Examining Charter School Performance

A Multilevel View of New York State Charters

Tyler W. Rinker

Research Questions


1. How do charter schools and districts compare in terms of mean grade 3-8 mathematics achievement above and beyond the effect of county, district poverty rate, and proportion minority effects in New York State?

2. How do charter schools and districts compare in terms of mean grade 3-8 English language arts (ELA) achievement above and beyond the effect of county, district poverty rate, and proportion minority effects in New York State?

3. How does county charter school rate effect average county mathematics achievement controlling for district level poverty rate and proportion minority in New York State?

4. How does county charter school rate effect average county ELA achievement controlling for district level poverty rate and proportion minority in New York State?

Charter School Context

  • Born 1988 by Ray Budde & Albert Shanker
  • Free of bureaucracy—innovative
  • Privately run public schools
  • Business community adoption

Bierlein & Mulholland (1994), Darling-Hammond (2010), Ravitch (2010)

  • Effectiveness
  • Harm public districts
  • Dis-empowers teachers
  • Increases racial segregation

Almond (2012), Bierlein & Mulholland (1994), Darling-Hammond (2010), Ni (2012),
Ravitch (2010), Silverman (2013), Winters (2012)


Charter vs. Public

In this study:

Charter - Charter schools
Public - Public non-charter schools

Rationale



Methods

Sample

plot of chunk unnamed-chunk-1


  • NYS Grade 3-8 assessment data
  • 2011-2012
  • NYSED's IRS school report card database
  • Level 1: 851 districts
    • 714 public
    • 127 charter
  • Level 2: 62 counties

Analysis

Variables

  • Outcomes
    • mathematics achievement
    • ELA achievement
  • Variable of Interest
    • sector (1-charter, 0-public)
  • Covariates
    • low SES
    • minority rate

Analysis

Multilevel Models

  • 2 level
    • District
    • County

  • 4 models

    Model Centering Hypothesis
    1 Unconditional
    2 Partly conditional Group Mean
    3 Fully conditional Group Mean 1 & 2
    4 Fully conditional Grand Mean 3 & 4

Enders & Tofighi (2007)



Results

Descriptives: Achievement

plot of chunk unnamed-chunk-2

Descriptives: Demographics

plot of chunk unnamed-chunk-3

Descriptives: Sector Distribution Achievement

plot of chunk unnamed-chunk-4

Descriptives: Sector Distribution Demographics

Descriptives: Demographics Scatterplots

plot of chunk unnamed-chunk-6

*press p to compare to choropleth demographics

Model 1

Fixed

\[\gamma_{00} = .646,SE= .008, t(61)=78.427, p.<001\]

  • 64% Average district level pass rate

Random

\[\tau_{00} = .0019, \chi^2(61)=151.69, p< .001\]

  • 7.4% of the variance between counties

Fixed

\[\gamma_{00} = .663,SE= .008,t(61)=68.262,p<.001\]

  • 66% Average district level pass rate

Random

\[\tau_{00} = .0019, \chi^2(61)=151.69, p< .001\]

  • 8.4% of the variance between counties

Model 2

Fixed

  • SES (\(\gamma_{10} = -.246, SE = .033, p < .001\))
  • minority rate (\(\gamma_{20}=-.190,SE=.048,p < .001\))

Random

\[\tau_{00} = .002, \chi^2(59)=231.821, p< .001\]

33.2% more between county variability explained

Fixed

  • SES (\(\gamma_{10} = -.251, SE = .038, p <.001\))
  • minority rate (\(\gamma_{20}=-.215,SE=.045,p<.001\))

Random

\[\tau_{00} = .002, \chi^2(59)=309.947, p< .001\]

43.8% more between county variability explained

Model 3

Fixed

  • Sector (\(\gamma_{03}=-.102, SE=.085, p =.237\))
  • Sector (\(\gamma_{30}=.187,SE=.034,p<.001\))

Charters show .187 unit increase in school ELA pass rate over public districts

Random

  • 41.3% of between county variability explained over unconditional model
  • 8.14% more than the model 2

*press p to see research hypothesis 1 & 2

Fixed

  • Sector (\(\gamma_{03}=-.258, SE=.062, p < .001\))
  • Sector (\(\gamma_{30}=.119, SE=.019, p <.001\))

Charters show .119 unit increase in school ELA pass rate over public districts

Random

  • 47.81% of between county variability explained over unconditional model
  • 4.01% more than the model 2

Model 4

Fixed

  • Sector (\(\gamma_{03}=-.173, SE=.078, p = .031\))

Counties with higher proportion of charter schools displayed significantly lower math achievement than counties with lower percentages of charters

Random

  • 36.96% of between county variability explained over unconditional model

*press p to see research hypothesis 3 & 4

Fixed

  • Sector (\(\gamma_{03}=-.254, SE=.066, p < .001\))

Counties with higher proportion of charter schools displayed significantly lower ELA achievement than counties with lower percentages of charters

Random

  • 47.58% of between county variability explained over unconditional model



Discussion

Summary


  1. Charters more effective in both mathematics and ELA achievement within their district

  2. Counties with higher percentages charters have a decreased average country math and ELA achievement

Explanation

Limitations







  • Aggregated school level data
  • Single time point
    • No growth
    • Reduced n

Future Study?


  • Longitudinal data
  • Growth model
  • Additional variables
    • Level 1
    • Level 2
    • Outcome

*press p to see possible additional variables of study




plot of chunk unnamed-chunk-8

References

Almond, M. (2012). The black charter school effect: Black students in American charter schools. Journal of Negro Education, 81(4), 354-365.

Bierlein, L. A. & Mulholland, L. A. (1994). The promise of charter schools. Educational Leadership, 52, 34-40.

Bifulco, R. & Ladd, H. F. (2007). School choice, racial segregation, and test-score gaps: Evidence from North Carolina's charter school program. Journal of Policy Analysis & Management, 26(1), 31-56. doi:10.1002/pam.20226

Chung, J. Y., Shin, I., & Lee, H. (2009). The effectiveness of charter school: Synthesizing standardized mean-changes. KEDI Journal of Educational Policy, 6(1), 61-80.

Darling-Hammond, L. (2010). The flat world and education: How America's commitment to equity Will determine our future. New York: Teachers College Press.

References

Enders, C. & Tofighi, D. (2007). Centering predictor variables in cross-sectional multilevel models: A new look at an old issue. Psychological Methods, 12(2), 121-138. doi:10.1037/1082-989X.12.2.121

Ni, Y. (2012). Teacher working conditions in charter schools and traditional public schools: A comparative study. Teachers College Record, 114(3), 1-26.

Raudenbush, S. W. & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage Publications.

Ravitch, D. (2010). The death and life of the great American school system: How testing and choice are undermining education. New York: Basic Books.

Silverman, R. M. (2013). Making waves or treading water? An analysis of charter schools in New York State. Urban Education, 48(2), 257-288. doi:10.1177/0042085912449840

References

Winters, M. (2012). Measuring the effect of charter schools on public school student achievement in an urban environment: Evidence from New York City. Economics of Education Review, 31(2). 293-301. doi:10.1016/j.econedurev.2011.08.014