Overall, 2021 saw a 7.6 per cent decline in the number of crimes registered, as compared to 2020. The crime rate per lakh population declined from 487.8 in 2020 to 445.9 in 2021. However, crime statistics do not always tell the full story, and lower crimes reported in an area do not necessarily mean it is safe.

How does the NCRB compile its report, and how should one read the data beyond the numbers?

Who publishes the NCRB report?

The NCRB was established in January 1986 with the aim of establishing a body to compile and keep records of data on crime. It functions under the Union Home Ministry. Apart from publishing annual reports, its functions include “Collection, coordination and exchange of information on inter-state and international criminals to the respective states”.

NCRB also acts as a “national warehouse” for the fingerprint records of Indian and foreign criminals, and assists in locating interstate criminals through fingerprint search.

How does the NCRB collect information for its report?

The NCRB report contains data received from the 36 states and Union Territories across the country. Similar data is furnished for 53 metropolitan cities, or those having a population of more than 10 lakh as per the 2011 census, by respective state-level crime records bureaus.

This information is entered by state/UT police at the police station/ district level, and is then validated further at the district level, then the state level, and finally by the NCRB.

How should the NCRB report be read?

By its own admission, the NCRB says there are limitations to its data. For one, “Since the publication caters to the ‘Principal Offence Rule’ for classification of crime, the actual count of each crime head may be under reported.”

The Principal Offence Rule states that in a case where multiple offences are registered, only the “most heinous crime”, carrying the most stringent punishment, will be considered when counting. For example, ‘Murder with Rape’ is accounted as ‘Murder’, leading to undercounting of the crime of rape.

Also, since the report only compiles data that are submitted at the local level, inefficiencies or gaps in data at that level have an impact. Vacancies or a shortage of police officers at the local level may hinder the collection of data.

Also, the data record the incidence of registered crime rather than of actual crime. So, when reported crimes against women in Delhi rose significantly in the aftermath of the 2012 bus gangrape case, it may have been a reflection of increased awareness about the need for registering crimes, both among those affected and the police, rather than an actual increase in the incidence of crime against women.

“‘Rise in crime’ and ‘increase in registration of crime by police’ are clearly two different things, a fact which requires better understanding”, says the NCRB, adding that ‘higher’ numbers for crimes may not always be a bad sign: “Increase in crime numbers in a state police data may in fact be on account of certain citizen-centric police initiatives, like launching of e-FIR facility or women helpdesks, etc.”

Also, since actual numbers will be mostly higher for the bigger states, a “crime rate” is calculated per unit of population. But again, the data used at present for determining the total population is old — from the 2011 Census.