Table of Contents

Misconception: There is a fixed number of districts, schools, and teachers for each effectiveness level.

The TVAAS value-added measures are based on one of two models, depending on the test. In both models, the expectation of growth is based on the statewide pool of test-takers for the subject subject/grade/year (or subject/year for EOC) of interest. More specifically, in both models, expected growth is based on the average amount of progress observed in the state. However, regardless of model, there is not a set distribution or fixed number of districts, schools and teachers for each effectiveness level.

TVAAS in Theory

Due to differences in tests and when students take each test, there are two different value-added models used to measure growth. While similar in concept, the models differ in the precise way that "growth" is determined. A short explanation of each is provided below:

  • For EOC and TCAP Social Studies assessments, expected growth is that students with a district, school, or teacher made the same amount of progress as students with the average district, school, or teacher in the state for that same year/subject/grade. If not all students are taking an assessment in the state, then it may be a subset.
  • For TCAP Mathematics and English Language Arts in grades 4 – 8, expected growth is that students maintained the same relative position with respect to the statewide student achievement from one year to the next in the same subject area. As an example, if students' achievement was at the 50th NCE in 2014 grade 4 math, based on the 2014 grade 4 math statewide distribution of student achievement, and their achievement is at the 50th NCE in 2015 grade 5 math, based on the 2015 grade 5 math statewide distribution of student achievement, then their estimated gain is 0.0 NCEs.

With either approach, the value-added measures tend to be centered on expected growth every year, with approximately half of the district/school/teacher estimates above zero and approximately half of the district/school/teacher estimates below zero.

However, it should be noted that there is not a set distribution of the value-added measures and being centered on the growth expectation does not mean half of the measures would be in the positive levels and half would be in the negative levels since many value-added measures are indistinguishable from the expectation when considering the statistical certainly around that measure.

TVAAS in Practice

The distribution of effectiveness level based on single-year 2019 TVAAS measures for teachers across subjects and grades is provided below, and this information illustrates that the distribution can vary by subject and grade, with the majority of teachers' students meeting or exceeding expected growth. The actual distribution depends on the extent to which there are measurable differences among districts, schools, and teachers, and the exact distribution in a subject/grade can vary from year to year.

As the figure below also illustrates, the percentage of reports considered either Level 1 or 2 is similar to those considered Level 4 or 5.

DISTRIBUTION OF TEACHER EFFECTIVENESS LEVELS BY SUBJECT AND GRADE