ADR UK Research Fellows: The first users of the Data First magistrates’ and Crown Court datasets

ADR UK is funding four Research Fellows for seven to 12 months to conduct analysis using the Data First magistrates’ or Crown Court linked datasets, or a linking dataset enabling analysis of relationships between the two. Both datasets contain case and defendant level data on criminal court use between 2011 and 2020.

They form the first cohort of Data First Fellows and the first funded users of the de-identified, research-ready datasets made available via the Data First programme. Data First is a ground-breaking data linkage programme led by the Ministry of Justice (MoJ) and funded by ADR UK to link and enable access to administrative data from across the justice system and beyond for research. 

Find out more about the Research Fellows and their projects below.

Dr Kitty Lymperopoulou

Dr Kitty Lymperopoulou, Senior Research Associate at Manchester Metropolitan University, will draw on both magistrates’ and Crown Court datasets to generate new evidence on the extent and drivers of identified ethnic inequalitie in the Criminal Justice System (CJS) in England and Wales. 

Taking advantage of the unique features of the Data First datasets, the project will examine the experiences and outcomes of minority ethnic groups in the court system; the relative importance of defendant, case, and court factors in explaining any ethnic differentials at different stages of the CJS; and make recommendations about ways of effectively addressing ethnic inequalities.

View project details

This project aims to explore the following research questions:

  • How do the experiences and outcomes of defendants in the magistrates’ court and Crown Court differ by ethnicity? 
  • Do ethnic disparities persist after controlling for defendant, case and court characteristics? 
  • What are the main drivers of any differences by ethnicity in court outcomes, and how much of the gap is attributed to defendant, case and court factors? 

The methodology used in this study:

  • Disproportionality analysis using the Relative Rate Index (RRI) will document the extent of disparities in court experiences and outcomes between white and minority ethnic groups.
  • Multilevel modelling will be used to examine the effect of offender, case and court characteristics on court outcomes, and to determine whether ethnic disparities persist after controlling for these characteristics.
  • Blinder–Oaxaca decomposition analysis will be used to decompose the ethnic gap into a compositional effect, due to group differences in the magnitude of offender, case and contextual characteristics, and a structural effect due to group differences in the estimated effects of these characteristics.

Duration: April 2021 – March 2022

Funding: £103,628.68


Dr Angela Sorsby

An investigation into racial bias in court case outcomes in England and Wales

Dr Angela Sorsby, Lecturer in Criminology at the University of Sheffield is primarily focusing on increasing understanding of disparities between ethnic groups within the CJS and whether any disparities are the same or different for men and women.

View project details

This project aims to explore the following research questions:

  • Are there differences between ethnic groups in:
    • the outcomes of cases (e.g. conviction/acquittal, severity of sentence) after controlling for other factors (e.g. age, offence type)?
    • the number of times individuals are prosecuted over a specified time period?
    • the proportion of convictions, compared to other outcomes such as acquittals, that a defendant receives in that time period?
    • the number of cases in which a defendant appears in the magistrates’ court prior to appearing in a Crown Court case?
  • Are differences between ethnic groups the same or different for men and women?
  • Are there differences between the magistrates’ and Crown Court in terms of outcomes for cases which can be tried in either court (triable either way), taking account of other factors such as offence type?
    • What is the effect of this if Black, Asian and minority ethnic defendants are more likely to elect to be tried in the Crown Court?

The methodology used in this study:

Regression analysis enables the relationship between two variables to be assessed while controlling for other variables in the analysis.

Regression analysis will be used to establish whether:

  • differences between ethnic groups in court case outcomes (whether the defendant is convicted and if convicted the severity of the sentence) remain after taking account of things such as age and offence type;
  • there are differences between the magistrates’ or Crown Court in the outcome of cases, taking account of other information available in the datasets.

Some people will appear in the dataset on multiple occasions. Aggregate data will be produced for individuals collating: the number of times each individual is prosecuted over a specified time period; the proportion of prosecutions that result in convictions; and the number of cases in which a defendant appears in the magistrates’ court prior to appearing in a Crown Court case.

Regression analysis will then be used to examine the relationship between these variables and ethnicity, taking account of other information in the dataset, such as the most serious offence over the time period.

In conducting the analyses, the interaction between gender and ethnicity will be examined to establish whether the findings are the same or different for men and women.

Duration: April 2021 – March 2022

Funding: £43,407.96


Dr Tim McSweeney

Understanding the nature, extent and outcomes of serious and organised crime cases heard before the Crown Courts in England and Wales

Dr Tim McSweeney, Senior Lecturer at the University of Hertfordshire, aims to better understand the nature, extent and outcomes of serious and organised crime (SOC) cases heard before the Crown Courts in England and Wales between 2013-2020. 

View project details

This project aims to explore the following research questions:

  • What is the nature and extent of SOC heard before the Crown Courts in England and Wales?
  • How much cumulative harm do these SOC cases account for in the higher courts’ caseload?
  • Is this harm equally distributed across (i) offence types, (ii) the different groups involved in SOC, and (iii) different locations?
  • Is there an association between cases involving SOC and the likelihood of court proceedings being discontinued or dismissed?
  • Among SOC cases, which factors (linked to defendant characteristics, group size, main offence, and location) were predictive of Crown Court proceedings being discontinued or dismissed?
  • Is there an association between involvement in SOC and repeat appearances before the courts?

The methodology used in this study:

  • Descriptive statistics will be used to report on the prevalence and incidence of SOC offending across the Crown Court caseload between 2013 and 2020 (Q1), and to describe the cumulative harm attributable to SOC cases during this period (Q2).
  • Analysis of variance (e.g. ANOVA) will test whether levels of harm varied by location, across offence types, or for different groups involved in SOC (Q3).
  • Chi-square tests will assess the extent to which proceedings were discontinued or dismissed (Q4), and rates of repeat use of the courts (Q6), differed between SOC and non-SOC cases. 
  • The project will use multiple logistic regression to determine whether a range of factors relating to demographics (for example, age, gender, ethnicity), SOC group size, main offence, and location were related to the probability of court proceedings being discontinued or dismissed for SOC defendants (Q5).
  • When examining the rate and frequency of court reappearances (Q6), the project will test the feasibility of matching SOC and non-SOC defendants on relevant demographic and offence variables using propensity score matching. This will allow the use survival or time-to-event analyses to then measure and quantify the impact of different factors (relating to demographics, main offence, location and SOC status) on the risk of a court reappearance within a defined follow-up period (for example, two years following release from custody).

Duration: April 2021 – October 2021

Funding: £48,606.35


Dr Becky Pattinson

Using linked magistrates’ and Crown Court data to explore defendant appearance(s) over time: specialisation, escalation or de-escalation?

Dr Becky Pattinson, Senior Research Associate at Lancaster University, will research the risk of repeat use of the magistrates’ and/or Crown Court to determine the scale and pattern of this risk in the time following the first experience of the magistrates’ or Crown Court, or index case. The project will utilise the data linkage between the magistrates’ and Crown Courts, which has the unique potential to match a case from the magistrates’ to the Crown Court.

View project details

This project aims to explore the following research questions:

  • What is the risk that a defendant will return to court?
  • How does this risk change in the time after the first observed court case (index)?
  • Is the risk of return for the same offence the same as for a different offence?
  • What are the risks of criminal careers such specialisation and escalation/de-escalation in offence type and/or seriousness of the offence(s)?
  • What influence did characteristics of the defendant and the index case have on the risks?

The methodology used in this study:

Following a general overview of the scale and pattern of defendants in repeat use of the courts, the project will apply methods of recurrent event analysis, including the Kaplan-Meier estimator and analysis of competing risks. These involve examining the time after defendants are observed in a first (index) case until the event of a second case; thereby entering a state of repeated use of the courts. Following analysis of the overall risk of repeated use, the project will focus on the different forms in which a second criminal case could occur, specifically any change in offence type or seriousness, in competing risks analysis.

Meanwhile, the effect that characteristics of the defendant and the index case have on the risk of repeat use will be analysed.

Duration: April 2021 – March 2022

Funding: £91,137.64

Categories: Research using linked data, ADR UK Research Fellows, ADR England, Office for National Statistics, Crime & Justice

Share this:

You are currently offline. Some pages or content may fail to load.