Over the last week, quite a bit has been written about the latest release of data by the National Crime Records Bureau (NCRB). While most commentary has focused on analysing whether crime in India has risen or fallen, and which cities are the worst affected, some have raised concern about the validity of the NCRB data itself.
It is useful to step back and ask what it is that we wish to do with the crime data. Our first response is likely to be: We want to know the law and order experience of people. Do people feel safe in their homes and in public spaces? Do they feel that the police is responsive to their complaints? How do they modify their behaviour in fear of crime? Only when we know the answers to these questions can we begin to quantify the extent and nature of crime, and police performance.
Why is it that reported crime fails to explain the answers to these questions? If citizens are too scared to go to the police, or feel like the police won't do anything anyway, or if the police refuse to report cases, then true crime is always going to be higher than reported crime, and we will never know. If citizens are scared and don't step out of their homes, then low crime may actually reflect fear for safety, than the existence of safety. If the police is responsive, then high crime reported may actually reflect that the police is doing a great job.
A way around this problem is to go ask citizens directly what their experience with crime was. Were they victims of specific crimes? Did they report these to the police? If they did, what was their experience? If they did not, why? Do they feel safe in their homes? Do they feel safe in public spaces at night? This will give us a survey-based measure of crime that can be compared with reported crime. This will also give the police and adminsitrators a sense of the "true" crime prevalent in society.
Such crime victimisation surveys are routinely carried out by governments in the US, UK and Australia, and the results benchmarked with reported crime. It is useful to draw an analogy with the surveys on learning outcomes carried out by Pratham - by consistently measuring skills of children, they have had remarkable impact on education policy. Similar efforts need to be made in the field of crime as well.
A second reason why we wish to know about prevalence of crime is to improve resource allocation. If from crime surveys we know that a particular kind of crime is on the rise, or a certain locality is vulnerable, then the police can take preventive measures through appropriate resource allocation. The police can use the results of successive surveys to see if their policy actions have had any impact on citizen experience. This provides a useful feedback loop to the police themselves.
One such survey was carried out by the Commonwealth Human Rights Initiative (CHRI) in Mumbai and Delhi in 2015. The survey is representative at the zonal level in each city. It asked, if households were victims of one or more of the following crimes in the last 12 months: theft, house break-in, sexual harassment, assault, criminal intimidation, unnatural death and missing persons. It also asked if they reported it to the police, and about their perceptions of safety.
The survey provides for some interesting insights. A full 13 per cent of households in Delhi and 15 per cent in Mumbai report being victims of one of the crimes, with theft being the most commonly experienced crime. Reporting of crime is low in both cities, at less than 50 per cent. The lowest reporting percentage is for sexual harassment - only 11.1 per cent cases were reported in Mumbai and 7.5 per cent in Delhi. People did not report because they didn't want to get caught up in police or court matters. Of those households who reported crime, roughly 36 per cent in Delhi and 51 per cent in Mumbai said they were satisfied with the first police response. Residents of Mumbai generally perceived the police in a more positive light, and felt safer than those in Delhi.
What is now required is the juxtaposition of the survey data with the NCRB numbers for Mumbai and Delhi in 2015. It might be useful for the police to ask - where the largest reporting gaps are. Which zones have a low ranking of safety perceptions? How can this data be used for more effective policing? At the same time, such surveys need to be carried out for other cities in India, so that a true picture of crime, and citizen's perceptions of safety may emerge.
The author is an academic at the Indian Statistical Institute, Delhi centre
Disclaimer: These are personal views of the writer. They do not necessarily reflect the opinion of www.business-standard.com or the Business Standard newspaper
