Sweden's latest National Safety Survey contains two findings that deserve to be read together. The share of people who said they were greatly concerned about crime in society fell from 54 percent in the 2025 survey to 45 percent in 2026. At the same time, newly reported results show that experiences of crime differ between groups. A national average can move in a welcome direction without describing every person's daily reality.
What changed in the 2026 survey?
On 14 October 2026, the Swedish National Council for Crime Prevention, Brå, reported a nine-percentage-point fall in the share of people greatly concerned about crime in society. Brå said the decline appeared across age groups and places of residence, and was slightly greater among men than women. Fewer respondents also thought the number of crimes had increased over the previous three years: 69 percent in 2026, compared with 83 percent in 2025.
These are measures of people's views and concerns. The 45 percent figure does not measure whether someone feels comfortable walking home, whether they have experienced harassment, or whether a public place has become safer. Brå separately reported continuing downward trends in self-reported exposure to several crime categories, including threats and harassment. Those findings should not be collapsed into one claim that everyone is now safer.
Whose experiences become visible?
For the first time in this survey, Brå asked about disability, sexual orientation, gender identity and trans experience. Among women with disabilities, 14.5 percent reported exposure to sexual offences, compared with 5.8 percent of women without disabilities. These are self-reported survey experiences, not a count of police-confirmed incidents. The survey's sexual-offence category covers different behaviours and should not be described as a rate of rape or serious assault.
The difference does not explain why these experiences vary, and it does not mean all women within either group have the same needs. It does show why a single figure about national concern cannot stand in for the experiences of women who have often been less visible in safety statistics. The new questions can help make those differences part of public discussion.
What should this mean for the conversation about safety?
It is reasonable to welcome a fall in concern. It is equally reasonable to ask whether changes in public spaces, services and support are reaching people whose experiences differ from the average. The practical questions belong with the organisations that design places and services: whose experiences informed a decision, how can people report a problem, and what happens after they do?
Survey data can tell us where to look, but not what any one person needs from the people around them. Listening to different communities, checking whether support is accessible, and following up on institutional decisions give the numbers a more useful place in the conversation. None of that should become another list of precautions women are expected to take.
Where Spher fits
Spher is a Swedish safety and community app for women. It helps women find communities and keep chosen contacts close through live-location sharing and alerts. Those connections can matter, but an app cannot prevent violence, guarantee a response or replace public responsibility. The 2026 survey is a reminder of why Spher's work sits alongside, rather than in place of, accessible services and safer public spaces.
Explore how Spher works if you would like to learn more about its community and safety features.
Sources and editorial note
Figures and survey descriptions were checked against Brå's National Safety Survey 2026 press release, 14 October 2026. The distinction between broad sexual-offence exposure and more serious offences is described in Brå's 2025 survey report. Brå says the 2026 survey used web and postal questionnaires sent to a sample of 200,000 people aged 16 to 84; that is the sample invited, not the number who responded. The full 2026 report and its detailed methodology were not accessible when this article was prepared. This article was prepared with AI assistance and fact-checked for publication under the Assi Spher editorial byline. The cover is AI-generated and does not depict survey participants.