If...done properly.
And doing any particular poll "properly" is ALWAYS debatable...thus it involves art as much as statistics or science.
There are literally millions of variables of any one poll that effects the results.
You are mixing the science of statistics with the variables of polling...and portraying the whole thing as science and statistics.
It's not.
kiln, you are on the right track, but i think that for the original study in question, the sample size was 146, but the census was 146. now, i dont account for lying when i calculate numbers, just data. therefore i would fail to reject the stat as it included the entire population. also, you hit on something that the public doesnt know. sure they tell you the sampling error, but what they dont tell you is the confidence interval. 95% is the most common, but 90% is sometimes used. 3% @ 95% can be huge margin of error.
as for sample size, i saw barman question that, sure sample sizes can be too small, but 1000 people of 20000 is a HUGE sample, unless you are using subgroups such as age, gender and race. at that point you may run into missing data and tiny sample sizes. just think, in presidential elections, a simple random sample of 3200 people is a decent size and good enough to do a test at 95%
the only problem that i have because of the continual nature of the results of these types of studies, they present correlation and causality, which would mean that media is hired because they are dems, and i dont think that to be true, maybe the dems are just more willing to talk or according to punter, nicer people. thats the reason for the donations and votes
Um, please tell me Kiln that you don't have a degree in any kind
of math and science. If you do, then I have even less faith in our
public school system than before.
Polling *is* not statistics.
Polling produces data. That data can vary from very flawed to
very accurate.
Statistics involves analyzing the data that was collected from the
polling.
Please read this through until you get it through your ignorant, thick
head.
Then, you can apologize to MisterMJ for being completely wrong in
your fundamental understanding on this basic issue - and in
your feeble, lame attempt at patronizing him in your complete
ignorance.
Thanks for offering a more coherent explanation than I've been able to muster. If you re-read the thread you'll see that Willie99 asserted that there were fewer people polled for a presidential election poll than for the poll regarding party affiliation of journalists. It was that error I was attempting to correct.
zit, i understand what you are saying. if you said that polling is part of statistics then yes. if polling is done completely, adhering to the definition of polling "A survey of the public or of a sample of public opinion to acquire information." as a broad defintion then that is correct also. like i said, its a gray area which kind of makes you both correct, but yes, i can see your point and annoyance at such a broad definition
after further consideration, (i drove to arby's) this is the best way i can explain it.
statistics IS planning, collecting, organizing, and drawing conclusions based on data,
27% is a statistic
67% and 46% ARE statistics
if there was a question on a test that asked if polling (done completely) is statistics, i would have to say true.
so yes kiln you are correct if you said polling IS a form of statistics OR polling IS statistics
statistics IS planning, collecting, organizing, and drawing conclusions based on data,
27% is a statistic
67% and 46% ARE statistics
if there was a question on a test that asked if polling (done completely) is statistics, i would have to say true.
so yes kiln you are correct if you said polling IS a form of statistics OR polling IS statistics
I was never unclear about what Statistics is/are but thanks for adding your corroboration. I think the egg on festeringZit's face was thick enough already, but the added pressure of a second opinion may compel him to do the right thing and apologize. Anyone who's checked in on this thread now knows that he's made a fool of himself. The question persists, however: is there any honor left for him? We'll all find out soon enough but, as I said before, I won't hold my breath. It means little to me but it might be good for him to make a public display of taking respnsibility for his actions.
No, it most certainly wasn't the first time.
But Zit will never admit he might have been wrong about anything. Actually I'm afraid this is true for most posters here.
I can't help but notice that festeringZit has found time to post in other threads but still hasn't found the time to show any dignity or class in this thread. Anyone surprised?
Preussen is completely wrong yet again, as I have absolutely no
qualms about admitting when I have been wrong. And, I can
find plenty of posts on here where I have admitted such.
In this case, I think by the letter of the law I am right, but the spirit
of what Kiln was saying is true, that polling by a broad definition is
part of a broad part of statistics.
I have taken multiple graduate statistics classes, and this
form of the definition never came up.
I will take the high road and issue an apology.
Kiln please accept my apologies, I was not aware of that broad
definition.
Seems kinda a qualified apology, considering-
<TABLE class=tborder id=post6473899 cellSpacing=0 cellPadding=6 width="100%" align=center border=0><TBODY><TR><TD class=thead style="BORDER-RIGHT: #fdde82 1px solid; BORDER-TOP: #fdde82 1px solid; FONT-WEIGHT: normal; BORDER-LEFT: #fdde82 0px solid; BORDER-BOTTOM: #fdde82 1px solid" align=right> #44 </TD></TR><TR vAlign=top><TD class=alt2 style="BORDER-RIGHT: #fdde82 1px solid; BORDER-TOP: #fdde82 0px solid; BORDER-LEFT: #fdde82 1px solid; BORDER-BOTTOM: #fdde82 0px solid" width=175>festeringZit<SCRIPT type=text/javascript> vbmenu_register("postmenu_6473899", true); </SCRIPT>
RX Senior
Join Date: Sep 2004
Location: Texas
Posts: 4,868
</TD><TD class=alt1 id=td_post_6473899 style="BORDER-RIGHT: #fdde82 1px solid"><!-- icon and title -->
<HR style="COLOR: #fdde82; BACKGROUND-COLOR: #fdde82" SIZE=1><!-- / icon and title --><!-- message -->Quote:
<TABLE cellSpacing=0 cellPadding=6 width="100%" border=0><TBODY><TR><TD class=alt2 style="BORDER-RIGHT: 1px inset; BORDER-TOP: 1px inset; BORDER-LEFT: 1px inset; BORDER-BOTTOM: 1px inset">Originally Posted by Kiln
If done properly, polling is statistics. I didn't realize you didn't understand that. Sorry.
</TD></TR></TBODY></TABLE>
Um, please tell me Kiln that you don't have a degree in any kind
of math and science. If you do, then I have even less faith in our
public school system than before.
Polling *is* not statistics.
Polling produces data. That data can vary from very flawed to
very accurate.
Statistics involves analyzing the data that was collected from the
polling.
Please read this through until you get it through your ignorant, thick
head.
Then, you can apologize to MisterMJ for being completely wrong in
your fundamental understanding on this basic issue - and in
your feeble, lame attempt at patronizing him in your complete
ignorance.
<!-- / message --></TD></TR></TBODY></TABLE>
