How TownRatings grades Canadian towns

Every grade in the Canada edition comes from public data, compares a place only with Canadian towns, and becomes a letter the same way everywhere. This page explains how, once, so each town page can stay about the town.

From numbers to letters

A place is a census subdivision, Statistics Canada’s unit for a municipality or its equivalent (a rural municipality, an unorganized area, a First Nations reserve), with 1,000 or more residents; the 24 biggest cities are split into neighbourhoods (below). For each subject we rank Canada’s towns of 1,000 or more on the public data behind it, and a place’s percentile is the share of those towns it does better than. Where a subject has several measures, each is ranked, the ranks are averaged, and the average is ranked again. The percentile becomes a letter on one fixed scale:

Bottom 5% · then steps of 10% · top 5%. There is no F: a D− town is still a real place where people live and make it work.

Towns set the scale. Only towns decide where each letter starts, never a city neighbourhood or a reserve. A neighbourhood or a reserve is placed on the towns’ scale and never moves a town, so “better than 70% of Canadian towns” means towns. Neighbourhoods often carry their city’s figures, its rent or its police force’s crime, and in one pool they would have pulled the typical town toward the cities.

Each place also shows its rank within its province, because the national letter and the local picture can differ. A province or territory with fewer than five graded places gives no province rank: Yukon, the Northwest Territories and Nunavut have none.

Which places get grades

• 2,500 residents or more: a grade for each subject the data allows, an overall grade, and ranks in Canada and in the province.
• 1,000 to 2,499 residents: subject grades, but no overall grade and no rank.
• First Nations reserves: subject grades where the data allows, but never an overall grade or a place in a ranked list (below).

Residents means one figure everywhere, the one every page shows: the place’s population in Statistics Canada’s latest estimate. Places under 1,000 have no page.

The overall grade

Seven subjects are graded. The overall grade blends them by weight: cost of living 18, safety 18, schools 16, health 14, jobs and income 12, walkability 8 and hazard risk 4. The blend is ranked again across Canada’s towns, so an A means the same share of towns everywhere. A subject a place has no grade for is left out of its blend, and the others’ weights are scaled up to fill it.

The gate depends on the province. A place needs 2,500 residents and grades for cost of living, safety and health everywhere. In Ontario, British Columbia and Alberta it also needs its Schools grade, as schools are graded there; everywhere else it needs jobs and income and walkability instead. Then it needs five of the seven. Lloydminster’s Alberta half, whose schools report to Saskatchewan, takes the rule of the provinces without school grades.

The national order is the data’s: the method is the same everywhere, and Quebec’s towns lead on cost and health more often than most, so no province’s towns are lifted or held back to even it out.

Safety

The Crime Severity Index of the police that serve the place, averaged over 2023, 2024 and 2025. Statistics Canada builds the index from every crime the police report, each weighted by how severe it is, so a robbery counts for far more than a theft; Canada as a whole was 100 in 2006. A town with its own force takes that force’s index; a town policed by the RCMP, the Ontario Provincial Police or the Sûreté du Québec takes its detachment’s, or that of the regional force or MRC it shares. Toronto’s and Ottawa’s neighbourhoods take their city’s index scaled by their own violent and non-violent crime rates; other cities’ neighbourhoods take the city’s.

Less sure, and the page says so. A town no force reports for takes the index of the nearest RCMP or provincial detachment within 60 km, and a neighbourhood of a city that publishes no neighbourhood rates carries the whole city’s. Both are graded, and both pages say the grade is less sure. The nearest-detachment towns read like their province’s other detachment towns, so the stand-in isn’t biased high or low.

Sources: Statistics Canada, Uniform Crime Reporting Survey, Crime Severity Index by police service, 2023 to 2025; Toronto Police Service and Ottawa Police Service open data.

Schools: Ontario, British Columbia and Alberta

Three provinces publish test results school by school: Ontario (EQAO’s grade 3, 6 and 9 tests and the literacy test), British Columbia (the Foundation Skills Assessment and graduation) and Alberta (the Provincial Achievement Tests). Each school’s results are scored within its province and measure, and a place’s grade averages the elementary and secondary schools in it or within 3 km.

The Schools grade is given only within Ontario, British Columbia and Alberta, each on its own: a town is compared only with the other towns of its own province, because each province publishes its own measures and they don’t share one scale. Ranked lists and comparisons follow the same rule, so towns from different provinces are never put in one order on this subject.

Everywhere else it is Not graded, and the town’s page says why:

• Quebec: the ministry publishes its success rates by school service centre, not school by school.
• Nova Scotia: the province reports its assessments by regional centre, and that series stops in 2019–20.
• Elsewhere: the province publishes no results school by school.

Sources: EQAO; British Columbia Ministry of Education and Child Care; Alberta Education.

Health

Five measures, four of them needed. From the Canadian Community Health Survey: self-rated health, chronic conditions and smoking, each standardized for age, so a town of retirees isn’t marked down for being old. Then life expectancy, and how far residents live from a hospital. The survey’s figures and life expectancy are published by health region, so a region’s towns share them; the distance to a hospital is each place’s own.

Sources: Statistics Canada, Canadian Community Health Survey and life tables by health region; Open Database of Healthcare Facilities.

Cost of living

Two measures, both needed, each against the place’s own median household income: the median value owners put on their homes at the 2021 census, and a year’s rent. The rent is CMHC’s surveyed rent where its survey covers the place, mostly places of 10,000 or more. Elsewhere it is the 2021 census rent scaled by its province’s ratio of CMHC rents to census rents, and the page says the figure is less sure.

No Cost grade where most homes aren’t on the market. In a town where under 30% of households own their home, most homes are public or employer housing, as in much of Nunavut and Nunavik, so neither the value nor the rent is a market price, and the town gets no Cost grade. City neighbourhoods are outside the rule: their renters pay a real market.

Sources: Statistics Canada, Census 2021 (dwelling values, shelter costs, household income); CMHC, Rental Market Survey.

Jobs and income

Three measures, two of them needed: the share of 15 to 64s in work, the median household income, and the average time to work, all from the 2021 census. Employment Insurance claims are shown on each page but not graded.

Source: Statistics Canada, Census 2021.

Walkability

Two measures, both needed: the share of residents within 800 metres of everyday places (a pharmacy for now; food shops, schools and post offices join it with our full OpenStreetMap extract for Canada), and the share of commuters who walk, cycle or take transit.

A site bus isn’t transit. Outside the metropolitan areas, where mining, quarrying or oil and gas employ a quarter or more of workers, the census counts the employer’s bus to the site as public transit. There the active share counts walking and cycling only, as in Wood Buffalo, Elkford, Tumbler Ridge and Fermont.

Sources: © OpenStreetMap contributors, Open Database Licence (ODbL); Statistics Canada, Census 2021, main mode of commuting.

Hazard risk

Two measures, both needed: Natural Resources Canada’s estimate of the share of building value lost to earthquakes in a typical year, and the share of homes Health Canada’s survey found above its radon guideline of 200 Bq/m³. Lower risk grades higher. Flood isn’t graded: where an official flood study covers a place, its page links it.

Wildfire is a ceiling. From the national burned-area records since 1986, we measure how much of the land within 25 km of a place’s homes burns in a typical year. In the low and moderate bands it changes nothing. In the high and very high bands Risk can be no better than the wildfire measure, however low the earthquake and radon risk; that is why Fort McMurray, Jasper, Slave Lake and the Okanagan read low on Risk.

Sources: Natural Resources Canada, Probabilistic Seismic Risk Model and National Burned Area Composite; Health Canada, Cross-Canada Survey of Radon Concentrations in Homes.

Civic life: shown, not in the overall

Civic life has a grade on each page but isn’t part of the overall. It reads library visits per resident, registered charities per 1,000 residents and the province’s volunteering rate, two of the three needed.

Small towns take their area’s charities. A charity registers at a mailing address, often in the service town of a wide rural area, which inflated small towns’ counts. A town under 10,000 therefore takes its census division’s rate of charities per 1,000 residents; bigger places keep their own.

Sources: Canada Revenue Agency, list of charities; provincial public library statistics; Statistics Canada, survey on giving, volunteering and participating.

Cities by neighbourhood

The 24 cities of 250,000 or more are split into neighbourhoods, traced from census tracts and named from each city’s own map or OpenStreetMap, and each neighbourhood is graded and ranked as a place. A city’s own page gathers its neighbourhoods; it has no grade of its own and is never ranked.

First Nations reserves

Each First Nations reserve of 1,000 or more has its own page, under its own name, with subject grades where the data allows. A reserve never gets an overall grade and never appears in a ranked list.

Places without a grade

• Places under 2,500 residents get subject grades but no overall grade; places under 1,000 have no page.
• No Cost Grade: a town where under 30% of households own (above), or where Statistics Canada suppressed the value or the rent.
• No Safety Grade: where no police force or detachment within 60 km reports for the place. Without Safety there is no overall grade.
• No Schools Grade: outside Ontario, British Columbia and Alberta. Inside them, a place with no school results within reach has none either, and so no overall grade.
• Campus neighbourhoods, where 15% or more of residents are aged 20 to 24, keep their grades but aren’t ranked or listed among the best places to live.

What the grades can’t tell you

• Bigger places record more crime per resident. The census is from 2021, and its work-from-home and commuting shares were taken in May 2021, during the pandemic.
• Official data, as published. Health survey figures belong to a health region and are shared by its towns; only the distance to a hospital tells them apart.
• How sure we are depends on the town. Home values are owners’ own estimates, not sale prices, and some rents are census rents scaled to CMHC’s.
• An average isn’t a street. Crime is counted where the police record it, by force: a town that draws visitors records crimes against them too.
• Grades refresh when the data does. Kingston’s police lost records from 2024 and 2025 to a cyber-attack, so its index may read low.

Adapted from Statistics Canada (Census of Population 2021, population estimates 2025, 2021 census boundaries) under the Statistics Canada Open Licence; this does not constitute an endorsement by Statistics Canada. Neighbourhood names in some cities © OpenStreetMap contributors (ODbL).