Social Media Intelligence for Law Enforcement
An exploratory study on the advantages and challenges of social media
monitoring for the Dutch National Police
D. A. van Veldhuizen (s1608703)
Master Thesis MSc Crisis and Security Management 13-01-2016
Supervisor: Prof. Dr. E. Bakker Second reader: Dr. R. S. Prins
2
Abstract
In recent years, the Dutch National Police has implemented a number of social media monitoring tools in order to extract intelligence from various social media platforms automatically and continuously. In order to advance research in this field, this exploratory study aims to find what the literature tells us about the advantages and challenges of social media monitoring in general, and to what extent these advantages and challenges are experienced in practice by the Real Time Intelligence Centre of the Dutch National Police. Through semi-structured interviews with expert users of three regional Real Time Intelligence Centres, this study found that the advantages are mostly congruent, but that the emphases on their importance differ. Based on these different emphases and differences regarding the challenges, this study argues that social media monitoring by law enforcement should be perceived as a distinct field of research that urgently needs more extensive research. Due to the trend of increased interweaving of the online and offline society, new research on the role of law enforcement in this new society is necessary.
3
Table of Content
Abstract ... 2
1. Introduction ... 6
1.1 Social Media Monitoring Research ... 7
1.2 Research question... 8
1.3 Structure of the research ... 9
2. Methodology ... 11
2.1 Research design ... 11
2.1.1 Extended case study design ... 11
2.2 Scope of the research ... 12
2.3 Research Methods ... 13
2.3.1 Literature study ... 14
2.3.2 Desk research ... 14
2.3.3 Interviews ... 15
2.4 Evaluation Criteria and interpretation ... 16
3. Body of Knowledge ... 18
3. 1 Advantages of Social Media Monitoring ... 18
3.1.1 Situational awareness ... 18
3.1.2 Detection and early warning ... 20
3.1.3 Emergency communications ... 22
3.1.4 Accessibility ... 23
3.1.5 Summarising the advantages of social media monitoring ... 24
3.2 Challenges of Social Media Monitoring ... 24
3.2.1 Privacy, consent, and dependency concerns ... 25
3.2.2 Organisational challenges ... 26
3.2.3 Semantic analysis challenges ... 26
3.3 Social Media Monitoring in the Dutch National Police ... 28
3.3.1 The Real Time Intelligence Centre ... 29
3.4 Expectations ... 31
4. Results ... 35
4.1 The use of social media monitoring by the Dutch National Police ... 35
4.1.1 Social media monitoring units ... 36
4.1.2 Social media monitoring methods ... 37
4
4.2.1 Situational Awareness ... 38
4.2.2 Detection & Early Warning ... 42
4.2.3 Emergency Communications ... 45
4.2.4 Accessibility ... 46
4.3 Challenges of Social Media Monitoring ... 48
4.3.1 Privacy, consent, and dependency concerns ... 48
4.3.2 Organisational Challenges ... 50
4.3.3 Semantic analysis challenges ... 53
4.4 Other observations ... 55
4.4.1 Police presence in the online world ... 56
4.5 Role of interviewees ... 57 5. Discussion ... 58 5.1 Expectations ... 58 5.1.1 Expectation 1 ... 58 5.1.2 Expectation 2 ... 59 5.1.3 Expectation 3 ... 60 5.1.4 Expectation 4 ... 61 5.1.5 Expectation 5 ... 62 5.1.6 Expectation 6 ... 64 5.1.7 Expectation 7 ... 64 5.1.8 Expectation 8 ... 65 5.1.9 Expectation 9 ... 66 5.1.10 Expectation 10 ... 66 5.1.11 Expectation 11 ... 67 5.1.12 Expectation 12 ... 68 5.1.13 Expectation 13 ... 68 5.1.14 Expectation 14 ... 69 5.1.15 Expectation 15 ... 69 5.1.16 Expectation 16 ... 70 5.1.17 Expectation 17 ... 71 5.1.18 Expectation 18 ... 71 5.1.19 Other observations ... 72
5.2 Answering the research question ... 73
5.3 Reflection... 74
5.3.1 Social media monitoring theory and practice discrepancy ... 74
5
6. Conclusion ... 77
6.1 Social Media Monitoring by the Dutch National Police ... 77
6.2 Comparing Social Media Monitoring Theory and Practice ... 78
6.3 Limitations ... 79
6.4 Policy recommendations ... 80
6.5 Further research ... 80
List of References ... 82
Appendix 2: transcribed interviews ... 88
Interview 1 ... 88
Interview 2 ... 100
Interview 3 ... 114
Appendix 3: Interview Categories ... 127
Category 1: Use of social media monitoring ... 127
Category 2: Accessibilty ... 130
Category 3: Emergency Communications ... 130
Category 4: Detection ... 131
Category 5: Situational Awareness ... 136
Category 6: Privacy, consent, and dependency concerns ... 141
Category 7: Organisational Challenges ... 142
Category 8: Semantic Challenges ... 147
Category 9: Role of the interviewees ... 152
6
1. Introduction
In the last decade, social media platform use is expanding among citizens in the Netherlands. This expansion has been driven by increased internet access and new mobile technologies. These platforms contain enormous amounts of data on the social lives and interactions of citizens, due to the massive amounts of user-generated content and consumer-generated media (Manovich, 2012). In recent years, interest in this data has surged in both the private and public sector. Through new technologies it has become possible to monitor social media platforms automatically and continuously, with the aim of disseminating relevant and actionable information for the interested party (Zeng, Chen, Lusch, & Li, 2010; Stieglitz, Dang-Zuan, Bruns & Neuberger, 2014). Monitoring tools have been designed to implement social listening to use the ‘wisdom of the crowd’ (Grubmuller, Götsch & Krieger, 2013; Zeng et al., 2010). Social media monitoring tools use natural language processing techniques that collect messages from social media through text mining with predetermined search queries. These queries are made up from predetermined keywords that are able to extract messages from social media that contain one or more of the keywords. In addition, the tools can apply filters on the collected messages in order to leave certain messages out or to categorise messages according to certain themes (Verma, Vieweg, Corvey, Palen, Martin, Palmer, Schram & Anderson, 2011). The central idea of social media monitoring is that citizens who use social media are used as passive sensors that provide information about their environment. For example, Japanese authorities successfully employ social media monitoring to detect earthquakes by using messages on social media to warn them when an earthquake occurred (Sasaki, Okazaki & Matsuo, 2010).
The potential value of user generated content on social media has been recognised by law enforcement in the Netherlands. In the ‘National Police Development Plan’ the Dutch police expresses the ambition to utilise social media for police operations (Ministry of Security & Justice, 2012). The monitoring tools that will support the operations of police officers will be used by the Real Time Intelligence Centres (RTIC) in all regional police departments. The use of social media for operational purposes varies between the regional police departments. From 2016 onward, all police departments will implement social media monitoring tools in all its RTIC’s departments (Van den Dries, personal communication, September 5, 2015). Currently, four of the police departments have been using social media monitoring tools for up to three years. Their experience with and perception of the tools can provide valuable information for a successful nationwide implementation in 2016. However, there is
7 currently no substantial research on social media monitoring of social media by police departments in the Netherlands.
1.1
Social Media Monitoring Research
The field of social media monitoring research as a whole is still undeveloped and to some extent separated. Until 2012, most research focused on the application of social media monitoring in crisis management (Kavanaugh Fox, Sheetz, Yang, Li, Shoemaker, Natsev & Nie, 2012). In recent years the field has been expanded to other aspects of social media monitoring. At the most abstract level, research can be divided into three areas: the application of social media monitoring, the technology and software for social media monitoring such as natural language processing, and lastly the challenges of social media monitoring.
The central idea in research on the utilisation of social media monitoring is the extent to which messages on social media can be valuable for interested parties. In other words: the advantages of social media. This field is divided into private and public sector research. In the private sector, monitoring research is aimed at marketing, communications, and other commercial applications concerning consumer social data management. In the public sector, research on the application of social media monitoring is primarily focused on monitoring by political parties, policy makers, and crisis management authorities (Stieglitz et al., 2014). Research on political parties and policy makers is primarily focused on social media monitoring for sentiment analysis on social media (Kavanaugh et al., 2012; Bekkers, Edwards & De Kool., 2013). Studies in the domain of crisis management focus on the application of social media monitoring in crisis communication, detecting crises and incidents, improving the situational awareness of responders, and strengthening the resilience of a community during a disaster. This body of research investigates the information value of messages on social media, the trustworthiness of the messages, the social activity and organisation contributing to resilience and crisis response, and other crisis management issues (Huang, Chan & Hyder, 2010; Lui, Palen, Sutton, Hughes & Vieweg, 2008; Mendoza, Poblete, and Castillo, 2010; Sakaki et al., 2010; Terpstra, Hartman, De Vries & Paradies, 2011; Terpstra, Stronkman, De Vries, Paradies, 2012; Vieweg, Hughes, Starbird, Palen, 2010; Truon, Caragea, Squicciarini, Tapia, 2014; Artman, Brynielsson, Johansson, Trnka, 2011; Lui, 2014; Palen & Starbird, 2011; Purohit, Hampton, Bhatt, Shalin, Sheth & Flach, 2014, Palen & Vieweg, 2008; Palen & Lui, 2007). Due to the novelty of the research, no definitive conclusions or theoretical frameworks on the value of social media monitoring have been designed. Most research in this field recognises the advantages of social media monitoring, but expresses the need for further research.
8 Research on the technology behind social media monitoring has primarily focused on the possibilities and limitations of natural language processing techniques. Researchers in this field have different backgrounds, ranging from social sciences to information technology and language studies. The research therefore varies from a technical approach, to an application and semantics driven approach to monitoring. Overall, research on social media monitoring in this sub-field attempts to further the development of sophisticated and intelligent text-mining technologies (Corvey, Vieweg, Rood & Palmer, 2010; Verma et al., 2011; Palen, Anderson, Mark, Martin, Sicker, Palmer & Grunwald, 2010).
In contrast to application and technology studies, research on the challenges of social media monitoring is focused on the downsides and obstacles of social media monitoring. Studies in the other research sub-fields recognise certain challenges as well. However, they are mostly concerned with the advantages and development of monitoring tools and do not focus on downsides of social media monitoring. Instead, research on challenges puts a focus on legal and ethical concerns, organisational obstacles, and problems concerning the output of social media monitors. Legal and ethical concerns pertain to issues surrounding privacy and consent, whereas organisational obstacles are comprised of issues with human resources, technology, policy, managerial will, and so on. Downsides regarding monitoring output pertain to the deficiencies of the messages that are extracted, such as noise, interpretation problems, and lack of inclusiveness (Bekkers et al., 2013; Huang et al., 2010; Stieglitz et al., 2014; Grubmuller et al., 2010).
1.2 Research question
No substantive research has been conducted that investigates the potential value of social media monitoring for law enforcement operations in the Netherlands, even though a number of regional police departments of the Dutch National Police have already been employing social media monitoring for some years. The information on social media can be valuable to operational police officers, as is shown by research on social media and first responders in crisis management (Vieweg et al., 2010). Technological developments enable monitoring of social media constantly and automatically. Social media monitor developers have successfully expanded the use of monitoring tools across both the private and public sector. However, it seems that this expansion in practice has primarily been driven by expectations, due to a lack of independent evaluative research. This thesis investigates the use of social media monitoring by the Dutch police through an exploratory research, with the aim of comparing social media monitoring theory with its use in practice by law enforcement in the Netherlands.
9 In order to investigate the potential value of social media monitoring for law enforcement in the Netherlands, the central research question of this research is twofold: a) What does the literature tell us about the advantages and challenges of social media monitoring in general? And b) To what extent are these advantages and challenges experienced in practice by the Real Time Intelligence Centre of the Dutch National Police? In order to find an answer to this research question the current body of literature on the advantages and challenges of social media monitoring in relation to law enforcement is investigated with the aim of extracting shared perceptions from the literature. Based on this investigation, a set of advantages and challenges of social media monitoring is formulated. In order to investigate how this relates to the Real Time Intelligence Centre of the Dutch National Police, the ‘National Police Development Plan’ is investigated in order to find how the police aims to utilise social media for operational purposes.
This research focuses on the Real Time Intelligence Centres (RTIC), considering the fact that they are already employing social media monitoring. The tasks and responsibilities of the RTIC will be mapped out to investigate how social media monitoring relates to the day-to-day operations of the RTIC. A number of expectations regarding the advantages and challenges of social media monitoring for the RITC’s specifically are formulated based on this observation. Subsequently, these expectations are compared to the social media monitoring practice by the RITC’. A number of interviews will be held with social media monitoring experts from three regions that have employed social media monitoring for over twelve months. This allows their experiences to be investigated in-depth. The outcome of these interviews are subsequently compared with the expectations. The result of this research can indicate to what extent the advantages and challenges of social media monitoring found in the body of literature agree with the experience of the investigated RTIC’s. Due to the limited scope of the research, the results should not be interpreted as more than an indication of the advantages and challenges of social media monitoring for law enforcement in the Netherlands. However, this thesis does not aim to provide definitive conclusions through explanatory research. Instead, this exploratory study is aimed at furthering research on social media monitoring by law enforcement through investigating the practical use of social media monitoring by the Dutch National Police for the first time.
1.3 Structure of the research
After this introductory chapter, the second chapter will introduce the research design and research methods used in this thesis. Due to the lack of a clear theoretical framework in this field of research, the literature study together with the analysis of the policy document ‘National Police Development Plan’ will constitute the current body of knowledge regarding social media monitoring by Dutch law
10 enforcement. The current body of literature on social media monitoring for crisis management is studied comprehensively, given the fact that research on the use of social media monitoring by law enforcement agencies is absent. These studies are relevant for this thesis because law enforcement is an important part of crisis management and first responders. Therefore, the advantages and challenges of social media monitoring in this field are most comparable to the situation of law enforcement. After the body of knowledge chapter, the fourth chapter presents and analyses the results of the interviews, after which their significance to the expectations is discussed in the Discussion chapter. Lastly, this thesis will conclude by summarising the primary results and implications of this research. The conclusion will also provide policy recommendations for social media monitoring by the Dutch National police, as well as recommendations for avenues of further research.
11
2. Methodology
In this chapter, the methodology used in this research is introduced. First of all, the research strategy and research design are introduced and justified based on the aim and characteristics of this research. Secondly, the scope of this research is explained by clarifying the unit of analysis, as well as the sample and sampling process. Finally, the research methods employed in this research are elucidated. Due to the characteristics of this research, alternative social research evaluation criteria are introduced.
2.1 Research design
This thesis synthesizes several arguments with regard to the advantages and challenges of social media monitoring from a substantial but underdeveloped body of literature for the first time. These arguments are used to build expectations that can be compared to the social media monitoring practice of the Dutch National Police. However, the use of social media monitoring in the Dutch police is fairly new . Only four police regions have been employing social media monitoring for a substantial period of time. In order to compare social media monitoring theory with practice, this research employs a qualitative research strategy to obtain an in-depth understanding of the social media monitoring experiences in the police regions. These experiences are used to investigate to what extent the use of social media monitoring in practice agrees with the expectations that are deduced from the body of knowledge. The case study design allows for an intensive and detailed analysis of a single case and is commonly used in a qualitative research strategy, although it can be used in both quantitative and qualitative research strategies. There is little consensus on what a case exactly is. It can range from a location to an organisation or other social phenomena in certain settings. However, one particular setting is intensively examined in case of a case study (Babbie, 2010; Bryman, 2008). This study employs the case study design due to its specific character of intensive and detailed analysis. This allows finding rich information regarding social media monitoring by the Real Time Intelligence Centres in the investigated police departments.
2.1.1 Extended case study design
Case study research in qualitative research is mostly used in an inductive approach to the relationship between theory and research, meaning that the findings from the case study are used to create new
12 theories. This research has an deductive approach and aims to find the similarities and differences between monitoring theory and practice. This research will therefore employ a variation on the traditional basic case study design, namely the extended case study design as suggested by Burawoy, Burton, Ferguson, Fox, Gamson, Gartrell, Hurst, Kurzman, Salzinger, Schuaffman, and Ui (1991). They argue that qualitative research should be performed based on extensive knowledge of theory, in order to rebuild or improve theory through inconsistencies between observation and the theory. Possible theoretical gaps can be found and existing theories can be modified through the use of extended case study design (Burawoy et al., 1991: 10). This variation on the basic case study design allows this research to find potential flaws in the expectations through a deductive approach. It is important to note that the generalisability of the extended case study design is commonly perceived as problematic, comparable to the basic case study design (Burawoy et al., 1991; Bryman, 2008; Babbie, 2010). These evaluation criteria and others are further discussed later in this chapter, when the research methods used in this thesis are explained.
2.2 Scope of the research
This thesis aims to obtain an in-depth understanding of the experience of the Dutch National Police with social media monitoring. The unit of analysis of this thesis is the Dutch National Police, in accordance with the definition of Babbie (2010) of the unit of analysis: ‘the what or whom being
studied’. In order to investigate the experience of the police, a sample of the police regions are
investigated. Purpose sampling is recommended in qualitative research, especially when the research method interviewing is employed. This means that the researcher samples based on relevance with regard to the research question (Bryman, 2008). The subjective nature of the sample impedes the generalisability of the findings. However, generalisability of the results is not the aim of this thesis. This thesis will therefore employ purposive sampling, meaning that only those police regions that can provide valuable information about their experience with social media monitoring are included in the sample. Valuable information is understood as information on the experience of using social media monitoring after the monitoring tool has been operational in the organisation for over twelve months. It is expected that during this period the monitor can be fully integrated in the operational process of the police and the experience with - and attitude towards social media monitoring have crystallised.
All police regions in the Netherlands will implement social media monitoring in 2016. Currently, six out of eleven police regions are using social media monitoring tools. However, two of those regions have
13 only used the monitoring tools for a few weeks. Their experience with social media monitoring is therefore not expected to be fully developed. Both regions will therefore not be included in the sample. The remaining four regions are North-Netherlands, Middle-Netherlands, Limburg, and Rotterdam. These four regions comprise the sample of this thesis. In the regions Rotterdam, Limburg, and Middle-Netherlands, the monitoring tools are being used in the Real Time Intelligence Centre. In North-Netherlands, the tools are used by the operational police officers under the supervision of the Open Sources police unit. However, the persons that will be interviewed are all social media monitoring experts in their police department, who are responsible for the functioning and evaluation of the social media monitoring tools. Even though their exact functions within the police may differ, the sample units are comparable due to their common relation to the monitoring tools.
Lastly, it is important to state that the four regions might not be objective in their perception of the social media monitors given the fact that they were the first regions to adopt the monitors. The experts of those regions may have a favourable attitude towards innovations such as social media monitoring, or they may be overly positive about their experiences in order to justify the investments made in the monitoring tool. The bias of the experts is unclear and will therefore be scrutinised during the interviews. In order to account for this bias, the role of the expert concerning the adoption and integration of social media monitoring is investigated. The results are presented in chapter 4.5 Role of
the Interviewees.
2.3 Research Methods
In order to increase the confidence in the findings of this theses, a triangulation of methods is used (Bryman, 2008). In this research three qualitative research methods are employed, namely desk research, a literature review, and semi-structured interviewing. In addition to these methods, Bryman (2008) recognises two other methods: ethnography and focus groups. Ethnography entails the observation of the research subject. For example, the experience of the regional police departments might be analysed through watching the users of the monitor during a period of time and observing how they would respond to the information presented in the monitor. This method would not be useful in this research however, due to the fact that the police’ perception of social media monitoring cannot completely be measured by merely observing. Instead, experts can share their perception easily through an interview. If the focus group method would be employed, an interview would be held with a group of users. This method cannot be employed however due to a lack of time and capacity at the side of the police department. Questionnaires could be filled out by Real Time
14 Intelligence Centre employees in order to investigate to what extent their experience with social media monitoring corresponds with the experience of the experts. However, this is not possible due to time constraints of the researcher. The size and scope of this thesis only allow for a synthesis of the body of knowledge and in-depth interviews with experts to create a deep understanding of their experience with social media monitoring. The three methods that are used in this thesis are further explained in the paragraphs below.
2.3.1 Literature study
The literature study constitutes the foundation of this research. The current body of literature of social media monitoring is studied in detail to find all potential advantages and challenges of social media monitoring. Currently, the body of literature is divided into different research fields. Research on social media monitoring focuses mostly on the communications dimension, the community resilience dimension, and the emergency responders dimension of social media monitoring. Monitoring by law enforcement can be perceived as part of the emergency responders dimension. The literature in the emergency responders dimension is divided into several areas. Some studies focus on the benefits of social media monitoring in crisis management. Other studies focus on the challenges of social media monitoring, for example with regard to privacy. However, no substantial research is specifically focused on the use of social media monitoring by law enforcement. This thesis synthesises the findings and arguments of the studies in the different fields. Based on this synthesis, a number of advantages and challenges of social media monitoring are formulated.
2.3.2 Desk research
In order to investigate how these benefits and challenges pertain to law enforcement in the Netherlands, the ‘National Police Development Plan’ is investigated using a desk research method (Ministry of Security & Justice, 2012). The policy document is investigated by finding which police units use social media monitoring. The tasks and responsibilities, the organisational position of these units and their departments, and their relation to police operations is studied. Subsequently, the extent to which the advantages and challenges of social media monitoring pertain to the core tasks and responsibilities of the relevant units and their relation to police operations is investigated. Based on this, a number of expectations are deduced. These expectations are either advantages or challenges of social media monitoring for the Dutch National Police. This is the first step from the body of literature that mainly focuses on social media monitoring by crisis management, to social media
15 monitoring by law enforcement. The literature together with the policy document forms the body of knowledge of this thesis. Due to the undeveloped nature of research on social media monitoring by law enforcement in the Netherlands, the body of knowledge is comprised out of both literature and a critical policy document. This body of knowledge constitutes the basis of this research providing the expectations that are investigated through semi-structured interviews.
2.3.3 Interviews
Semi-structured interviews will constitute the final research method of this thesis. In order to investigate to what extent the expectations agree with the social media monitoring practice of the Real Time Intelligence Centres (RTIC), semi-structured interviews with experts on social media monitoring from the respective regional police departments are held. These experts have been actively involved in the adoption and use of social media monitoring in their respective regional police departments. Through these interviews, data on user experience is collected and analysed. As explained by Bryman (2008), a semi-structured interview is conducted with only a list of fairly specific topics that need to be covered. A semi-structured interview allows for asking follow-up questions on interesting remarks by the interviewee instead of following the strict regime of a structured interview. The topics of the semi-structured interviews are based on the expectations deduced in the body of literature. One of the main advantages of this form of interviewing is its flexibility. Instead of answering rigid predetermined questions, interviewees can share their view on the topics. This can lead to new insights and may even provide hunches for new expectations and theories. Semi-structured interviews can provide rich information that can be used both to investigate predetermined expectations due to the undeveloped nature of the research on social media monitoring by law enforcement, as well as potentially find new insights. Furthermore, this method gives the interviewee’s point of view a prominent place which can provide excellent data for comparison with the expectations. Since this study is very specific, for this period of time replicability is less interesting. More quantitative measures can be developed as the amount of social media monitoring RTIC’s grows.
For every police region one expert has been interviewed. Every interview contains three open questions: why the police region used social media monitoring, what the advantages of social media monitoring are according to their experience, and lastly, what the challenges or downsides of social media monitoring are according to the interviewees. Later, more specified questions on situational awareness and organisational challenges are asked as follow-up questions. Comparability between the interviews will be sought by asking similar follow-up questions in every interview. Due to the spontaneous nature of the answers to the first open question, these answers are perceived to be more
16 valuable than those given to more specified questions. In addition to this differentiation, the results are structured according to the frequency of the specific answers. An answer to a certain question that is given by all interviewees is perceived to be more valuable than an answer provided by just one interviewee. This however does not mean that answers that are given just once are not valuable. Their singular nature may be a consequence of a region specific trait, thus potentially providing valuable and interesting results.
During the interviews, an substantial amount of time and energy is invested with the aim of decreasing the subjectivity of the results. For example by asking to what extent employees in the department of the interviewee can agree with their answers, but also by emphasizing the independent character of this research and the fact that there are no desirable answers they have to conform to. Also, it is expected that the anonymised interviews will contribute to the objectivity of the results. It is recognised that the answers may contain a certain bias the interviewee is unconscious about. Even so, these three interviewees can be seen as expert users of social media monitoring with regard to police operations in their region, and are therefore the only logical choice to interview about the experiences of their region with social media monitoring.
2.4 Evaluation Criteria and interpretation
Social research is commonly evaluated along three criteria: reliability, replicability and validity (Babbie, 2010; Bryman, 2008). The reliability of a study pertains to whether the outcomes of the study can be repeated by other researchers. It questions whether the measures that are used for concepts are in fact measuring those concepts. Replicability means that other researchers have to be able to replicate the research process. Therefore, researchers must explain the research process accurately. Validity is concerned with the integrity of the conclusions of a research, and can be divided into internal and external validity. Internal validity is concerned with whether there is a good match between the researchers observations and the conclusions. External validity pertains to the extent to which they can be generalised across other contexts (Bryman, 2008: 31 – 33).
However, semi-structured interviewing in an extended case study design aims to find an deep understanding of one particular setting, and not to find generalizable results. Therefore, alternative criteria for evaluating qualitative research and qualitative interviewing have been designed. Auerbach and Silverstein (2003) argue that traditional criteria are unfit for qualitative research, due to the fact that the criteria suppose objectivity in research. Qualitative research is focused on the in-depth
17 understanding of a situation through interpretation, thus inherently allowing subjectivity in the research. Instead, they propose to replace the criteria reliability and validity with the following criteria:
Transparency: other researchers understand the interpretation process through transparency
about the data analysis procedure.
Communicability: the themes and theoretical constructs in qualitative data analysis make
sense to other researchers.
Coherence: the theoretical constructs must fit together and allow a coherent story.
Transferability: theoretical constructs can be described in abstract patterns that can be found
in different contexts. However, the specific content of those patterns depends on the context.
This research is designed according to these evaluation criteria. In order to ensure transparency, communicability and coherence, the semi-structured interviews are transcribed and the interpretation of the results is explicated. The transcribed results will be divided into categories in accordance with the advantages and challenges of social media monitoring as described in the body of literature. This means that the results will be interpreted along the advantages and challenges found in the body of literature. In addition, a category for results that do not correspond with one of the advantages or challenge will be created. The transcribed interviews are added in Appendix 2. The categorised results are presented in Appendix 3. The categorisation of the results will be further elaborated in the Body of Knowledge and Results chapters.
Recording and transcription of the interviews will take place with consent of the interviewees. Additionally, the interviews are anonymised due to researchers’ employment at the provider of the Twitcident monitoring tool. The interviewees might be hesitant to share their actual experiences if the answers can be related back to them, for that may put tension on their relationship with the provider of the monitor. Anonymised interviews are expected to provide more accurate results, even though this may impede the replicability of the research. Due to the small N in this research however, that effect is limited. In order to account for the earlier mentioned bias that may be present in the sample, the interview guide will include a question on the favourability of the subject to innovations such as social media monitoring. The interview guide is attached in Appendix 1.
18
3. Body of Knowledge
In this chapter, a number of expectations are deduced from the literature and Dutch ‘National Police Development Plan’ policy document. These expectations are compared to the outcome of the semi-structured interviews of this thesis, in order to answer the research question. The expectations are developed based on the current body of literature in the field of social media monitoring and the goals and tasks of the regional Real Time Intelligence Centres of the Dutch police. First, the advantages of social media monitoring are investigated. Secondly, the challenges of social media monitoring are explored. Thirdly, this chapter investigates the goals and tasks of the Real Time Intelligence Centres, with the aim of finding similarities between these goals and the advantages and challenges of social media monitoring. Finally, a number of expectations are deduced based on the similarities .
3. 1 Advantages of Social Media Monitoring
In the paragraphs below, the advantages of social media monitoring found in the body of literature that pertain to the operations of law enforcement are synthesized into four topics: situational awareness, detection and early warning, emergency communications, and accessibility.
3.1.1 Situational awareness
The first advantage of social media monitoring is its potential impact on the situational awareness of police officers. In order to understand this process, situational awareness must first be defined. One of the most influential authors in the field of situational awareness is Mica R. Endsley (Salmon , Stanton, Walker, Jenkins, Ladva, Rafferty and Young, 2007). He defines situational awareness as: ‘the perception
of the elements in the environment within a volume of time and space, the comprehension of their meaning, and the projection of their status in the near future’ (Endsley, 1995: 36). Even though there
are other definitions of situational awareness, Smith and Hancock for example define situational awareness as adaptive and externally directed consciousness (Smith and Hancock, 1995: 138), this thesis will use the definition of Endsley.
In addition, Endsley distinguishes three levels of situational awareness. The first level is the perception of elements in the environment. This entails comprehension of the environment at the most basic level; the police officer is aware of the elements in his surroundings. Level two situational
19 awareness pertains to the comprehension of the situation by the police officer. In the second level, the police officer is able to recognise patterns and to form a comprehensive understanding of the environment. The third and final level is the projection of future status. On this level, the police officer is able to determine the preferred course of action with regard to the objectives through comprehension of the situation, and dynamics of the elements in the environment, as well as a holistic understanding of the situation (Endsley, 1995).
3.1.1.1 Real-time information and situational updates
Monitoring social media automatically and in real-time allows law enforcement to disseminate and analyse actionable information from social media about a particular environment, elements in that environment, and the dynamics of that environment. This contributes to the situational awareness of police officers that operate in that particular environment (Corvey et al., 2010; Huang et al., 2010; Terpstra et al., 2011; Terpstra et al, 2012; Vieweg et al., 2010; Verma et al., 2011; Truon et al., 2014). At the most basic level, social media can provide text-based information about a particular environment and the changes that occur in that environment. For example, Vieweg et al. (2010) found in their research on the Oklahoma Grassfires and the Red River Floods in 2009 that the social media platform Twitter contained large amounts of information about fire locations, flood levels, weather, visibility, road conditions, and many other elements.
Because social media is monitored in real-time, it can provide situational updates to responders in the area about the elements in their environment. Additionally, social convergence during an incident can be witnessed on social media. Therefore, responders do not only receive information about the dynamics of the threat, but also about the response of the community involved in the incident (Lui et al., 2008; Kavanaugh et al., 2012). Situational updates can contribute to all levels of situational awareness by providing information about the elements in the environment, as well as providing information about possible future alterations of those elements. This allows the police officer to create a comprehensive understanding of his environment,
3.1.1.2 Visualisation of information
In addition to text-based information, social media can provide visual information of an environment with photos and videos of a situation that are being shared by people witnessing the situation. This visual information can provide police officers with an direct impression of the elements and their dynamics in the environment. The visual character of this information contributes to great extent to
20 sense-making of the situation by police officers, and therefore has an important impact on their situational awareness (Terpstra et al., 2011; Kavanaugh et al., 2012, Lui et al., 2008).
Social media monitoring can provide other important visual information as well. Messages can also contain geo-location information such as addresses, street names, intersections, city names and landmarks. Location information can also be integrated in the messages, as well as in the profile of the author. Through searching for geographical elements in messages on social media, a broader understanding of the environment and situation can be attained. The geo-location information can subsequently be plotted on a map to visualise this information, enhancing its impact on the situational awareness of the police officers (Vieweg et al., 2010; Terpstra et al., 2011; Terpstra et al., 2012; Kavanaugh et al., 2012).
Lastly, social media messages can be categorised in different themes through a filtering process. Messages are collected and divided according to predetermined themes, such as public unrest or violent crimes. Similarly to geo-location plotting, these messages can be plotted on a map in order to show how certain issues are spread and developing over an area, thus visualising the information and enhancing the situational awareness of police officers (Terpstra, 2012; Kavanaugh et al., 2012). Other advantages of filtering with regard to detection and accessibility are explained later in this chapter.
In summary, social media monitoring contributes to the situational awareness of police officers by providing situational updates about the elements in an environment and the dynamics of that environment. This enables them to make sense of their surroundings as well as providing them information to determine a future course of action. Additionally, social media monitoring can provide important visual information through photos, videos, geo-location plotting, and thematic plotting. This improves the sense-making process of their environment by police officers. Social media monitoring therefore contributes to all levels of situational awareness as described by Endsley (1995).
3.1.2 Detection and early warning
The second advantage of social media monitoring is the ability to detect security issues, which can provide law enforcement with an early warning of these issues. Security issues that are detected can be divided into acute and non-acute security issues. Due to the constant and real-time monitoring of social media, both acute and non-acute security issues can be detected as a result of particular filtering strategies. Both types of security issues and filtering strategies are explained in this paragraph.
21
3.1.2.1 Detection of acute security issues
Acute security issues are, for example, natural disasters and social convergence (Vieweg & Starbird 2010; Sasaki, et al., 2010). With social convergence, events with a large population density and with cause for heightened security measures are meant (Kavanaugh et al., 2012). These acute security issues can be detected through spikes in social media volume with regard to a certain topic. For example, Sasaki, Okazaki and Matsuo (2010) investigated social media behaviour just before, during, and after an earthquake occurred in Japan. They found that they could detect earthquakes with a probability of 96% through social media monitoring, as a result of recognising specific spikes in earthquake-related messages on social media (Sasaki et al., 2010). Social convergence can be detected in a similar fashion, only with different search key-words and filters. Law enforcement is able to detect the social convergence as soon as it is mentioned on social media if a social media monitor is programmed to monitor for demonstrations or riots in a particular area. Monitors can be programmed to provide a warning when a predetermined amount of messages around a certain topic in a predetermined timeframe are detected (Kavanaugh et al., 2012). Law enforcement can therefore detect acute security issues immediately, as long as the issue at hand has a certain impact on its surroundings and creates a spike in social media volume. A substantial body of research has shown that acute security issues are accompanied by spikes in social media volume, therefore enabling law enforcement to detect these issues through social media monitoring (Terpstra et al., 2011; Terpstra et al., 2012; Sasaki et al., 2010; Kavanaugh et al., 2012; Huang et al., 2010; Vieweg & Starbird, 2010, Vieweg et al., 2010).
3.1.2.2 Detection of non-acute security issues: early warning
In addition to the detection of acute security issues, social media monitoring with certain filters can detect and monitor non-acute security issues. These issues are for example social convergence planning, sentiments around certain issues, and insights in trends (Kavanaugh et al., 2012; Terpstra et al., 2011). The issues do not necessarily require direct action of law enforcement. However, they can provide valuable information with regard to an early warning of potential future security issues. Furthermore, constant monitoring can also detect offenders in the aftermath of a incident. For example, photos and videos that have been collected by a monitor can contain evidence of offences that were not detected during the incident itself (Kavanaugh et al., 2012). Social media monitoring can therefore detect security issues and offenses before, during, and after an incident.
22 In summary, monitoring social media through specific filters and warning systems can assist law enforcement in detecting acute security issues, such as natural disasters or social convergence, as well as non-acute security issues. Monitors can be programmed to recognise spikes in social media volume or to detect important individual messages that contain early warning information.
3.1.3 Emergency communications
The third advantage of social media monitoring can be epitomised as the benefits of a new communications channel between law enforcement and the public. Huang et al. (2010) describe how during the Morakot typhoon in Taiwan in August 2009, traditional emergency communication channels did not have the capacity to handle all the calls and information that was send to them. This caused a lack of information among both the emergency response and the public at large, given the fact that they could not properly be informed by the ill-informed authorities. Due to the increased use of mobile communications, a larger part of the public is able to inform the authorities of an incident or emergency (Huang et al., 2010). Problems may occur if traditional emergency communications channels are not organisationally prepared for this. By using social media monitoring, authorities such as law enforcement could utilise social media as an additional communications channel in times of emergency. In that case the public can still reach authorities via social media when traditional systems fail. Social media monitoring could potentially ensure that all the relevant messages are detected by the authorities.
3.1.3.1 Information volume and velocity
In addition to providing an extra communications channel, monitoring social media is becoming increasingly essential for an effective response by authorities in case of an incident. Studies by Kavanaugh et al. (2012), Terpstra et al. (2011) and Sakaki et al. (2010) show that in the early stages of an incident, social media provide more, faster and better information than official sources. This is especially true in the first hour after an incident has taken place. This may not be surprising, given the fact that social media has an enormous reach due to the great amount of users that easily access their social media accounts via mobile phones to provide valuable information (Kavanaugh et al.,2012). In addition to providing an extra communication channel, social media monitoring is therefore essential for law enforcement when responding to large or small incidents due to the amount of valuable information that is present at social media platforms.
23
3.1.4 Accessibility
The first three paragraphs of this chapter all argue that valuable information is present on social media, for both acute and non-acute security issues. Social media users provide information that can improve the situational awareness of police officers on the ground, as well as reveal trends over space and time that would otherwise have gone unnoticed. However, this information would not be accessible without social media monitoring. Due to the enormous amount of messages that are posted on a seemingly endless variety of topics every moment, disseminating the relevant messages would be impossible without systematically monitoring social media (Corvet et al., 2010; Huang et al., 2010; Kavanaugh et al., 2012; Terpstra et al., 2011; Verma et al., 2011; Truon et al., 2014). Therefore, the most important aspect of social media monitoring is the monitoring tool itself. Manually searching for single keywords on social media would be an extremely inefficient and ineffective method. In order to identify and disseminate relevant information, social media monitors use natural language processing techniques that allow them to scan social media automated and in real-time and structuralise the findings according to predetermined themes (Corvey et al., 2010; Verma et al., 2011; Truon et al., 2011).
The indispensableness of automated and systematic monitoring is threefold. Firstly, the amount of messages posted on social media increases dramatically during times of mass social convergence and disasters. Without systemic, thematic monitoring the information on social media would be inaccessible, just when it is most needed (Verma et al., 2011, Corvey et al., 2011; Terpstra et al., 2011). Vieweg and Starbird (2010) investigated to what extent they could find messages with original information, relevant to emergency responders, on social media during the Red River flooding in 2009. They used a set of search key-words in order to find relevant messages. They found that less than 10% of the messages they analysed contained original information. The large majority of messages on social media were a reproduction of earlier information, and therefore far less relevant to responders such as law enforcement. Automated, systematic monitoring tools can filter the messages in such a way that they only show original messages.
Secondly, manual continuous monitoring of social media on multiple themes across a variety of social media platforms in order to detect any security issues, is highly inefficient and ineffective compared to automated motoring (Corvey et al., 2010; Kavanaugh et al., 2012). Social media monitoring allows law enforcement to monitor social media platforms efficiently and effectively on a variety of themes due to the filtering system of the monitoring tools.
Lastly, social media monitors can detect trends spanning time and space, thus providing information that would not be clear if the messages would be processed individually. Also, social media
24 monitors are able to plot geo-location information and thematic information on maps, which can improve the situational awareness of police officers (Truon et al, 2014; Terpstra et al., 2011).
In summary, social media monitoring can provide law enforcement with thematised, precise, and enriched information by disseminating the information through automated, monitoring tools. These tools extract specific information, as well as provide meta-analyses of large social media data sets.
3.1.5 Summarising the advantages of social media monitoring
The advantages of social media monitoring can be divided into four groups. First of all, social media monitoring can provide law enforcement with thematised, precise, and enriched information by disseminating the information through automated, systemic monitoring tools. This information would otherwise be inaccessible due to the sheer number of messages. Secondly, social media monitoring contributes to the situational awareness of police officers through providing situational updates about their environment. Additionally, social media monitor can provide important visual information through photos, videos, geo-location plotting, and thematic plotting, thus improving the sense-making process of their environment by police officers. Thirdly, social media monitoring social media can detect acute and non-acute security issues through specific filters that detect spikes in social media volume. Lastly, social media monitoring provides an additional communications channel between the public and law enforcement that provides the police officers with important information, even if traditional communication channels would fail.
3.2 Challenges of Social Media Monitoring
Three types of social media monitoring challenges are investigated in the following paragraphs,. First, privacy, ethical and dependency concerns are explored. Secondly, organisational challenges are investigated. What are important organisational issues for implementing social media monitoring? Lastly, challenges concerning semantic analysis are explored. How can semantic analysis ensure reliability, inclusiveness, and objectivity?
25
3.2.1 Privacy, consent, and dependency concerns
The first set of challenges pertains to concerns about the privacy of social media users, consent, and dependency of third parties.
3.2.1.1 Privacy
Users may not be aware of the public nature of their messages due to the blurred boundaries between the public and private domain on social media. The fact that parties in both in the private and public sector are using the information in their messages for their own goals may not be clear to every social media user (Vieweg & Starbird, 2010). Grubmuller et al. (2013) and Bekkers et al. (2014) both argue that privacy concerns are particularly relevant when the social media messages are used by public organisations. They state that citizens are more concerned about their privacy when dealing with the government than they are when dealing with private sector parties. Vieweg and Starbird (2010) propose that the messages on social media could be anonymised. In that way, the information in the messages can remain valuable, but the author can’t be identified.
3.2.1.2 Consent
Even though social media platforms are designed to spread the messages of their users to anyone who is interested in them, users may not perceive their account and messages to be completely public. The line between the public and private domain is therefore not as clear on social media as it may be in real life. And even if users understand that their messages are public, they may not know that third parties are using, analysing, and archiving their messages (Grubmuller et al., 2013; Stieglitz et al., 2014). Social media users agree to the terms and conditions of the platform they use, but may very well not be aware of agreeing to third party use of their messages. Vieweg and Starbird (2010) argue that in most cases there is no active consent of the users allowing law enforcement to use their tweets. Privacy concerns are especially important in the case of governmental agencies according to Bekkers et al. (2013). They argue that transparency is needed on social media monitoring by law enforcement.
3.2.1.3 Third party dependency
Some authors also express concerns about dependency of third parties in addition to privacy and consent concerns,. First of all, law enforcement becomes dependent of social media platforms. For example, the platforms might not be able to cope with the amount of users in case of a disaster or they might not allow the monitor tool to access their entire database of messages. Furthermore, the
26 platforms could decide to change their policy and terminate their collaboration with law enforcement agencies (Huang et al., 2010; Stieglitz et al., 2014). This dependency of social media platforms may reduce the appeal of social media monitoring by law enforcement. Secondly, monitoring tools are dependent of internet infrastructure, such as cabling, routers, electricity, and so on. This means that law enforcement agencies are dependent of telecommunication companies as well. Another problematic aspect of this issue is the fact that infrastructure often is disrupted or destroyed in case of a disaster, rendering social media monitoring useless in the areas in which it is most needed (Huang et al., 2010). Lastly, law enforcement agencies are dependent of citizens to provide them with useful messages on social media. However, some areas may have limited internet or social media access. Furthermore, some parts of the public may be underrepresented on social media, thus creating an image that may not represent the actual situation.
3.2.2 Organisational challenges
The second set of challenges of social media monitoring pertains to organisational challenges. Both Huang et al. (2010) and Terpstra et al. (2012) argue that implementing social media monitoring in an organisation demands investments in both technology and human resources. Additionally, it requires managerial will to successfully implement social media monitoring tools into existing operations. Terpstra et al. (2012) state that the implementation of social media monitoring may be frustrated by these barriers and that both organisational and technical solutions are needed in order for professionals to use social media monitoring. However, Huang et al. (2010) warn that it should not be perceived as a silver bullet. According to them, organisations may expect too much of it and depend too much on it. Both can be harmful in the case of law enforcement. Social media monitoring should therefore be viewed as an addition to the current information channels.
3.2.3 Semantic analysis challenges
The last type of challenge are issues surrounding the concept of semantic analysis. Social media monitoring depends on semantic, or text-based, analysis. This type of monitoring poses several problems, such as interpretation, noise, inclusiveness, and trustworthiness.
3.2.3.1 Inclusiveness
Semantic analysis does not guarantee the processing of all the relevant and important data that can be found on social media, according to Grubmuller et al. (2013). A selection of social media sources is
27 made when a monitor is programmed. Not all social media platforms may be included, which means that not all relevant messages may be included. Furthermore, the words in the search query may not be all-inclusive and therefore not include all the relevant messages present on social media, even when all social media platforms are being monitored. In addition, the search query is inherently subjective. It is built according to the builder’s perception of the topic. Therefore, the outcome is also inherently subjective and may not include all the relevant messages (Grubmuller et al., 2013). Stieglitz et al. (2014) add to this issue that messages on social media have a textual content that can be context-rich, includes emoticons, abbreviations, slang, sarcasm, irony, and so on. Semantic monitoring may not understand the context of the messages and therefore miss relevant messages or select irrelevant messages.
3.2.3.2 Noise
Stieglitz et al. (2014) provide an example of how monitoring tools can select irrelevant messages, also referred to as ‘noise’. Social media contains much noise that needs to be filtered in order for social media monitoring to be efficient and effective. Monitoring tools aim to select the right type messages through comprehensive search queries and the right filters to reduce noise as much as possible. However, Kavanaugh et al. (2012) and Stieglitz et al. (2014) argue that semantic filtering is a complicated process and question the capability of monitoring tools to do so. Filtering noise is an important part of the monitoring tool. The value of the tool would decrease drastically if the tool would show too much noise.
3.2.3.3 Interpretation
In addition to the subjectivity of the output of the monitor itself, the processing of the output is subjected to subjectivity as well. Grubmuller et al. (2013) argue that the output of the monitor is interpreted by the person who reads and processes the information. That person can interpret the message in another way than the author meant it to. This is especially the case with sarcastic or ironic messages. If those type of messages are read without context, they might very easily be misinterpreted. Additionally, a person could try to start a rumour and spread misleading messages. Without the context of the messages, interpreting the messages may be very difficult and could eventually lead to wrong conclusions. This can make the output of social media monitoring unreliable in some cases.
28
3.2.3.4 Trustworthiness
Lastly, social media monitoring is concerned with the issue of trustworthiness of messages. How can a monitor validate messages and determine to what extent a message is authentic and accurate? (Huang et al., 2010). Another concern pertaining to trustworthiness is the question to which extent the participants on social media are representative of a larger group. Even though social media use is widely spread in the Netherlands, not everyone may feel the need to share certain topics on social media. Misrepresentation decreases the overall trustworthiness of messages on social media in itself (Bekkers et al., 2013). Furthermore, Grubmuller et al. (2013) and Bekkers et al. (2013) argue that rumour, false information and noise can spread fast on social media. This may lead to law enforcement acting on false information or cause distrust of the monitor itself.
Mendoza et al. (2013) disagree with this argument. They found in their study on the reliability of Twitter messages that the reliability of those messages can be measured. They argue that people question information on social media that has not yet be validated, and share information that is validated by for example governmental agencies. This shows that social media users do not blindly spread false information and that social media monitoring could determine to some extent whether a message is reliable or not. However, the risk of acting on false information remains present, potentially impeding the trustworthiness of social media monitoring.
3.3 Social Media Monitoring in the Dutch National Police
In this paragraph, the intended use of social media by the Dutch police will be investigated. According to the development plan of the Dutch National Police (Ministry of Security & Justice, 2012) social media is being used by two different parts of the organisation: the communications and media department, and the Real Time Intelligence Centre (RTIC). The word ‘social media’ appears fourteen times in the policy document: twelve times concerning a communications department, and twice concerning the RTIC. No other departments are linked to social media use whatsoever. The RTIC monitors social media to find information that can be shared in real-time with police officers that are operational on the ground. This research focuses on social media monitoring with regard to the operations of police officers, not on the communicative dimension of social media. Therefore, the RTIC will be central to this research. In order to compare the social media monitoring theory with practice, this section will map out the tasks and responsibilities of the RTIC and investigate to what extent similarities can be found between those responsibilities and the advantages of social media monitoring. Those
29 similarities, combined with the challenges described in this chapter, will form the expected advantages of social media monitoring by the RTIC in practice. This is investigated later in this research through semi-structured interviews. Only the advantages of social media monitoring are compared to the responsibilities of the RTIC due to the fact that the advantages described in the paragraphs above pertain to social media monitoring in crisis management. The challenges of social media have a universal character; they are applicable to all settings of social media monitoring by a public organisation. Therefore, the advantages are compared to the tasks and responsibilities of the RTIC in order to investigate which advantages pertain to social media monitoring by law enforcement in the Netherlands.
In the paragraphs below, the RTIC and its position in the police organisation will first be explored. Subsequently, the tasks and responsibilities of both the RTIC and the Basic Team police officers will be investigated.
3.3.1 The Real Time Intelligence Centre
The Dutch National Police consists of three operational levels: the national level, the regional level, and the local level. The national level focuses on national and specialist police issues. The regional levels are fully responsible for the operational activities in their region. Each region is divided into districts and each district has a number of so called ‘Basic Teams’. These Basic Teams form the foundation of the law enforcement organisation and focus on a specific, local area, for which they are responsible for the day-to-day police operations. More specialist police tasks such as criminal investigation or inter-regional cooperation are carried out by other divisions of the national and regional police organisation (Ministry of Security & Justice, 2012). Each regional police department has a Real Time Intelligence Centre (RTIC), which provides the Basic Teams with relevant operational information. The Dutch National Police consists of eleven regional police departments. In trans-regional cases or special operations, the National department may be operationally active in the regions. However, every regional department has the capacity to provide comprehensive police operations to their region. In order to do so, every department has a Regional Operations Centre (DROC) as well as a Regional Information Organisation (DRIO). The DRIO provides information to the police department and police operations. The DROC is responsible for the operational management of police officers on the ground. The Real Time Intelligence Centre is officially part of the DRIO but is physically located within the DROC. This allows them to provide real-time information to police officers, in addition to information that is already provided by the emergency room. The position of the RTIC in the DROC is illustrated in figure 1.
30
Figure 1
3.3.1.1 RTIC tasks and responsibilities
The main responsibility of the RTIC is to provide police officers of the Basic Teams with relevant operational information, both on request and non-requested. For example, after police officers have received information from the emergency room, the RTIC provides the police officers with additional information about the object, environment, and potential threats. However, the RTIC is not restricted to emergency situations for providing information. It can provide real-time intelligence 24/7 using internal police sources, open sources and social media. The RTIC analyses the information and provides police officers with crucial, actionable information. The ‘National Police Development Plan’ (Ministry of Security & Justice, 2012) describes the core tasks of the RTIC as follows:
1) The RTIC continuously develops a comprehensive up-to-date operational perception of upcoming events, potential threats and the operational agenda;
2) The RTIC provides real time intelligence, requested and non-requested, to support operations containing acute security issues, such as emergency response, detection, protection, and supervision, as well as to police officers who face less acute security issues, to improve their situational awareness;
3) The RTIC monitors additional sources in real-time, and establishes relationships between different information sources and incidents, to inform and advice police officers;
4) The RTIC preserves information that is not taken into account by police officers during operations.
