Table of Contents

Misconception: Teachers of small classes are disadvantaged with PVAAS.

The PVAAS Teacher Value-Added report provides a value-added (or growth) measure as well as a standard error. The standard error provides information about the amount of evidence or data used in the growth measure; the two metrics are used together to assess whether there is significant or moderate evidence that the teacher's group of students have met, exceeded, or fallen short of the growth standard. The standard error is based on the number of students linked to the teacher as well as the variability in those students' test scores. Although there might be concern that teachers of small classes are disadvantaged by PVAAS, they are actually protected by using a value-added estimate and standard error together.

PVAAS in Theory

Students in all class sizes have the ability to show growth, and the standard error simply provides a confidence band around each estimate. With smaller amounts of data (meaning fewer students), there is less evidence in each estimate, so the standard error tends to be larger than teachers linked to a large number of students. However, although teachers of small classes might have larger standard errors than other teachers, they are also more likely to have a larger gain—either positive or negative. Thus, the two metrics even out, and teachers of small classes are not disadvantaged.

PVAAS in Practice

Actual data may be the most readily apparent way to demonstrate that small classrooms show similar growth as large classrooms. The chart below plots the number of students used in each teacher's PVAAS Value-Added report against the teacher growth index (the value-added estimate divided by its standard error) for PSSA Mathematics in grade 5 in 2019. Each dot represents one teacher, and verified rosters were used where available. The chart demonstrates that teachers serving both small and large numbers of students can show both high and low growth, as measured by PVAAS. Although current state policy requires that teachers are linked to at least 11 individual student scores in order to receive a Teacher report, the chart below shows that even teachers of very small classrooms are not disadvantaged with a sophisticated value-added approach, like PVAAS. In the chart below, the actual correlation between the growth index and number of students is -0.03, which is negligible.

PENNSYLVANIA GROWTH INDEX VERSUS NUMBER OF STUDENTS LINKED TO TEACHER