Thank you, Alexandra!

I am also really passionate about teaching economics! #TeachEcon
I served as the Graduate Pedagogy Fellow for my department at UCSC, working with teaching teams to design scaffolded, active-learning based activities and effective assessments for high-enrollment courses.
See more: bit.ly/gs_teaching

Beyond my JMP, I’m working on several other projects in education. One of these (with
@RobertFairlie6
, Saurabh Khanna, and Prashant Loyalka) is forthcoming at Journal of Political Economy: Microeconomics.

You can check them out here: bit.ly/research_gag...

(11/n)

Link to JMP: bit.ly/jmp_gagandeep

I am immensely grateful for the support of my advisors and committee members: Laura Giuliano, George Bulman, and Robert Fairlie, as well as the incredible #econ department at UC Santa Cruz
(10/n)

Relying solely on test-score VA to evaluate teachers can miss the teachers who are most effective at improving non-cognitive skills, which matter for boys’ trajectories. This suggests we need better ways to measure and integrate non-cognitive VA into accountability systems.
(9/n)

What does this mean for policy?
Test-score VA measures a real dimension of teaching, but it’s (a) not all that teachers do, and (b) has stronger impacts on girls—who are not the group falling behind.
(8/n)

As a result, I show that the gender-differentiated impacts of teachers are not stemming from teachers teaching boys and girls differently, or from teacher-like-me effects—but from teachers' individual strengths interacting differently with students' relative skill gaps.
(7/n)

Applying this framework to the data, I show that gender gaps in observed outcomes imply gender gaps in unobserved skills—specifically a relative deficiency in non-cognitive skills for boys and in cognitive skills for girls (even with girls outperforming boys).
(6/n)

To interpret these results, I propose a framework wherein teachers' strengths interact with the relative skill mixes of their students. I show that teachers who improve a given skill have stronger impacts on the group of students with a relative deficiency in said skill.
(5/n)

(a) Teachers who improve cognitive skills improve girls' test scores and course grades more than boys' outcomes.
(b) Teachers who improve non-cognitive skills improve boys' outcomes more than girls' outcomes
(c) They are typically not the same teachers!
(4/n)

Prior work measures teacher effectiveness (aka value-added) in both test scores and grades, interpreting them as teacher impacts on cognitive and non-cognitive skills respectively. I build on this, by studying teachers’ gender-differentiated impacts in each dimension. Here’s what I find👀:
(3/n)

My JMP investigates gender gaps in standardized test scores and teacher-assigned grades as students progress through school—and the role that teachers play in this context. Boys are typically behind on both these metrics, and these gaps persist/grow with time.
(2/n)