Diversity Monitoring in the Library

My article Diversity Monitoring in the Library: Categorisation Practices and the Exclusion of LGBTQ Library Users is now available to read (open access) in The International Journal of Information, Diversity and Inclusion.


The collection of data about the identity characteristics of library users is the latest development in a long history of contested categorisation practices. In this article, I highlight how the collection of data about lesbian, gay, bisexual, trans and queer (LGBTQ) people has implications for the undertaking of diversity monitoring exercises in academic and public libraries. Based on experiences in the United Kingdom, I argue that recuperative efforts to ‘fix’ categorisation practices are not enough and overlook how categories of gender, sex and sexuality are constructed through the practice of diversity monitoring, how categories are positioned in time and space, and who is involved in decision-making about who to include and exclude from the category of ‘LGBTQ’.

Queer data, evidence and inaction

Paper presented at the University and College Union LGBT+ Liberation Conference on 3 November 2021.

Much of what I’m going to cover today is based on arguments in my book Queer Data: Using Gender, Sex and Sexuality Data for Action that explore the role of evidence in documenting LGBTQ lives.

Prior to joining the University of Glasgow, I worked as Head of Knowledge and Research for the higher education organisation Advance HE – leading on some of their outputs on equality, diversity and inclusion. So I hope that what I present today straddles theory and practice, academic and professional services, and gives us all some radical prompts on the use of data for action.

When writing my paper I kept returning to the metaphor of pushing on a closed door. Sara Ahmed (2021) uses the metaphor of doors in her recent writing on complaint and diversity work, I intend to continue this metaphor and question the role that data plays in providing an evidence base to show something is wrong.

Whatever the weight of force trying to push open the door, when it is locked it will only budge with a key. To what extent can force – or the weight of evidence – present an alternative solution to the use of a key? The metaphor of pushing on a closed door has implications for those who are LGBTQ as well as work about LGBTQ lives – how is knowledge produced, what types of knowledge and what purposes does it serve?

Black background with white text that reads 'Diversity management does not equal LGBTQ liberation'.

Crucially, my critical reading of data pushes back against timid, conservative and often harmful approaches to ‘diversity management’ that have proliferated in colleges and universities since at least the introduction of the 2010 Equality Act. Here energy focuses not on changing exclusionary structures but on making sense of a ‘problem’ – a term I use to describe something understood by those in power as unwelcome that needs to be dealt with.

To put it simply – what I want to call into question is the assumption that more data (or more detailed data) improves the lives of LGBTQ people.

What do we know?

One of the most common uses of data is to describe or make sense of a problem. Problems might include pay and benefits, precarious employment, lack of inclusion in the curriculum, mental health, underrepresentation in particular fields or roles, or bullying and harassment.

Data collection activities including those associated with HESA, UCU, UCAS, Athena SWAN and – on a national level – the census are key players in the gathering of data to make sense of problems. Yet, how problems are defined, what is brought into view and what problems are understood as ‘worth solving’ are not natural features of the sector but are the product of the interests and biases of those already in positions of power.

Thinking about the relationship between data and problems invites two questions:

  • What labour is required to undertake this work?
  • What stories are being told with this data?

I’m going to address these two questions in turn.

In terms of labour, a need to know the scale of a problem before work can take place to address the problem poses the question: who is expected to undertake this groundwork and at what cost?

To take one example, recent years have witnessed a stream of consultations in Scotland and the UK on reform of the Gender Recognition Act. The UK Government consulted on the issue in 2018, alongside separate consultations by the Scottish Government in 2017 and 2020, and Select Committee investigations in 2016 and 2021. The Scottish Government plans to introduce a reform Bill in February 2022.

In the most recent UK Government consultation, although over 100,000 individuals and organisations responded, with the majority backing meaningful reform, the government decided not to act on this result.

For LGBTQ people, why are we expected to collect, analyse and present data as a prerequisite for change?

In this example, do we need more data about the ‘problem’? And, even though a vast amount of data has already been collected, what (if anything) is the impetus for those in positions of power to actually do something?

Writing about data on racial injustice in the US, sociologist Ruha Benjamin (2019) argues that the problem is not ‘a lack of knowledge’ and that ‘demanding data on subjects that we already know much about is a perversion of knowledge’. Minoritised groups are invited to accumulate yet more data about injustices, burning through finite resources, to provide further proof of the problem. But what if this time, money and energy were used in other ways? This is therefore a question of labour. What if this labour was spent on the development of mental health services, the provision of community-level legal support, or monetary help with overdue bills and outstanding rent payments?

Gathering more data on the problem, when enough evidence already exists, can therefore contribute to the development of damage-centred or deficit narratives about the group under investigation. As observed by Catherine D’Ignazio & Lauren F Klein (2020), data accumulated tends to focus on ‘problems’ as removed from a more rounded account of people’s strengths, creativity and agency. This accumulation of data can maintain the status quo and, for D’Ignazio & Klein (2020), ‘minoritized individuals and groups should not have to repeatedly prove that their experiences of oppression are real’.

Intended audiences

These reflections beg the question: to whom are minoritised individuals and groups expected to present this information? For LGBTQ people, why are we expected to collect, analyse and present data as a prerequisite for change? As I explore in my book Queer Data, information about LGBTQ lives is often intended to change the hearts and minds of a cis, heterosexual audience. However, this cis, heterosexual audience may or may not take action, depending on the quality of the proof provided. The presentation of data therefore functions as a means to convince others to do something.

As you might imagine, any reliance on already powerful individuals and organisations to take action is problematic. Principle 21 of the Feminist Data Manifest-No (2019) highlights how projects intended to address inequity cannot rely on the goodwill of institutions to fix problems. The Manifest-No rightly highlights a missing step between the production of knowledge about the ‘problem’ and the rationale for responding to the findings presented.

This framing of data shifts the onus from the quality of data collected to who or what is expected to listen and take action.

Rather than a force to produce something new – whether good or bad – data also operates as a jam in the system that perpetuates the status quo and protects current ways of working.

We see this gap between knowledge about a ‘problem’ and the rationale for taking action when we consider practices of complaint. Like many others here, I am angry at the upsurge in anti-trans coverage across many media outlets and the anti-trans institutional responses from several UK universities. While sitting down to draft my strongly worded email to the BBC – I thought to myself, who is this email intended to reach and to what extent is the infrastructure that exists for me to express my anger actually designed to manage and placate my concerns.

It reminds us that even with the most robust, inclusive and comprehensive dataset – or well written complaint – we still need to counter the challenge that ‘good data’ only enables us to push harder on a closed door.


Reliance on the goodwill of powerful individuals and organisations to ‘do something’  is an unsustainable foundation for LGBTQ liberation. Efforts to amass more and more data, with the hope that the brute weight of evidence will push open the door, requires yet more and more labour from those already minoritized and disadvantaged by existing structures. More insidiously, it also requires minoritised groups to gather further data about themselves as being a ‘problem’, which can solidify damage-centred or deficit narratives.

Rather than a force to produce something new – whether good or bad – data also operates as a jam in the system that perpetuates the status quo and protects current ways of working.

Data is central to LGBTQ liberation but its use for action still requires evidence to fall upon the ears of individuals who wish to listen and take action in response.

What we need is a paradigm shift where we dedicate less energy to the production of knowledge about problems and more energy to the production of knowledge about solutions. I offer no guarantee that a focus on solutions will solve the complex challenges we see in colleges and universities. But it gives us hope that we might find the key and no longer need to push on closed doors.

Inclusive data is not enough

The UK Statistics Agency’s Inclusive Data Taskforce this week published their report Leaving no one behind – How can we be more inclusive in our data?

There’s a lot of good in the report and the eight detailed recommendations presented, providing high-level guidance and direction to those engaged in strategic decisions about data in the UK.

However – in terms of specific recommendations about gender, sex and sexuality data – the Taskforce’s focus on ‘inclusivity’ scratches at the surface of the problem and fails to meaningfully address challenges produced by power imbalances and data injustice.

For example, Recommendation 3.4 notes:

Sex, age and ethnic group should be routinely collected and reported in all administrative data and in-service process data, including statistics collected within health and care settings and by police, courts and prisons.

The thrust of the recommendation is fine, though how ‘sex’ and ‘ethnic group’ are conceptualised and counted will likely differ across data collection exercises.

These differences aren’t a problem, per se, as it’s important that data collectors ask questions that return data they require to answer their research questions. However, a failure to acknowledge the contextual and nuanced approaches of data collectors might wrongly suggest there is only one way to collect data about sex, for example.

As the Office for National Statistics have demonstrated in their research into the design of the sex question for the 2021 census in England and Wales, there exists at least five ways to collect data about sex in a national census.

Recommendation 3.4 therefore left me wondering if this actually illuminates our understanding of gender, sex and sexuality data or wrongly suggests an easy answer that does not currently exist?

Furthermore, Recommendation 5.4 notes:

Data producers and analysts should ensure that the language used in the collection and reporting of all characteristics is clear. For example, clearly distinguishing between concepts such as sex, gender and gender identity; or ethnic identity and ethnic background. This would help to avoid ambiguity and confusion among respondents and data users, which can undermine data and analytical quality, as well as belief in the validity and reliability of data.

Again, the Taskforce expect too much from data producers, analysts and those sharing their data. The concepts ‘sex’, ‘gender’ and ‘gender identity’ are ambigous and confusing among respondents and data users, there is no quick fix here.

It is particularly frustrating to see the Taskforce refer to data about the protected characteristic of gender reassignment as data about ‘gender identity’.

The choice of language is most likely intended to model the wording adopted for the question used to count trans and cis populations in England and Wale’s 2021 census, which describes itself as a ‘gender identity’ question but only collects data about the gender of people whose gender differs from their sex registered at birth.

The majority of respondents likely use the terms ‘sex’ and ‘gender’ interchangeably in everyday life and will not register any difference between the concepts when disclosing information about themselves. However, use of the tautological phrase ‘gender identity’ – in place of a term such as ‘trans status/history’ (as used in Scotland’s 2022 census) – is a slippage that only adds to this confusion. We don’t collect data about someone’s ‘religious identity’ or ‘racial identity’ or ‘sex identity’ – why ‘gender identity’? The ‘identity’ suffix is intended to differentiate the concept from ‘gender’ but as the Taskforce understand ‘gender identity’ to relate to the protected characteristic of ‘gender reassignment’ (and not ‘gender’, per se) this makes the recommended approach extremely hard to follow.

This slippage of language also aids the efforts of trans-exclusionary campaign groups to drive a wedge between concepts of sex and gender in data collection activities. Groups opposed to the counting of trans people believe that data collection agencies should capture sex data (as something only biological) and gender data (as something only social), overlooking the overlap and interconnectedness of the two concepts. This approach is not reflective of the social world nor how individuals share information about themselves in data collection exercises.

Although promising much, the Inclusive Data Taskforce fail to go far enough. Future work by the UK Statistics Agency needs to move beyond ‘being inclusive’ as an aspiration and ensure their outputs decrease rather than increase confusion about gender, sex and sexuality data.