A compiled reference of 34 verified statistics on fires in dormitories, fraternity and sorority houses and other campus housing, from how many fires a year and how the count has moved since 2014, to the cooking share, the hours and months fires cluster in, the deaths, injuries and dollar losses, how far fires actually spread, and the federal fire log and annual fire safety report every school with on-campus housing has to keep. Every figure is drawn from a named public source (NFPA, the US Fire Administration, Campus Firewatch, the Code of Federal Regulations) and independently checked against the original before publishing.
Key statistics
6 highlights from this report
Key statistics
Key takeaways
Campus housing fires are a cooking problem with a calendar. Roughly three quarters start in a kitchen, most are small enough to stay inside the object that caught, and they cluster in the evening, on weekends, and in the first months of the academic year. The deaths, though, do not follow that pattern: they concentrate in bedrooms and, on the national fatality record, in off-campus houses that no residence life team inspects.
US fire departments answer about 3,198 dorm, fraternity and sorority fires a year.
About 3 in 4 campus housing fires start in the kitchen or cooking area.
Over half of dorm fires start between 4 p.m. and midnight.
September alone carries 12 percent of the year's university housing fires.
87 percent of tracked student fire deaths happened off campus, not in the dorms.
Federal rule: every campus housing fire logged within two business days.
How we built this report
Every figure on this page was compiled in September 2026 from named public sources and checked against the original document before publishing.
- Compiled from primary sources
National fire incident analyses, federal fire administration topical reports, a national campus fatality tracker, the Code of Federal Regulations, and a published university annual fire safety report.
- Only explicitly stated figures
We include only numbers a named source states directly, and attribute each inline at the point it appears, with the period the number covers.
- Disagreements shown, not averaged
The two national datasets count campus housing differently and cover different years. Where they give different numbers we show both periods rather than blending them into a single false figure.
- Independent review
Written by one co-founder, reviewed by the other before publishing.
Scope caveat: these are national estimates built from fire department incident reports, not a census of every campus fire. Reporting is voluntary in parts of the system, confined fires carry far less detail than nonconfined ones, and the fatality tracking is compiled from media monitoring. Individual figures cover different year ranges, stated with each stat, and should be confirmed against the specific source before use.
Campus housing fires, by the numbers
All 34 figures, grouped by theme, each from a named public source and independently verified against the original document.
How many fires, and which way the count is moving
According to the National Fire Protection Association (nfpa.org), US fire departments responded to an estimated annual average of 3,198 structure fires in dormitories, fraternities, sororities and other related properties each year during 2020 to 2024. Its earlier five-year analysis, covering 2017 to 2021, put the figure at 3,379 a year and noted that the annual count peaked at 4,226 in 2014 and has declined since, with a sharp drop in 2020 and 2021 that NFPA attributes to colleges operating remotely. The US Fire Administration (usfa.fema.gov), working from the same national incident data but with a slightly different property definition, estimated 3,800 university housing fires a year for 2007 to 2009.
Statistic 1
US fire departments responded to an estimated annual average of 3,198 structure fires in dormitories, fraternities, sororities and related properties.
NFPA, 2020 to 2024 average
Statistic 2
The comparable annual average over the earlier five-year window was 3,379 structure fires.
NFPA, 2017 to 2021 average
Statistic 3
The annual number of fires in these properties peaked in 2014 at 4,226 and has steadily declined since.
NFPA, 2023 report
Statistic 4
An estimated 3,800 university housing fires occurred each year in the United States, roughly 1 percent of all residential building fires.
US Fire Administration, 2007 to 2009
Statistic 5
83 percent of these fires occurred in unclassified dormitory-type occupancies, 13 percent in barracks or dormitory properties and 5 percent in sorority or fraternity houses.
NFPA, 2017 to 2021
Statistic 6
Dormitories and dormitory-type residences accounted for 94.9 percent of university housing fires, sorority and fraternity houses for 5.1 percent.
US Fire Administration, 2007 to 2009
What this means: the direction is good and the base rate is still high. A portfolio of residence halls is fielding thousands of fire department responses a year nationally, and the drop since 2014 tracks sprinklered new construction and remote-learning years more than it tracks student behavior changing.
It is a cooking problem, not a candle problem
The single most durable finding across both national datasets is that campus housing fires are kitchen fires. NFPA (nfpa.org) reports that approximately three out of four fires in dormitory-type properties began in the kitchen or cooking area and that cooking equipment was involved in nearly 8 out of 10 fires in 2020 to 2024, rising to 86 percent in its 2017 to 2021 analysis. The US Fire Administration (usfa.fema.gov) put cooking at 87.5 percent of all university housing fires. Candles, the hazard most campus policies are written around, accounted for half of 1 percent. The same pattern shows up in general residential data: see our compiled cooking fire statistics and electrical fire statistics.
Statistic 7
Approximately three out of four fires began in the kitchen or cooking area, accounting for 58 percent of civilian injuries and 14 percent of direct property damage.
NFPA, 2020 to 2024
Statistic 8
Cooking equipment was involved in nearly 8 out of 10 fires in dormitory-type properties.
NFPA, 2020 to 2024
Statistic 9
Cooking equipment was involved in 86 percent of reported fires over the earlier five-year window.
NFPA, 2017 to 2021
Statistic 10
Cooking was the cause of 87.5 percent of all university housing fires, ahead of intentional at 2.2 percent, heating at 1.7 percent and open flame at 1.3 percent.
US Fire Administration, 2007 to 2009
Statistic 11
Candle fires, long a focus of campus housing policy, constitute half of 1 percent of all university housing fires reported to the national system.
US Fire Administration, 2007 to 2009
Statistic 12
Unattended equipment was a factor in 24 percent of all fires, the single most common factor contributing to ignition.
NFPA, 2017 to 2021
Statistic 13
A heat source too close to combustible material was a factor in 7 percent of fires but was responsible for 27 percent of the civilian injuries.
NFPA, 2017 to 2021
Statistic 14
Among the larger nonconfined fires, the leading causes were intentional at 15.7 percent, other unintentional or careless actions at 12.4 percent, open flame at 10.8 percent and other heat at 10.3 percent.
US Fire Administration, 2007 to 2009
Statistic 15
Intentionally set fires accounted for 27 percent of all direct property damage in dormitory-type properties.
NFPA, 2017 to 2021
What this means: the fire prevention that would actually move the number is unglamorous. It is the state of the shared kitchen, whether the range hood and microwave are clean, and whether anything combustible is sitting near a heat source. Those are conditions you can see in a photo, which is why they belong in a routine room check rather than in an annual walkthrough, the same argument we make for student housing operations generally.
When campus fires happen
Campus housing fires follow the academic calendar and the dinner hour, not the winter heating season that drives ordinary home fires. NFPA (nfpa.org) found the peak time of day was between 4 p.m. and midnight, when over half of the fires occurred, and that Saturday and Sunday were the peak days. The US Fire Administration (usfa.fema.gov) found the year peaks in September at 12 percent of all fires, with the fall term months carrying a third of the annual total and the summer months falling away as buildings empty.
Statistic 16
The peak time of day was between 4 p.m. and midnight, when 54 percent of the fires occurred.
NFPA, 2017 to 2021
Statistic 17
Fires were least likely between midnight and 8 a.m., but those fires accounted for disproportionate shares of the injuries and direct property damage.
NFPA, 2017 to 2021
Statistic 18
University housing fires peak from 7 to 9 p.m., with those two hours carrying 8.5 percent and 8.3 percent of the day's fires, and bottom out from 5 to 6 a.m. at 1.2 percent.
US Fire Administration, 2007 to 2009
Statistic 19
Saturday and Sunday were the peak days, at 15 percent and 16 percent of fires respectively, but weekday fires generally accounted for larger shares of injuries.
NFPA, 2017 to 2021
Statistic 20
University housing fires peak in September at 12 percent of the year's fires; September, October and November together account for 33 percent.
US Fire Administration, 2007 to 2009
Statistic 21
January, February, September, October and November were the peak months for fires in dormitory-type properties.
NFPA, 2020 to 2024
Statistic 22
Direct property damage was disproportionately high relative to the share of fires in January, May and August, when dormitory-type properties have lower occupancy.
NFPA, 2017 to 2021
When university housing fires happen
Share of the year's university housing fires by month, US Fire Administration, 2007 to 2009. Bars are drawn to scale from the published percentages.
The fall term carries the year. September, October and November together account for 33 percent of all university housing fires, while June and July, when most halls are empty, together account for 10 percent.
What this means: the risk window is the same window as the busiest operational period of the year. Move-in and the first weeks of term are when kitchens get used hardest by people using them for the first time, which is also when residence life staff are least able to look at every room. The same seasonality shows up in the workload data in our student housing turn season statistics.
Deaths, injuries and dollar loss
On-campus housing itself is comparatively safe. NFPA (nfpa.org) recorded an annual average of 2 civilian deaths, 20 civilian injuries and 20 million dollars in direct property damage for 2020 to 2024, and in its 2017 to 2021 analysis noted that deaths were too few to support further breakdown. Per-fire losses run far below ordinary residential fires, per the US Fire Administration (usfa.fema.gov). The fatal pattern sits elsewhere: bedrooms rather than kitchens, and, on the national student fatality record kept by Campus Firewatch (campus-firewatch.com), off-campus houses rather than residence halls.
Statistic 23
Fires in dormitory-type properties caused annual averages of 2 civilian deaths, 20 civilian injuries and 20 million dollars in direct property damage.
NFPA, 2020 to 2024
Statistic 24
Over the earlier window the same fires caused an annual average of 23 civilian injuries and 12 million dollars in direct property damage, with an average of two civilian fatalities per year.
NFPA, 2017 to 2021
Statistic 25
Average dollar loss was 1,970 dollars per university housing fire, against 17,090 dollars per residential building fire, and just 80 dollars per confined university housing fire.
US Fire Administration, 2007 to 2009
Statistic 26
University housing fires produced 5.2 injuries per 1,000 fires, against 28.7 per 1,000 for residential building fires generally.
US Fire Administration, 2007 to 2009
Statistic 27
Bedrooms at 3 percent and other sleeping areas at 2 percent made up 5 percent of fires but 26 percent of civilian injuries and 16 percent of direct property damage.
NFPA, 2017 to 2021
Statistic 28
175 college and university related fire fatalities were recorded from January 2000 to October 2018: 153 off campus, 10 in residence halls, 10 in Greek housing and 2 other.
Campus Firewatch, October 2018
Statistic 29
86 percent of college and university related fire fatalities since January 2000 occurred in off-campus housing, where roughly two thirds of students live.
Campus Firewatch, October 2018
Statistic 30
Since January 2000, 18 significant multiple-fatality fires killed 63 people, most of them off campus.
Campus Firewatch, October 2018
Statistic 31
Approximately 25 percent of all campus fire fatalities follow a party.
US Fire Administration, campus fire safety guidance
What this means: the numbers argue for two different programs. Inside the buildings you own, the exposure is frequent small kitchen fires with modest loss, managed by condition and housekeeping. The fatal exposure is in the off-campus houses the institution does not control, where the lever is education rather than inspection.
Alarms, and how far fires actually spread
The reason campus housing loss numbers stay low is containment. NFPA (nfpa.org) found approximately 9 out of 10 fires were confined fires that did not extend beyond the object of origin, and the US Fire Administration (usfa.fema.gov) reached 92 percent once confined and nonconfined fires are combined. Detection is close to universal in this property type, though the same data shows what happens in the minority of rooms where the alarm is absent or fails.
Statistic 32
Approximately 9 out of 10 fires, 88 percent, were classified as confined fires that did not extend beyond the object of origin.
NFPA, 2017 to 2021
Statistic 33
87.4 percent of university housing fires were confined fires, with cooking fires confined to a container alone making up 81.3 percent of all fires; 92 percent of all university housing fires stayed within the object of origin.
US Fire Administration, 2007 to 2009
Statistic 34
Smoke alarms were reported present in 82 percent of nonconfined university housing fires and alerted occupants in 85.3 percent of confined fires; no alarm was present in 9 percent of nonconfined fires.
US Fire Administration, 2007 to 2009
What this means: detection is doing the work, and the gap is device-level, not building-level. A hall can hold a fully compliant alarm system and still have individual rooms where the device was covered, pulled or blocked. That is a per-room condition, and it is only ever caught by someone looking at the room. Our compiled smoke alarm and carbon monoxide detector statistics cover detection failure modes in more depth.
The federal reporting regime: fire log and annual fire safety report
Campus housing is one of the few residential property types with a standing federal fire disclosure duty. The Higher Education Opportunity Act of 2008 added it, and the rule lives at 34 CFR 668.49 (govinfo.gov): any institution maintaining on-campus student housing must publish an annual fire safety report, keep a written fire log, and report its fire statistics to the Secretary of Education. American University (american.edu) describes the practical shape of compliance in its published annual fire safety report: the document is posted to the university website by October 1 each year, and mandatory supervised fire drills are conducted at least once during both the spring and fall semesters for all residence halls.
Requirement 1
Every institution maintaining on-campus student housing must prepare an annual fire safety report, a duty in force since October 1, 2010.
34 CFR 668.49(b)
Requirement 2
Statistics must be reported for each on-campus housing facility for the three most recent calendar years: number of fires and cause of each, fire-related injuries treated at a medical facility, fire-related deaths, and value of property damage.
34 CFR 668.49(c)
Requirement 3
The fire log must record the nature, date, time and general location of each fire, with entries made within two business days of receiving the information.
34 CFR 668.49(d)
Requirement 4
The most recent 60-day period of the fire log must be open to public inspection during normal business hours, with older portions produced within two business days of a request.
34 CFR 668.49(d)
Requirement 5
A fire-related death includes any person who dies within one year of injuries sustained as a result of the fire.
34 CFR 668.49(a)
Requirement 6
The report must also describe each housing facility's fire safety system, the number of fire drills held in the previous calendar year, rules on portable electrical appliances, smoking and open flames, and evacuation procedures.
34 CFR 668.49(b)
Requirement 7
In practice: one university's published report is posted to its website by October 1 each year, with mandatory supervised fire drills conducted at least once in both the spring and fall semesters for all residence halls.
American University annual fire safety report, 2025
What this means: the fire log is a two-business-day clock on a record the public can ask to see, and the annual report needs three years of per-building history. Both are only as good as what the front-line staff wrote down at the time of the incident, which makes the capture step, not the reporting step, where campus fire disclosure usually breaks. September is Campus Fire Safety Month, per NFPA, which makes it the natural point in the year to check that the log and the rooms agree with each other.
Cite this study
Academic or press use: copy a ready-made reference. RapidEye is the publisher.
Quick FAQ
How many dorm fires happen each year in the US?
US fire departments responded to an estimated annual average of 3,198 structure fires in dormitories, fraternities, sororities and other related properties during 2020 to 2024, according to NFPA (nfpa.org). The same body counted an annual average of 3,379 such fires over the earlier 2017 to 2021 period, with the annual count peaking at 4,226 fires in 2014.
What causes most dorm fires?
Cooking. NFPA reports that approximately three out of four fires in dormitory-type properties began in the kitchen or cooking area and that cooking equipment was involved in nearly 8 out of 10 fires in 2020 to 2024. The US Fire Administration (usfa.fema.gov) put cooking at 88 percent of university housing fires in its 2007 to 2009 analysis, and unattended equipment is the single most common factor contributing to ignition.
When do campus housing fires happen?
Evenings, weekends and the start of the academic year. NFPA found the peak time of day was between 4 p.m. and midnight, when over half of fires occurred, with Saturday and Sunday the peak days. The US Fire Administration found university housing fires peak in September at 12 percent of the year's fires, with September, October and November together accounting for 33 percent.
How many people die in campus fires each year?
NFPA counted an annual average of 2 civilian deaths in dormitory-type properties during 2020 to 2024, alongside 20 civilian injuries and 20 million dollars in direct property damage. That count covers on-campus style occupancies only. Campus Firewatch (campus-firewatch.com), which tracked student-related fatal fires nationally, recorded 175 college and university related fire fatalities between January 2000 and October 2018, of which 153, or 87 percent, occurred in off-campus housing.
Are colleges required to report fires in student housing?
Yes. Under 34 CFR 668.49, any institution that maintains on-campus student housing must publish an annual fire safety report, maintain a written fire log, and report fire statistics to the Secretary of Education. The report must cover the three most recent calendar years for each on-campus housing facility, log entries must be made within two business days, and the most recent 60-day period of the log must be open to public inspection during normal business hours.
Is September Campus Fire Safety Month?
Yes. NFPA states that September is Campus Fire Safety Month and that it works with The Center for Campus Fire Safety on the awareness campaign. Campus Firewatch, which originated the program, reported that since it started in 2005 over 410 proclamations have been signed by the nation's governors, along with resolutions in the US Congress.
Data sources
Every figure on this page traces to one of these named public sources, each checked against the original document before publishing.
