Student Welfare Optimization in College Matching

Source: student_welfare_matching.md


Student Welfare Optimization in College Matching

Student-Optimal Deferred Acceptance Empirically

Theoretical Foundation

The Gale-Shapley Deferred Acceptance (DA) algorithm (1962) produces a student-optimal stable matching when students propose: each student receives their most-preferred partner consistent with stability. The key properties:

Empirical Implementations

NYC High School Match (2003)

Boston School Choice (2005)

NRMP Medical Residency Match

Known Limitations

  1. Not Pareto efficient: DA does not maximize total student welfare. Abdulkadiroglu, Pathak, and Roth showed that the inefficiency "can potentially be severe," and empirical findings from the NYC match corroborated this
  2. Proposer advantage: Students get optimal stable matching, but this can still be far from their first choices at highly selective institutions
  3. No mechanism is both stable and efficient: Stability and Pareto efficiency are fundamentally incompatible (Roth, 1982). Gains for some students from breaking stability always create justified envy for others
  4. Tiebreaking matters: When colleges are indifferent among students, different tiebreaking rules lead to different matchings with different welfare properties

Relevance to College Admissions

U.S. college admissions does not use DA. Instead, it operates as a decentralized market with:

This decentralized structure introduces information frictions, strategic complexity, and welfare losses that a centralized DA mechanism would partially address.


Alternative Mechanisms for Student Welfare

Mechanism Comparison

Mechanism Strategy-Proof Stable Pareto Efficient Used Where
Student-proposing DA Yes (for students) Yes No NYC schools, Boston, NRMP
College-proposing DA No (for students) Yes No Theoretical
Top Trading Cycles (TTC) Yes No Yes Theoretical; some kidney exchange variants
Boston/Immediate Acceptance No No No Pre-2005 Boston, China (variants)
Serial Dictatorship Yes N/A Yes Simple assignment problems
Decentralized (current U.S.) N/A No No U.S. college admissions

Top Trading Cycles (TTC)

Boston/Immediate Acceptance Mechanism

Consistent Pareto Improvement over DA

Implications for Simulation

The decentralized U.S. college admissions market is none of these mechanisms -- it lacks strategy-proofness, stability, and efficiency. This creates space for modeling:


Existing ABM Simulations of College Admissions

Reardon, Kasman, Klasik, and Baker (2016) -- Stanford CEPA

"Agent-Based Simulation Models of the College Sorting Process" Published in Journal of Artificial Societies and Social Simulation (JASSS), Vol. 19, Issue 1.

Model Architecture:

Key Parameters:

Parameter Value Source
Resource-caliber correlation 0.3 ELS:2002
Quality reliability 0.7 + 0.1 x resources Plausible estimate
Caliber enhancement +0.1 x resources Test prep literature
Application count 4 + 0.5 x resources ELS:2002

Information Model:

Admission Model:

Key Findings:

  1. Resource-caliber correlation is the dominant driver of sorting inequality (eliminating it reduced the 90th-10th percentile gap from 20x to 4x)
  2. Information disparities, application enhancement, application count inequality, and utility preferences each produce modest individual effects but collectively create "non-trivial" stratification
  3. Model reached equilibrium by year 10-20
  4. Validated against IPEDS 2010-2011 data: selectivity and yield patterns matched real institutional data

Relevance: This is the closest published model to the college-sim project architecture. Key differences from our simulator: Reardon et al. use continuous distributions rather than archetype-based student generation, and a simpler two-attribute student model.

Assayed and Maheshwari (2023) -- Jordan Medical Colleges

"Agent-Based Simulation for University Students Admission: Medical Colleges in Jordan Universities"

Assayed and Al-Sayed (2025) -- Survey Paper

"Student Behaviors in College Admissions: A Survey of Agent-Based Models" Published in International Journal of Emerging Multidisciplinaries.

Sirolly (2023) -- Toy Model

"A Toy Model of College Admissions"

Other Notable Models


Undermatching / Mismatch Literature

Hoxby and Avery (2012) -- The Foundational Paper

"The Missing 'One-Offs': The Hidden Supply of High-Achieving, Low-Income Students" NBER Working Paper 18586.

Key Findings:

Student Typology:

Hoxby and Turner (2013) -- The ECO Intervention

"Expanding College Opportunities for High-Achieving, Low-Income Students"

Intervention Design:

Results:

Implication: Information intervention alone dramatically reduces undermatching. The problem is primarily informational, not financial or academic.

Determinants of Mismatch (NBER Working Paper 19286)

Key Findings:

Lincove and Cortes (2016) -- Automatic Admissions

"Match or Mismatch? Automatic Admissions and College Preferences of Low- and High-Income Students" NBER Working Paper 22559.

Bastedo and Flaster (2014) -- Methodological Critique

"Conceptual and Methodological Problems in Research on College Undermatch"

Mizala et al. (2026) -- International Evidence

"Bright but Poor: Undermatching in the Access to Postsecondary Education" American Educational Research Journal.

Welfare Consequences of Undermatching

Empirical evidence on outcomes:

  1. Graduation rates: Students who undermatch graduate at lower rates than peers at better-matched institutions
  2. Earnings: Attending a more selective institution is associated with higher lifetime earnings, particularly for low-income and minority students (Dale and Krueger, 2014)
  3. Graduate school access: Selective college attendance increases probability of graduate/professional school enrollment
  4. Network effects: Peer quality, alumni networks, and institutional resources compound over careers

Key Parameters for Simulation

Based on the literature, these are the critical parameters for modeling student welfare in a college admissions simulation:

Student-Side Parameters

Parameter Literature Value Source
Resource-caliber correlation 0.3 Reardon et al. (ELS:2002)
Information quality (low-resource) 0.7 base Reardon et al.
Information quality (high-resource) 0.7 + 0.1 x resources Reardon et al.
Application count (low-resource) 4 applications Reardon et al. (ELS:2002)
Application count (high-resource) 4 + 0.5 x resources (up to ~7) Reardon et al. (ELS:2002)
Caliber enhancement from resources +0.1 x resources Test prep literature
Undermatching rate (low-income, high-achieving) ~92% income-typical behavior Hoxby & Avery (2012)
Information intervention effect +46% peer enrollment Hoxby & Turner (2013)

College-Side Parameters

Parameter Literature Value Source
Yield estimation window 3-year running average Reardon et al.
Admission volume adjustment Based on prior year fill rate Reardon et al.
Quality metric Weighted average enrolled caliber Reardon et al.
ED yield boost Binding commitment ~90%+ yield Common knowledge

System-Level Parameters

Parameter Description Typical Range
Stability % of matched pairs with no blocking pair 85-95% in decentralized markets
Pareto efficiency % of students who could improve without harming others DA achieves ~85-90% of optimal
Undermatching rate % of students at institutions below their caliber 20-40% depending on definition
Strategic behavior prevalence % of students who misrepresent preferences 10-30% under non-strategy-proof mechanisms

Information Asymmetry Parameters

  1. Student knowledge of own caliber: How accurately students assess their competitiveness (signal noise)
  2. Student knowledge of college quality: How well students perceive fit and resources (correlated with SES)
  3. College knowledge of student quality: Admissions offices observe noisy signals (GPA, SAT, essays) of true ability
  4. Strategic sophistication: Proportion of students who optimize application portfolios (higher in high-SES)

Recommendations for College Simulator

1. Add Information Asymmetry Layer

The current simulator uses deterministic scoring. The literature strongly suggests adding:

Implementation suggestion: Add a perceptionNoise parameter to each student archetype. Elite prep school students get low noise (0.05-0.1); rural/under-resourced students get high noise (0.3-0.5). This single parameter captures much of the Reardon et al. information asymmetry finding.

2. Model Undermatching Explicitly

Based on Hoxby and Avery:

Implementation suggestion: When generating application lists for students from under-resourced high schools, apply a "consideration set filter" that removes colleges the student has never heard of (probability based on distance, marketing reach, and school counselor quality).

3. Track Welfare Metrics

Add post-simulation welfare analysis:

4. Implement Yield Management Feedback

Colleges should adjust behavior over simulation runs:

5. Model Strategic Behavior Heterogeneity

Not all students are equally strategic:

6. Consider Adding a DA Benchmark Mode

For research validity, implement an optional mode where:

This would allow measuring the "price of decentralization" in student welfare terms.

7. Calibration Targets

Validate the simulation against known empirical patterns:


References

  1. Gale, D. & Shapley, L.S. (1962). "College Admissions and the Stability of Marriage." American Mathematical Monthly, 69(1), 9-15.
  2. Roth, A.E. (2008). "Deferred Acceptance Algorithms: History, Theory, Practice, and Open Questions." International Journal of Game Theory, 36, 537-569.
  3. Abdulkadiroglu, A., Pathak, P.A., & Roth, A.E. (2005). "The New York City High School Match." American Economic Review P&P, 95(2), 364-367.
  4. Abdulkadiroglu, A., Pathak, P.A., Roth, A.E., & Sonmez, T. (2006). "Changing the Boston School Choice Mechanism." NBER Working Paper 11965.
  5. Abdulkadiroglu, A. & Sonmez, T. (2003). "School Choice: A Mechanism Design Approach." American Economic Review, 93(3), 729-747.
  6. Hoxby, C.M. & Avery, C. (2012). "The Missing 'One-Offs': The Hidden Supply of High-Achieving, Low-Income Students." NBER Working Paper 18586.
  7. Hoxby, C.M. & Turner, S. (2013). "Expanding College Opportunities for High-Achieving, Low-Income Students." Stanford Institute for Economic Policy Research Discussion Paper 12-014.
  8. Reardon, S.F., Kasman, M., Klasik, D., & Baker, R. (2016). "Agent-Based Simulation Models of the College Sorting Process." Journal of Artificial Societies and Social Simulation, 19(1), 8.
  9. Pathak, P.A. & Sonmez, T. (2008). "Strategy-Proofness versus Efficiency in Matching with Indifferences: Redesigning the NYC High School Match." American Economic Review, 98(5), 1636-1689.
  10. Erdil, A. & Ergin, H. (2008). "What's the Matter with Tie-Breaking? Improving Efficiency in School Choice." American Economic Review, 98(3), 669-689.
  11. Bastedo, M.N. & Flaster, A. (2014). "Conceptual and Methodological Problems in Research on College Undermatch." Educational Researcher, 43(2), 93-99.
  12. Assayed, S.K. & Maheshwari, P. (2023). "Agent-Based Simulation for University Students Admission: Medical Colleges in Jordan Universities."
  13. Assayed, S.K. & Al-Sayed, S. (2025). "Student Behaviors in College Admissions: A Survey of Agent-Based Models." International Journal of Emerging Multidisciplinaries.
  14. Kloosterman, A. (2020). "School choice with asymmetric information: Priority design and the curse of acceptance." Theoretical Economics.
  15. Mizala, A. et al. (2026). "Bright but Poor: Undermatching in the Access to Postsecondary Education." American Educational Research Journal.