Who is harmed, by whom, where and when

Suppose you're trying to decide what to worry about for the people in your house. A son who's twenty and goes into town most weekends. A mother in her fifties who works from home. An older relative who's stopped going out after dark. Each of them has a fear, and so do you on their behalf. This lesson gives you the numbers that say who is actually harmed, by whom, where and when, in the US and in England and Wales, so that you can set each fear beside a measured risk without dismissing either. You'll find that the most feared harm and the most likely harm are usually different things, and that "who harms whom" has several right answers, each belonging to one crime, one group of people and one source.

Every figure below carries the five labels lesson 2, "Reading a crime number", taught you to ask for: who counted, what, whose harm and where, when, and per how many. That's not fussiness. In this lesson the labels are the answer, because the same question gets different numbers from different sources, and all of them are correct.

Age, and the steepest line in the data

The Bureau of Justice Statistics publishes the National Crime Victimization Survey each autumn. Its bulletin for 2024, Criminal Victimization, 2024, gives rates per 1,000 people aged 12 or older for nonfatal violence: rape or sexual assault, robbery, and aggravated and simple assault, whether or not anyone told the police.¹ Here they are by age.

Violent victimisations per 1,000 persons age 12 or older, United States, 2024, from BJS, Criminal Victimization, 2024, Table 3. Nonfatal violence only; the survey can't count homicide. All bars are drawn to one scale.

Two things about that chart before you use it. First, 2024 was the year the survey changed its questionnaire, and the bulletin reports only the half of the sample that got the old one, so these estimates are less precise than usual, and figures from the new questionnaire will not line up with them.¹ Second, a rate per 1,000 is a population figure. It says that in a thousand people of that age, that many violent victimisations were reported to the survey in a year. It does not say who in the thousand, and one person can be victimised more than once, which becomes important later in this lesson.

The line falls with age after the mid-twenties, and it falls a long way. The group with the most fear in many surveys, older people, has the lowest measured rate. England and Wales show the same shape. The Office for National Statistics' article on violent crime for the year ending March 2024 found the highest rate among people aged 16 to 24: 2.2% had been a victim of violence from a stranger in the year, and 1.3% of violence from an acquaintance.² That article gives no figures for older people by who the offender was. Its edition for the year ending March 2018 gives every age group, for violence with or without injury from anyone: 4.0% of 16 to 24 year olds had been a victim in that year, against 0.4% of people aged 65 to 74 and 0.3% of those 75 and over. Across all adults it was 1.7%.⁸

Derek, from lesson 1, is seventy and lives in England, so the 0.4% is his age group's figure: the percentage of adults aged 65 to 74 in England and Wales who were victims of violence with or without injury in the year ending March 2018. Two cautions come with it. It is eight years old. And it leaves out robbery, which is what the mugging near his club was: that article says it does not include robbery.⁸ This course found no England and Wales robbery figure for his age group. The older relative in the household at the start of this lesson gets the same figures, for whichever of those age groups fits. A low rate does not make Derek's fear a mistake, either. It is one input to a decision that's his to make.

How you live, and where

Age is not the only line. The same US bulletin gives rates by marital status and by the kind of place people live, all for 2024 and all per 1,000 people aged 12 or older:¹

Separated people had the highest rate in the table, 74.7 per 1,000, but it is also the least precise number in it. The separated group is a small part of a half-sample, and the same survey put its rate at 43.1 in 2023. What holds in both years is that it sits well above the married rate.¹ England and Wales point the same way: in the year ending March 2018, "Separated adults (2.7%) were more likely to be victims of violence than adults that were married (0.9) or widowed (0.4%)."⁸ Neither source, in the parts this course read, explains the difference, or says how much of it age or anything else accounts for. This course will not supply a reason the sources did not give. Lesson 7, "Domestic abuse and stalking", comes back to separation from a different direction, with a study of killings by partners, and it starts with the helplines.

In England and Wales, the ONS article for the year ending March 2024 names groups that a "who is at risk" lesson must not leave out:²

• Disabled people: 1% had been a victim of violence from an acquaintance in the year, against 0.4% of people without a disability; 15.3% had experienced harassment, against 7.5%.
• Gay, lesbian and bisexual people: harassment was reported by 21% of gay and lesbian people and 35% of bisexual people, against 8.4% of heterosexual people.
• The young, for stalking and harassment: stalking peaked at ages 16 to 19 (8.8% in the year), and so did harassment (19.8%).

For most of these groups the commonest measured harm is harassment, not violence.

The survey's figures for that year carry a label worth knowing. The regulator had suspended the survey's accreditation in 2022 over quality concerns, including a falling response rate, and ONS marks the years ending March 2023 and March 2024 as "not designated as accredited official statistics". Accreditation was reinstated in 2024, as the regulator's statement on the survey sets out.³ The numbers are still the best survey estimates available for those years; the label tells you to hold them a little more loosely.

None of these figures is a statement about blame. A higher rate for people who've separated, for disabled people or for the young describes where harm has fallen. Responsibility for each of those crimes lies with the person who committed it, and nothing a group does or is makes it otherwise.

Men, women, and who harms whom

Men and women

The ONS article puts the first half in its own words: "a higher proportion of men were a victim of violence with or without injury where the perpetrator was a stranger (1%) than women (0.4%)". The other figures in the answer above come from the same article, and it adds harassment: 11.2% of women and 6.6% of men.² Each is a percentage of adults aged 16 and over living in households in England and Wales, in the year ending March 2024.

Priya, from lesson 1, is twenty-six, lives in a city in England, and says she knows the numbers put her more at risk from people she knows. For women in England and Wales the figures bear her out: domestic abuse, stalking and sexual assault (6.6%, 4.0% and 3.4%) were each far commoner than violence from a stranger (0.4%), which is the one harm here where men's rate is higher. If what she meets on the walk home is being followed or shouted at, the nearest measured harm is harassment, the 11.2%.² No source this course read gives a figure for women of her age together. The 2018 article gives the two separately, for violence with or without injury in the year ending March 2018: 1.2% of women of all ages, and 2.4% of people aged 25 to 34, men and women together.⁸

So "who is more at risk, men or women?" has no answer until you say which crime. In the US, the victim survey for 2024 found rates of 24.7 per 1,000 for males and 21.9 for females.¹ The bulletin does not test that difference, so this course will not call it either a real gap or a tie; the two are close. That does not mean the sexes meet the same harms. It means that, added together across all nonfatal violence, the totals came out near each other.

One question, several right answers

Here's the question people argue about most: is violence mostly by strangers, or mostly by people you know? Each row below is a real answer from a source this course read. Read them one at a time, with the measure, before you compare any two.

Sources for the rows, in order: ¹, ⁴, ⁵, ⁶, ⁷.

Now take the rows apart. The US survey's first row is a direct challenge to the slogan that violence is mostly by people you know: nearly half of nonfatal violent victimizations in 2024 were by strangers.¹ The FBI's police data for the same country and year show strangers for only 13.5% of assault victims.⁴ Both are correct. They count different things. The survey counts what people tell an interviewer, reported to police or not; the police data count offences known to police, where the assault category includes intimidation, and where large shares of cases are filed as "known to victim and other" or, for homicide, "all other". The FBI explains that last category: about a third of homicide victims (32.7%) had relationships "categorized as “all other,” meaning the victim(s) did not see the offender to determine whether they knew them, was a mutual combatant of the offender, or had a combination of other undefined relationships with multiple offenders."⁴ Those are reasons the two can differ; which of them does the most work, nobody this course read has measured.

In the US police data, robbery is the crime where strangers lead: "the most common relationship type reported by robbery victims in 2024 was stranger (42.5%)."⁴ Killings of women are the opposite. The ONS Homicide Index puts it plainly: "Adult female victims (defined as those aged 16 years and over) were most commonly killed by a partner or ex-partner (40%) or a family member (16%) in year ending (YE) March 2025. For adult males, the suspected killer was more commonly a friend or acquaintance (20%) or a stranger (18%)".⁶ The CDC's page on intimate partner violence cites its death reporting system for 27 states in 2015: the data suggest about one in five homicide victims were killed by an intimate partner, and over half of female homicide victims by a current or former male partner.⁵ That's part of the country, ten years ago, so read it as a pattern rather than a current level.

There's one more England and Wales figure that looks as if it belongs in the table and does not. The crime survey for the year ending March 2026 estimated that about 0.7% of people aged 16 and over had been a victim of violence from a stranger in the year, and 0.4% from an acquaintance.³ Those are percentages of all adults, not shares of violence. Put 0.7% in the same column as the US survey's 46% and you've compared a slice of the population with a slice of the crimes: two different measures, which is the kind of mix-up lesson 2 was about.

Testing a leaflet's sentence

Now the method at work. A charity leaflet says: "Most violence is committed by someone the victim knows." Test it against each row in turn, and write down what you'd say for each.

For killings of women, it is right on the figures this course has for both countries: 56% of adult female homicide victims in England and Wales were killed by a partner, ex-partner or family member, before you add friends and acquaintances,⁶ and the CDC's page, citing death records from 27 US states in 2015, reports over half of female homicide victims killed by a current or former male partner.⁵ For rape, it holds too, as far as the figures go: in England and Wales, 16.3% of victims of rape or assault by penetration since age 16 were attacked by a stranger, fewer than one in five, in the survey years to March 2020 and March 2025 combined.⁷

For nonfatal violence in the US as a whole, the sentence hides more than it shows. Strangers account for about 46% in the survey. Whether people the victim knew make up all of the rest, this course can't say: the bulletin gives figures for strangers and for partners and family, and none for everyone else. Either way, "most" is at best a narrow majority, and a reader would come away thinking strangers are rare.

Where and when

The place follows the relationship. For rape or assault by penetration, the ONS article for the year ending March 2025, pooling the survey years to March 2020 and March 2025, found that of victims since age 16, "over one-third (38.2%) were victimised in their own home and 8.0% were victimised on the street, in a car park, park, or another open public space"; the perpetrator's home came next, at 28.5%.⁷ The street is a real part of the picture and a small one. That does not make advice about walking home wrong; it means most of the risk of rape sits somewhere a street-safety tip does not reach, which is why this course gives domestic abuse a whole lesson.

For violence generally, the most recent ONS narrative on when and where is old, from the year ending March 2018, so read it as a pattern rather than a current level. That year, "more than half of violent incidents (54%) occurred in the evening or during the night", and victims believed the perpetrator was under the influence of alcohol in 39% of violent incidents.⁸ Location followed the relationship: 78% of domestic violence happened around the home, while violence from strangers happened most often around work (30%), in the street (21%) or in or near a pub or club (19%). The survey's "around work" means work premises, inside or outside, and their car parks; the pub or club category is separate.⁸

The US has an older special report on strangers, covering 2005 to 2010. In those years, "strangers committed 62% of violent victimizations occurring in public places", and about 9% of violent victimizations in the victim's home.⁹ Men were victimised by strangers at about twice the rate of women in 2010, 9.5 against 4.7 per 1,000.⁹ The levels have changed since; use the pattern, not the numbers.

Put together, the two countries' data point the same way on place: strangers in public, and more often against men; people known to the victim at home, for the harms that fall most on women. The evening and alcohol figures above are for violent incidents of every kind, not for strangers alone.

How killings start

The ONS Homicide Index for the year ending March 2025 counted 522 victims in England and Wales, 8.6 per million people, "the lowest rate since 1977".⁶ Its account of how they happened matters most for lesson 8: "Around half (51%, 266 offences) of all homicide cases were thought to have resulted from a quarrel, a revenge attack or a loss of temper". That share was 59% where the suspect was known to the victim and 44% where not. Only 5% were in furtherance of theft or gain, which means in the course of stealing or trying to profit. A sharp instrument was the method in 39%, most often a kitchen knife. Of female victims, 81% were killed in or around a house; of male victims, 46% were, and 32% in a street, footpath or alleyway.⁶

So killings rarely begin with a theft, and about half were thought to begin in a quarrel, a revenge attack or a loss of temper, a larger share where the suspect was known to the victim. Lesson 8, "Talking it down", takes up threatening encounters in public and at work; that link is this course's step, not ONS's. It is not advice for an abusive relationship, which is lesson 7's subject.

ONS attaches a caution to its figures on who is killed, and this course applies the same caution to every table here: "Demographic factors discussed in this section are not necessarily independently related to homicide."⁶ It points to socioeconomic factors, for individuals and for neighbourhoods. A rate for a group tells you where harm fell. It does not tell you that the group's age or sex caused it.

What the surveys undercount

Lesson 1 set out an "undercount" reading of the fear paradox. Here is the evidence behind it, from the statisticians themselves. ONS says the survey's interviewer questions underestimate domestic violence, because victims may not want to disclose it with an interviewer present, and it asks about domestic abuse, sexual assault and stalking in a separate self-completion section instead; the 6.6%, 4.0% and 3.4% for women above come from those private questions.²³ In the US, a National Research Council panel concluded that the victim survey "does not measure the low incidence events of rape and sexual assault with the precision needed for policy and research purposes", and judged "that it is likely that the NCVS is undercounting rape and sexual assault victimization."¹⁰ So two of the harms that fall most on women are ones these surveys have trouble counting. That supports the undercount reading. It is one of lesson 1's four readings, and the other three still stand.

How often violence means serious injury

Most violence does not end in serious injury. The England and Wales crime survey, as reported in the ONS article for the year ending March 2024, found: "The CSEW estimated that victims sustained a physical injury in 42% of violent incidents in YE March 2023." The commonest injury was minor bruising or a black eye, in 25% of violent incidents; cuts were 7% and broken bones 2%.² Notice the year: the injury figure is for the year ending March 2023, though the article is the next year's.

In the US report on strangers, for 2005 to 2010, "about 22% of victims of violence committed by strangers were injured, compared to 31% of victims who knew the offender."⁹ Violence from someone you know was more likely to injure than violence from a stranger, in those years and in that survey.

None of this makes violence minor to the person it happens to. The figure is useful for a narrower job: it corrects the picture in which every incident ends in hospital, and it lets you see that most of the harm you're planning against is the ordinary kind.

The commonest crime is not violence

If you asked the England and Wales crime survey what crime a household is most likely to meet, the answer for the year ending March 2026 was fraud, among the crimes in its headline count. That count leaves out domestic abuse, sexual victimisation, stalking and harassment, and harassment is about as common: 7.8% of adults met non-sexual harassment that year.³ Lessons 1 and 2 gave you the headline figures: 4.5 million fraud incidents of the 9.6 million the survey counted, nearly half, and 7.8% of adults defrauded in the year. One figure they didn't give: violence with or without injury was 989,000 incidents, about a tenth.³ In the US, lesson 1 met the Federal Trade Commission's 3 million fraud reports and $15.9 billion of reported losses for 2025. The same testimony to Congress gives the FTC's own estimate, allowing for the fraud nobody reports: the cost to consumers in 2024 "could be as high as $195.9 billion".¹¹ Those are reports and an estimate, not a survey, so they don't line up with the England and Wales figures; each tells you fraud is common in its own country.

This course keeps fraud short, because other courses already teach it well. Lesson 5, "Home, the door and the doorstep", covers the con that turns up on your doorstep. Personal Finance Fundamentals lesson 10, "Defending it", teaches the three things nearly every scam has to do and the payment rail as the gate, and Digital Literacy lesson 11, "Recognising a con, and the ten minutes after you clicked", covers the messages. The point here is only the weighting: a plan for a household that covers the street and not the phone is planning for the rarer harm.

After the first time

Ken Pease's review for the Home Office in 1998 pulled together what was known, and its first four conclusions run: "victimisation is the best single predictor of victimisation; that when victimisation recurs it tends to do so quickly; that high crime rates and hot spots are as they are substantially because of rates of repeat victimisation; that a major reason for repetition is that offenders take later advantage of opportunities which the first offence throws up".¹² That last clause is the reason, and it's Pease's, not this course's: the first crime shows someone the unlocked window, the route, the account, and it's still there next week.

You may have heard a number attached to this, that 4% of people suffer 44% of crime. Pease himself warns against leaning on it: "The conclusion that 4% of people suffer 44% of crime is the sound-bite from Farrell and Pease (1993) which seems to stick in people’s memory, but this kind of calculation is dependent upon various counting conventions, and varies by place and time."¹² Quote the idea, not the number.

The home security study that lesson 5 relies on repeats the point for burglary, citing earlier research: "previous burglary victimisation is the single highest predictive factor of current risk".¹³

So here is the one practical conclusion this lesson draws, and it's this course's step from those findings rather than a finding itself. After something happens, the weeks straight after are the time to close the opening the offender used: mend the lock, change the route, secure the account that was breached. In this course's labels it's sensible, untested: this course found no trial comparing people who changed the lock that week with people who waited, but the research on repeats says the risk is highest then, and says why. It is not a judgement on what anyone did before. The first crime was the offender's choice; the change afterwards is aimed at the opening it showed.

One family, two sets of numbers

Go back to the household from the start, and suppose they live in England. The son is twenty and goes into a city centre most weekends. His mother is 52 and works from home. She worries most that he'll be stabbed and killed on a night out. He does not worry much about anything.

Start with him. The measured risk that fits his life is violence from a stranger: in the year ending March 2024, 2.2% of people aged 16 to 24 in England and Wales were victims of violence from a stranger, the highest of any age group, and men more often than women (1% against 0.4% across all ages).² In 2018, stranger violence happened most often around work, in the street and in or near a pub or club, and violent incidents of every kind were more often in the evening or night (54%) than in the day.⁸ If it happens, it most likely will not mean serious injury: across violent incidents of every kind, 42% caused any physical injury in the year ending March 2023, and broken bones 2%.²

Now the harm his mother fears. Homicide in England and Wales in the year ending March 2025 was 12.3 per million males, about one in 81,000 in a year, and that rate is for males of every age.⁶ A sharp instrument was used in 39% of all killings. This course has no figure for stabbings that do not kill, so the homicide rate is the far end of the scale, not the whole of it. The fear is serious because the harm is the worst there is; the chance is small. His likelier risks are a stranger's assault, as above; fraud, which 7.8% of adults met in the year ending March 2026;³ and harassment, which 6.6% of men reported in the year ending March 2024.² The fraud figure is for all adults. Personal Finance Fundamentals lesson 10 covers what the FTC's data show about age and fraud.

Then her. For women aged 16 and over in England and Wales, in the year ending March 2024, the survey found violence from a stranger at 0.4%, harassment at 11.2%, and domestic abuse, stalking and sexual assault at 6.6%, 4.0% and 3.4%.² Those are figures for all women, not for her, and they say nothing about her household or her marriage. The example stops there, because that's where the evidence about her stops. Like her son, she faces fraud (7.8% of adults) and harassment, at the higher women's rate.

What the family learns is that their fears and their risks are shaped differently. The most feared harm, a killing, is the rarest. The likeliest harms are ordinary: a scam and harassment for both of them, and an assault in town for him. That's not a reason for her to stop worrying about him. It's a reason for the household's planning to cover the phone and the bank account as well as the night out.

What people get wrong

"Most violence is by strangers," and its mirror, "most violence is by people you know." Each is true for some crimes and victims and false for others. Killings of women and rape: mostly by people known to the victim. Robbery: strangers the commonest relationship in US police data for 2024 (42.5%). Nonfatal violence in the US survey for 2024: nearly half by strangers. Violence from strangers in England and Wales: more often against men than women. Say which crime, which victims and which source.

"The person most at risk on the street at night is a woman alone." Measured by violence from strangers, men, and people aged 16 to 24, are more at risk than women and than older people.² That does not make a woman's fear on the street irrational: harassment is common, and 11.2% of women reported it in the year ending March 2024. It means the street is not where most rapes happen,⁷ and the commonest killer of adult women is not a stranger but a partner or ex-partner.⁶

"Older people are most at risk." They're the least victimised age group in the US survey, at 7.5 per 1,000 in 2024 against 34.8 for 18 to 24 year olds.¹ In England and Wales in the year ending March 2018, 0.4% of people aged 65 to 74 were victims of violence, against 4.0% of 16 to 24 year olds.⁸

"Violence usually means serious injury." In England and Wales in the year ending March 2023, fewer than half of violent incidents caused any physical injury, and broken bones were 2%.²

"Lightning does not strike twice." Being victimised is the best single predictor of being victimised again, and repeats come quickly.¹²

Practice

Connections

This lesson is lesson 2's labels put to work: every row of the stranger table is a different source, measure and population, and the table only makes sense once you ask for all five labels. It also gives lesson 1's Derek and Priya the figures for people like them, and shows where the evidence runs out, and it adds the National Research Council's judgement on rape to lesson 1's "undercount" reading of the fear paradox.

It leads to three lessons. Lesson 4, "Opportunity and attention", turns from who is harmed to what you can change. Lesson 7, "Domestic abuse and stalking", is about partners and ex-partners. In the figures above they were the perpetrators for 43.2% of victims of rape or assault by penetration since 16 and the suspected killers of 40% of adult women killed, and that is why it gets its own lesson. Lesson 8, "Talking it down", starts from the ONS finding that about half of killings in England and Wales were thought to begin in a quarrel, a revenge attack or a loss of temper.

Go deeper

• BJS, Criminal Victimization, 2024. Read in full for this course. The tables by age, sex, marital status and place are short and worth reading for yourself; the next edition comes out each autumn.
• ONS, The nature of violent crime in England and Wales: year ending March 2024. Read in full for this course. The clearest single page on who is at risk in England and Wales, split by crime and sex.
• ONS, Homicide in England and Wales: year ending March 2025. Sections 1 to 8 read for this course. How killings happen, and ONS's own caution about reading its demographic figures.
• Ken Pease, Repeat Victimisation: Taking Stock, Home Office, 1998. This course read the summary and section 2. The founding review of why crime repeats, and the most practical idea in this lesson.

Sources

1. Bureau of Justice Statistics, Tapp and Coen, Criminal Victimization, 2024, NCJ 310547, September 2025 . Read in full. Rates by age, sex and marital status from Table 3, by place from Table 8, relationship to offender from Table 1; the 46% is worked out from Table 1's figures (3,091,500 of 6,671,640), not stated by BJS. 2024 estimates are from the legacy half of the sample. The bulletin does not test the difference between the male and female rates.
2. Office for National Statistics, The nature of violent crime in England and Wales: year ending March 2024, 26 September 2024 . Read in full. Crime Survey for England and Wales, adults 16 and over; the injury figures are for the year ending March 2023; the years ending March 2023 and 2024 are not accredited official statistics.
3. Office for National Statistics, Crime in England and Wales: year ending March 2026, 23 July 2026 . Read in full. Headline crime and what it leaves out, fraud, harassment, stranger and acquaintance prevalence, domestic abuse measurement. Accreditation history from the Office for Statistics Regulation's statement of 2 April 2026, read in full.
4. FBI, UCR Summary of Reported Crimes in the Nation, 2024, August 2025, and its FAQs . Read in full. NIBRS victim-offender relationships, pp. 6 to 9; the percentages are shares of victims.
5. CDC, "About Intimate Partner Violence", read as Internet Archive capture 20260921054056 of cdc.gov/intimate-partner-violence/about. **The page read in full; the data brief behind it was not read.** The homicide figures are the page's summary of the National Violent Death Reporting System; the page's reference for them is the system's report on 27 states for 2015, which this course did not read.
6. Office for National Statistics, Homicide in England and Wales: year ending March 2025, 5 February 2026 . Sections 1 to 8 read in full. Home Office Homicide Index. The one-in-81,000 is this course's arithmetic from the male rate of 12.3 per million. Places of killing from Figure 8's data.
7. Office for National Statistics, Nature of sexual assault by rape or penetration, England and Wales: year ending March 2025, 4 November 2025 . The main points and the perpetrator and location sections read. Victims of rape or assault by penetration since age 16; the perpetrator and location figures pool the survey years ending March 2020 and March 2025.
8. Office for National Statistics, The nature of violent crime in England and Wales: year ending March 2018. The sections on age, marital status, alcohol, timing and location read, with the data for Figure 10 (age and sex) and Figure 13 (location, with its notes defining each location), read on 25 September 2026. The article covers violence with or without injury and does not include robbery. The pub or club figure (18.9% of stranger violence) is from Figure 13's data.
9. Bureau of Justice Statistics, Harrell, Violent Victimization Committed by Strangers, 1993 to 2010, NCJ 239424, December 2012 . First four pages read (highlights, Table 1, trend text).
10. National Research Council, Estimating the Incidence of Rape and Sexual Assault, 2014 . Summary chapter read in full. Conclusion 7-1, and the panel's judgement on undercounting from the same chapter.
11. Federal Trade Commission, prepared statement to the Joint Economic Committee, 25 March 2026 . Pages 1 to 3 read.
12. Ken Pease, Repeat Victimisation: Taking Stock, Home Office Crime Detection and Prevention Series Paper 90, 1998 . Summary and section 2 read.
13. Tseloni, Farrell, Thompson, Evans and Tilley, "Domestic burglary drop and the security hypothesis", Crime Science 6:3, 2017 . Full text read. The sentence on previous burglary is in the discussion, citing Farrell and Pease (1993) and Osborn and Tseloni (1998).
14. Support lines in the content note, each read on its service's own page on 25 September 2026: Rape Crisis England and Wales, Victim Support, Rape Crisis Scotland, Victim Support Scotland, the Domestic and Sexual Abuse Helpline, Victim Support NI, RAINN and VictimConnect.