Urban Data

By the end of this lesson, students will be able to (per CED LO IMP-6.E and Suggested Skill 3.E):

• Explain how qualitative and quantitative data are used to show the causes and effects of geographic change within urban areas (per LO IMP-6.E).
• Identify standard quantitative sources — decennial census, American Community Survey, Eurostat, national statistical offices, and Demographic and Health Surveys — and describe what they measure (per EK IMP-6.E.1).
• Identify standard qualitative sources — ethnographic fieldwork, interviews, photo documentation, oral histories, and community narratives — and describe what they reveal (per EK IMP-6.E.2).
• Interpret a combination of quantitative and qualitative urban data to characterize change in a neighborhood.
• Apply Skill 3.E by explaining what a dataset or map implies about urban processes, rather than simply describing it.
• Identify the limits of each data type and justify the value of mixed methods and triangulation.

Topic 6.9 is the methods lesson of Unit 6. Per EK IMP-6.E.1 and EK IMP-6.E.2, the CED names two complementary data traditions and expects students to use both to explain urban change. Skill 3.E elevates the task: students do not merely read the data, they must explain what the data imply about geographic principles, processes, and outcomes.

Two data traditions, two kinds of questions

Quantitative and qualitative data answer different questions about the same place. Quantitative data answer how many, how much, how fast. Qualitative data answer how does this feel, why do people act, what do residents value. Neither tradition alone produces a full account of urban change. The best urban research uses both.

• Quantitative (EK IMP-6.E.1): measures population composition and size — ages, races, incomes, housing, commuting.
• Qualitative (EK IMP-6.E.2): captures individual attitudes — hope, grief, resistance, adaptation, belonging.
• Skill 3.E: students must explain what data imply about causes and effects, not merely describe values.

Per EK IMP-6.E.1, quantitative data from census and survey data provide information about changes in population composition and size in urban areas. Several sources dominate in practice.

What quantitative sources measure, in aggregate

Pooling across these sources, urban geographers obtain reliable measures of: total population, age structure, race and ethnicity composition, household income, poverty rate, housing occupancy and tenure, commuting patterns and mode choice, languages spoken at home, educational attainment, and migration in and out of each census tract or district. These measures allow year-over-year and tract-over-tract comparisons that reveal the composition-and-size changes named in EK IMP-6.E.1.

Quantitative data become useful when they are mapped, indexed, or plotted across time. Urban researchers rely on several recurring products:

• Heat-island maps: surface-temperature rasters, built from satellite thermal bands and correlated with census-tract tree-canopy and impervious-surface measurements, reveal where asphalt-heavy neighborhoods reach 4 to 7°C hotter than tree-shaded ones in the same city.
• Gentrification indices: composite scores that combine rising median rent, rising college-educated share, falling minority-population share, and rising median household income per census tract. Governing magazine, the Urban Displacement Project, and academic teams publish these indices regularly.
• Eviction-filing dashboards: the Eviction Lab at Princeton University, led by Matthew Desmond, maps eviction-filing rates by census tract and has documented that Black majority tracts in United States cities face eviction filings at double the rate of white majority tracts at the same income level.
• Census-tract dot-density maps: one dot per 100 residents, colored by race or language, reveal patterns of segregation and integration at a glance.
• Age-structure pyramids per district: highlight greying suburbs, childless downtowns, and student-dense districts.

Worked example — reading early-stage gentrification

Consider a Washington, D.C. census tract in the Shaw neighborhood with the following ACS five-year changes between 2015 and 2023:

What do these numbers imply? Rising rents combined with a falling Black population share, a rising college-educated share, rising household income, and falling child share together imply classic early-stage gentrification. The tract is being re-sorted from long-term Black family residents to short-term college-educated newcomers. The quantitative data do not prove displacement — families could be moving for other reasons — but they strongly imply it. Skill 3.E is the practice of making and defending that inference.

Per EK IMP-6.E.2, qualitative data from field studies and narratives provide information about individual attitudes toward urban change. Where quantitative data count, qualitative data listen.

• Ethnographic fieldwork: a researcher embeds in a neighborhood for months or years, observing daily life, taking detailed field notes, and building relationships. William H. Whyte's Street Life Project (conducted 1971 to 1980 in New York City) filmed and observed how people actually used plazas and sidewalks. His film The Social Life of Small Urban Spaces reshaped urban design.
• Resident interviews: structured or semi-structured conversations that capture attitudes, histories, and worries. The questions are open-ended; the data are transcripts.
• Photo documentation and photo-elicitation: residents or researchers photograph the neighborhood; interviews discuss the photographs.
• Oral histories: Studs Terkel's Chicago-based recordings gathered first-person accounts of Depression-era, Great Migration, and postwar Chicago neighborhoods. Oral histories make historical urban change personal.
• Mass Observation diaries: the United Kingdom's Mass Observation project (founded 1937) has collected thousands of everyday diaries that document attitudes toward urban and suburban change across generations.
• 311 service-request narratives: residents phoning or texting city nonemergency services leave narrative complaints about potholes, noise, abandoned buildings, and rats. The text of these requests is a qualitative signal at municipal scale.
• Community forums: Nextdoor posts, neighborhood subreddits, and local Facebook groups are soft qualitative sources. They carry selection bias — older homeowners are overrepresented — but they reveal attitudes that no census will capture.
• Venue reviews: TripAdvisor, Yelp, and Google reviews are soft qualitative sources that researchers mine for sentiment toward commercial change, the arrival of tourist-oriented businesses, or neighborhood decline.

Quantitative data can tell you that a highway was built through a neighborhood and that the population fell by 40 percent. Qualitative data tell you how residents felt about the highway: grief at the loss of their church, anger at officials who did not listen, resilience through the formation of a new block association across the highway. Only qualitative methods capture attitude.

Two landmark works

Two books anchor the qualitative tradition in urban geography.

• Jane Jacobs, The Death and Life of Great American Cities (1961): drawing on her own Greenwich Village fieldwork and observations of Boston's North End and Chicago neighborhoods, Jacobs argued that mixed-use, short blocks, and old buildings produced the street-level vitality that master-planner demolitions destroyed. Her book was primarily qualitative; it reshaped postwar urban planning.
• Matthew Desmond, Evicted: Poverty and Profit in the American City (2016): Desmond lived in two Milwaukee neighborhoods for over a year, tracking eight families facing eviction. He then paired the narratives with quantitative eviction-filing data from the Eviction Lab. The book won the Pulitzer Prize for General Nonfiction in 2017. It is the modern canonical example of mixed methods applied to urban poverty.

Both Jacobs and Desmond made strong claims about urban change; both grounded their claims in first-person observation. Attitudes and values, the CED notes at the Enduring Understanding level (IMP-6), shape the built landscape. Qualitative data are the only way to access those attitudes and values at the resolution of individual residents.

Strong urban research combines the two data types. A claim that a neighborhood is gentrifying needs both:

• Quantitative evidence (EK IMP-6.E.1): rising rent, rising education, falling minority share, rising income — the composition-and-size change.
• Qualitative evidence (EK IMP-6.E.2): interviews with long-term residents describing new coffee shops replacing the corner store, grandparents priced out, the sense that the block is no longer theirs — the attitudinal change.

Without the quantitative evidence, the claim is anecdote. Without the qualitative evidence, the claim is abstraction. Desmond's Evicted is the model: the eviction-filing rate in a ZIP code is a number, but the reader understands displacement only after following Arleen and her children through eight moves in a single year. Researchers call this combination mixed methods; they call the validation of findings across data types triangulation.

Skill 3.E is the verb that matters. The College Board does not ask students to list the numbers; it asks students to explain what the numbers imply. Below is a worked exercise.

This is the central move of Skill 3.E: quantitative data narrow the possibilities; qualitative data identify which possibility actually occurred. Students who can execute this move can earn the top band on any Skill 3.E free-response question.

Both data types have weaknesses. Good urban geographers name them openly.

• Census undercount: the United States decennial census undercounts undocumented residents, homeless residents, transient populations, young children, and rural minorities. The 2020 census missed an estimated 3.3 million Hispanic residents and 2.5 million Black residents.
• Sampling error in surveys: ACS five-year estimates at the census-tract level carry wide margins of error, especially for small subgroups. Researchers must report margins of error alongside point estimates.
• Small-N selection bias in qualitative work: an ethnographer who spends a year with eight families cannot generalize to a city of one million. Qualitative findings are deep but narrow.
• Self-selection in community forums: Nextdoor users skew older, whiter, and more homeownership-oriented than the underlying neighborhood. Venue reviews on Yelp skew toward extremes.
• Administrative data gaps: 311 requests come disproportionately from English-speaking, civically-engaged residents, not from the neighborhoods with the worst conditions.

The standard response is triangulation: combine multiple imperfect sources so that their weaknesses do not overlap. A claim supported by census data, ACS estimates, ethnographic fieldwork, and 311 narratives is far stronger than a claim supported by any single source.

Four urban systems with distinct quantitative and qualitative traditions.

United States · Chicago

Decennial census + ACS + Terkel oral histories

Chicago pairs the rich quantitative record of the United States decennial census and American Community Survey with a qualitative tradition anchored by Studs Terkel's oral histories and the University of Chicago School of urban ethnography. Recent academic work layers heat-island satellite data onto tract-level ACS income data to show that historically redlined neighborhoods are still both hotter and poorer. The pairing answers both EK IMP-6.E.1 and EK IMP-6.E.2.

Country page →

United Kingdom · London

ONS census + Charles Booth's 1890 poverty maps

The United Kingdom's Office for National Statistics runs a decennial census and continuous labor-force surveys. London is also the birthplace of foundational mixed-methods urban mapping: Charles Booth's 1886 to 1903 Inquiry into Life and Labour in London paired block-by-block color-coded poverty maps (quantitative classification) with his team's detailed policemen's-beat narratives (qualitative observation). Booth's maps are a direct ancestor of the modern ACS census-tract map.

Country page →

Brazil · Rio de Janeiro

IBGE decennial census + favela-resident narratives

Brazil's IBGE (Instituto Brasileiro de Geografia e Estatistica) conducts a decennial census with a special favela-enumeration protocol that still undercounts informal-settlement residents. Qualitative work — the documentaries Bus 174 and Favela Rising, and academic oral histories from the Rocinha and Complexo do Alemao favelas — provides attitudes and lived experience that the census cannot capture. Triangulating the two reveals a population both larger and more resilient than the official numbers suggest.

Country page →

Japan · Tokyo

JSTAT census + near-real-time district surveys

Japan's national census (Kokusei Chosa) is conducted every five years by the Statistics Bureau. Tokyo's special wards and municipalities supplement it with frequent district-level household surveys, some conducted near-annually. Qualitative supplements include Japan Broadcasting Corporation (NHK) community documentaries and university-led ethnographies of shotengai shopping-street decline in inner wards. The result is one of the world's most densely-measured urban environments.

Country page →

1. A census tract shows a 50 percent increase in median rent and a 25 percentage-point decline in Black population share over 8 years. Per Skill 3.E, what do these data imply, and what additional qualitative data would you need to confirm or refute that implication?
2. Per EK IMP-6.E.2, qualitative data reveal individual attitudes toward urban change. Identify three urban changes in your own community and name a qualitative source that would reveal residents' attitudes toward each.
3. The United States decennial census systematically undercounts undocumented residents. What does this undercount imply for policy decisions that use census data to allocate federal funds and legislative seats? Per Skill 3.E, explain the implication, do not simply state the fact.
4. A Nextdoor forum and a census-tract ACS dataset give different portraits of the same neighborhood. How should a geographer reconcile them? Which source would you trust more for which kind of question, and why?
5. Matthew Desmond combined eviction-filing rates (quantitative) with ethnographic narratives (qualitative) in Evicted. Applying Skill 3.E, explain what the combination of the two data types implies that neither could imply alone.

Draft alignment — pending educator review. AP HG codes correspond to the official College Board Course and Exam Description (Effective Fall 2020, V.1). Statements below are paraphrased in CountryReports' own voice; refer to the College Board's published CED for verbatim wording.

AP® and Advanced Placement® are registered trademarks of the College Board. The College Board was not involved in the production of this material and does not endorse it. Standards statements above are paraphrased; codes refer back to the official College Board CED, the NCSS C3 Framework, the Common Core State Standards, and other cited frameworks.