The Administrative Data Research Network was an ESRC-funded project that ran from October 2013 to July 2018. It is currently at the end of its funding cycle and is no longer taking applications. Administrative data research will be taken forward in a new project, which was launched at the end of 2018.

Visit the Administrative Data Research Partnership for further information. 

This archival website reflects the state of play at the end of the project in July 2018. All content has been frozen and may not be up to date.

open menu

It’s a criminal waste: How using administrative data about crime could better inform public policy

By Professor Susan McVie, ADRC-S Scotland Co-Investigator, Chair of Quantitative Criminology, School of Law, University of Edinburgh


The idea of collecting data and measuring crime in a systematic way is not new.  As far back as the 1830s, counting crime became prominent in France where it was promoted by so-called ‘moral statisticians’ as part of their mission to apply scientific principles to the study of the social world.  This idea of measuring bad behaviour fitted well with the aims and practices of centralised bureaucracies that were being established all across Europe at that time in an effort to bring more elements of public life under central government control.  As a result, the collection and analysis of crime data soon became – and still is - part of the standard work of government statisticians within the UK and many other countries.   

In Scotland, a variety of organisations that are collectively referred to as the ‘criminal justice system’ collect routine administrative data about crime and justice, which is then published centrally by the Scottish Government.  A phenomenal amount of information is made available through the government’s Crime and Justice Statistics web pages, from high level statistical trends to detailed datasets.  These data provide an excellent overview of the number of crimes recorded by the police, the number of people prosecuted and then convicted in our courts and the number of people that serve time in our prisons, to give but a few examples.  These data are extremely valuable, but they have limitations in terms of informing public policy making.  Why?  Because they largely provide a measure of the activity of the organisations themselves.  In other words, we see crime through the lens of the criminal justice system, but we do not necessarily see the ‘big picture’ of crime.

Data held by the criminal justice system could, in theory, be linked to many other datasets.  For example, it could be linked to information held on offending by a variety of other agencies, such as the Children’s Hearings System, Social Work Departments and Third Sector organisations.  It could also be linked to rich social survey data, such as school based studies, national victimisation surveys, and local studies of offending.  And it could be linked to other forms of administrative data, such as information held on health, education, family life, employment and mortality.  The potential for linking such datasets together to allow us to see the ‘big picture’ exists, but it is limited by a range of factors.  We do not have unique identifiers that enable us to link data routinely and we do not generally seek people’s permission to link their data in any case.  Compared to other jurisdictions (such as the Nordic countries), we are lagging behind in our research capacity and our ability to make joined up public policy decision making.

Let’s consider some examples.  Influential criminologists have suggested that parenthood and employment are important factors that lead to desistance from offending.  However, analysis of linked administrative data in Norway found that while parenthood had a preventive effect on offending amongst adults, teenage parents are at higher risk of offending.  Furthermore, Norwegian researchers found that most people in employment who desisted from offending had done so prior to starting employment, rather than as a result of getting a job.  In other words, policy interventions encouraging more young people to have children or more offenders to get jobs may not have had the intended consequences!  In Sweden, a study linking crime data and hospital records found a high level of mental illness and hospitalisation amongst those who went on to commit sexual offences, which was of significant benefit in developing an ‘early warning’ system.   And an Australian study linking death records with youth justice data found that young people sentenced to custody were at significantly increased risk of early preventable death. 

Such studies highlight the significant benefits of linking crime-related data to other forms of information about individual lifestyles and circumstances.  However, in Scotland there are only a small number of studies that have benefited from this type of administrative data linkage.  Most notably, the Edinburgh Study of Youth Transitions and Crime collected data on a cohort of around 4,300 young people over a period of almost two decades, including survey data and official records.  By linking together various sources of data, this study learned four key lessons:

  • Young people who get involved in serious offending in their teenage years are amongst the most vulnerable, deprived and socially marginalised groups in society.
  • Trying to identify the children who are ‘at-risk’ of getting involved in serious offending in their early years is not an exact science and runs the risk of labelling and stigmatizing individuals, resulting in worse outcomes for these children
  • Rather than focusing on the early years, there are critical transition points in the teenage years that hold the key to leading young people away from crime
  • Managing these transition points successfully involves keeping young people out of formal systems of justice, as this helps young people to desist from offending

The results of the Edinburgh Study have had a profound effect on influencing legislative change and strategic policy development around youth justice in Scotland.  For example, it influenced the development of the Getting it Right for Every Child (GIRFEC) model of youth welfare, which became national policy in 2010.  It formed a critical part of the evidence base for the development and national implementation of the Whole Systems Approach to Youth Justice in 2011.  And it has been extensively referred to in the debates about the increase in the age of criminal responsibility in Scotland, which has yet to be decided upon.

There are many other areas of public policy in Scotland where the linkage of administrative data on crime could inform better policy development, evaluate effectiveness and save money.  For example, will the Scottish Government’s commitment to reducing the education attainment gap have any impact on crime rates in Scotland? Will increasing the minimum unit of alcohol price result in an increase in offending amongst some people from low income households?  And what crime reduction interventions have the greatest benefit and save the Scottish public the most money?

The Administrative Data Research Centre in Scotland crime and justice research strand is keen to hear from anyone with ideas about these types of data linkage.  The potential for identifying and linking administrative datasets exists; however, it needs the will and interest of the research community to bring such projects to fruition.  I encourage researchers with an interest in this area to bring project ideas to our attention.  Otherwise, the lack of research in this area really does represent a criminal waste of perfectly good data!

Written by Susan McVie from ADRC-Scotland and published on the ADRN blog under Creative Commons license CC BY-NC-SA 4.0

Published on 15 November 2016

Page last updated: 27/07/2018