenditures

Table 3. Effects of Charter Expansion on TPSD Performance on State Tests

Instructional expenditures in TPSDs increase, the percent of revenues from federal sources

increases and from state sources decreases.

The percent of all TPSD students passing all subjects of states

tests increases, as does the pass rate for Hispanic and FRPL

students.

The percent of black students in TPSDs decreases and the percent

of white and ESL students increases.

Figure 3. Demographics in Charter vs. Non-Charter Districts

Figure 4. Percent Passing Annual Assessments in Charter vs. Non-Charter Districts

13%

44%39%

56%

10% 13%

6%

31%

61%

53%

12%6%

0%

10%

20%

30%

40%

50%

60%

70%

% Black % Hispanic % White % FRPL % SPED % ESLCharter Non-Charter

84%80%

65%69%

82%

66%

86%81%

64%69%

81%

69%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

ELA (AllGroups)

Math (AllGroups)

All Tests(Black)

All Tests(Hispanic)

All Tests(White)

All Tests(FRPL)

Charter Non-Charter

Figure 5. Charter Sector Growth and Deviation from Time Trends in Houston ISD*

FINANCE

PERFORMANCE*Figure 5 highlights resultsfrom Houston ISD, the district serving the largest number of charter school students.

The solid lines show the fitted values from fully specified models, demonstrating the timetrend in outcomes as thecharter sector continues togrow.

The dashed lines show hypothetical fitted valueswhen the percent charterenrollment is held at 0%. Thegrowing gap between thelines demonstrates how thegrowth of the charter sector changes the district’s trajectory over time.

Arsen, D., & Ni., &. (2012). The effects of charter school competition on school districtresource allocation. Education Administration Quarterly, 49(1), 3-38.

Center for Research on Education Outcomes. (2017). Charter school performance inTexas. Retrieved from https://credo.stanford.edu/pdfs/Texas_report_2017.pdf.

Booker, K., Gilpatric, S. M., Gronberg, T., & Jansen, D. (2008). The effect of charter schools ontraditional public school students in Texas: Are children who stay behind left behind?Journal of Urban Economics, 64, 123-145.