 |
|
Capt Dizle
|
|
Oh, and collective punishment. Probably over the line but then it's a tough call.
At some point you have to make the people who support terror pay a price.
|
|
|  |
 |
|
Tripledoc
|
|
quote: Originally posted by jimmytrick
Oh, and collective punishment. Probably over the line but then it's a tough call.
At some point you have to make the people who support terror pay a price. |
Recall that today there was a story in the British newspaper The Independent which said that malnutrition in the Palestinian occupied territories had reached 'sub-Saharan standards'.
I guess they are waiting for those checks from that underling at Iraqi foreign dept. fourth floor.
Nevertheless I think that the situation is now moving into territory which would really demand intervention of some kind or another.
|
|
|  |
 |
|  |
 |
|
CharlesBHoff
|
|
el paso texas
May 2002 time: 05:33
|
|
quote: Originally posted by asleepathewheel
take a stats course. |
First it is easies to lie or commit perjury with stats and the size of your sample does matter a great deal. I play alot of pokar the royal flush is very rare there is only 1 change out of 300,000 of getting than royal flush for each hand being deal to you. Does that mean it you have 300,000 desks of pokar card that have one hand deal that one of then will be with than 100 % change of being than royal flush. Most likely none will be than royal flush.
|
|
|  |
 |
|
CharlesBHoff
|
|
el paso texas
May 2002 time: 05:33
|
|
The next president will have to raise taxes or end the Bush tax cuts to bring sanely to the budget. Taxes are the income which allow than government to run an do thing for the people. The prosit 13 wreck Calif fine public education system certain very wealth people donot like to pay they fair share in taxes. Than government need than fair tax rate that is neither too hight or too low to run without huge budget debit.
|
|
|  |
 |
|
Spiffor
|
 |
CPA - Evil Clone brigade
Nov 2001 time: 06:33
|
|
quote: Originally posted by CharlesBHoff
First it is easies to lie or commit perjury with stats and the size of your sample does matter a great deal. |
It is easy to lie with stats, whatever the size of the sample. You can even lie with stats encompassing the whole population (such as election results). The key to lie with stats is to provide a faulty interpretation of the results, it doesn't lie in the results themselves.
quote: I play alot of pokar the royal flush is very rare there is only 1 change out of 300,000 of getting than royal flush for each hand being deal to you. Does that mean it you have 300,000 desks of pokar card that have one hand deal that one of then will be with than 100 % change of being than royal flush. Most likely none will be than royal flush. |
Samples of 1000 - 2000 people are enough to measure societal trends when they are well done, because it has been often proven that larger samples repeat the same patterns. You can get a better accuracy with a higher sample (read: a smaller error margin), but that's it. General trends remain the same. The margin of error of a poll with 1000 people is about 3%. You need 4 times more people to halve the error margin. 4000 people make sense. 16000 people only in specialized polls. More than that is stupid.
In your example, royal flush would be deemed statistically insignificant, and as such, its complete absence from the results wouldn't make the trends any less true.
The only reason why you'd want to have a big sample is when you have many variables, and you want to analyze each of them with high accuracy.
For example, if you're polling on European politics, you only need a 1000 sample over Europe. However, if you want to analyze the differences per country, you need 1000 questionees per country (that's 15000 already). If you want to analyze the opinions of people based on their nationality AND their social class, you need a sample of 1000 per social class within each country. If you kept to the original 100 across Europe, to back you analysis of "what do wealthy Belgians think of Europe", you could only rely on very little amount (maybe 5 or 10) of wealthy Belgians questioned. Such a minuscule sample has virtually no relevance whatsoever.
|
|
|  |
All times are GMT. The time now is 05:33. Apolyton Time is 00:33. |
top of page
|
| archivepost |
|
Forum Rules:
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
|
HTML code is ON
vB code is ON
Smilies are ON
[IMG] code is ON
|
|
|
|
|
|