This UI blog looks at how recording practices in the Scottish criminal justice system can impact on the quality of statistics produced by public agencies and the implications of this for research on inequalities. Tapping into a wider debate on sex and gender-identity, recently intensified by the passage of the Census Amendment (Scotland) Bill through the Scottish Parliament, I argue that the introduction of recording practices based on gender self-identification, rather than biological sex, creates a lack of clarity over what is being measured by criminal justice agencies and what some types of official data represent. This has a profound relevance to our understanding of crime and justice inequalities in the criminal justice system.
Below, I argue that clear recording standards and principles are needed which recognise that sex and gender identity are separate concepts. I also argue that data on biological sex should not be replaced with data on gender identity, given the profound relevance of the former to people’s lives. This is particularly important for data on crime and justice because there are specific sex-related factors that impact on people’s experience of victimisation and offending, and on how cases are processed and recorded by the criminal justice system. Instead, data on both characteristics should be collected separately to better understand how these function independently and together, and how they impact on people’s experience of inequality.
Statistics are an essential part of government policy-making. They provide those involved in delivering public services and tackling social problems with information to develop and monitor policies and decisions, and to allocate resources. Statistics are also essential for research purposes to better understand how social and economic factors impact on people’s lives. The Understanding Inequalities research programme makes use of a wide range of data sources to explore how inequalities are created, perpetuated and reduced within Scottish society. One key area of our research programme concerns crime and justice inequalities, where we draw on recorded crime data, population survey data and convictions data from official government sources, as well as population records from the National Records of Scotland.
Within the field of criminological research, official crime data (such as recorded crime and victim surveys) enables stakeholders, policy-makers and researchers to analyse trends and patterns in both victimisation and offending, which in turn better our understanding of what can be done to reduce crime. This allows interventions and services to be developed, and resources to be targeted, more effectively and efficiently.
A good example of how recorded crime data is used can be found in the work of the Understanding Inequalities team. Here we are examining the relationship between recorded crime and non-crime incidents, controlling for area deprivation, and exploring the issue of hidden inequalities in exposure to crime. While recorded crime can only provide a partial picture of offending behaviours, these data can nonetheless offer valuable insights into those populations at greatest risk of crime; for example, in relation to age, sex or geography, which can be analysed with other data sources.
To ensure that research using criminal justice data is accurate and credible, it is essential that published statistics are valid, reliable and consistent, and that the data criteria of what is being measured are clearly defined. However, for the purposes of collecting data on a person’s sex, this no longer appears to be the case. Within the criminal justice system, as well as other policy fields, it appears that agencies have replaced some types of data based on sex, with data based on gender identity or self-declared sex. For example, Police Scotland states that it records incidents based on self-declared identity, unless evidentially relevant to the crime. The Cabinet Secretary for Justice has also stated that the Scottish Courts and Tribunal Service record by ‘self-declared gender’, rather than biological or legal sex.
Recording based on gender identity alone is likely to have significant implications for the accuracy and reliability of criminal justice statistics, given the sharp differences in both victimisation and offending patterns by sex. For example, the inclusion of a very small number of natal males who identify as female could easily skew female sexual offending statistics, or any other category for which the prevalence of female offending is particularly low. Criminal Proceedings in Scotland data show that only three females were convicted of sexual assault in 2017/18, compared to 299 males, while 82 females were convicted of attempted murder and serious assault, compared to 1,086 males. Yet whether these data accurately reflect the sex of the convicted person is no longer clear. Nor is it possible to determine whether any variation in female offending over time is due to the inclusion of self-identification cases in some years, but not in others. In addition, any police data about the victims of crime based on self-declared sex would not enable us to determine whether there was inequality in the extent to which certain groups (based on either sex or gender identity) were at greater risk of crime than others.
Data that conflates sex and gender identity is also not consistent with the legal definition of sex in the Equality Act 2010, which is biological (although having a Gender Recognition Certificate impacts on a person’s legal status).
It is not clear why or when criminal justice agencies introduced a policy of recording based on self-declared sex only, or whether due diligence was carried out in relation to the impact on data quality. For example, there appear to be no credible arguments to support the assumption that robust data on sex, whether legal or biological, is no longer needed. Nonetheless, the unregulated introduction of these practices by public authorities, without proper democratic debate, means that it is no longer clear what is being measured in relation to sex and gender identity. This could potentially have a profound effect on any research on inequalities that focuses on the criminal justice system.
In terms of next steps, it is recommended that public authorities collect data on both sex and gender identity separately. The more data we gather on both characteristics, the better placed we are to understand how these operate independently and together, and how they impact on people’s experience of inequality. Given the current confusion, it is also suggested that the Scottish Government review how all public authorities and agencies record sex and/or self-declared gender identity, and set out clear data recording criteria and principles.
Image c/o Istock/Matjaz Slanic