On 30-Dec-06 kornbrot wrote:
> People appear to be looking for a statistical reason for the
> difference when there is a MUCH simpler psychological explanation
> Psychological research has consistently show thar the correlation
> between attitudes expressed in questionniares is, at best, very
> loosely correlated with behaviour.
> Reasons? 1. people may not wish estimate, 2. people may not be
> able to estimate Seasonal scenario
> Questionnaire on IDEAL number of family arguments at Christmas
> Dinner/New Year's eve party or any other seasonal celebration
> Observastion of ACTUAL number of family arguments at Christmas
> Dinner/New Year's eve party or any other seasonal celebration
> Who believes that any differences are due to defects in STATISTICAL
> methodology
> My takehome lesson?
> Marketing type questionnaires are pretty useful for predicting
> behaviour!
>
> Happy New Year
>
> Diana
Fair points! And a Happy New Year to you and to all!
Now to be unseasonally slightly grumpy. Or perhaps it's
a contribution to the Ideal Number of Arguments at this time.
Diana's email landed with a 73KByte thump. Hmmm.
20 Dec:
John Barker's long original, with HTML attachment: 18KB
Strip HTML: 9.5KB
Mike Brewer's reply, no inclusions, no HTML: 4.5KB
Paul Spicker's reply, no inclusions, with HTML: 11KB
Strip HTML: 5KB
Paul Lambert's reply, no HTML,
includes Paul Spicker above Mike Brewer above John Barker: 18KB
Strip inclusions: 10KB
30 Dec:
Diana Kornbrot's reply, with HTML
includes Paul Lambert above Paul Spicker above Mike Brewer
above John Barker: 73KB
Strip HTML: 21KB
Strip inclusions: 4.8KB
Total weight of thread as received: 124.5KB
Weight after stripping HTML: 58KB
Weight after further stripping inclusions: 33.8KB
("Weight" includes message headers, so the savings relative to
real message content are substantially greater)
Comments
A: Messages like the above are essentially plain-text in content.
Allowing the "HTML copy" to be attached as well adds nothing
except bloat -- over 100% on average, and often much more.
Consider Diana's, where the HTML occupied 52KB out of 73KB.
B: Very often, the inclusion of preceding messages in a reply
happens as a result of the default "stacking" Reply format
generated by software, and the writer in fact does not use
the inclusion as a reference for points in their reply.
So it is superfluous.
Where the writer does need to refer explicitly to passages
in preceding mails, this can be done by editing these so as
to include only the relevant passages and interleaving the
comments.
Therefore (A) avoiding the "HTML copy", and (B) judiciously
editing, or entirely omitting, the "stacked inclusions", can
save a lot of bytes when people reply. Nearly 100KB in this
short series of 5 messages.
The basis for some seasonal Good Resolutions?
Or for a Family Argument? (And don't ask me what the ideal
target for that is -- but perhaps I can try to start one
and see how keen people are).
Best wishes to all,
Ted.
--------------------------------------------------------------------
E-Mail: (Ted Harding) <[log in to unmask]>
Fax-to-email: +44 (0)870 094 0861
Date: 30-Dec-06 Time: 14:28:47
------------------------------ XFMail ------------------------------
******************************************************
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.
*******************************************************
|