[ILUG] Advertising in GMail

[Kevin Brennan]

Colm MacCarthaigh wrote:
> On Tue, Nov 21, 2006 at 11:05:02AM +0000, Dave O'Connor wrote:
>   
>>> Make up your mind.
>>>
>>> It was "pretty accurate" and now it's become a "good guess"
>>>
>>> Either it's accurate or it's not and we can see that it's not
>>>       
>> Gawd. Now I remember why I unsubscribed.
>>
>> Read what I said, numbnuts.
>>     
>
> The problem is that you are abusing statistics. Although it's possible
> to make general statements about the level of accuracy of statistically
> aggregated data it is *never* appropriate to make inferences from that
> to any one particular datapoint.
>   
wouldn't the point of the statement "good guess" mean that it's not 
intended to me an Inference.

> So, although I can say that 95% of ILUG subscribers are male, that
> doesn't mean "particular ILUG subscriber is male" is a "good guess",
> hell it's not even appropriate to say that "there's a 95% probability
> that this particular ILUG subscriber is male". 
Well if you randomly pick an ILUG subscriber from the list then what is 
the probability that 'it' will be male.
> Those statements are
> meaningless. The person either is or isn't male, there's no intermediate
> state.
>
>   
You will find this last statement inaccurate if you follow some of the 
spam links.

> So, you either are in or around Nass or you're not, and although you can
> say that a GeoLocation service gets it right - in the aggregate - NN% of
> the time, that still says precicely nothing about the "quality" of any
> particular guess.
>
>   

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.