Researching children and terrorism: the pros and woes of building an original dataset

NCRM news
Dr Gina Vale, University of Southampton
A boy sitting on concrete stairs looking at a mobile phone (photo by Gaelle Marcel on unsplash)A boy sitting on concrete stairs looking at a mobile phone (photo by Gaelle Marcel on unsplash)

In September 2021, Cressida Dick, then-Metropolitan Police Commissioner, warned of children forming a "new generation of extremists" in the United Kingdom. Her comment came in response to Home Office data that revealed the highest number of arrests of minors for terror offences since records began – a number that has since further increased. However, research on this emerging trend and "new" terrorist actor has been stymied by lack of case information behind these abstract statistics.

Over the past two years, I have worked with Hannah Rose – Hate and Extremism Analyst at the Institute for Strategic Dialogue – to track, compile and analyse these cases. In our new Childhood Innocence Project, we present the first in-depth research into criminal justice responses to child terrorist activity through construction of a unique dataset of children convicted of terrorism offences in England and Wales since 2016.

In addition to the resulting analysis, our goal was also methodological: to create and publish the first disaggregated dataset that can enable better understanding of youth-specific terrorism offending and its response. However, we quickly discovered why no one before us had embarked upon this task.

Data blockages and blockades

The field of terrorism studies has always faced data availability issues. Those involved in extremist violence are not usually the most accessible or cooperative research participants. In order to track and understand motivations behind terrorist activity, scholars and analysts often turn to documentary sources. These can take the form of propaganda or communications produced by the individual or associated group or network, or – relevant for our study – legal case records for those processed through the criminal justice system of their respective national jurisdiction. In England and Wales, access to court transcripts requires a lengthy and financially expensive application process, creating a significant barrier for research. Fortunately, for high-profile or significant cases, the judge’s sentencing remarks are often published in the public interest, or detailed quotations from media coverage can be pieced together to provide a partial transcript to begin building a case profile for analysis.

However, adding juvenile offenders into the mix complicates matters further. Whereas adult terrorism trials are almost exclusively held in a crown court, where hearings are logged for transcription, for those tried and sentenced as minors – before their 18th birthday – cases are often heard in a magistrates’ court and never recorded. Add this to the fact that juvenile offenders are mostly anonymised by automatic reporting restrictions, and the result is great difficulty in identifying and distinguishing individual cases in the first instance, let alone accessing official, reliable records. Our study therefore required a painstaking process of open-source data searches and triangulation.


Beginning in February 2022, Hannah and I drew from multiple sources to compile our dataset – the Crown Prosecution Service, police services, legal resource repositories and fixed-term Google searches. Our focus on child terrorism offenders required us to impose specific parameters – inclusion and exclusion criteria – for our data, each presenting its own obstacles.

First, in order to study children’s involvement in terrorist activity as children, cases were only included if the offender was aged under 18 years at the time of their (first, if multiple) offence. For cases that were clearly publicised as involving juvenile offenders, we cross-referenced multiple data points – such as location, age, indictment period and sentence – in order to separate anonymous teenagers with otherwise similar offending profiles. For other cases, news coverage prioritised reporting of age at conviction or sentence, requiring us to calculate backwards the age at offence.

Second, we chose to only include "successful" convictions of a substantive offence under the UK Terrorism Act 2000 and subsequent terrorism legislation. This, therefore, excluded cases of acquittal or dismissal, as well as convictions for "terrorism-relevant"  offences, such as explosives or hate crimes. Theoretically straightforward, but in reality, public reporting of the title of offence was inconsistent. This highlighted the importance of police reporting and press releases to accurately communicate terrorism cases for reliable comparative analysis.

Third, whilst we ideally would have analysed cases since 2000 in line with the introduction of the above legislation, repositories and publications hosting more historic cases proved either inconsistent or incomplete. Eventually, we chose to collect convictions from 2016 onwards in line with the point at which the Crown Prosecution Service began to publish data on terrorism convictions.

These parameters were, at times, frustrating. We are fully aware of remarkably interesting cases of teenagers’ involvement in terrorism that needed to be excluded. For example, several high-profile convictions preceded our date range and, in one notable case, the offender was arrested shortly after their 18th birthday but public reporting cannot (yet) verify that the offending initiated during legal childhood.

However, despite these limitations, the resulting dataset contains 43 cases of convicted child terrorism offenders since 2016. We estimate that we have collected the majority of relevant cases recorded by the Home Office during the period of study. Ours is the first dataset to include detailed information of each case: child’s location and gender; ideological affiliation; offences, pleas and sentences; connections with co-defendants. From this, our analytical report, titled Childhood Innocence?: Mapping Trends in Teenage Terrorism Offenders, identifies and unpicks trends in teenagers’ offence types, warning signs of compounding criminality and potential recidivism, sentencing patterns, and inconsistencies in rehabilitative opportunities.

Only the beginning…

The current wave of child engagement in terrorism in England and Wales shows no sign of slowing down. Therefore, Hannah and I have taken the decision to continue to update the dataset as and when new cases are confirmed. It is kindly hosted as a live resource by the International Centre for the Study of Radicalisation (ICSR), our project publisher.

It is also important at this point to mention research ethics and data security issues. Hannah and I applied for ethics approval at both the University of Southampton (my institution) and King’s College London (base of the ICSR and the institution where Hannah is a current PhD candidate). Given the scale and sensitivity – concerning minors and illicit activity – of the data, the process of receiving ethics approval and completion of a Data Protection Impact Assessment from both institutions lasted more than five months. Despite using only open-source data, we took the decision to anonymise all individuals in our dataset and analysis, irrespective of available public reporting. Our reason was two-fold: to protect the children themselves, and to protect ourselves. Each datapoint is footnoted for data transparency and to fulfil the goal of our dataset: to render these cases accessible for further study. However, with increasing discussions of researcher safety and wellbeing in mind, we felt that blanket anonymisation would provide a degree of separation between us as researchers and our data subjects.

Our hope is that through our analysis, and the public availability of the live dataset, the Childhood Innocence Project will enable researchers, practitioners and policymakers to better understand and respond to the roles played by young people in extremist ecosystems and their radicalisation pathways. In an era of rising extremist youth activism, we urge stakeholders to support data availability in order to inform better response planning by counter-terrorism scholars and practitioners.

About the author

Dr Gina Vale is a Lecturer of Criminology at the Department of Sociology, Social Policy and Criminology at the University of Southampton. She is also an Associate Fellow of the International Centre for Counter-Terrorism (ICCT) and a member of the Vox-Pol Network. Gina’s research takes a feminist, intersectional approach to studying terrorism and extremist violence. She primarily focuses on terrorist governance, genocide, extremist propaganda, and the organisational roles and experiences of women and children. Email: Twitter/X: @GinaAVale.