WILL Policy Brief: Apples-to-Oranges?

Published on: February 9, 2015

Too often, the media, politicians, and interest groups try to form conclusions by lumping schools together according to “sector” – traditional public schools in the Milwaukee Public School (MPS) system, private schools in the Milwaukee Parental Choice Program (MPCP), and independent public charter schools.  They compare average WKCE test scores according to each sector, as if they are one homogenous unit.  But, given the differences in private schools in the choice program and public charter schools – and the limitations of the WKCE testing, this is a flawed approach to comparing schools in Milwaukee.  WILL’s newest policy brief (found here) shows a better way to form conclusions from the flawed WKCE data.

With Governor Scott Walker’s proposed budget calling for the legislature to lift the cap on the statewide Wisconsin Parental Choice Program, there will be a renewed focus on the performance of the Milwaukee voucher program.  If people insist on making comparisons between MPS and MPCP, WILL’s latest policy brief gets closer to an “apples-to-apples” comparison when analyzing WKCE test scores by taking into account a school’s religious affiliation, composition of African American students, and composition of low-income students.  Publically available student data allows comparisons between schools in Milwaukee that serve predominantly low-income, African American students (aka “80/80” schools). When looking at these schools across all sectors, we make the following conclusions:

1. MPCP students in Catholic or Lutheran schools outperform MPS students. The percentage of proficient students in Catholic or Lutheran schools was 3.0 percentage points higher in reading and 6.8 percentage points higher in math. MPS schools, on the other hand, have an advantage over MPCP schools that are not Catholic or Lutheran.

2. In addition, among the “80/80” schools, independent charter schools perform significantly better than MPS schools. The percentage of students in independent charter schools scoring at least proficient in reading and math was 2.4 and 7.5 percentage points higher, respectively, than in MPS schools.

3. When we examine the top-20 schools in Milwaukee, per the imperfect WKCE test score measures, there are no schools, from any sector, that are predominately low-income, African-American schools.  About half are MPS and the other half are MPCP or independent charter schools.

4. When we examine the top-20 schools in Milwaukee that serve low-income, African American students, we see that there is a variation of school type. Three of the top four schools in math are Lutheran.  Ten of the top-20 schools are MPS schools, six are private schools in the MPCP, and four are independent charters.

It would serve public policy better to improve the information provided to parents, identify high performing schools and figure out how to expand their seats, and identify chronically low-performing schools and figure out how to turn them around.  In light of this, one should question the reason for making comparisons across sectors when clearly both high-performing and low-performing schools operate in each sector.  Nonetheless, comparisons will likely continue to be made, and this report offers a better (though still not ideal) alternative to the usual ones.

Click here for the full policy brief.