Thu, 10 Nov 2011

DNA database: analysis of offending figures

GeneWatch UK has published an amazing analysis of the latest data on the National DNA Database released by the Home Office in September. In just five pages (including one page of references), the DNA database: analysis of offending figures (pdf) briefing debunks not only the misleading comments made following the release of the Home Office data, but also the misunderstanding of the individual cases mentioned to justify retention when they offer no such support. If you have the time to read only one document on DNA retention, this is one not to miss. Here are a few excerpts:

In September 2011, new data was released by the Home Office regarding the likelihood of future offending by persons who have been arrested for the first time. These figures have been used by critics of the Protection of Freedoms Bill to claim that “every year, 23,000 people, who under Labour’s system would be on a DNA database will, under government plans go on to commit further offences” and that this will allow 23,000 people to become victims of crime in the future. This claim is incorrect because it fails to take account of the limited role of the DNA database in solving crimes.

[...]

Using the higher figure of 36,000 persons a year estimated to be rearrested and sanctioned for a recordable offence following a first arrest with no sanction, it is possible to estimate how many of these crimes might have been solved using individuals’ DNA profiles, were they to be retained on the National DNA Database.

[...]

This amounts to about 28 convictions a year (including 8 for ‘Scottish List’ offences), because only about half of DNA detections lead to a conviction. Of these estimated 28 convictions a year, the majority would be delayed not lost since any future arrest of the individual would lead to a match being made between their DNA profile and the relevant crime scene DNA profile (which would be stored indefinitely if it did not match an individual’s profile when it was loaded onto the database).

[...]

Numerous individual cases have been cited in support of retaining innocent individuals’ DNA on the National DNA Database. However, closer inspection of these cases has repeatedly revealed that most would not be affected by proposals in the Bill. Over the ten year period since legislation was introduced to retain innocent people’s DNA profiles there have been no examples of murder cases cited in parliament or the press that would have remained unsolved had innocent people’s DNA profiles been taken off the database.

[...]

These cases suggest that “widening the net” to retain innocent individuals’ DNA profiles on the DNA database has been the wrong priority compared to taking DNA from known suspects for a crime.

Here's the summary of the DNA retention – Analysis of arrest-to-conviction data submission from the Home Office Economics and Resource Analysis Group to the Parliamentary Under-Secretary of State for Crime Prevention that is analysed by GeneWatch UK:

[The Economics and Resource Analysis Group] analysis of arrest-to-conviction data obtained from the PNC indicates that the time taken for the risk of conviction of individuals with no previous convictions who are arrested but not sanctioned to fall to the level observed in the general population is approximately three years. This assumes that DNA profiles are retained on arrest for all offences. If the scope of the provisions is restricted to ‘Scottish list’ offences with retention on arrest, the time taken for conviction risk to fall to the population level is 3¾ years. If scope is restricted to ‘Scottish list’ offences with retention only on charge, the time taken is 4¾ years (although this result is subject to significant uncertainty due to the small sample size).

We have also made provisional estimates of the outcomes of profile retention under each regime, assuming retention for three years. These suggest that restricting the scope of the retention provisions (from all offences on arrest to ‘Scottish list’ on charge) significantly reduces the number of ‘innocent’ profiles retained. The proportion of those profiles expected to receive a sanction before the end of the retention period does not change as scope is restricted, while the proportion of sanctions that are received which are serious (as defined by the ‘Scottish list’) rises slightly in absolute terms. Thus, the Scottish model appears to be effective in offering protection to individuals who are arrested but not sanctioned for any offence. The extent to which it is effective in ensuring that only the profiles of more ‘serious’ potential offenders are retained is debatable.

The Protection of Freedoms Bill is now in the Lords. For more details see:

websiteblogblog archivenews feedfeedback