Base strength predicted outcome improvement better than growth in aggregate, and pooled regression estimates exceeded school fixed effects estimates¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r3 · q2
Currency · newest study 2020 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Most outcome improvement was, in aggregate, better predicted by strength in a measure coming into the school year than by growth during the year. → Hart 2020 q2 i? Measures generally predicted larger outcome changes in the pooled regression model than in the more conservative school fixed effects model. → Hart 2020 (2)

Evidence¶

Hart 2020¶

Hart, H., Young, C., Chen, A., Zou, A., & Allensworth, E.M. (2020). Supporting school improvement: Early findings from reexamination of the 5Essentials survey. Chicago, IL: University of Chicago Consortium on School Research. https://consortium.uchicago.edu/publications/supporting-school-improvement

q2 · i? · associational · r3

Results overview comparing base strength and growth predictors across outcomes. The report gives the example that "base strength in Student-Teacher Trust predicted improvement in college enrollment better than growth in Student-Teacher Trust."

"We also observed that most outcome improvement was, in aggregate, better predicted by having strength in a measure coming into the school year, rather than by growing over the school year in question."

Hart 2020 (2)¶

Hart, H., Young, C., Chen, A., Zou, A., & Allensworth, E.M. (2020). Supporting school improvement: Early findings from reexamination of the 5Essentials survey. Chicago, IL: University of Chicago Consortium on School Research. https://consortium.uchicago.edu/publications/supporting-school-improvement

q2 · i? · associational · r3

Comparison of the two statistical models. The report explains the pooled regression is a more liberal test while the fixed effect model is more conservative, since schools with consistent measure scores and outcome growth contribute no within-school variation to estimate.

"In general, measures predicted a larger change in outcomes in the pooled regression model than the fixed effects model."

Discussion¶

Related Claims¶

• 5Essentials Survey measures predict school improvement in elementary and high schools, but not all measures relate to all outcomes — related
• Elementary attendance improved more in schools that were strong or growing on 5Essentials Survey measures, in both pooled and within-school models — related