Value added is the use of statistical technique to identify the effects of schooling on measured student performance. The value added model uses what data are available about students–past test scores and student demographics in particular–to control for prior student knowledge, home and community environment, and other relevant factors to better measure the effects of schools on student achievement. In practice, value added focuses on student improvement on an assessment from one year to the next.
This report presents value-added results for Madison Metropolitan School District (MMSD) for the two-year period between November 2007 to November 2009, measuring student improvement on the November test administrations of the Wisconsin Knowledge and Concepts Examination (WKCE) in grades three through eight. Also presented are results for the two-year period between November 2005 to November 2007, as well as the two-year period between November 2006 to November 2008. This allows for some context from the past, presenting value added over time as a two-year moving average.
Changes to the Value Added Model
Some of the details of the value-added system have changed in 2010. The two most substantial changes are the the inclusion of differential-effects value-added results and the addition to the set of control variables of full-academic-year (FAY) attendance.
In additional to overall school- and grade-level value-added measures, this year’s value-added results also include value-added measures for student subgroups within schools. The subgroups included in this year’s value-added results are students with disabilities, English language learners, black students, Hispanic students, and students who receive free or reduced-price lunches. The results measure the growth of students in these subgroups at a school. For example, if a school has a value added of +5 for students with disabilities, then students with disabilities at this school gained 5 more points on the WKCE relative to observationally similar students across MMSD.
The subgroup results are designed to measure differences across schools in the performance of students in that subgroup relative to the overall performance of students in that subgroup across MMSD. Any overall, district-wide effect of (for example) disability is controlled for in the value-added model and is not included in the subgroup results. The subgroup results reflect relative differences across schools in the growth of students in that subgroup.
Much more on “Value Added Assessment”, here.