The College Town Effect

A lot of places in America look like this in the ACS:

• very young
• surprisingly “poor” (high poverty rate)
• often not actually cheap (rent and home values can still be high)

That combination can be real distress.

But sometimes it’s something else: students.

When a county has a big university, the headline stats can get skewed in a predictable way. It doesn’t mean the data is wrong — it means the composition is different.

This is the college town effect.

Key takeaways

• Median age drops in college counties because student populations are concentrated in a narrow age band.
• Poverty rate rises because many students have low personal income (even if their real living standard is supported by family, aid, or loans).
• Income looks “low” in a way that can be misleading if you read it as “local jobs are terrible.”
• Housing can still be expensive because demand is real (and constrained housing supply is common in college towns).

The pattern: young + high poverty is a clue

Here’s the simplest way to see it: put every county on a scatter plot:

• X-axis: median age (lower = younger)
• Y-axis: poverty rate (higher = more people below the poverty line)

College-heavy counties tend to show up in the upper-left: young and high-poverty.

A high poverty rate in a college town does not automatically mean “everyone is struggling.”

It means the county includes a large population whose income looks low on paper, by design (students).

That said: this is not a “poverty isn’t real” argument. College towns have service workers, renters, families, and people on fixed incomes too. The point is that the headline rate becomes harder to interpret without context.

A simple “college town effect” score

To make this concrete, I built a simple index that looks for counties that are:

• younger than typical
• higher-poverty than typical
• (optionally) more college-educated than typical (if the metric is available)

This is not a “best” list. It’s a pattern detector — a shortlist builder for places where the poverty rate needs extra context.

If you’re scanning a list and something looks odd — young + high-poverty + not cheap — this table shows the counties where that pattern is strongest.

Four examples (and what to check next)

These are classic “college town effect” counties. Each one has a real local economy — but the ACS headline stats are heavily shaped by student population.

1) Watauga County, NC (Boone)

If you only see “high poverty,” you might miss the obvious context: Watauga is extremely young.

2) Tompkins County, NY (Ithaca)

3) Washtenaw County, MI (Ann Arbor)

4) Dane County, WI (Madison)

In all of these, the “right” question isn’t “is poverty real?” — it’s:

What does the poverty rate represent in this place?

My quick checklist when I see the pattern

1. Median age: if it’s unusually low, student composition is likely part of the story.
2. Rent burden: college towns can be expensive even when income looks low.
3. Compare to neighbors: if the county stands out but nearby counties don’t, it’s a clue.
4. Look at the place page: use USAviz to sanity-check the basics and the peer medians.

What this is (and isn’t)

This is:

• A way to avoid misreading ACS poverty rates in places with large student populations.
• A pattern scan that points you to places where extra context matters.

This isn’t:

• A claim that poverty in college towns is “fake.”
• A substitute for looking at neighborhoods, wages, housing stock, or local policy.

Explore your county

• Browse counties: /counties
• Or search any ZIP/county/metro: /search

If you find a place that looks “young + poor + expensive,” there’s a decent chance you’ve found the college town effect in the wild.

Methodology & sources

• Core demographics and economic metrics: U.S. Census Bureau, ACS 2020–2024 5-year estimates.
• Poverty rate refers to ACS table S1701 “Percent of population below poverty level” (all people; S1701_C03_001E). Family poverty is a different measure and can downweight the student effect.
• Geography: counties (FIPS).
• Filtering for charts: counties with population ≥ 50K (to reduce tiny-county noise).
• “College town effect” score: weighted percentile ranks for youth (lower median age), poverty (higher poverty rate), and education (higher bachelor’s+ share, if available in the dataset build).

Caveat: Small-area estimates can be noisy, and poverty is a threshold-based measure. Treat small differences as noise and look for big patterns.

Closing thought

The ACS is one of the best tools we have for seeing local America — but it’s not a personality test.
In places with lots of students, the headline stats can look contradictory until you remember who’s in the denominator.