At 17:24 30/01/2014 +0000, Martin Rathfelder wrote:
>Would anyone like to hazard a guess as to the sort of bias which would
>occur if a load of lefties, but only them, remove themselves from the data set?
>Do we know anything about the pattern of illness by political belief?
The important point is surely that we don't want to have to (hazard a)
guess? I certainly don't know the answer and, in any event, am sure that
it will/would not be _only_ 'lefties' who opted out. One can certainly
imagine that those, of whatever political beliefs, who do not wish
information about their health/illness to be 'shared' might well have
patterns of health/illness which differ from in at least some respects from
the population in general - but, again, ideally we don't want to have to be
guessing. It seems essentially the same issue as the bias that results
from any other sort of self-selection (e.g.non-responders to surveys), and
there is, to the best of my knowledge, no way to properly address that, or
to characterise/quantify the resultant bias.
Kind Regards,
John
----------------------------------------------------------------
Dr John Whittington, Voice: +44 (0) 1296 730225
Mediscience Services Fax: +44 (0) 1296 738893
Twyford Manor, Twyford, E-mail: [log in to unmask]
Buckingham MK18 4EL, UK
----------------------------------------------------------------
******************************************************
Please note that if you press the 'Reply' button your
message will go only to the sender of this message.
If you want to reply to the whole list, use your mailer's
'Reply-to-All' button to send your message automatically
to [log in to unmask]
Disclaimer: The messages sent to this list are the views of the sender and cannot be assumed to be representative of the range of views held by subscribers to the Radical Statistics Group. To find out more about Radical Statistics and its aims and activities and read current and past issues of our newsletter you are invited to visit our web site www.radstats.org.uk.
*******************************************************
|