RESEARCH & MODELING

MEASURING WHAT TEST SCORES MISS

Which forms of potential are devalued by the current outcome-driven system?

When an employer sees a candidate from a prestigious university with a high GPA, how much of that signal reflects real ability, and how much reflects a family’s capacity to pay for tutoring, coaching, and admission to a brand-name school?

Empirical estimation in progress


We do not define well-roundedness as the absence of a spike.

Olympiad Records & Academic Outcomes

Credentials, Ability & Family Resources

01 / EMPIRICAL RESEARCH · WORKING DRAFT


We ask whether, conditional on comparable educational opportunity, well-rounded scholars achieve research and compensation outcomes statistically equivalent to those of spiky scholars.

We ask whether, conditional on comparable educational opportunity, well-rounded scholars achieve research and compensation outcomes statistically equivalent to those of spiky scholars.

WELL-ROUNDED

Documented breadth across domains, with no verified competition record and a breadth index above the sample median.

SPIKY

A verified national or international competition record.

DO OLYMPIAD MEDALS STILL MATTER AFTER ADMISSION?

We do not define well-roundedness as the absence of a spike.


This comparison is limited to scholars who have already become assistant professors.

COMPARE WITHIN Undergraduate institution tier · Undergraduate major · PhD cohort · Current academic field

EXAMINE Field-adjusted research outcomes · Compensation where public salary records are available

COMPARE WITHIN Undergraduate institution tier · Undergraduate major · PhD cohort · Current academic field

EXAMINE Field-adjusted research outcomes · Compensation where public salary records are available

WHAT WOULD MAKE THE COMPARISON CONVINCING?

COMPARE WITH CARE

Compare scholars within the specified educational and career groups, using research measures that account for differences across fields.

VERIFY AWARDS

Check competition records against published rosters by name, country, and year. A missing line on a CV is not enough to establish that someone never won an award.

TEST EQUIVALENCE

Set a practically meaningful margin before estimation. A statistically insignificant difference alone does not establish equivalence.

02 / ECONOMIC MODELING · EMPIRICAL ESTIMATION IN PROGRESS

When an employer sees a candidate from a prestigious university with a high GPA, how much of that signal reflects real ability, and how much reflects a family’s capacity to pay for tutoring, coaching, and admission to a brand-name school?

When an employer sees a candidate from a prestigious university with a high GPA, how much of that signal reflects real ability, and how much reflects a family’s capacity to pay for tutoring, coaching, and admission to a brand-name school?


WHAT DOES A CREDENTIAL REALLY TELL US?


The model uses neural networks to approximate decisions across a working life. Empirical estimation is ongoing, using the Panel Study of Income Dynamics. The aim is to understand how employer learning, the detection of credential inflation, and family wealth affect the return to an inflated signal.

COLLEGE

Admission and prestige shaped by academic strength and financial resources.

HIGH SCHOOL

Ability, family resources, and investment in academic credentials.

WORKING LIFE

Employers update their beliefs as they observe work; individuals continue developing skills.

DANG NGUYEN SY HOANG

Academic Portfolio · 2026–2027

mail: hoangndsy7@gmail.com