Fit locks on doors and windows, install a motion-sensor light outside, and actually lock up when leaving: these help prevent burglary; an alarm alone shows no clear benefit

A door lock requiring a key to deadlock, window locks, and an outside motion-sensor light typically cost tens to hundreds of yuan. Replacing the whole security door costs over a thousand yuan. The hard part is actually locking doors and windows every time you leave, including remembering open windows in summer.

The Crime Survey for England and Wales asked 37416 households between 2008 and 2012 which security measures they had, then compared them with burgled households. Homes with no measures represented only 4.9% of all households but 36.6% of households where burglars entered and stole. The authors calculated “security protection factors”: how much lower burglary odds were than in homes with no measures. Door plus window locks had a factor of 12.54, reducing the odds to about one-twelfth. Door locks here meant double locks or deadlocks whose bolts require a key to retract. Door locks, window locks, and an outside sensor or timer light had a factor of 34.41. Adding an indoor timer light brought it to 49. With only one measure, outside lights had a factor of about 3 and door locks about 2.8. Window locks alone showed no discernible difference. Having five measures was no better than having the right three. Homes with only an alarm and no other measures actually had higher burglary odds than homes with no measures. Homes displaying dummy alarm boxes also had higher odds of attempted but unsuccessful entry. Locks count only when locked: from 1993 to 2006/07, burglaries involving forced locks, broken windows, or removed door panels fell from 31 to 7 per thousand households, a 77% decrease. Entry through unlocked doors or windows stayed at 7 per thousand, with no decline (England and Wales).

Tseloni A, Thompson R, Grove L, Tilley N, Farrell G (2017). The effectiveness of burglary security devices. Security Journal 30(2):646–664. https://doi.org/10.1057/sj.2014.30；Tilley N, Thompson R, Farrell G, Grove L, Tseloni A (2015). Do burglar alarms increase burglary risk? A counter-intuitive finding and possible explanations. Crime Prevention and Community Safety 17(1):1–19. https://doi.org/10.1057/cpcs.2014.17；Tseloni A, Farrell G, Thompson R, Evans E, Tilley N (2017). Domestic burglary drop and the security hypothesis. Crime Science 6:3. https://doi.org/10.1186/s40163-017-0064-2；国家市场监督管理总局、国家标准化管理委员会 (2022). GB 17565-2022 防盗安全门通用技术条件. https://openstd.samr.gov.cn/bzgk/gb/newGbInfo?hcno=A75C1C66CB989D0A40BFA7A262C93290；全国人大 (2020). 民法典（第七百一十五条）. https://www.spp.gov.cn/spp/fl/202006/t20200602_463888.shtml

Disputed: there is contrary evidence for the alarm findings. The same authors analyzed 1992–1996 data and found that adding an alarm did reduce burglary odds then. They also advise caution about the unexpected result; one possible explanation is that households install alarms after being burgled. All figures are from the UK and compare households with and without measures rather than randomized groups, so the grade is B. These figures concern burglary only; there are no data showing that door locks prevent home-invasion robbery. If someone is in your home, back away and call the police first; see Item 10 (call the police before fighting when a conflict starts). If you cannot retreat, see Item 11 (you may fight back against an unavoidable attack). Renters should agree with the landlord before changing locks or adding window locks. Under the Civil Code, if tenants add or modify fixtures without consent, landlords can require restoration or compensation. When buying a security door, check the mandatory national standard GB 17565-2022, effective January 1, 2024. The main beneficiaries are you and your family.