Online Social Networks and Police in

India - Understanding the Perceptions,

Behavior, Challenges
Niharika Sachdeva, Ponnurangam Kumaraguru
Cybersecurity Education and Research Center (CERC), IIIT-Delhi
[email protected], [email protected]
Abstract. Safety is a concern for most urban communities; police departments bear the
majority of responsibility to maintain law and order and prevent crime. Police agencies
across the globe are increasingly using Online Social Network (OSN) (such as Facebook
and Twitter) to acquire intelligence and connect with citizens. Developing nations like
India are however, still exploring OSN for policing. We interviewed 20 IPS officers and 21
citizens to understand perceptions, and explored challenges experienced while using
OSN for policing. Interview analysis, highlights how citizens and police think about
information shared on OSN, handling offensive comments, and acknowledgment
overload, as they pursue social and safety goals. We found that success of OSN for
policing demands effective communication between the stakeholders (citizens and
police). Our study shows that OSN offers community-policing opportunities, enabling
police to identify crime with the help of citizens. It can reduce the communication gap and
improve coordination between police and citizens. We also discuss design opportunities
for tools to support social interactions between stakeholders.

Introduction
Police departments across the globe are increasingly using Online Social
Networks (OSNs) to connect with citizens and share law and order1 related
1 In this paper, we use law and order synonymous with social issues like theft, crime, trafc

information. For instance, in 2011 riots, UK Police used Twitter to provide

localized information, dispel rumors, reassure citizens and to find looters. After
the riots, looking at the impact of OSN (Twitter) on policing, the UK government
emphasized the need for each police force to develop an OSN communications
plan (Home Affairs Committee (2011)). Thousands of citizens in developed
countries post content and follow police departments such as Boston police and
UK police. Citizens provide useful content to police, e.g., situational information,
participate in online beat programs2 and identify victims, accused, missing
people, lost or stolen vehicles (GMP police).
OSNs have demonstrated massive potential to reform policing; however, it
introduces various challenges (Davis et al. (2014)). Firm legal restrictions on
public disclosures and police impression as coercive arm of the state, make it
difficult to communicate with citizens (Denef et al. (2013)). Further, past events
show that the police misunderstood the citizens expectation resulting in
reputation loss of the department (Los Angeles News (2014)). OSN attributes like
openness, volume, and velocity at which information spreads also introduce
unique challenges for the police. For example, during Vancouver riots (2010),
police had no procedure in place to collect citizens tweets with possible suspect
clues (mentioning @VancouverPD) (The COPS Office and the Police Executive
Research Forum (2013)).
Research shows benefits, effectiveness, and challenges of using OSN by police
in developed world (Cobb et al. (2014); Palen, L. and Vieweg, S. (2008); Semaan
and Mark (2012)). However, need for OSN mediated communication,
collaboration, and connectedness between citizens and police in developing
nations is largely unexplored. With the increased penetration of OSN in
developing countries like India, police are exploring the effectiveness of OSN as
a communication channel to maintain law and order (Nayak (2014)). Unlike the
developed world, police organizations in India lack adequate police personnel.
According to the United Nation guideline, 270 280 police personnel are
recommended per 100,000 citizens (Express News Service (2013)). The policing
department in India has only 130 personnel per 100,000 citizens and there is only
one IPS (Indian Police Services) officer for every 359,953 citizens Ministry of
Home Affairs (2010). In contrast, the US has 233 police officers per 100,000
people (Wu et al. (2012)). This lack adequate police personnel results in many
under-policed areas in India. Police have felt the need to obtain community
collaboration to accomplish its increasingly vast duties and is exploring use of
technology like OSN to reach community (Express News Service (2013)).
However, police personnel have limited exposure to technology (The Economics
Times (2014)). These limitations (inadequate number of police personnel and
technology exposure) make it difficult to adopt findings of OSN use by police in
developed world to facilitate policing needs in developing countries like India.
To address these gaps, we study (largely unexplored) needs, challenges, and
2 In Police Beat programs, individual police ofcer is held responsible for communitys policing
needs in different geographical area.

preparedness of police and citizens for using OSN in a developing nation like
India. In this work, we adopt multi-stakeholder approach to examine the OSN use
for community policing in India. Our approach includes interviews of 21 citizens
and 20 IPS officers who lead and command various police and intelligence
organizations in India. Our work provides an insight about how OSN can help
Indian police to build a community and communicate with citizens to achieve
community-policing goals. This knowledge can help improve policing services
and facilitate community-policing efforts.

Research Objective
In this work we analyze, whether OSN based technology can be adopted to
support communication and collaboration for making safer society in developing
countries like India. To analyze our research objective, we study the following
supporting aspects: Why police and citizens can use OSN for improving policing
and preventing crime? Which are the different kinds of OSN platforms police is
exploring to use? What is the target audience that the police want to reach through
OSN? We also analyze the challenges police and citizens think they might face
while using OSN for communication and collaboration. The work particularly
aims to analyze the role of OSN in supporting collaboration between police and
citizens to fight crime, recover from ongoing threats, and maintain law and order.

Our Contributions
Our research builds upon prior knowledge of OSN use by citizens, first
responders, and organizations for effective collaboration (Cobb et al. (2014);
Shklovski, I., Palen, L., and Sutton, J. (2008); Stoll et al. (2012); Voida et al.
(2012)). This research is essential to devise appropriate communication strategies,
collaboration methods, and laws and regulations. Our findings are:
Citizen participation on OSN increases the human resource available with
police to identify offenders.
OSN can help to reduce communication gap and improve coordination
between police and citizens.
Four challenges that can hinder use of OSN for community policing. These
include maintaining meaningful communication, information verification
overload, immediate acknowledgement of information shared, and lack of
technical expertise and policies to handle information generated on OSN.
Identify how technological innovation and CSCW research can help support
better community policing on OSN.

Related Work
Many studies show how OSN has played an effective role during events involving
law and order issues like the Boston bombings, Sichuan earthquake (2008), Haiti
earthquake (2009), Oklahoma grassfires (2009), and Chile earthquake (2010)
(Gupta and Kumaraguru (2012); Gupta et al. (2013); Mendoza, M., Poblete, B.

and Castillo, C. (2010); Qu, Y., Wu, P. and Wang, X. (2009); Starbird, K. and
Palen, L. (2011); Vieweg, S., Hughes, A., Starbird, K., and Palen, L. (2010)).
These studies demonstrate that OSN provides critical real time information and
reduces the misinformation during crisis events. Research shows that citizens use
OSN for public coordination during a crisis situations; researchers have
categorized public response and shown different communities which developed
during crises on OSN (Gupta, A., Joshi, A., and Kumaraguru, P (2012); Hughes,
A., L. Palen, J. Sutton, S. Liu, and S. Vieweg. (2008)). Research also show that
police organizations need effective communication strategy to provide timely
information to citizens (Chermak, S. and Weiss, A.; Denef et al. (2013); Heverin
and Zach (2010)). These studies provide insights on different strategies and
activities that police perform on OSN. Recent studies show that OSN is a
plausible resource for police forces to reach citizens (Denef,S., Kaptein, N.,
Bayerl, P. S. (2011)). Police in developed nations have realized the effectiveness
of OSN in various activities such as investigation, crime identification,
intelligence development, and community policing (Denef, S., Kaptein, N.,
Bayerl, P. S. (2011); IACP (2013); Lexis Nexis Risk Solutions. (2012)). Few
studies explore technology for community interaction and collaboration to
prevent crime (Heverin and Zach (2010); Lewis and Lewis (2012)).
Benefits of OSN are also accompanied by various challenges like interactivity
and pace of information diffusion (Denef, S., Kaptein, N., Bayerl, P. S. (2011)).
Police departments in developed countries have made reasonable efforts and
progress to adopt OSN. Developing nations like India are also influenced by OSN
and police (in these regions) are still evolving skills to use OSN for policing
(Plane (2013)). Few studies in India show that OSN was used to spread
misinformation and public agitation during crisis events such as Mumbai terror
attacks (2011), Muzzafarnagar riots and Assam disturbance (2012) (Gupta and
Kumaraguru (2011); Kumaraguru (2013)). These studies report that in these
events, panic was spread through fake images, messages, and videos on OSN.
Surveys have shown that OSN introduced challenges for police officers such as
fake / impostor accounts which target law enforcement agencies, security and
privacy concerns, civil liabilities and resource constraints like time and technical
skill of the staff (IACP (2013); Lexis Nexis Risk Solutions. (2012)). Another
challenge was easy accessibility of OSN to malicious people who can modify or
spread rumors, making sharing information a complex task (Denef et al. (2013)).
These studies provide little insight about rationale and expectations of police
behind these actions and acceptance of these actions by citizens.
To best of our knowledge, it is the first study that examines police and citizens
behavior and expectation regarding OSN use for community policing in India.
HCI and CSCW research can improve the communication and collaboration
between the two stakeholders (police and citizens) for better policing. For this,
insight into the technological interactions between these stakeholders is required.
Our research expands on the existing knowledge of OSN for law and order
situation by providing a focused study that begins to address these specified
research gaps. We believe that the insights from our study will provide
opportunities to develop better communication strategy for police and citizens.

Background
To understand OSN role in collaboration and communication to fight crime, we
discuss interaction strategies of police and community policing approach.

Community Policing Approach

Community policing aims to achieve following: a) offering decentralized
decision-making that empowers field officers to identify crime, b) prioritizing the
problem with the help of local citizens, and c) introducing transparency in the
policing. To achieve these objectives two prime components are recognized
community partnership paradigm and problem-solving approach (Community
Policing Consortium (1994)). We discuss each of these components below.
Community partnership involves maintaining orderliness and safe
neighborhoods with the help of citizens. For this partnership, Indian police
introduced provisions such as neighborhood watch committee; join beat
constables to perform night beats, and arrange social and cultural get-togethers to
understand neighborhoods. These activities of community partnership help
develop trust between the police and the community. Police need to develop
community trust before involving citizens in decisions-making process. With
increased trust, police are able to get valuable information available with the
citizens about their neighborhoods. Therefore, community partnership involves
developing relationships that increase bond of trust by maintaining community
contact and frequent communication with citizens.
Problem solving approach involves four aspects: Information exchange,
problem identification, problem solving, and Trust. It involves identifying
common concerns for citizens and the police. The prioritization happens with help
of citizens, e.g., a police team might think that burglary is the biggest problem in
their locality, however, community might consider women assault and harassment
as a bigger concern. Once the problems are prioritized, police work at providing
solutions to the issues raised by citizens. Problem-solving approach considers
those solutions as best that can satisfy community members to improve safety,
reduce anxiety, and strengthen the ties between police and citizens.
The provisions so far adopted for community policing are mostly nontechnology based. Realizing the potential of OSN and to involve citizens in
policing activities, Indian police have made its presence on various OSNs such as
Twitter and Facebook. Table I describes Twitter followers and Facebook likes of
the popular Indian police pages. In this work, we use the knowledge of
community policing components and analyze role of OSN to facilitate policing.
We also identify the design challenges that need to be addressed to increase OSN
role in community partnership and problem solving. We found that for almost all
the police departments, number of followers on Twitter is less in comparison to
Facebook likes. Given that the number of likes is more on Facebook than
followers on Twitter (for most of the departments), wherever appropriate, we
provide examples from Facebook.

Table I: Twitter followers and Facebook (FB) likes on police pages. ** shows both FB and
Twitter profiles were not verified. * shows Twitter page was verified. Post shows if others were
allowed to post on FB. Joined shows year in which the page came in existence.
Police Departments
Bangalore City*
Bangalore Traffic*
Chennai**
Delhi Traffic**
UP Police PR**

Likes
105,463
249,968
50,979
202,858
8,304

Followers
12,100
8,045
1,108
2,59,000
4,585

Post
Yes
Yes
Yes
Yes
Yes

Joined
2011
2012
2013
2011
2013

Methodology
In this research, we conducted a multi-stakeholder study consisting of 41 semi
structured, in-depth interviews. We recruited all police officers through word-ofmouth and mailing lists dedicated for IPS officers; IPS is one among the three All
India Service (Ministry of Home Affairs (2010)). We completed 20 individual
interviews with IPS officers, each of about an hour. In our study, 95% participants
were male and 5% were female; 20.00% were in the age group 25 34 years,
10.00% were 35 - 44 years, 55% were 45 55 years and 15% were 55 65+
years. They provided services in different states and three officers served special
branches. They were of the rank ADG (Additional Director General), DGP
(Director General of Police), and above. Interview questions comprised of topics
such as the need for police to use OSN, how OSN has been helpful so far to
police, understand OSN usage policies followed by departments, and challenges
in adoption of OSN. Among eight officers who used OSN for official purposes,
only two officers had used OSN for more than two years. Rests were planning to
use OSN for official purposes.
Next, we interviewed 21 citizens to understand their perspective on the
presence of police on OSN. Citizen participants demography was diverse (e.g.
age group, education, and occupation). Interview questionnaire for citizens
comprised of different topics such as, ways in which citizens will like police to
help through OSN, and preferred OSN for communication with police. Most
citizen participants reported using OSN like Facebook for the last three to five
years, whereas police officers reported recent use (almost an year) for policing
activities. Ten citizens were aware of Facebook police pages; among these, six
had heard or visited police pages on OSN and three mentioned that they had
visited these pages to communicate with police. Citizens in the interviews
consisted of 57.14% male and 42.86% female; 23.81% were between the ages of
18-24 years, 42.86% were 25-34 years, 14.29% were 35-44 years, 9.52% were
45-54 years, and 9.52% were 55-65+ years. Participants were from a broad range
of educational background: teaching / research (25.53%), fashion designing
(11.76%), MBA (5.88%), computer / IT (29.41%), and other fields (29.42%).
These interviews (both citizens and police) were conducted by one of the
authors and she met the interviewee one on one. Few interviews were conducted
through telephone. We chose to conduct interviews in English as it is the common

business language used in the country. All interviews were recorded and
transcribed for analysis. We used randomly generated numbers to identify the
subjects in our notes so as to maintain subjects privacy. Participants were shown
consent information, after they agreed to participate in our study. Recruitment
approach followed in this work has also been used in other CSCW studies
(Semaan and Mark (2012)).
Qualitative Analysis: We applied limited grounded theory analysis on
interview responses. A single analyst iteratively assigned open codes, axial codes
and dimensional codes. Grounded theory method allows coding by a single
analyst because of the intense labor involved in the coding process (Charmaz).
Initially, 25 codes were developed, which were condensed to four categories
why OSN, preferred OSN, target audience, and limitation of OSN for policing.
Over successive iterations of coding, 3 categories why OSN, preferred OSN and
limitation of OSN categories became independent dimensions with their own
subcategories. Participants gave mixed and nuanced responses for target
audience; therefore we combined tentative subcategories as one category.
During the process, we developed various memos based on the incidences and
cases shared by the participants and relationships observed between the
categories. We used conceptual memos to record our thinking about the meaning
of codes and recorded when events happened, and what were their consequences.
Authors are aware that the core aspect of grounded theory method is an emerging
theory from comparisons of codes and successive iterations. However, in this
paper authors do not report a full theory analysis and use these categories in
combination with quantitative results to understand the policing scenario on OSN.
This limited use of grounded theory, for interpreting data is an established
approach in CSCW and HCI (Charmaz; Muller et al.; Muller and Kogan (2012)).

Results
In this section, we report participants perception, evaluate the current status of
OSN for policing and the associated challenges.

Why are OSN needed?

Three themes emerge from our analysis of the interviews: 1) OSN can support
policing by increasing citizen participation to identify crime, 2) OSN can reduce
the communication gap, and 3) OSN can improve co-ordination between the
police and citizens. We now discuss these themes and support these inferences
through evidences from content analysis of Facebook pages. We refer interviewed
IPS officers as P1 . . . P20, and citizens as C1 . . . C21. To analyze the situation on
actual position on OSN, we provide anecdotal evidences from Facebook pages,
which substantiate the perceptions of IPS officers and citizens on use of OSN.

Increase citizen participation to identify crime

Officers express that OSN can help increase the personnel (citizen volunteers)
available for police to identify offenders. According to officers, using OSN,
citizens can easily report defaulters and lawbreakers. P7 says, OSN can be used
to create a community of people who will be using it (OSN) [to identify crime].
Another officer states that before the introduction of OSN, only traffic police on
roads were responsible for catching traffic violators, but now anybody can use
OSN pages to post pictures of traffic violators. P17 says, Delhi Traffic Police
page involves public [citizens] in finding traffic violators, on the basis of which
challans [fine] are issued. So our limited resources are increased, and we can
catch many more people by seeking help of public [than before]. Officers believe
that using OSN, citizens can provide real time content to inform about crime. For
example, P9 states, Young people are using OSN on the mobile devices. They
can upload real time content [from the crime location].
Consistent with police officers, citizens also agree that OSN pages can help
report issues related to traffic and crime. These issues include traffic congestion,
accidents, beggars, corruption, and other issues with public transport. Most
citizens remark that they will like to use OSN pages to report traffic issues like
unruly behavior of taxi or autorickshaw drivers. For example, C1 says, I use
public transport like autorickshaw; they [auto drivers] either overcharge or harass
passengers by refusing to go. I will like to report these cases. According to
citizens, OSN can also provide opportunities to engage with police to report crime
such as theft, neighborhood issues, and crime against women. C13 states, if I see
a girl in a vulnerable situation I might give this information [on the Facebook
page of police], and they [Police] can take instant action. C12 says that she can
instantly geo-tag the crime location in the Facebook post to help officers receive
quick information about exact location of the crime.
Facebook content analysis shows that citizens use these pages to report
activities such as the use of tinted dark glasses in cars, crimes against women, and
neighborhood issues like drunk men on streets. For example, A citizen complains
about drunk men and posts a video as an evidence, I am posting a video as well
of people drinking openly at XXX wines, sec 1X-1Z road. I had posted pics from
the same location on 13th Aug as well (see Figure 1). These observations show
that citizen participation on OSN can increase the personnel available for
identifying crime and law offenders.

Figure 1: Citizen posted an image on Facebook to report illegal use of streets for alcohol
consumption.

Reduce the communication gap

We find that OSN can reduce the communication gap between police and citizens
in three ways: 1) OSN can help understand the public opinion about various law
and order issues, 2) OSN can help get feedback from citizens, and 3) OSN allows
citizens to overcome social pressures to approach the police.
Understand citizen opinion: Officers mention that OSN can provide effective
understanding of public sentiments and opinion. P3 states, Whenever we think of
OSN, we think it in terms of what are the aspirations of the public [citizens], what
they want from us, what are their grievances and how we can address them.
Officers report OSN to be better than existing media (such as television) to gauge
public opinions. P3 states, OSN is more representative; police delivery [of
services] can be more up to mark. On TV we are not sure of how many people are
interviewed, they show only 2 - 3 people. This number cannot be used to show
mass opinion. According to officers, understanding the opinion of citizens is
important for early crime detection, crime prevention, and for reducing
communication gap between police and citizens. Similar to police officers, few
citizens also agree that police can know about public opinion using OSN. C6
states, Your status speaks everything. They [Police] can know what is going in
mind of citizens and take action accordingly.
Feedback from citizens: Officers believe that OSN based feedback (both
positive an negative) from citizens can bridge the gap between police and public.
We asked officers, what will be their reaction to the message such as Scenario A,
I will like to thank the police to improve law and order. P9 says, make it reach as
many people as possible and post it. Helps to show a positive image of your
organization. Good feeling among the community and for people. P12 says,
further propagate such message; these tweets are hardly to be seen. Definitely
they will boost up the moral. Officers considered it important to reply to
feedback posts to keep the connection with citizens and show that citizens views
are respected. P8 says, convey thanks for posting; it should be taken cognizance
of. The other person should feel that somebody is taking note of his thoughts.
Officers said that they will also appreciate negative feedback and take appropriate
actions. We showed officers a post, Scenario B, in which a citizen had
complained about unavailability of police on a highway where a girl was being
beaten. P1 says, we will check if actually it is happening and take action;
subsequently will post about the action. Officers believed that OSN can help
explain the reasons for such lapse so that people can understand the situation better.
Adding, further officers mentioned that in the absence of OSN, police depend
upon exiting media such as television and newspaper to explain their decisions.
On these media, the police views are presented as understood by the journalists.
However, with OSN police officers can directly communicate with citizens and
express their views (without influence of journalists). P4 says, I would like to use
OSN to give official version of any information. Because they [citizens] always
hear the story from what media [television or newspaper] has to say which always
might not be correct. [Using OSN] They [citizens] can hear what we have to say
directly from us. Police version is the correct version and people should start

trusting it [police version] instead of media [television or newspaper].

We showed the above-discussed scenarios A and B to citizens. Few citizens
said that they will like to spread / reply to positive messages. C1 says, There are
certain situations when police really give their best. [I will share this post to]
Encourage police and let others know that police is doing something. More
citizens mention that they will share or like a post in Scenario B than Scenario A,
so that police are more informed about the issue. Citizens believe that like or
share utility might let police understand the gravity of the situation. A citizen
mentions that if many people complain about the same issue police might take
some action.
In Facebook posts, we find that citizens appreciate police for the good work
and share what troubles them. For example, a citizen complains about an
undisciplined drunk police constable (see Figure 2). Such posts can help police
understand citizens concern, which play a significant role in reducing
communication gap. Understanding the citizens need for anonymity, police say
in a post that citizens need not reveal their identity if they want to help solve
violent crimes. Thus, content analysis of Facebook pages shows that OSN can
reduce the communication gap between citizens and police by encouraging
frequent exchange of information and feedback.

Figure 2: Post showing police acknowledging citizens concern regarding undisciplined police
constable. This post received 102 likes.

Provides anonymity: The third method which helps reduce the communication
gap is OSNs ability to facilitate anonymous communication between police and
citizens. Many citizens mention that they will prefer an anonymous platform to
communicate with police. C7 says that he will like to make an anonymous
complaint on OSN against local Member of Legislative Assembly (MLA,
politician), and illegal activities of a local shopkeeper, which otherwise is not
possible as these influential people can harm him. Citizens report that they are
fearful and concerned about their security, therefore, use fake-profiles to share
information with police. For example, C16 states, I created a fake profile to post
pictures of traffic offenders, so that law offenders are punished, but they do not
get to know that I complain about them. Some officers also acknowledge that
citizens may not feel comfortable revealing their identity on Facebook, and may
use pseudonyms similar to undercover informers. These observations show that
OSN helps overcome social fears and pressures while communicating with police.

Improve coordination
In this section, we show that OSN can improve coordination between police and
citizens by keeping citizens informed and delivering targeted messages.
Keep citizens informed: Officers mention that the police can use OSN to keep
citizens informed about various arrangements. For example, P15 states, OSN
help in disseminating information and forewarn [citizens] about areas that might
see a conflict. Further, P2 states, OSN can help inform citizens about jewellery
snatching cases, time it can happen, and precautions to be taken. Officers show
interest in posting wide variety of advisories (an official announcement or
warning) through OSN such as crime alerts, safety tips for women, children and
senior citizens, places to avoid during major events and natural calamities like
cyclone or floods. An officer states, A cricket match and a popular festival
happened on the same day; we were expecting a large gathering for both the
events. These crowds were required to cross each others route to reach their
destinations. To avoid traffic, we issued instructions through public post on OSN;
this proved helpful to manage the crowds and avoid traffic on roads. Citizens
also view OSN as a means to obtain timely alerts from police. C7 says, Suppose
there is a threat in Delhi, and they [Police] give me alerts on it [OSN] then I
would have liked to follow it [police Facebook page]. Such alerts can keep
citizens informed and improve coordination.
Targeted communication: Officers acknowledge the need to communicate
appropriate advisories to appropriate audiences for improving coordination with
them. These advisories are communication-intensive, highly social, cultural, and
influence social networks in the real world. Officers in different regions have
different needs and priorities for issuing these advisories. For instance, P1 states
that OSN can be used to spread awareness regarding social evil practices of
witch-hunting in northeast India. 3 Officers from Haryana (a state in North India)
mention that they will like to create awareness about women, children and senior
citizen safety as one agenda. Few citizens also mention that they will like or share
an advisory only when it is relevant to their network. This suggests that OSN can
help in spreading targeted messages to improve coordination.
Consistent with the interview analysis, we find that the police use Facebook
pages to keep citizens informed about policing arrangements. A police post
informs citizens which roads will be closed for general traffic from 5.00 AM to
9.00 AM (see Figure 3). Another post informs citizens about criminals and states
Nataraja@Mallu, a notorious rowdy involved in assault, extortion, kidnapping &
attempt to murder booked under Goonda Act today. To target specific audiences,
we find that Delhi police maintains a dedicated page for North Eastern Indian
citizens who were recently victimized in the city.

Figure 3: Image posted by Delhi Traffic Police on Facebook informing citizens about traffic
diversion for Independence day celebrations.

Which OSN police prefer to reach the citizens?

We asked officers, which OSN do you use for your day-to-day activity to
maintain law and order? We find that police organizations are using or planning to
use a variety of OSN platforms. Among these, Facebook and WhatsApp are the
most preferred networks. Some platforms are used to communicate with citizens
(Facebook and Twitter) and some are used only within the organization
(WhatsApp and Youtube). Sixteen officers prefer Facebook over other OSN for
activities involving citizen participation; among these sixteen, 6 officers use only
Facebook. P4 said: Facebook is more interesting. Personal viewpoint can be
expressed and it is more interactive. People [citizens] prefer to give their inputs
through Facebook. Some officers advocate the need of dedicated Facebook pages
for policing tasks. P19 says, Depending upon the need of the department they
can have it [a Facebook page] for specific purposes like womans extortion cell,
child abuse cell, and extortion cell. Few officers mention using Twitter; these
officers believe that Twitter requires less time as it is used by few people. P15
says, Actually it is the matter of how much time you can devote. Facebook gets
too big and involves many more people [whereas on] Twitter you make a small
comment and it is done.
For personal and within organization use, officers use mobile-based
applications like WhatsApp and BlackBerry Messenger. Many officers feel that
WhatsApp is better than Facebook to get instant information about the current
situation on the ground. P13 says: During a gathering or law and order situation,
the usual question is how many people [are there]. Then they [officers in control
rooms] start guessing, if officers on duty have a phone they can click the picture
and upload it, so that in the control room, we have an idea. Officers further add
that WhatsApp showed when the person was last seen online and therefore cannot
deny receiving the message. P12 mentions, Nobody can deny that he or she have
received a message. While comparing Facebook with WhatsApp, P18 says,

With WhatsApp, we can send an audio file as well. We can transfer photos,
images instantly with WhatsApp. I feel Facebook is terrible in this case. Suppose
you want to upload an image to Facebook, it will take a lot of time. Officers feel
that policing requires instant response, as time plays a crucial role and can make
or break things.

Who is the target audience?

We asked officers, who are the target audience you want to reach through OSN.
We find that there was no clear understanding of target audience on OSN among
the officers. P6 says, every citizen is our target audience, but audience is limited
to account holders [on OSN]. Similarly P4 remarks, persons who are using it
[OSN], affluent classes, students younger generation who can afford to use the
Internet. Few officers who feel that younger population and students are
common audience on OSN believe that OSN can help reach and connect with
youth better. P3 says, Youth is sometimes available only on OSN; what they
have to say is sometimes not available on traditional media / sources of
information. They have very strong ideas which are otherwise missed out.
Few officers say that currently OSN was restricted to urban areas and did not
have good reach in rural areas. P9 says, Urban sector, yes we can find some
information. Rural areas, it is not of priority [to reach through OSN]. We find
similar views among few officers regarding the use of mobile phone applications
like WhatsApp; P13 believes that there are few WhatsApp users in India. The
limited reach restricts efforts to use OSN for policing by some agencies in urban
areas only; but in the past, we find that OSN effect percolated to rural areas in
crises situations like riots in Muzzafarnagar and Assam Hindustan Times (2013);
Press Trust of India (2013). This shows that officers have limited understanding
of OSN reach and audience it impacts. Officers do not mention any distinction in
the population on different platforms and believe that OSN can be used to connect
with anybody.

OSN Challenges and Overheads

We now discuss the following four challenges that can hinder communitypolicing efforts using OSN: maintain meaningful communication, verification
overhead, acknowledgment overhead, and lack of technical capabilities.
Maintain meaningful communication
Participants acknowledge the need for maintaining information exchange through
OSN that is meaningful to improve policing decisions and arrangements (here
after referred as meaningful information). We identify two factors that can inhibit
this meaningful information exchange on OSN: 1) unclear and generic
information from police and 2) violent and abusive content from citizens.
Citizens remark that advisories during riots help take appropriate decisions but
think that some police posts are too ambiguous and may not help take appropriate

decisions. For instance, citizens find that police advisory released on Facebook
asking citizens not to trust any rumors posted on OSN during Muzzafarnagar riots
(See Figure 4) was too generic to be enforced. For example, C4 states, This
information is useless. If my friend is tweeting or sharing [Figure 4] how will I
know that it is not fake. The post says not trust any tweet, why should I trust this
post. However, most police officers feel that it is useful to share such
information.

Figure 4: Advisory shown to participants posted by UP police on Facebook Please, do not trust
any rumor, video clipping shown on Facebook, Twitter, YouTube, WhatsApp. Nuisance creators
are being controlled sternly. Post received 26 likes.

Officers fear that on OSN, citizens may use inappropriate expressions and
informal language to communicate their thoughts. P10 states, Expectation of
people how police should use OSN is not based on the clear understanding of law
enforcement agencies. They think that the way they use informal style language
on Facebook accounts, law enforcement will also do the same. That is not
possible. An officer mentions that Twitter profile and Facebook wall have the
potential to create havoc. Officers also express inability to remove or control the
spread of abusive content from the official pages. They believe that removing
inappropriate comments may give the impression that the police are censoring the
page and may obstruct community-policing initiatives. Similar to interview
analysis, online posts on Facebook pages show that citizens post abusive content.
For example, in reply to a police post informing citizens about arrest of a
criminal, a citizen posts comments like Shoot all bloody rapists, bastard, and
That bastard private parts should be cut so it sets an example for others too.
Verification Overhead
Officers believe that information obtained from OSN is a starting point, and it
is essential to verify the information. P17 states, Any information is just
information as long as it is there. After that it has to be verified, correlated with
other information, for this piece of information to go to the category of
intelligence. Officers also mention about various techniques they may use to
verify the information available on OSN. One approach is to look at the number
of responses (multiple posts) posted related to an issue. P3 states, More are the
number of responses that we check, more we are close to the truth. Another
approach that officers mention for virtual verification is to check the source of the
information. Citizens also agree that information obtained from OSN is not
completely trustworthy and requires verification. However, most citizens say that
if information comes from police pages, it will be trustworthy and will need no
verification. Citizens (similar to police) rely upon the source of the information to

judge its reliability. C12 states, I do not trust the information, I verify the source
of it. Analysis of behavioral data shows that police may ask citizens for
additional information for verification. For example, one of the police pages
posted, Thanks, Kindly re-post the pictures with clearly visible R/C [Vehicle
Registration] No., time & place also so that proper action can be taken against the
police personnel. In some cases, police request citizens to contact an officer who
can verify the details and take action. For example, Dear [name], Please visit at
[XXX] Police Station and lodge a complaint with the details, and they will take
necessary action in this regard. Thank you; shows police need to verify
information shared on OSN.
Acknowledgement overload
OSN users often expect a quick response when they post a complaint or a
request on OSN (Kelly (2014)). We analyze citizens view on how long police
can take to respond to a citizens request. Citizens expectation varied from few
minutes to a week. C5 states, if police has to use OSN, then there should be a
team who should be checking it every second. For some citizens, type of
complaint and volume of complaints received on OSN influence frequency of
acknowledgement. C14 states, That depends upon the issue, traffic problems can
be answered in 1 or 2 days. Children / people who need immediate help like
accident cases should get an immediate response. Very few citizens state that the
police should send an immediate acknowledgement that they have received the
complaint. According to police officers, responses cannot be provided
immediately. Few officers say that they can respond once in a day and for some
officers it depends on the need. This expectation difference between police and
citizens can inhibit the active use of OSN for policing. Analysis of Facebook
pages shows that the police do acknowledge citizens post, for example in a post
police states, Dear [name], Your post has been conveyed to High Ground Police
Station (080-229XX587/XX83), They will assist you in this regard. Thank You.
Lack of technical teams and policies
Officers think that lack of manpower who are capable of handling OSN pages of
police inhibits social media adoption for community policing. They also mention
about the need for guidelines and policies, which can lead to uniformity and
clarity in decisions while using OSN.
Most officers believe that police departments require trained people who can
help leverage the benefits of OSN. P9 states, Team of better people as of now, it
is a resource constraint for us who can help us use it better. Officers think that
the OSN team should be acquainted with police decisions and be very specific
about content they post on OSN. P3, while discussing characteristics of the person
handling OSN page states, Person, who can be very specific about what he
speaks; for now, he will share very limited type of information.
Officers mention the need for policies that can help understand the pros and
cons of using OSN. For example, P4 states, It [OSN for policing] is still in an

experimental phase but to leverage its benefits, policy has to be built, keeping in
mind OSN usage, pros, and cons. Like Facebooks servers are outside India, so
how information [confidential] can be saved [stored] on them. Officers mention a
need for guidelines on content to be posted. They believe guidelines on content
are important to generate information that can keep citizen communities
interested during lean periods (time when there is no crisis to deal with). Few
officers think that policy can help define boundaries and extent to which OSN
should be used by police. P5 states, Centralized guidelines should be there on
how to use Facebook, Twitter and YouTube otherwise everybody will start using
it in his or her ways. Thus, the current absence of OSN policies and guidelines
limit the use of OSN for policing.

Discussion
In this work, we explore how OSN can facilitate community policing in India.
Consistent with the previous research in the developed world, we find that OSN
can help involve citizens in policing through various ways: increasing citizen
participation in neighborhood watch, reducing communication gap and improving
coordination between grassroots members of community and police department
(Heverin and Zach (2010)). Researchers in the CSCW field are uniquely
positioned to develop tools necessary to support police-citizen relationship on
OSN. Current CSCW research investigates OSN use by first responders and
citizens primarily during crisis situations, which are event-driven and have
specific goals to be achieved in a short term (Denef et al. (2013)). It is unclear
how this knowledge can be used for day-to-day communication in policing
activities that are diverse and are not time bound. We find that effective policing
through OSN requires identifying the needs, planning activities, creating useful
content to be shared, and constantly communicating with citizens. We now
discuss OSN role to facilitate frequent contact and ease information exchange as
required in different components of community policing in day-to-day policing.
OSN Role in Community Partnership Paradigm: Lack of communication is
frequently cited as a major problem resulting in lack of trust in services offered by
police (Lewis and Salem (1981)). Our study shows that OSN provides
opportunities for police to keep citizens informed about policing arrangements
and decisions. However, officers in our study think that they need to understand
the content and characteristics of interaction that can happen on OSN during lean
periods (when no major event is happening of law and order interest). This can
help keep the community involved and get persistent visibility to police. We think
that persistent presence on OSN can also help police to get quick visibility during
high impact events like blasts where communication need has been felt the most.
Our study shows that police believe OSN can help reach the desired set of
population and can provide opportunity to have a constructive dialogue and

exchange feedback among police and citizens. However, police say that constant
communication needs to be constantly guided through specialized 24X7 OSN
teams, policies, and guidelines. We find that though OSN facilitates quick
information exchange between police and residents, designing appropriate nudges
that educate about the legal and social implications of abusive content can reduce
the misuse of OSN. This can help improve effective communication between the
two stakeholders. Defamatory content can be a major block for successful
adoption on OSN by police. Most citizens during interviews mention that abusive
content should not be used on OSN while expressing disagreement on police
actions. We find that, in recent events, defamatory content posted against a
politician on Facebook and WhatsApp led to violent protests in Mumbai, India.
Police filed a complaint against the accused, but lacked concrete means to educate
citizens about implications (legal and social) of such content.
Citizens often fear police and feel that revealing facts about crime may expose
them to criminals; these apprehensions hinder police community association
(Community Policing Consortium (1994)). Prior work in developed world
suggests that citizens may not trust community-policing technology that does not
keep their submission anonymous (Lewis and Lewis (2012)). Similarly, in our
study, we found that citizens prefer to use OSN as it provides anonymity that can
help citizens reach police without social pressures and fear; thus, OSN facilitates
community contact and partnership. OSN provides a platform where citizens can
hold police accountable for taking action against crime but not expose themselves
to criminals or other social pressures. However, anonymous posts involve
legitimacy issues that make it difficult for police to take actions. We believe that
technological solutions require amalgamation of security and CSCW domain such
that the proposed technology can provide anonymity to citizens, but also keep
minimal checks to authenticate information if needed.
OSN for facilitating problem solving: We find that both citizens and police
believe that using OSN, citizens can inform the police about vulnerable streets in
the city, report geo-tagged posts that can give instant information about the
location of the crime to the police. Citizens can also post images and videos of
unsafe neighborhoods. This information exchange can facilitate two important
aspects of community policing: a) the problem-identification and b) prioritization
based on citizens input. Problem-solving approach suggests that best solutions
are those that satisfy the community. Police after examining OSN generated
content can identify priorities of different communities and use this information
to make appropriate judgment to satisfy the safety needs of citizens. This can help
make community feel safer and can generate trust in police. Police Officers in our
study mention that OSN can help explain reasons for lapse in policing
arrangements directly to citizens without involving journalists (as needed in other
mediums like television and newspapers). This can help citizens understand the
situation at hand and also encourage information exchange to address the lapse.
We find that for success of OSN, police will need to acknowledge the
information exchanged through OSN; our work shows this can be an overhead for
the police. Therefore, the police departments need a technology that can imbibe

an effective acknowledgment system on OSN platforms. We find that citizens

preference for response time on the post varied depending upon the case. An
analysis of communication patterns can reveal expectation of citizens regarding
response time to complain. This knowledge can be used to design a semi
automated acknowledgement system. Social tools can also help classify the
content posted by citizens on OSN in various categories, and police departments
can set flags to send an acknowledgement based on the category. This can help
reduce the overhead and enable frequent OSN use.
We also find that legitimacy of information exchanged through OSN can
impede the information exchange during problem-solving process. Setting up
OSN profiles and pages, can get a large number of posts for police to address.
Legitimacy of this information is also a concern for police and they feel it is
challenging to verify all the information posted on OSN. However, police feel
that multiple responses on the topic can be a good indicator that the information is
legitimate and verified. Tools and applications can be built to ease this verification
and validation. These tools can provide cross-platform verification and get
responses from different OSN, resulting in increased pace at which police obtains
responses from different platforms on the issue. Alerts can be generated to inform
police about the verification being done on the content; this can expedite the
police decision.
As we conclude this paper, we believe that this work outlines the requirements
(as extracted from our interviews) for a socio-technical solution that can facilitate
two important component of community policing community partnership and
problem-solving process. This study creates an understanding about policing
requirements from OSN. This study also provides designers and researchers with
insights for designing technology that empowers and encourages police - citizen
communities to address social issues. Community policing so far has been
addressed using non-technology tools but community oriented platforms like
OSN can act as a facilitator and catalyst to policing.

Limitations
This study provides insights on perceptions, behavior and challenges for OSN use
in community policing, however, there are some limitations to this work. We only
study users from urban and suburban areas where OSN influence is high. It will
be interesting to study a broader space. Other communities for e.g. OSN users in
rural areas may prefer different technologies like voice based solutions for
community policing. The number of male policemen in our study is dominant,
however male and female ratio in our study, is representative ratio of the genders
in IPS (Joshi (2012)). Another potential limitation is of the interview methods that
analyze perceptions of the participants and cannot measure the behavior of
subjects. We present preliminary evidences of actual behavior using Facebook
content but think that behavior can be understood through further analysis in
future studies.

Acknowledgements
We would like to thank all participants for sharing their views with us. We would
also like to thank TCS for funding the project through Ph.D. fellowship. Our
special thanks to Mr. Nandkumar Sarvade who helped us connect with IPS
officers. We would like to thank CERC and Precog members for supporting us
throughout the project; special thanks to Aditi Gupta and Siddhartha Asthana.

References
Charmaz, K., Constructing grounded theory: A practical guide through qualitative analysis.
Thousand Oaks, CA, USA: Sage.
Chermak, S. and Weiss, A. Maintaining legitimacy using external communication strategies: An
analysis of police-media relations.. Journal of Criminal Justice 2005.
Cobb, C., T. McCarthy, A. Perkins, A. Bharadwaj, J. Comis, B. Do, and K. Starbird (2014):
Designing for the Deluge: Understanding and Supporting the Disturbed, Collaborative Work
of Crisis Volunteers. CSCW.
Community Policing Consortium (1994): Understanding Community Policing. A Framework for
Action. Technical report, Bureau of Justice Assistance.
Davis, E. F., A. A. Alves, and D. A. Sklansky (2014): Social Media and Police Leadership:
Lessons From Boston. Technical report, NIJ and Harvards Kennedy School of Government.
Denef, S., P. S. Bayerl, and N. Kaptein (2013): Social Media and the Police - Tweeting Practices
of British Police Forces during the August 2011 Riots. In: Proc. CHI 2013. pp. 3471 3480.
Denef, S., Kaptein, N., Bayerl, P. S. (2011): ICT Trends in European Policing.
Express News Service (2013): Is community policing need of the hour?
http://www.newindianexpress.com/states/karnataka/article1430481.ece.
GMP police, GMP on social media.
http://www.gmp.police.uk/content/section.html?readform&s=BF065D2949849972802579610
0424FED.
Gupta, A. and P. Kumaraguru (2011): Twitter Explodes with Activity in Mumbai Blasts! A
Lifeline or an Unmonitored Daemon in the Lurking?. Technical report, IIIT-Delhi.
Gupta, A. and P. Kumaraguru (2012): Misinformation on Twitter during Crisis Events.
Encyclopedia of Social Network Analysis and Mining (ESNAM). Springer publications.
Gupta, A., H. Lamba, and P. Kumaraguru (2013): $ 1.00 per RT #BostonMarathon
#PrayForBoston: Analyzing Fake Content on Twitter.. eCrime Research Summit.
Gupta, A., Joshi, A., and Kumaraguru, P (2012): Identifying and Characterizing User
Communities on Twitter during Crisis Events. Workshop on UMSocial, Co-located with
CIKM.
Heverin, T. and L. Zach (2010): Twitter for City Police Department Information Sharing. In
Proc. ASIST 2010.
Hindustan Times (2013): Video that sparked communal violence in Muzaffarnagar reaches other
districts.http://www.hindustantimes.com/India-news/lucknow/Video-that-sparkedMuzaffarnagar-clashes-makes-its-way-to-other-districts/Article1-1120683.aspx.
Home Affairs Committee (2011): Home Affairs Committee concludes police failed to appreciate
magnitude of riot task. http://www.parliament.uk/business/committees/committees-az/commons-select/home-affairs-committee/news/plsd-report-publication/.
Hughes, A., L. Palen, J. Sutton, S. Liu, and S. Vieweg. (2008): Site- Seeing in disaster: An
examination of on-line social convergence. ISCRAM.
IACP (2013): 2013 IACP Social Media Survey.
http://www.iacpsocialmedia.org/Portals/1/documents/2013SurveyResults.pdf.
Joshi, S. (2012):Only 442 women police stations across India:Police research data.

http://www.thehindu.com/todays-paper/tp-national/tp-newdelhi/only-442-women-policestations-across-india-police-research-data/article4236877.ece.
Kelly, G. (2014): Social media marketing requires prompt response
times.http://www.businesstexter.com/social-media-marketing-requires-prompt-responsetimes/.
Kumaraguru, P. (2013): Riots in Muzaffarnagar (Uttar Pradesh), India. September, 2013..
Technical report, IIIT-Delhi.
Lewis, D. A. and G. Salem (1981): Community crime prevention: An analysis of a developing
strategy. Crime & Delinquency, vol. 27, no. 3, pp. 405421.
Lewis, S. and D. A. Lewis (2012): Examining Technology That Supports Community Policing.
In: Proc. CHI 2012. pp. 13711380, ACM.
Lexis Nexis Risk Solutions. (2012): Survey of Law Enforcement Personnel and Their Use of
Social Media in Investigations. www.lexisnexis.com/investigations.
Los Angeles News (2014): NYPD Twitter campaign backlash spreads to LAPD.
http://abclocal.go.com/kabc/story?section=news/local/los angeles&id=9514478.
Mendoza, M., Poblete, B. and Castillo, C. (2010): Twitter under crisis: Can we trust what we
RT?. 1st Workshop on Social Media Analytics SOMA.
Ministry of Home Affairs (2010): Indian Police Service (IPS). http://mha1.nic.in/ips/ips
home.htm.
Muller, M., W. Geyer, T. Soule, , and J. Wafer. Geographical and Organizational Commonalities
in Enterprise Crowdfunding. In Proc. CSCW 2014, pp. 778 789.
Muller, M. J. and S. Kogan (2012): Grounded Theory Method in HCI and CSCW. In J. Jacko
(ed.), HumanComputer Interaction Handbook, CRC Press.
Nayak, V. (2014): 92 Million Facebook Users Makes India The Second Largest Country
[STUDY]. http://www.dazeinfo.com/2014/01/07/facebook-inc-fb-india-demographic-users2014.
Palen, L. and Vieweg, S. (2008): The Emergence of Online Widescale Interaction: Assiatance,
Alliance and Retreat. CSCW.
Plane,
B.
(2013):
[Infographic]
How
Police
Departments
Use
Twitter.
http://connectedcops.net/2013/04/23/infographic-how-police-departments-use-twitter/.
Press Trust of India (2013): Social media if misused becomes a threat to social security: Shinde.
http://tech.firstpost.com/news-analysis/social-media-if-misused-becomes-a-threat-to-socialsecurity-shinde-215320.html.
Qu, Y., Wu, P. and Wang, X. (2009): Online Community Response to Major Disaster: A Case
Study of Tianya Forum in the 2008 China Earthquake.. In Proc 42nd Hawaii Intl Conf. on
System Sciences.
Semaan, B. and G. Mark (2012): Facebooking Towards Crisis Recovery and Beyond:
Disruption as an Opportunity. In Proc. CSCW 2012, pp. 2736.
Shklovski, I., Palen, L., and Sutton, J. (2008): Finding Community through information and
communication technology in disaster response. CSCW.
Starbird, K. and Palen, L. (2011): Voluntweeters: Self- organizing by digital volunteers in
times of crisis.. In Proc. CHI 2011, pp. 10711080.
Stoll, J., kirsten Foot, and W. K. Edwards (2012): Between Us and Them: Building
Connectedness Within Civic Networks. In Proc. CSCW 2012, pp. 237240.
The COPS Office and the Police Executive Research Forum (2013): Social Media and Tactical
Considerations For Law Enforcement. Technical report.
The Economics Times (2014): NCRB to connect police stations and crime data across country in
6 months.
Vieweg, S., Hughes, A., Starbird, K., and Palen, L. (2010): Micro-blogging during two natural
hazards events: What Twitter May Contribute to Situational Awareness. In Proc. CHI 2010.
Voida, A., E. Harmon, and B. Al-Ani (2012): Bridging Between Organisation and the Public:
Volunteer Coordinators Uneasy Relationship with Social Computing. In: Proc. CHI 2012.
Wu, Y., E. G. Lambert, B. W. Smith, S. Pasupuleti, K. Jaishankar, and J. V. Bhimarasetty (2012):
An Exploratory Comparison of Policing Views Between Indian and U.S. College Students.
Intl Criminal Justice Review 2012.