Teacher value-added estimates are sensitive to the test scale on which they are computed¶

Claim · All claims
Evidence · 1 study · associational r? · q2
Currency · newest study 2017 · not yet checked against its sources · not reviewed by a person · overdue for a recheck since 2020: tool and model accuracy changes quickly

Subclaims¶

q2 i? Estimates of teacher value added change when the underlying test scale is transformed, even mildly. → Soland 2017

Evidence¶

Soland 2017¶

Soland, J. (2017). Is teacher value added a matter of scale? The practical consequences of treating an ordinal scale as interval for estimation of teacher effects. Applied Measurement in Education, 30(1), 52–70. https://www.nwea.org/research/publication/is-teacher-value-added-a-matter-of-scale-the-practical-consequences-of-treating-an-ordinal-scale-as-interval-for-estimation-of-teacher-effects/ doi:10.1080/08957347.2016.1247844

q2 · i? · associational · r?

The study estimated teacher value added on an original scale, applied nonlinear transformations, and re-estimated, finding that "value added is sensitive to the scale used." No effect size is printed in the available text.

"Results show that value added is sensitive to the scale used."

Discussion¶

Related Claims¶

• Mild departures from the original scale can change a teacher's odds of being classified high- or low-performing by a factor of 5 — a narrower finding that bears on this claim
• Rank orderings of teachers do not change radically under mild scale transformations — related
• The choice of which assessment to use as the outcome measure matters more for value-added estimates than any student or peer control-variable specification choice — related