Comments on: Polling Data http://www.snowflakesinhell.com/2007/11/21/polling-data/ Firearms Policy and Politics in Pennsylvania Mon, 15 Mar 2010 05:53:19 -0400 http://wordpress.org/?v=2.9.2 hourly 1 By: Just an FYI http://www.snowflakesinhell.com/2007/11/21/polling-data/#comment-15553 Just an FYI Wed, 05 Dec 2007 05:38:23 +0000 http://www.snowflakesinhell.com/?p=1975#comment-15553 [...] Just an FYI I talked about this here a few weeks ago. The questions are so loaded it's not even funny. In short, this poll is pretty [...] [...] Just an FYI I talked about this here a few weeks ago. The questions are so loaded it’s not even funny. In short, this poll is pretty [...]

]]> By: Linoge http://www.snowflakesinhell.com/2007/11/21/polling-data/#comment-15003 Linoge Thu, 22 Nov 2007 22:28:51 +0000 http://www.snowflakesinhell.com/?p=1975#comment-15003 I was not trying to say that any particular side is blameless in any pollster situation, rather that people should not buy into any poll numbers without first seeing the actual questions asked of the polling body... And that goes for just about any poll. I was not trying to say that any particular side is blameless in any pollster situation, rather that people should not buy into any poll numbers without first seeing the actual questions asked of the polling body… And that goes for just about any poll.

]]> By: teqjack http://www.snowflakesinhell.com/2007/11/21/polling-data/#comment-15001 teqjack Thu, 22 Nov 2007 21:37:15 +0000 http://www.snowflakesinhell.com/?p=1975#comment-15001 How about asking convicts "have you ever drunk tap water?" The result would approximate 100%, so clearly drinking tap water should be outlawed. Still. I admit that the bias is usually not conscious. The "[Should we] Require handgun owners to immediately report lost or stolen guns to the police" query does not look bad at a glance, until you substitute something else (eg `garden gnomes`, `returnable cans`). How about asking convicts “have you ever drunk tap water?”
The result would approximate 100%, so clearly drinking tap water should be outlawed.
Still. I admit that the bias is usually not conscious. The “[Should we] Require handgun owners to immediately report lost or stolen guns to the police” query does not look bad at a glance, until you substitute something else (eg `garden gnomes`, `returnable cans`).

]]> By: PN NJ http://www.snowflakesinhell.com/2007/11/21/polling-data/#comment-14997 PN NJ Thu, 22 Nov 2007 15:09:32 +0000 http://www.snowflakesinhell.com/?p=1975#comment-14997 It's very hard to find any gun polls where the questions aren't biased one way or the other. This seems to be true for both pro-gun and anti-gun poll sponsors, although the anti-gun crowd is much more blatant. For example, consider the following three variations on a common kind of question: (1) Do you believe private citizens should be allowed to own handguns? (2) Do you believe private citizens should be allowed to own handguns to protect themselves and their family? (3) Do you believe private citizens should be allowed to own handguns even though they might be used to kill someone or commit a crime? Most gun polls will give you some version of (2) or (3). Another trick is restricting poll responses to a small set of outcomes (eg, new gun control laws) while excluding other possible outcomes contrary to poll sponsor interests. In general, we should all be highly skeptical of gun polls, whether favorable or not. It’s very hard to find any gun polls where the questions aren’t biased one way or the other. This seems to be true for both pro-gun and anti-gun poll sponsors, although the anti-gun crowd is much more blatant.

For example, consider the following three variations on a common kind of question:

(1) Do you believe private citizens should be allowed to own handguns?

(2) Do you believe private citizens should be allowed to own handguns to protect themselves and their family?

(3) Do you believe private citizens should be allowed to own handguns even though they might be used to kill someone
or commit a crime?

Most gun polls will give you some version of (2) or (3).

Another trick is restricting poll responses to a small set of outcomes (eg, new gun control laws) while excluding other possible outcomes contrary to poll sponsor interests.

In general, we should all be highly skeptical of gun polls, whether favorable or not.

]]> By: Linoge http://www.snowflakesinhell.com/2007/11/21/polling-data/#comment-14994 Linoge Thu, 22 Nov 2007 07:20:39 +0000 http://www.snowflakesinhell.com/?p=1975#comment-14994 *sigh* The abuse of statistics is such a pathetic tactic, but one of amazing power. The average human has an almost mystical viewpoint of statistics, polls, and the like, and is, more often than not, willing to buy into them without bothering to do any research into how they were gathered or what they actually indicate. Thanks for exposing the acutal questions asked... certainly shows the agenda of those doing the asking. *sigh* The abuse of statistics is such a pathetic tactic, but one of amazing power. The average human has an almost mystical viewpoint of statistics, polls, and the like, and is, more often than not, willing to buy into them without bothering to do any research into how they were gathered or what they actually indicate. Thanks for exposing the acutal questions asked… certainly shows the agenda of those doing the asking.

]]>