Education policy is frequently in the crosshairs of ideological disagreement. This dissertation analyzes three controversial policies over which elected school boards often have control: charter schools, suspension bans, and ability groups.
How do charter schools impact district academic growth? Researchers typically focus on large districts with many charter schools, but the most common experience is an average-sized district shifting from no charters to one. A difference-in-differences design analyzing a decade of charter expansion in California reveals that impact is contingent on charter type: locally funded charters (i.e. affiliated with the district) lead to either static or decreased growth while directly funded charters (i.e. independent of the district) lead to higher academic growth.
Many policymakers have banned or limited suspensions for all but the most serious offenses. The 2013 suspension ban in Los Angeles Unified School District provides a natural experiment; it led to a substantial, 0.2 standard deviation decrease in academic growth among middle schools that had previously issued the banned suspensions. Four subsequent suspension bans – in San Francisco, Pasadena, Oakland, and (grades K-3) all of California – also appear to have harmed academic growth. Simultaneously, suspension bans have an uncertain relationship with dropout rates, the primary mechanism by which bans are meant to impact the school-to-prison pipeline. Instead of banning suspensions, policymakers should carefully test other efforts that decrease suspension and dropout rates without harming academic growth.
Finally, educators have utilized between-class ability grouping – sorting students in one grade into different classes by prior ability – for over a century. Proponents rely on a previously untested mechanism: decreased classroom dispersion in prior academic ability allows teachers to target their instruction more narrowly. This final paper measures classroom dispersion directly for the same students over four trimesters. Multivariate regressions and multilevel models evaluate the relationship between classroom dispersion and academic growth while controlling for other classroom characteristics as well as student, teacher, and school effects. Analyses reveal that English classrooms with less dispersion in prior ability experience slightly less growth. However, there is a trade-off: between-class ability grouping improves equity at the expense of overall academic growth.
| Identifer | oai:union.ndltd.org:bu.edu/oai:open.bu.edu:2144/34781 |
| Date | 16 February 2019 |
| Creators | Zarecki, Dominic |
| Contributors | Kriner, Douglas |
| Source Sets | Boston University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
| Rights | Attribution 4.0 International, http://creativecommons.org/licenses/by/4.0/ |
Page generated in 0.0024 seconds