2009/5/20 Tom Steinberg <[email protected]>:
> I'm a touch dubious about the sample bias - 94 MPs have partial
> expenses published, leaving 500 others.

You are entirely right to be dubious. The point has been raised and
acknowledge by Mark Reckons but he doesn't seem to have understood it
properly.

We know that the Daily Telegraph has been highly selective (and
systematic) about the order of its publication. There's nothing
surprising about that - they were quite clear that was what they were
doing.

What that means is that the graph/data we have shows that there is a
strong and clear correlation between being in a safe seat and being
selected by the Daily Telegraph for publication and that, I am afraid,
is almost certainly all the data tells us.

Note: this is not the standard objection to a statistical correlation,
viz that correlations don't imply causation. That is, I am not suggest
that A and B correlate because they are both caused by C. Such a
criticism is something we should always bear in mind.

No, the data doesn't even get that far. We don't have access to
"sleaziness" or whatever, rather we have access to publication by the
DT. Only if you assume that the DT randomly sampled all MPs and only
published those against whom there was some level of scandal is this
data any use, but we know that is not the case.

>
> I'd suggest redoing this research when the data is available without
> being filtered through the highly particular lens of the Daily
> Telegraph.

Absolutely! Though there might be some difficulty with separating the
good from the bad. Its clearly something the MPs have trouble with.
Hazel Blears, whose behaviour appears to have been straightforwardly
criminal is still (as I write) a cabinet minister. The speaker, whose
hand in the till conduct would have entitled any reasonable employer
to dismiss him summarily with little fear of a successful challenge in
an employment tribunal, is still an MP (and has been praised, not
criticised) by party leaders.

Sadly, any attempt to actually use this data is going to hit the wall
of subjectivity. I live in a moral universe where fiddling expenses is
wrong. I suspect that most MPs (and most people who earn more than
£40k per annum) live in a different one.

Francis

>
> Tom
>
> 2009/5/20 Stefan Magdalinski <[email protected]>:
>>
>> http://markreckons.blogspot.com/2009/05/mps-expenses-and-safe-seats-correlation.html
>>
>> stef
>>
>> --
>> /*
>> Stefan Magdalinski
>> +447769 666528 (phone)
>> smagdali (IM/twitter/flickr/dopplr/skype/etc)
>> */
>>
>>
>> _______________________________________________
>> Mailing list [email protected]
>> Archive, settings, or unsubscribe:
>> https://secure.mysociety.org/admin/lists/mailman/listinfo/developers-public
>>
>
> _______________________________________________
> Mailing list [email protected]
> Archive, settings, or unsubscribe:
> https://secure.mysociety.org/admin/lists/mailman/listinfo/developers-public
>



-- 
--
Francis Davey

_______________________________________________
Mailing list [email protected]
Archive, settings, or unsubscribe:
https://secure.mysociety.org/admin/lists/mailman/listinfo/developers-public

Reply via email to