Domestic abuse: statistics recorded by the police in Scotland - 2016/17

Annex 3: Auditing of Data by Her Majesty's Inspectorate of Constabulary in Scotland (HMICS)

HMICS Crime Audit 2013

6.25 Following the creation of Police Scotland in April 2013, HMICS carried out a review of incident and crime recording for a number of crime types including crimes with a domestic abuse aggravation flag. HMICS sampled a number of records across the then 14 divisions in Police Scotland to establish the accuracy of the reporting and to assess compliance with the Scottish Crime Recording Standard ( SCRS). The review involved qualitative research with a range of people involved in the crime recording process, including interviews and focus groups with officers and staff. Thirty incidents in each of the 14 divisions were sampled. The sample size was not large as it was only intended to serve as an indicator of recording practice across Scotland.

6.26 Compliance rates varied across division and crime type; however crimes with a domestic abuse flag achieved a high compliance rate of 99%. The report states: "the compliance rate for Domestic Abuse incidents (99%) illustrates what can be achieved when a focussed and robust approach is taken to attending, investigating and recording a particular crime type".

6.27 The full report, including key findings, recommendations and improvement actions, can be accessed from the HMICS website: https://www.hmics.scot/publications/review-incident-and-crime-recording

6.28 It should be noted that the audit was carried out before the implementation of the VPD to all Police Divisions. However the audit does imply that the front line officers and call handlers were well trained and knowledgeable when it came to recording domestic abuse issues, and that users of domestic abuse statistics can have reasonable confidence that crimes associated with domestic abuse are being recorded correctly.

6.29 It should also be noted that as the audit is based on a sample survey of incident and crime records (rather than all records), the true value may differ slightly from the results in the audit. This is because sampling in this way is always subject to a range of quantifiable and non-quantifiable error.