Data-matching will be of limited use to the government in introducing individual electoral registration, especially in identifying potential electors, says a report by Parliament's political and constitutional reform committee. The document, titled Individual Electoral Registration and Electoral Administration, recommends that …
I have several different addresses, depending on who you ask. I'm sure I'm not the only one. Since I own a house, obviously I have that address (but don't live there). But since I work in another location, I stay near work, so some of my registrations are there. In addition to avoid loosing things in the gaps where I move from one location to another, I have some things registered at a P.O. box, and others at my parants'.
The reason for most of the differences are because of the different requirements of the data holder, e.g. It's the address of the property where the utility is delivered (various), it must be where your car is normally parked (insurance); it must be where you sleep at night (doctors); It can't be a P.O. box, but also can't be a different hotel every 2 weeks (DVLA), etc.
I'll admit that I am not in the majority of the population, but since they are talking about this as a way of getting 100% of the population, using a method where people show-up at different addresses is not going to help. Of course another point is that if someone does not want to vote, why insist on them being registered in the first place (or is it a BB issue?)
I'd be very interested to find out more about the methods used for matching addresses. I suspect that for most local authorities, the leading enterprise software solutions were way out of their price range and it sounds like they defaulted to manual labour to eyeball the data. I also suspect that the free/low cost/open source software options are not mature enough to deal with particularly cruddy data.
Even so, address matching is available from a number of bureau service providers who could probably clean this much more cheaply than a handful of people can. I know of both providers and software options that can offer a 95% + match rate for residential addresses (business addresses are a different matter, but that wasn't the point of the exercise - assuming that voters don't live in business properties).
So what is meant by "data matching"???
The article implies the method for data matching was a wholly manual, labour-intensive exercise requiring authorities to boost staff numbers. But this is "data" right?
I ask myself, what ever happened to using computers and algorithms? Since local authorities rely on package solutions for all their IT i am not surprised at the low 75% match rate. This sounds like something from the 1950's not the 21st century when the likes of Google can match you anything.
Examining the article closely I cannot believe 25% of residential addresses are named only with no house numbers, and what about the postcode? Sure DW&P use postcodes?
Perhaps dear El Reg, an FOI request could find out exactly what these problem addresses are!
Why did they kill CORE?
Surely DW&P and the Electoral Register need to match their addresses to Eric Pickle's new database first? http://www.nlpg.org.uk/
CORE was killed off by parliament last July. Why? Authorities that used specialist matching software or services typically got a 95%-99.9% match to the NLPG which underpins Eric's new database. Surely if the DW&P were made to similarly reconcile their data to the very database that the entire public sector is supposed to be using now, then authorities shouldn't be so unfairly burdened!
At least this is only a trial, so perhaps now parliament will see sense and realise that authorities are ill-equipped to solve data-problems with human resources alone.
The Electoral Registers could be 100% accurate and the only thing that would change would be the Electoral Registers would be 100% accurate. (Okay, and telemarketers/junk mailers who buy copies of the public version would have a vastly improved database of people to harrass.)
The people who do vote are already on the register by definition. The people who aren't on the register don't vote, and have the right and ability to get themselves registered if they gave a damn, which it seems fair to assume that they don't.
There are countless more worthwhile projects than getting councils to spend their already tight budgets on what is essentially a tarted up Yellow Pages of people.
Pointless? That depends on why you want a database listing everyone old enough to vote.
If it's to send them voting forms; then yes pointless as you have said. (In fact it's not just that people can easily add themselves to the register, it's hard to not; they keep bugging you, sending people to your house to fill out the form etc.)
If you want a list of everyone who is old enough to vote (and pay taxes, etc.), then maybe not pointless...
"In Southwark, 25% of DWP records could not be matched to properties in the borough."
Could it be massive benefit fraud? Seems plausible in Southwark....
Ha ha yes!
So it should be the DWP doing the work and not the local authority!
What will Mr Eric Pickles make of this i wonder? Surely this is a leftie parliamentarian plot to instigate a benefit takers' democracy to take on the Tory strongholds and oust the coalition? Hmmm i like it.
Is increasing voter turnout always a good thing?
You could argue that increasing voter turnout is not always a good thing. We're better off if fewer stupid people vote, if fewer racists vote, if fewer Sun readers vote ... wait, there's a large intersection of those sets. So why try to push them?
Yeah, but that still might not help; you only get to vote for this lying sh*t, or that crooked w**ker...
Two colored thumbs are the only way to go.
One color for the district in which you live, another after you vote. Everything else fails, sometimes miserably.