Research in the UK has shown that police forces in Britain are continuing to add the DNA – and incidentally the fingerprints, although this is never mentioned – of innocent people to the DNA database despite the European Court of Human Rights ruling that it was illegal (and the government’s promise to accept the ruling). According to The Guardian newspaper today, 90,000 innocent people have been added to the National DNA database (NDNAD) since a the court ruling and the Association of Chief Police Officers (ACPO) – incidentally, a private organisation – is still telling chief constables to continue with this collection. On the other hand the process of removing individual profiles has been painfully slow: only 611 DNA profiles of innocent people have been removed, and all as a result of individual challenges in court. It seems that the police are determined to drag their feet as long as possible and, in fact, break the law quite openly. Hardly a good example…
ubisurv 
> and incidentally the fingerprints, although this is never mentioned
Even more rarely mentioned are palm prints. Both fingerprints and palm prints are captured in police stations on IDENT1 Livescan terminals (bought from Northrop Grumman) where your hand is held on a glass pane and scanned (i.e., no more ink). I believe that Livescan terminals are now installed in all stations. (See http://www.npia.police.uk/en/10504.htm for a bit more info.)
Another recent news is that the detection rate is going down as the size of the NDNAD increases. http://gizmonaut.net/blog/uk/2009/10/ndnad_more_and_less_of_the_same.html
A good resource to help with the process of requesting removal of one’s DNA profile (and destruction of the DNA sample, and deletion of the fingerprints, palm prints and associated PNC record) is http://reclaimyourdna.org/
Thanks, David. The detection rate decline doesn’t surprise me, but it would probably need a data specialist to take a look and explain exactly why…
David, I’d be interested to hear why the detection rate decline does not surprise you, whether it matches the hunch I mentioned in my write up or whether you have a different thoughts about it? I don’t think there’s enough data available for a data specialist to reach any conclusion, though a specialist, at this stage, would help figure out what other data we should ask for to get a better picture.
I have a similar hunch, that’s all! I have some colleagues who work on these kinds of things, so I will ask them!
A colleague, Jon Bright, says this:
“I think the problem here is the difference between actual and relative stats. what you are looking at here is a fall in actual numbers (on p37), however this doesn’t necessarily indicate a fall in the detection rate. it could simply indicate a fall in the number of crime scene profiles being loaded (i.e. the number of crimes where a forensic examination of the crime scene takes place). this is what the report itself briefly claims on p.30.
In the report, they claim that the detection rate is actually rising – on page 34 they show the crime scene to subject match as having increased since 2005 to over 58%. most of the matches occur, apparently, when new crime scene profiles are loaded on to the database, rather than when new individuals are loaded on. an increase in individuals present on the database should therefore increase the % of matches which occurs when a new crime scene is loaded on, but will not itself generate many new matches. if crime scenes being loaded are falling, the absolute number of detections should fall as well, regardless of the size of the database.”