[WS:] You need many "samples" or more precisely many observations (i.e. a large sample) only if the population from which you sample is diverse - to ascertain adequate representation. If the population is homogeneous - a small sample will do.
More importantly, however, the metaphor I used pertained to the nature of the substance being sampled, not the sampling methodology.
Wojtek
On Sun, Oct 31, 2010 at 7:16 PM, Catherine Driscoll
<catherine.driscoll at sydney.edu.au> wrote:
> Wojtek writes:
>
> 'Martin: "Since television is often the most accessible avenue to the
> research of social science via it's products (advertising and pop
> culture entertainments), etc."
>
> [WS:] When you study bowel movement, taking a small sample is
> sufficient - you do not need to take the full load on a daily basis.
> Ditto for television and pop culture.'
>
> Actually, you and many other people need to have studied a vast number and array of bowel movements in order to know how that particular small sample should be understood.
>
> Not that I'm buying the argument that TV is justifiable because it's a sample of society. TV *is* society, as much as this list, the apparently endless renovation of the house across the street from me, or this stupid paper I should be writing and for some reason have decided a return to LBO will be a distraction from.
>
> Catherine
>
> On Sun, Oct 31, 2010 at 5:57 PM, martin schiller <mschiller at pobox.com> wrote:
>>
>> On Oct 31, 2010, at 1:32 PM, Angelus Novus wrote:
>>
>>> I was just irritated by the
>>> suggestion that those who choose not to watch television are joyless,
>>> self-denying hairshirts.
>>
>> Since television is often the most accessible avenue to the research of social science via it's products (advertising and pop culture entertainments), those who watch are the 'hairshirts', tracking the division and decline of civilization.
>>
>> ___________________________________
>> http://mailman.lbo-talk.org/mailman/listinfo/lbo-talk
>>
> ___________________________________
> http://mailman.lbo-talk.org/mailman/listinfo/lbo-talk
>
> ___________________________________
> http://mailman.lbo-talk.org/mailman/listinfo/lbo-talk
>