Research
What the data actually says
Original analysis built on the same data that powers every school profile on this site — UC's own school-by-school admissions counts, federal education data, and 30 years of trend history. Every brief publishes its methodology and reconciles against official published figures.
Free for editorial use with attribution to High School Reach · Reporters: custom data cuts on request
UC admissions · the $276M compact · August 2026
California paid $276 million to swap out nonresident students. The swap is unwinding.
The state paused the deal's funding in 2025-26 — and in the same cycle, first-year classes at Berkeley, UCLA and San Diego went back to their 2019 nonresident share. The admit files show it was a decision, not a yield accident, and the compliance metric won't register it for years.
Read the brief →UC testing review · August 2026
“Show me the data”: the school-level numbers behind the UC math debate
Newsom asked for data; 3,000+ UC faculty asked for a test. For 1,121 California high schools, both halves of the question are already public. The join: proficiency and UC outcomes rise together (r = 0.72) — yet 20 of the UC's 100 biggest feeder schools post grade-11 math proficiency under 50%.
Read the brief →UC admissions · campus yield · August 2026
UC admits a lot of out-of-state students. Outside Berkeley and UCLA, almost none of them come.
Campus-by-campus first-year yield by residency. UC Davis gives most of its offers to non-Californians, but Californians are 80% of the students who actually enrol — out-of-state yield is 4.8% there and 2.8% at Riverside, against roughly 29% at the two flagships. Full nine-year table as a CSV.
Read the brief →UC admissions · new data · August 2026
Getting into a UC and getting into the UC you wanted are two different things
UC admitted a record 102,031 California first-year students for Fall 2026. The campus-by-campus table — applicants, admit rates from 10.5% to 94.7%, and the admitted-GPA bands — plus why the two flagships now differ sharply on in-state share.
Read the brief →UC admissions · July 2026
The Berkeley Exception: at the same GPA, only one UC favors Californians
Controlling for applicant GPA, an out-of-state student is admitted at a higher rate than an identical Californian at every selective UC except Berkeley — at Irvine, Davis, and Santa Barbara by roughly 30 points.
Read the brief →UC admissions · 30 years · July 2026
The 30-year arms race: applications up 6.8×, admit rates 62% → 27%
In 1994, UC Santa Barbara admitted four of every five applications and UCLA admitted half. The full 1994–2026 series, campus by campus, from UC's own counts — and why parents' intuitions about 'getting in' misfire.
Read the brief →College costs · July 2026
The $134K blind spot: the federal cost data ends where the median college family begins
The government's net-price brackets stop at $110K+ — below the ~$134K median income of college-parent-age families — and average only aid recipients. Across 800 privates, the published figure runs 42% below the no-aid price.
Read the brief →Test scores · July 2026
Do test scores still matter? The colleges' own numbers say yes
At selective test-optional colleges, score submitters were admitted at 1.2 to 3.2 times the rate of non-submitters — derived from each school's own published data. The table, the reinstatement wave, and how a family should decide whether to test.
Read the brief →College mobility · July 2026
The colleges that move kids up: the mobility leaders are regional publics, not famous names
Opportunity Insights measured which colleges take students from the bottom of the income ladder to the top. The leaders are regional publics — six CUNY campuses and Cal State LA among the top twelve — with mobility rates about six times the Ivy-Plus median. The leaderboard, and the access-versus-success math behind it.
Read the brief →AI · Colleges · July 2026
The AI-ready colleges: which schools are actually building for the AI economy
There is no official AI ranking, so we measured three primary-source signals across 1,268 four-year colleges: the AI and computing degrees they confer, their AI research output, and whether they host a federal NSF AI Institute. The leaders differ by lens — and a bachelor's degree in AI itself is still remarkably rare.
Read the brief →National enrollment · July 2026
Where America's high schools are growing — and where they're not
National high-school enrollment is flat — but the flat line hides a sixteen-point spread: 24 states growing (Oklahoma +8.4%), 26 shrinking (Hawaii −6.9%). The state-by-state table, and what it means for the administrators on each side.
Read the brief →Enrollment · July 2026
The public-private divergence: California's high schools are shrinking — except the ones that aren't
Public high-school enrollment fell 1.7% from 2020 to 2025 while private enrollment grew 4.3% — shrinking is now the majority experience for public schools, and the statewide senior-class peak is already behind us. School-by-school winners and losers.
Read the brief →The county series · 20 counties
Where California's high school grads get into the UC system, county by county
Every high school in the county with published 2025 UC outcomes — public and private — ranked by UC Reach Score, with the story behind the table. Covers every California county reporting at least fifteen schools.
In the pipeline
The GPA arms race (what a 4.0 bought in 1994 vs. today) · Does your high school matter? What school-average GPA does — and doesn't — predict · Test-optional in practice: submitter vs. non-submitter admit rates from colleges' own data · District-level tables for the largest unified districts. New briefs monthly — follow a school to hear when they land.