This project stems from Karma Nirvana’s strategic priorities for 2021-2024, one of which is to “Develop a national dataset to better understand the scope, scale and prevalence of – and appropriate responses to – Honour Based Abuse (HBA). This is vital, because there is currently no prevalence estimate for HBA, or data-informed understanding of the national, regional or local scope and scale of the problem. Even where organisations are collecting data, therefore, we cannot know what proportion of the total number of real cases they are identifying.
My research confirms what was probably already known in the sector: currently, data on HBA is incomplete, very piecemeal, and not very reliable. This means we cannot use the current data to meaningfully inform understandings of scope and scale, or estimate prevalence, because of its incompleteness, piecemeal nature, and unreliability.
In this blog, I cover some initial findings emerging from the research. I have surveyed quantitative and qualitative data available in the public domain, and also spoken to 22 people in four key sectors: the police, healthcare, local authorities, and the third sector.
What Data?
Data is currently collected by (at least):
- Police forces across the UK.
- The Forced Marriage Unit, relating to their helpline.
- Family Courts, relating to Forced Marriage Protection Orders and FGM protection orders.
- The Crown Prosecution Service (CPS).
- Healthcare providers, because of the statutory duty to report FGM.
- Karma Nirvana, from their helpline (published by the Office for National Statistics).
- Some local authorities.
- IDVAs (in some cases)
- MARACS (in some cases)
- Schools (in some cases)
- Other service providers (e.g. refuges, by-and-for service providers, domestic abuse charities etc.)
Creating a national database is not as simple, however, as amalgamating all this data, even if that was practically possible.
There are several issues with this data, many of which are recognised across the sector. These include:
- Not all the anonymised data which is being collected being made available to other organisations, researchers and/or the general public.
- Gaps in the data (e.g. in the CPS data, between 2019 and 2024) making longitudinal understanding of data difficult.
- Organisations changing the way they collect or present data (e.g. the Forced Marriage Unit) making longitudinal comparisons difficult.
- Organisations only collecting data on a specific sub-set of HBA (e.g. forced marriage) and/or for a very specific region (e.g. one local authority).
- Data being organised differently for analysis. For instance, some organisations, (e.g., the FMU) collect data on an annual year basis, and other organisations (e.g. Karma Nirvana and the police) on a financial year basis. Some organisations use police force areas to geographically group data, others use local authority areas, and others use NHS commissioning areas, and these do not directly overlap.
- Different software systems recording HBA in different ways (and/or allowing HBA to be directly tagged, or not).
- No clear process for flagging or tagging something as HBA (e.g., there is no official offence code for HBA in the police).
- Limited time and resource for people on the frontline to properly identify HBA.
- Limited understanding of how to properly tag something as HBA on an organisation or agency’s systems.
- Inaccuracies and/or over-counting of incidents, e.g. a contact with police being tagged as “HBA”, “forced marriage” and “FGM”, when it may have been only one, or two, of these things, not all three.
- Data-collation and/or analysis being done by people with no specialist training in HBA (e.g. staff in IT departments), meaning some qualitative data might be missed.
- Organisations using different definitions of HBA.
- Lack of current and/or mandatory training on HBA, leading to a lack of confidence from people on the frontline in identifying HBA.
All of these issues create barriers for good data collection and data-sharing, and for the creation of evidence-based policy and understanding of the best response to HBA in the UK. If the data going into any database or analysis is not robust, then any findings, or policies built on those findings, won’t be robust either.
What Happens?
Data being collected on HBA is currently being used by stakeholders to:
- Meet statutory obligations to record and report data on certain issues
- Making the case for the need for funding
- Create “problem profiles” to improve localised understanding (e.g. in the police) and/or specific, local case studies for training and learning
- Be transparent
- Inform campaigning and media work
Very few people I interviewed said data would directly inform their organisation’s policy on HBA. Similarly, very few said they shared data for reasons to do with informing evidence-based policy, apart from informing the funding policies by local and national government or other bodies – i.e. evidencing need.
Similarly, few respondents said they had a clear idea of what other organisations did with any data they shared. This included when they sent statistics to government departments, aside from where those were published publicly for reasons of transparency.
Some noted that there would be barriers to sharing data, especially where that data is owned by the providers of specialist software, but many other interviewees expressed themselves as keen to share anonymised, high-level data if they knew who with, and for what purpose.
What Next?
One reason Karma Nirvana are campaigning for a statutory definition of HBA is that it would help overcome at least one issue identified above – that of different organisations using different definitions, and therefore collecting data which is not necessarily comparable as it might include different things. A sector-wide definition has recently been agreed, and the Home Office recently surveyed the sector on what definitions they used, and whether or not there should be a statutory one. We will have to wait and see what they decide based on the outcomes of that survey.
Of course, a statutory definition would make little difference if it was not also backed up by government requirements for organisations to report their data, and collect it robustly. Many people I interviewed for this project explained how their organisations were relatively reactive when it came to data collection and processing: they would collect and share what they were asked to (e.g. by government or funders).
Interestingly, no data on HBA has been requested as part of the local authority return on how many victims of domestic abuse they supported as part of the Support in Safe Accommodation Data Monitoring Collection. This means that the need for specialist support for victims of HBA is not necessarily being mapped by local authorities, although HBA is recognised as a form of Domestic Abuse. In turn, this suggests not as much specialist provision as might be needed is being provided.
Overall, data-collection practices need to be improved, and support (including financial resource) to do so provided. Some barriers to comparison could be ironed-out by sector-wide agreement on how to present and collate data. Others will take a lot longer to fix, as the underlying issues are more fundamental.
Forced Marriage Research 