FAQ

The main aim of the EDT is to create a unique and large dataset to examine specific risk and protective factors for terrorism.

The EDT was established during the European Union funded DARE-project (Database and Assessment of Risk of violent Extremists). The DARE project started in 2017 and ended in 2019. Thereafter the EDT obtained funding from the Dutch National Coordinator for Counterterrorism and Security, the Dutch Ministry of Justice and from the European Commission (DG Justice).

The project launched in October 2017 in collaboration with research institutes and judicial organizations from several EU Member States. The Netherlands Institute of Forensic Psychiatry and Psychology (NIFP) coordinates and leads the research that started in the EU-funded DARE project (Database and Assessment of Risk of Violent Extremists) together with the Institute for Interdisciplinary Research on Conflict and Violence (IKG) of the University of Bielefeld, and the Psychosocial Service of the Belgium Penitentiary Institutions. The prison services of different German Member States, the Austrian Prison Service and the Swedish Prison and Probation Service also participate.

The coded EDT data consists of personal and contextual offender information from judicial files of the participating Member States. These files include comprehensive documentation from the police, public prosecution, detention, forensic mental health assessment reports and probation reports.

Based on judicial information sources, a large number of potential risk factors, protective factors, and indicators for terrorism are coded in the EDT. These encompass demographic data, childhood circumstances, trigger factors, ideologies, motives, mental health issues, and the nature of the terrorist offence and type of sentence. Information from forensic mental health assessments is included to provide insight into mental health issues as possible risk factor for terrorist acts. The EDT items are divided over the following 16 domains:

EDT domains

1. Compilation Case File

This domain is used for administrative purposes. This section involves the criminal case number, police registration number, court registration number or, if available, the name of the criminal case or group file.

2. Demographic Data

This domain includes information about geographical origin, socioeconomic status, education, belief when young, profession, family circumstances and kinship, non-stable employment status, being an expatriate, being alienated and social bonds.

3. Aspects crime

This domain includes information about specific aspects of the committed crime such as location, modus operandi, possible people involved and the victims and subjects emotional condition. Furthermore, information about the conviction and judicial consequences is mentioned.

4. Criminal History

In this domain the criminal history of the subject is documented. Specifications of former crimes, previous convictions, sentences and compliance with former judicial interventions are included.

5. Personal History

In this domain several aspects of the personal history of the subject are involved. Vicitimization and trauma entails information about (witnessed) violence at home or outside the family situation. Moreover information about negative experiences with authorities, misrecognition, discrimination and meta-dehumanization is given. In this domain neglectful parenting and early disruption in care are included. Furthermore, problems with work and school are coded. Somatic illnesses and mental disorders from the subject and their family are documented.

6. Personality Traits

This domain covers the personality disorder and traits that are diagnosed by a psychiatrist or psychologist before (anywhere in history), during or after the crime. Furthermore, other personality traits are coded as well, such as poor regulation of aggression, problems with relationships, expressed feelings of distrust and expressed feelings of anger.

7. Psychiatric Symptoms  

This domain covers the psychiatric symptoms of the subject that are diagnosed by a psychiatrist or psychologist before (anywhere in history), during or after the crime. Previous hospitalization, previous treatments and previous suicide attempts  and their specifics are coded as well.

8. Prior To Crime: Incidents

This domain covers important events that happened in the year prior to the moment that the crime is committed. The incidents that are coded are: had or awaited a negative relational loss or degrading event at work or school, degrading financial problems or other negative incidents in year prior to crime.

9. Prior To Crime: Personal Acts

This domain covers actions of subject in the year prior to the moment that the index crime is committed. It involves for example self-isolation, farewell acts to friends/family, increase in energy, mentioning the crime, aspects about the crime and outlets used for mentioning the crime.

10. Preoccupation with Weapons & Violence

This domain is about wethere the subject had a preoccupation with weapons, violence and military aspects in the years prior to the crime.

11. Radicalization /Ideology

This domain covers aspects of the subjects radicalization. This includes change of clothing, social context and location of radicalization, attitude parents towards radicalization and direction of radicalization.

12. Beliefs and Attitudes (VERA-2R)

These items relate to the VERA-2R indicators. Several aspects of the beliefs and attitudes of the subject are covered. For example: expression of grievances about perceived injustice and the extent, expression of anger, moral outrage or hatred in response to perceived injustice, rejection of democratic society and the extent and indication of a human target being dehumanized and the frequency.

13. Social Context & Intention (VERA-2R)   

These items relate to the VERA-2R indicators. This domain covers the social context and its impact on the process of radicalization to violence and an individuals intention to plan and carry out an act of violence. It entails information about the search and the development of violent extremist materials, the target and the plans for the attacks, personal contact with violent extremists or terrorists and  susceptibility and ability to influence.

14. History, Action & Capacity (VERA-2R)

These items relate to the VERA-2R indicators. Several aspects of the history, action and capacity of the subject are covered. Exposure to political violence, family/friends involved in violent extremism or terrorism, previous convictions of violent acts are mentioned. Military and extremist ideological training and being a trainer/leader/influential figure in the ideology are covered. Lastly, the ability to facilitate extremists violent acts and the extent is documented.

15. Commitment & Motivation (VERA-2R)

These items relate to the VERA-2R indicators. The subject can have different motivations to take part in acts of violent extremism. The motivations that are coded are: religious obligation and/or glorification, criminal opportunism, the desire to belong to a group, moral principles and/or moral superiority, excitement and adventure, gaining status, search for meaning or significance in life.

16. Protection and Risk Mitigation (VERA-2R) 

These items relate to the VERA-2R indicators. This domain covers the protective factors in the subjects life. Some items are: a change in ideological values, the rejection of violence as a mean to achieve ideological goals, the availability of professional care and willingness to participate, support from family/friends/community/others and other potential protective factors.

Currently, the EDT includes terrorist and violent extremist offenders from the Netherlands, Belgium, Germany, Austria and Sweden who were convicted from 2012 onward. The selection of cases was made by the judicial organization in each EU Member State. All kinds of terrorist and violent extremist deeds are included, such as, for example, jihadism and other religious-based forms of terrorism, as well as nationalistic, right- and left-wing, and single-issue terrorism. These cases also include convictions for involvement in terrorist acts, even in the absence of ideological motives of the offenders themselves. To avoid potential stigmatization, those persons who are indicted or suspected of terrorist or violent extremist crimes, but are not (yet) convicted, are not included.

Radicalization, terrorism and violent extremism are continually evolving concepts, which in part accounts for the heterogeneity of the available definitions. In order to fully comprehend the meaning of the results of the prospective studies that will use the EDT data, it is above all important to know which definitions were used to obtain the data. Therefore, we will now discuss the definitions of key concepts that underpinned the data collection upon which the EDT is based.

Radicalization  Radicalization is often used to describe the process of adopting an extremist belief system that may result in the acceptance, legitimation and/or use of violence. In the EDT codebook, radicalization is coded when meeting the following description:

“The radicalization domain in the EDT covers events that happened in the years prior to the moment that the crime is committed. Radicalization is: The active pursuit of and/or support for fundamental changes in society that may endanger the continued existence of the democratic order (aim), which may involve the use of undemocratic methods (means) that may harm the functioning of the democratic order (effect) (The Dutch Intelligence Service (AIVD) and The National Coordinator of Terrorism and Security (NCTV). Social bonds may be of essential importance for radicalization and participation in terrorist groups, and social and emotional support and development of a shared identity appear to play a role in the radicalization to violence process (Bjørgo, 2011).  Jihadi terrorists in Europe appeared to have a dogmatic, ideological rigidity and a simplistic and utopian world view (Bakker (2006). Note: immersion into religious practice can result in radicalization, but on itself should not be considered as radicalization. Source: EDT Codebook.

Terrorism  There are many definitions of terrorism. Terrorism can be driven by a range of (political, social, religious or other) ideologies or motivations, can take different forms and can be associated with different types of individuals and groups. One unequivocal feature of terrorism is that it is always a premeditated act (rather than a brief period of anger or impulsivity). Although regularly assumed that for an act to be classified as a terrorist act, it must be driven by a political, social, religious or other ideologically based motivation, this may not always be the case. For example, motivations can also be driven by monetary gain, status seeking, group belonging or excitement. For the EDT, we followed the definition of the NCTV, which defines terrorism as: “the threat of, the preparation of, or the committing of, serious violence based on ideological motives against people, or deeds aimed at causing socially disruptive material damage with the goal to cause social change, to instil fear among the population or to influence political decision-making.”

The decision to use this specific definition means that all types of terrorism can be included in the EDT. For all articles regarding terrorism in each included Member State, see EDT codebook, Appendix 5.

Violent Extremism  Violent extremism is often considered as unlawful violence in furtherance of a religious, political, social or other ideology. It can be described as the beliefs and actions of people who either support or themselves use violence to achieve ideological, religious or political goals. This is why the American FBI defines terrorism as “the unlawful use of force and violence against persons or property to intimidate or coerce a government, the civilian population, or any segment thereof, in furtherance of political or social objectives.” We took these definitions into account when developing the EDT, including the definition of the United States Agency for International Development (USAID), which defines violent extremism as: “advocating, engaging in, preparing, or otherwise supporting ideologically motivated or justified violence to further social, economic or political objectives.”

To select violent extremist cases (CF_GROUP=3), the EDT instructs coders to use a list of each country if available, specifying convicted persons engaged in ideologically motivated crimes. Hate speech, propaganda, blackmailing, and sexual offences are excluded.

The EDT dataset allows for the inclusion of individuals (under research group 3-Violent Extremists) who would commonly be considered as perpetrators of hate crime – that is, relatively spontaneous violent or threatening offences directed toward another individual on the basis of their gender identity, ethnicity, religious affiliation, or sexual preference.

No, the EDT data are limited to only those individuals who displayed positive signs of violent extremism and terrorism (i.e., they adopted extreme ideologies or engaged in extremist ideologically driven behaviors). Without a control group consisting of comparable non-(violent) extremists, the data cannot be used to “predict” involvement in violent extremism and terrorism. Researchers should also avoid using the data to devise checklists of demographic, and/ or personal characteristics that may indicate that someone is radicalizing or has the potential to radicalize. Moreover, de EDT is developed to identify risk factors for violent extremist and terrorist acts. Although radicalization can result in terrorism, risk factors for radicalization may differ from risk factors for violent extremism and terrorism.

Since the EDT contains personal data and mental health information, which is collected in the judicial domain and, hence, not readily available to outside researchers, it is of great importance to protect this data as much as possible. Processing personal data and, in particular, information about individual mental and physical health places these individuals at significant risk in terms of the potential harm caused by either a data breach or the abuse or misuse of their personal information.

To protect the personal data included in the EDT, we ensured that the database both meets the security requirements for IT services set by the Dutch government and complies with the new European privacy legislation, including the General Data Protection Regulation (GDPR). The number of text fields was reduced as much as possible to minimize the identifiability of the individual person behind the data. Access to the database is strictly limited to those researchers involved in the project, and, even then, researchers only have direct insight into the data which they entered themselves. Prior to data entry, a confidentiality statement was signed by the participating organizations and researchers.

Based on the European GDPR, one of the safety measures employed in this project is the encryption of personal data by a Trusted Third Party. This means that personal data fields are not accessible or visible after having been entered. The database information can only be updated by means of a two-way encryption procedure, which allows the pseudonymized personal data to be decrypted from the database to request new (case) information about the concerned data subject if required. The NIFP project administrator periodically checks all entries for the purposes of monitoring the quality of the data based on distributions, outliers, inconsistencies, missing values and logical errors.

In order to assess the extent to which the data processing is compliant with GDPR and the EU Directive 2016/680, a Privacy Impact Assessment (PIA) was carried out before the data entry started. The document was discussed by a committee of Dutch privacy and security advisors and presented to the Data Protection Officer of the Dutch Ministry of Security and Justice. Furthermore, a risk analysis was carried out by the Dutch National Coordinator for Security and Counter Terrorism (NCTV), who concluded that the study design posed no disproportionate risks to the privacy of the subjects.

No, jihadist terrorist cases from the Netherlands and Belgium are overrepresented in the current sample.

Explanation:
The EDT includes terrorist and violent extremist offenders from the Netherlands, Belgium, Germany, Austria and Sweden, who were convicted from 2012 onward. Due to extra fundings, cases of terrorist offenders who were convicted in the Netherlands and Belgium could be entered during a longer periode. Therefore, the EDT database is not, and should not be treated as, a comprehensive set of all individuals who have committed terrorist crimes in Europe.
Besides, the target group of EDT research consists of persons who have been convicted of a terrorist or violent extremist offence. However, terrorism legislation differs between countries. For example, right-wing extremists in Germany might not be convicted under existing terrorism legislation. To include all violent offenders acting on extremist and/or ideological views and motives within each EU Member State, a separate research group of violent extremists was added based on convicted persons engaged in ideologically motivated crimes. However, it appeared difficult to identify violent extremist offenders, because criminal codes for these crimes are lacking. As a result, jihadist terrorist offenders are overrepresented in the EDT. The aim is to increase the number of violent extremist offenders in the EDT in future research.

The information in the EDT originates from qualitative judicial files from different EU Member States. It includes data from the police, public prosecution, detention, forensic mental health assessment reports, and probation reports. With relatively few exceptions, the same information sources were available in each of the participating EU Member States. However, it is important to mention that we do not have access to the Belgian police files.

Also differences exist in the criminal codes for terrorism between EU Member States. As a result, the type of terrorist crimes between different States can vary.
To clarify potential differences between the types of terrorist offenders across EU Member States, we started to compare the types of violent extremist and terrorist offenses criminal codes. In future research it would be useful to compare motives for the crime, and ideology. and potential differences between Member States in the presence of persons committing group or single offenses, because these may be accompanied with different risk factors (for example with respect to the presence of psychopathology).

At this point, 23 different coders from the different member states have entered data into the database. It is important that different coders interpret all the different variables in the same way. Therefore, interrater reliability (IRR) between the different coders is tested. The IRR is tested by coding test cases. In those test cases the degree of agreement between the score of the coder and a final score is calculated. Since the distributions of the observed ratings often fell under one category of ratings, the estimate of Cohen’s kappa turned out to be unrepresentative. Therefore an alternative kappa based on the percentage of agreement was calculated. The percentage of agreement is corrected for agreement based on chance. To interpret the percentage of agreement the classification of Landis and Koch (1977) was used: κ ≤ .20 = light, .20 < κ ≤ .40 = fair, .40 < κ ≤ .60 = moderate, .60 < κ ≤ .80 = good and .80 < κ ≤ 1.00 = excellent.

The average IRR of all coders on all variables is .75, which is good.

There are however some individual variables with low IRR. These variables should therefore be interpreted with caution:

1. PH4 Subject mentioned negative experiences with authorities in general	14. BA7a Degree of limited empathy and understanding for those outside own group
2. PH5 Subject mentioned misrecognition	15. HAC6 Has organization skills and/or acces to funding or resources to facilitate extremist violent acts
3. PH6b Specify context perceived discrimination	16. CM2 Motivated by criminal opportunism to take part in violent extremist acts
4. PH7 Subject mentioned meta-dehumanization	17. CM8 Motivated by a search for meaning or significane in life to take part in violent extremist act
5. PT15 Expressed feelings of anger	18. CM8a Extent subject is motivated by a search for meaning or significance in life
6. PCI2 Had or awaited a degrading event at work or school in year prior to crime	19. P6 Subject has support from community for relinquishing the use of extremist violence
7. PCA1 Mentioning crime before committing crime in year prior to crime	20. P7 Subject has support from family members to relinquish the use of extremist violence
8. PCA1b Type of aspects mentioned about crime	21. P7a Specify influence support from family members to relinquish use of extremist violence
9. PCA5 Subject mentioned suicide/death wish in year prior to crime	22. P9 Subject has support from others to relinquish use of extremist violence
10. PW1 Subject had a preoccupation with weapons, violence and military aspects in the years prior to crime	23. P11 Subject has external judicial control during resocialization
11. R4 Social context of radicalization	24. P12 Subject has non-violent extremist interests
12. BA2a To what extent subject expresses grievances about perceived injustice	25. GC3 Status other person(s)/ group member(s) directly involved in crime(s) subject
13. BA5a Extent subject expressed anger, moral outrage and/or hatred in response to perceived injustice	26. SC7 Subject wore combat or fighting clothes during crime

A value of -99 in the EDT dataset indicates that researchers were unable to find information in the sources for the individual and variable under review (i.e., value is unknown).

A value of -96 indicates that for a specific observation, the value is not applicable. For example, if an individual is not a known member of an extremist group, variables related to the individual’s role in an extremist group would be coded as -96. Users wishing to sum or average values in the data should be aware of these coding conventions, and take appropriate steps to recode or remove observations with these values.

Missing data is a challenge that all researchers confront, but particularly for the EDT given the nature of the data that were collected and the methods that were used to collect them. A number of variables in the EDT, particularly those representing private and sensitive information, such as mental health history and childhood family dynamics, were especially challenging to obtain. Besides, items representing difficult concepts like meta-dehumanization have a relatively large number of missings. The amount of missing data was also likely increased by our coding guidelines, which instructed researchers to be conservative and record values as missing instead of absent (i.e. as a missing code of “-99” instead of a value representing “No”) whenever sources failed to report values for most variables.

There are some variables with a high percentage of missings. 32% of the variables has more than 20% missing values. 11% of the EDT variables has more than 40% missing values. Due to the large number of missings, we may remove a number of variables from the dataset.

Users should be aware that a large percentage of missing data decreases the external validity. This means that the results of items with many missings does not perse apply to the whole European terrorist offender population. Several methods for handling missing data are detailed in Jensen, et al., Empirical Assessment of Domestic Radicalization. We recommend that users of the data familiarize themselves with these, and other, approaches to handling missing data prior to performing statistical analyses using the EDT data.

No, there are other datasets of terrorist offenders in Europe (see Bowie, 2020 [1]). When making inferences about terrorism, users should look for commonalities across multiple data sources.

[1] Bowie, N. G. (2020). A New Inventory of 30 Terrorism Databases and Datasets. Perspectives on Terrorism, 14(1), 54–66. URL: https://doi.org/10.1080/10576100802339185;

In order to distinguish between the risk factors for terrorism and risk factors for violence in general, besides convicted and deceased terrorst offenders, a control group of convicted violent offenders is included in the EDT. To compare the risk indicators in both groups, differences in background characteristics between both groups have to be reduced. Therefore, the control group is matched. This means that we equate (or “balance”) the distribution of covariates in the research group and the control group. We match the groups on the following variables:

1. Gender (male/female)
2. Presence of forensic mental health assessment (Yes/No) 
3. Most severe type of crime
4. Age

The variables gender, type of crime and age are chosen since these factors may influence the persons risk for (extremist) violence. The matching on FMH assessment is important since an unequal distribution of available reports among the groups, may have impact on the chance to find any psychopathology.

The matched control group is entered in Belgium and the Netherlands.

It is important to be aware that due to a large number of missings, some variables are not comparable between the research group and control group.

Yes. There are several questions of interest that the EDT data are not designed to answer.

• First, as noted above, the EDT data may not be representative of characteristics and risk and protective factors of terrorist offenders for the period before 2012. tTherefore, users should be cautious when using the data to assess aggregate trends over time.
• Second, the EDT data cannot be used to predict who will radicalize and will be involved in violent extremist and terrorist acts, because of the low base rate of terrorism and because EDT data only allows for the identification of associations and not for causal designs. 
• Third, users should avoid drawing conclusions from the present or absent rates of individual variables in the EDT data without also assessing how often those variables are present or absent in the general population. For example, users who are interested in the immigration rates of the individuals in the EDT data or the prevalence of psychopathology should not draw conclusions about these prevalence rates without first considering prevalence rates among the general population (or among criminal offenders).

The EDT data can be used to explore a number of important aspects of violent extremism and terrorism in the included European Member States. These include comparisons of ideological and sub-ideological groups, group-based and lone actors, and violent and non-violent extremists. The data can also be filtered by exposure date, age, gender, location, ideology, group, and more to address specific aspects of terrorism in European States. Users should be aware, however, that filtering the data reduces the number of valid cases for exploration and may render statistical tests ineffective. Furthermore, the EDT allows for research into the following topics:

• Identification of Critical Risk and Protective Factors for Terrorism:

When enough control group cases of violent offenders have been entered into the EDT to enable statistical analyses, it will be possible to examine whether the assumed risk and protective factors for terrorism are able to distinguish between extremist violence and ordinary violence.

• Terrorism and Role of Psychopathology:

The EDT dataset can be used to examine psychopathology as potential risk factor for engagement in terrorism, considering contextual risk factors. This could not only involve research into mental disorders, but also the diagnosis of underlying traits or symptoms that lack a specific mental disorder diagnosis. It is important to take these traits and symptoms into account, also because a differentiated psychopathological approach is considered best practice for forensic experts.

• Group Comparisons:

Comparisons between different terrorist offender groups can be analysed. Offender groups in the EDT differ regarding their ideology (i.e., left, right, Islamist, ethno-nationalist, idiosyncratic individual causes), membership to groups (lone actor versus group member; leaders versus followers), age or gender. Analysis can focus on the comparison of risk and protective factors for engagement in terrorist and violent extremist offences between these different offender groups.

• Risk Patterns:

Interactions between risk factors can be examined, which can lead to the identification of ‘risk patterns’: clusters of risk factors, related to subgroups of the terrorist offender group. Another distinction in terms of risk patterns can be made with respect to the type of terrorist crime committed and the role of the offender in the terrorist group. One important research direction will be to distinguish between the various risk patterns that may be related to different types of terrorist crimes and to distinguish significant from non-significant factors on an empirical basis.

• Terrorism Pathways:

The EDT dataset allows for the examination of global pathways of risk factors related to childhood circumstances, the period of radicalization, as well as the year prior to the criminal act. This allows us to gain better insights into the importance of different biographical and socialization risk factors at different periods of a person’s lifecycle.

• Increasing the Evidence-base of Risk Assessments for Terrorism:

The EDT dataset can be used to increase the evidence-base of terrorism and violent extremism risk assessment instruments like the Violent Extremist Risk Assessment instrument (VERA-2R). This tool is also used in Belgian prisons and Flemish justice houses to estimate the risk of recidivism.

Research to evaluate and validate the risk factors for violent extremism is needed to improve the ability of the VERA-2R to assess and manage the risks for violent extremism and to make risk assessments more accurate for specific subgroups.

• Effectiveness Judicial Interventions for Terrorist Offenders:

The design of the EDT allows for longitudinal research to enable monitoring recidivism and disengagement. This can clarify effectiveness of interventions targeted at the terrorist offender group. This is highly relevant to the Ministries of Justice, antiterrorism agencies, and secret services of participating Member States to develop standardized approaches for violent extremism risk assessment with evidence-based risk prioritization, risk targeting, and risk-based rehabilitation of terrorist offenders.

After completing the necessary data quality checks, an anonymized dataset comprising a selection of items can be made available for the purpose of answering research or policy questions, publication, replication or validation. To comply with the privacy regulations, data requests will be assessed based on their societal, scientific, or policy relevance. Only a selection of the EDT dataset will be shared, depending on the specific research goals and subsequent approval of the EDT project board.

In the case of data transfer, the receiving party must complete and sign a data sharing form, comprising an overview of the purpose of the research and intended usage of the received data, alongside security safeguards.

All conditions for data-sharing are described in the: Working Procedure Data Sharing EDT

Yes. Please use the following citation if using the EDT data in your own research/publication:European Database of Terrorist offenders (EDT) – extraction date: [dd-mm-yyyy]. Alberda, D., Duits, N., van den Bos, K., Ayanian, A. H., Zick, A., and Kempes, M. (2021). The European database of terrorist offenders (EDT): development usability and options. Perspect. Terror. 15, 77–99.

Inquiries about the EDT data and/or related research, can be sent to: Europa@dji.minjus.nl