| Subject: |
|
Re: Re: Numberwatching the Lib Dems |
| Name: |
|
Grant |
| Date Posted: |
|
Feb 6, 07 - 2:19 PM |
| Email: |
|
gperkc@aol.com |
| Instant Messenger: |
|
http://,1 |
| Message: |
|
Interesting.
The numbers clearly show a higher recent trend BUT as with all numbers collected for govt. statistics one wonders what other factors might be behind them.
For example if the criteria for deploying the barrier had been slightly changed on the basis of the precautionary principle the number of deployments could easily be significantly increased for a very small change in operational standing orders.
Now I have no idea about whether or not that or something akin to it may have happened to influence the numbers but unless the possibilities can be explicitly excluded (never happened) or included (did happen but there is a conversion factor available for oranges to apples) the numbers by themselves are of little help.
Most govt. stats suffer from this sort of problem which, overall, may not be very significant to the net results. But where components of the data are used in a subset to inform policy the anomalies can become important and obfuscate the issue. To the point, in most cases I have looked at, where the unresolvable questions are likely to have more of an influence than the numbers recorded. But they remain unresolvable.
On this basis society seems happy to have people with personal prejudices make policy decisions for it.
Huhne's c.v. suggests he knows all about these sorts of things. If not the problems are greater than I could imagine.
Grant |
|
|
|
|
|
|
Reply
Home
bravenet.com