Frequently asked questions

I’ve heard that the student achievement component in Dallas ISD's Teacher Excellence Initiative (TEI) is fair. Is this true?


Proponents of TEI say:

  • Three metrics are used, and the teacher gets credit for the best of the three.
  • Even if the student isn’t on grade level the teacher will get credit for academic “growth” shown. ​
Student achievement in TEI is based on standardized test scores. Listed below are the three metrics used to rank and pay teachers in DISD's TEI, and why they shouldn't be. STAAR pass rate – The higher the socioeconomic status (SES) of your students the better they will perform on standardized tests. This statement is uncontroversial as it has been well documented for decades. Teachers cannot control the SES of their students and thus should not be paid accordingly. Furthermore, rapid improvement in pass rates on high-stakes tests has been shown to be due to test prep not content mastery. ( New York, Texas, Texas) Value-added model (VAM) – It is an unreliable growth model in education.
  • No less an organization than the American Statistical Association raises serious concerns about the lack of reliability with VAM for evaluating teachers.
  • Peer-reviewed research, that included Dallas ISD 4th and 8th grade teachers, found weak to nonexistent relationships between VAM measures and the quality of teachers’ instruction. The author said, “Given the growing extent to which states are using these measures for a wide array of decisions, our findings are troubling."
  • The Intensive Partnership for Effective Teaching was a six year national study that mirrored the TEI model and found no improvement in student outcomes and no improvement in teacher effectiveness.
Student Growth Percentile (SGP) – It is an unreliable measure of academic growth. DISD’s SGP is called Academic Peer Groups. SGPs were approved in Race-to-the-Top grants in 2010 which led to wide use of SGPs. Unfortunately it’s statistical properties had not been well researched at that time. Research conducted since that grant concluded that the SGPs are too unreliable to be valid.

While there is information to be gleaned from standardized tests, current tests were not designed for nor are they capable of what is called high-stakes use, such as grade advancement, participation in advanced coursework, salary determination, and failing and closing schools. And when a test-based metric is given such importance, the focus of instruction becomes test centered, reducing or eliminating subject areas that aren’t tested.