Specifically ...

IWETHEY v. 0.3.0 | TODO

1,095 registered users | 0 active users | 1 LpH | Statistics
Login | Create New User

Welcome to IWETHEY!

Post #414,799 by drook 10/24/16 12:34:31 PM 10/24/16 12:34:31 PM Reply	Specifically ... Research, microtargeting & polling projects - Over-sample Hispanics - Use Spanish language interviewing. (Monolingual Spanish-speaking voters are among the lowest turnout Democratic targets) - Over-sample the Native American population The lower the expected turnout of a particular demographic, the higher the margin of error if you survey that demographic in proportion. Remember the story recently of the single black student who skewed a poll by 3% every time he responded? -- Drew
Post #414,801 by malraux 10/24/16 12:37:34 PM 10/24/16 12:37:34 PM Reply	Yes, that's the LAT poll I mentioned. His opinion was weighted at something over 300 if I recall. Regards, -scott Welcome to Rivendell, Mr. Anderson.
Post #414,803 by drook 10/24/16 12:40:06 PM 10/24/16 12:40:06 PM Reply	Yes, that one If I understand things correctly, you oversample so that you can underweight and reduce volatility. That poll went the other way, just take who they got and overweight to make up the difference. -- Drew
Post #414,806 by malraux 10/24/16 12:40:31 PM 10/24/16 12:40:31 PM Reply	Exactly. Regards, -scott Welcome to Rivendell, Mr. Anderson.

hey nother serious question - (boxley) - (20) - Oct. 24, 2016, 12:29:32 AM EDT

From a casual glance - (hnick) - Oct. 24, 2016, 05:49:11 AM EDT

Taking things out of context isn't way to enlightenment. - (Another Scott) - (5) - Oct. 24, 2016, 07:32:30 AM EDT

not taking anything out of context, was looking at the Atlas project info - (boxley) - (4) - Oct. 24, 2016, 12:04:28 PM EDT

I don't see what Wikileaks has to do with polling, but YMMV. - (Another Scott) - (3) - Oct. 24, 2016, 12:18:37 PM EDT

Serious question. - (mmoffitt) - (2) - Oct. 24, 2016, 12:31:26 PM EDT

Depends on the context. ;-) -NT - (Another Scott) - Oct. 24, 2016, 12:40:13 PM EDT

The BBC and PBS are pretty good as sources. -NT - (a6l6e6x) - Oct. 24, 2016, 12:45:06 PM EDT

Over-sampling isn't nefarious - (malraux) - (11) - Oct. 24, 2016, 12:26:13 PM EDT

Thank you. -NT - (Another Scott) - Oct. 24, 2016, 12:30:54 PM EDT

Specifically ... - (drook) - (3) - Oct. 24, 2016, 12:34:31 PM EDT

Yes, that's the LAT poll I mentioned. - (malraux) - (2) - Oct. 24, 2016, 12:37:34 PM EDT

Yes, that one - (drook) - (1) - Oct. 24, 2016, 12:40:06 PM EDT

Exactly. -NT - (malraux) - Oct. 24, 2016, 12:40:31 PM EDT

Also too. - (Another Scott) - (1) - Oct. 24, 2016, 12:39:38 PM EDT

I totally missed that - (drook) - Oct. 24, 2016, 12:43:31 PM EDT

thats what I was looking for, thanks -NT - (boxley) - Oct. 24, 2016, 02:15:13 PM EDT

More from The Atlantic - (malraux) - Oct. 24, 2016, 06:31:44 PM EDT

It's almost as if... - (pwhysall) - (1) - Oct. 25, 2016, 04:55:46 AM EDT

:-)! -NT - (a6l6e6x) - Oct. 25, 2016, 03:25:16 PM EDT

PK weighs in: Oversampling at the Fed! - (Another Scott) - Oct. 24, 2016, 07:01:15 PM EDT

"There are two major products that come from Berkeley: LSD and UNIX. We don't believe this to be a coincidence." -- Jeremy S. Anderson
178 ms