@article{masters2012, title={Measuring and rewarding school improvement}, author={Masters, Geoff N}, year={2012} }
schools have limited control over many of the measures that could be proposed to evaluate their effectiveness. p3
School Practices as:
− whole-school improvement agenda − systematic performance monitoring − focused use of school resources − collaboration to improve teaching − curriculum clarity and coherence − support for individual needs − use of proven teaching strategies − commitment to professional learning
...combined with Conextual Influences as:
− cultural and family aspirations − socioeconomic backgrounds − geolocation of school − language spoken at home − family and community support − school ownership and governance − school financial resources
...drive Student Outcomes as: (p3)
− engagement and participation − literacy and numeracy skills − academic achievement − employment-related skills − school completion and attainment
Because student outcomes are the result of both school practices and contextual influences, it is not possible to infer a school’s effectiveness directly from familiar student outcome measures such as mean scores and proficiency levels on standardised tests. p3
Many schools look much better on performance [outcome] measures than they do upon inspection of their practice. Schools in general, and high-performing schools in particular, produce a large part of their performance with social capital, not with instruction. (Elmore, 2008, 54)
This paper considers two approaches to measuring school improvement:
- inferring school improvement from improvements in student outcomes; and
- inferring school improvement from improvements in school practices known to be correlated with student outcomes. p4
In the United States, most states and school districts use test scores as measures of school performance. This practice became widespread following the No Child Left Behind Act of 2001 which required schools to demonstrate ‘adequate yearly progress’ in reading and mathematics. More than thirty US states now provide rewards or sanctions to schools based on the test performances of their students. In England, student test scores are interpreted as measures of school performance after being adjusted for a range of contextual influences. Each school’s resulting contextualised value-added (CVA) score is assumed to be directly comparable with the CVA score of every other school in the country. p7
There is also the issue of differential effectiveness - schools affect student outcomes differently dependending on student.