National Records of Scotland

Preserving the past, Recording the present, Informing the future

Unreliability of 'Ground for Divorce' Statistics

Unreliability of 'Ground for Divorce' Statistics

It has been discovered that the breakdowns of the numbers of Divorces by Ground for Divorce, which appeared in former Vital Events Reference Tables 8.1 and 8.3 on this website, are not reliable. Work is in hand to address this and improved reporting is available for data processed with effect from the last quarter of 2008. 

For a number of years, a significant proportion of the DR1 forms sent by the Scottish Court Service (SCS) to the National Records of Scotland (NRS) were incomplete. The box which should contain the code for the 'Ground for Divorce' had been left blank. Unfortunately, in such cases, NRS staff neither sought the missing information from SCS, nor allocated a special "not known" code - instead, they simply entered the code for one of the known Grounds for Divorce.

As a result, it is not possible to produce the correct figures from the datasets that NRS holds, and the figures shown in the two tables are not reliable. Apparent trends shown by the published figures may be unreliable, because they could be affected by a change in the proportion of "not known" cases. 

This came to the attention of the NRS Vital Events Statistician and the Scottish Government (SG) Civil Justice Statistician in September 2008. NRS subsequently allocated a special "Ground not known" code for use when the DR1 form's boxes have been left blank, so it is possible to produce figures which show the numbers of "not known" cases separately for the final quarter of 2008 onwards and SCS reissued guidance to court staff to remind them to complete all the fields on the DR1 form. 

In Autumn 2010, SCS and the SG introduced new procedures to collect and transfer all the required information on divorce electronically. This should improve both the accuracy of the data and also the efficiency of the data collection process.

The only way to recover accurate data on the grounds for divorce would be to examine the paper records held in each of the 49 sheriff courts. However, it would be prohibitively resource-intensive to do so, and therefore accurate data will not be available until the publication of results from the first full year's data that were collected using the new procedures.

We greatly regret any inconvenience caused by this data problem.

All statistical publications