Common Data Set & Admissions Data: Top 20-50 Selective Colleges

data_top20_50_cds.md

Common Data Set & Admissions Data: Top 20-50 Selective Colleges

Research compiled from 2023-24 Common Data Sets and institutional admissions reports.

Data Table: All Schools

Notes on SAT Data

• UCLA: Does not use SAT/ACT in admissions (UC system test-free policy). No SAT percentiles reported in CDS.

• Test-optional schools: Most schools above were test-optional for 2023-24. SAT ranges reflect submitters only (typically 40-60% of enrolled students), creating upward bias.

• Michigan: Wide SAT range (1340-1530) reflects large public university with in-state/out-of-state mix. Overall acceptance rate ~18%; in-state ~39%, out-of-state ~18%.

• Georgetown: Requires SAT/ACT submission (not test-optional), so SAT data is more representative of the full admitted class.

Acceptance Rate Tiers (for simulation calibration)

Ultra-Selective (<6%)

Very Selective (6-10%)

Selective (10-20%)

Feeder School Data Availability

Schools That Publish or Have Published Feeder Data

No colleges in this list officially publish feeder school matriculation data in their CDS or admissions reports. Feeder school data comes from external sources:

1. Harvard Crimson (2024 investigation): Identified 21 schools that collectively send ~1 in 11 accepted Harvard students. Top feeders include Phillips Exeter, Phillips Andover, Boston Latin, Stuyvesant. This is the most comprehensive public feeder data for any elite school.

2. National Student Clearinghouse: Publishes aggregate High School Benchmarks Report with high-school-to-college enrollment data. Does not break out by specific college destination at the individual school level publicly.

3. Michigan School Data Portal: Michigan publishes state-level data showing which colleges Michigan high school graduates attend (mischooldata.org).

4. College-Specific Sources:

5. Princeton's Class of 2028: ~60% from public high schools (published aggregate, not school-specific)

6. Yale's Class of 2027: ~60% from public high schools (published aggregate, not school-specific)

7. Brown: Publishes aggregate public/private split but not individual feeder schools

8. Third-Party Databases:

9. IvyLeagueFeeders.com: Crowdsourced data on feeder school placements

10. Crimson Education research: Lists of top 20 feeder schools to Ivies

11. Niche.com: School profiles sometimes include college matriculation data reported by schools

Feeder Data Availability by School

Key Finding

No elite college publishes official feeder school lists. The best feeder data comes from:

• Investigative journalism (Harvard Crimson 2024)

• State education departments (Michigan, California UC system)

• Crowdsourced platforms (IvyLeagueFeeders.com)

• Individual high school college counseling reports (e.g., Naviance data shared by schools)

Data Gaps Assessment

High-Quality Data Available (Simulation-Ready)

These data points are reliably available from CDS for all 25 schools:

• Overall acceptance rate

• Total applications received / admitted / enrolled

• SAT/ACT score percentiles (except UCLA)

• GPA distributions

• Class rank distributions

• Geographic diversity (in-state/out-of-state/international)

• First-generation student percentage

• Pell Grant recipient percentage

Moderate Data Gaps

• ED/EA acceptance rates: Some schools report ED numbers in CDS Section C (e.g., Carnegie Mellon ED: 12.5%), but many do not break out by round. This is critical for simulation round modeling.

• Hook multipliers: No CDS reports explicit admit-rate boosts for athletes, legacies, or donors. Must be inferred from investigative reporting (e.g., Harvard trial data, Duke/Princeton legacy studies).

• Yield rates: CDS provides enrolled/admitted ratios, but not by round or demographic segment.

• Waitlist conversion rates: CDS Section C9 has waitlist data for some schools, but reporting is inconsistent.

Significant Data Gaps (Not in CDS)

• Feeder school matriculation: No official data. See Feeder section above.

• Legacy admit rates: Not published. Estimates from studies: ~30-40% at Ivies (vs. 3-10% overall).

• Athlete admit rates: Not published. Estimates: recruited athletes admitted at 2-4x the base rate at most selective schools.

• Donor/development case rates: Not published. Extremely limited public data.

• Essay/EC scoring rubrics: Proprietary. Some insight from Harvard litigation discovery.

• Regional/school-type quotas: Not published. Evidence of geographic balancing from CDS demographic data.

Recommendations for Simulation Parameters

1. Acceptance rates: Use CDS data directly. All 25 schools have reliable 2023-24 data.
2. SAT ranges: Use CDS 25th/75th percentiles. Note test-optional bias (submitter pool is self-selected higher). For UCLA, use GPA-only model.
3. ED/EA boosts: Estimate from schools that report (Carnegie Mellon, Dartmouth, Brown). Typical ED boost: 2-3x the RD rate.
4. Hook multipliers: Use Harvard trial data as baseline: legacy ~5-6x, athlete ~4x (ALDC data), first-gen ~1.5x.
5. Yield rates: Calculate from CDS enrolled/admitted. Range: ~45-70% for HYPSM+, 30-50% for Ivy+, 20-40% for selective.
6. Feeder school effects: Model as a quality multiplier based on school tier rather than specific feeder pipelines, since school-specific data is unavailable.

Source Summary

Official CDS Documents (Direct PDF Links)

• Columbia CDS 2023-24

• Brown CDS 2023-24

• Northwestern CDS 2023-24

• UChicago CDS 2023-24

• Johns Hopkins CDS 2023-24

• Rice CDS 2023-24

• Notre Dame CDS 2023-24

• Carnegie Mellon CDS 2023-24

• WashU CDS 2023-24

• Emory CDS 2023-24

• UCLA CDS 2023-24

• Michigan CDS 2023-24

• Williams CDS 2023-24

CDS Portal Pages

• UPenn CDS

• Dartmouth CDS

• Cornell CDS

• Vanderbilt CDS

• Georgetown CDS

• Boston College CDS

• UVA CDS

• Amherst CDS

Feeder School Data Sources

• Harvard Crimson 2024 Feeder School Investigation

• Michigan School Data Portal

• National Student Clearinghouse Benchmarks

• UCLA Academic Planning & Budget