January 20, 2019
Playing with Polls for Fun and Profit
Timothy Birdnow
Daren Jonescu, my good friend in South Korea, discusses the recent Washington Post/ABC News poll showing the public blames Trump and the Republicans for the government shutdown.
Daren makes the following points:
I haven't looked at the internals of this poll, but it mirrors something I found about the Trump border wall address to the Nation.
The Left-wing website Mediaite is crowing about a recent Quinnipiac Poll trying to determine how the American People saw Donald Trump's speech on the Border wall:
When asked "Did President Trump’s recent televised address to the nation change your mind about building a wall along the border with Mexico, or not?†only 2 percent responded "yes,†while 89 percent said no, and the remaining 9 percent hadn’t heard enough or didn’t answer.
The poll did not ask how respondents had changed their minds, leaving open the possibility that Trump actually talked a few Americans out of supporting a wall. The poll also had a margin of error of plus or minus 3.3 percent, which means Trump could have changed as few as zero minds, or as many as 5.3 percent.
Respondents also believed Senate Minority Leader Chuck Schumer’s (D-NY) and House Speaker Nancy Pelosi’s (D-CA) rebuttal over Trump’s address by a margin of 46 to 36 percent.
According to Quinnipiac’s press release, Americans were very clear in their views of the wall, and those views were overwhelmingly negative:
American voters are negative in every question about the wall, saying:
59 – 40 percent that it is not a good use of taxpayer dollars;
55 – 43 percent that the wall would not make the U.S. safer;
59 – 40 percent that the wall is not necessary to protect the border;
52 percent say the wall is against American values as 41 percent say the wall is consistent with American values.
Finally, the poll found that Americans oppose shutting down the government over funding for the wall by a two-to-one margin, and 56 percent blame Trump and the Republicans for the shutdown.
End excerpt.
Sounds pretty bad, si? But, as always, this is messaged data and who was asked is as important as what.
But here is the methodology employed:
This RDD telephone survey was conducted from January 9 – 13, 2019 throughout the nation.
Responses are reported for 1,209 self-identified registered voters with a margin of sampling error of +/- 3.3 percentage points, including the design effect. Margins of sampling error for subgroups are available upon request.
Surveys are conducted in English or Spanish dependent on respondent preference with live interviewers calling landlines and cell phones.
All data was collected and tabulated by the Quinnipiac University Poll.
PARTY IDENTIFICATION QUESTION WORDING - Generally speaking, do you consider yourself a Republican, a Democrat, an Independent, or what?
REGISTERED VOTERS
PARTY IDENTIFICATION
Republican 27%
Democrat 31
Independent 35
Other/DK/NA 8
End excerpt.
Huh? four percent more Democrats than Republicans and a whopping thirty five percent "independents (which means either people strongly supportive of Democrats or people farther left wing.) This thing was clearly weighted strongly in favor of opponents of the Wall. Also, notice they conducted many of these surveys in Spanish; hardly a way of finding people who want border security.
Also, please note that it took them the better part of a week to conduct this survey. Why? The longer it went the more likely the average, uninformed person would accept the media narrative on this.
This wouldn't be the first time Quinnipiac was caught with biased polling.
It should also be pointed out that as late as the midterm elections :
The vast majority of Republican voters and more than 53 percent of swing voters in the recent midterm elections support President Donald Trump’s most central promise: The construction of a border wall along the U.S.-Mexico border.
Exit polling conducted by Zogby Analytics for the Federation for American Immigration Reform (FAIR) reveals the vast majority of Republican voters — more than 80 percent — support the construction of a border wall along the southern border. Less than five percent of Republicans oppose a border wall.
Meanwhile, a majority of about 53.3 percent of swing voters in the midterm elections support Trump’s proposed border wall to stop illegal immigration into the country. A plurality of more than 47 percent of all midterm election voters said they support the construction of a border wall.
End excerpt.
and as of December:
1 Dec 2018
Nearly 2-in-3 supporters of President Trump and Republican voters say they support shutting down the federal government in order to secure full funding for a border wall on the United States-Mexico border.
The latest Marist Poll reveals that 65 percent of Trump supporters and 65 percent of GOP voters say the president should not compromise on his proposed border wall, even if it means a government shutdown.
Among those who call themselves "strong Republicans,†nearly eight-in-ten say Trump should shut down the government to secure the border wall funding.
A plurality of about 47 percent of Trump’s base, the white working class, say the president should stay strong on the border wall and not compromise, even with the possibility of a government shutdown.
Non-college educated white men, 50 percent; white evangelical Christians, 53 percent; and Americans living in rural communities, 51 percent, are the most supportive demographic groups of shutting down the government to secure the $25 billion needed to build the border wall.
When asked if building the border wall to stop soaring illegal immigration at the southern border was an "immediate priority,†about 67 percent of Trump supporters and 63 percent of Republican voters said yes, the border wall is an immediate priority for the nation.
A plurality of the white working class, 42 percent, said the border wall is an immediate priority, as well as 45 percent of non-college educated white men and 42 percent of Americans in rural communities, who said the same.
As Breitbart News reported, Trump said on Tuesday he is fully committed to shutting down the federal government if Democrats do not help fund his border wall, although Sen. Chuck Schumer (D-NY) said he and fellow Democrats would not waver in refusing to fund the wall.
End excerpt.
But now we are to believe that the public hates the wall, hates Trump and the Republicans for demanding a scant five billion for it, and everyone is happy with gangsters and terrorists coming into the country.
Interestingly enough, the Quinnipiac poll showed:
There is a security crisis along the Mexican border, American voters say 54 - 43 percent, and voters say 68 - 26 percent there is a humanitarian crisis.
End excerpt.
How do you square that with this lack of support for Trump and the Wall? You can't.
In other words, recent polls are worthless on this issue.
As Rassmussen pointed out on January 11:
In the midst of a government shutdown over disagreements about building a border wall, two-out-of-three voters still think illegal immigration is a serious issue, but nearly half of voters think the government isn’t working hard enough to stop it.
A new Rasmussen Reports national telephone and online survey finds that 48% of Likely U.S. Voters think the government is doing too little to stop illegal immigration, up five points from 43% who felt the same way in August. Twenty-eight percent (28%) believe the government is doing too much to stop illegal immigration, down from 34%, while 17% think the government’s level of action is about right. (To see survey question wording, click here.)
The survey of 1,000 Likely Voters was conducted on January 8-9, 2019 by Rasmussen Reports. The margin of sampling error is +/- 3 percentage points with a 95% level of confidence.
End excerpt.
How do you do more? Clearly, even if people think the Wall wouldn't be all that effective it would be an attempt to do something. Why would the public NOT support that?
A recent Marist poll has Trump's numbers up 19% with Latino voters as of January 17. And while Trump's overall approval is down from forty two to thirty nine percent, one wonders if that is not a statistical rounding error. Most polls are plus or minus three percent.
Considering the endless drumbeat of negativism by the media, it is amazing Trump is where he is. The fact is, Trump is winning just by not being down near zero support. These polls are designed to push people, not to inform. I wouldn't trust them in the least.
Daren Jonescu, my good friend in South Korea, discusses the recent Washington Post/ABC News poll showing the public blames Trump and the Republicans for the government shutdown.
Daren makes the following points:
Here is the poll’s lead question, and the one being cited, misleadingly, for headlines all over the moronosphere:
Q: As you may know, the federal government has been partially shut down because Donald Trump and Republicans in Congress and Democrats in Congress cannot agree on laws about border security. Who do you think is mainly responsible for this situation?
The results of the poll, unsurprisingly, showed Trump and the Republicans as 53% responsible, the congressional Democrats as 29% responsible, both equally responsible at 13%, "neither†at 2%, and "no opinion†at 4%.
First of all, if you asked this exact question in a vacuum, without the respondents having been primed in any way, or provided any context, I guarantee the result would show "Donald Trump and the Republicans†as mainly responsible. Why? Because Donald Trump is the only individual person named in the question, and of course he is an extraordinarily famous person.
Furthermore, by lumping Trump together with "Republicans in Congress†as a single option, Trump’s name recognition, as well as his being the only actual person named, forces respondents to provide the optically convenient and obviously desired poll result: Trump and Republicans to blame; Democrats not to blame.
Notice, further, that the accompanying story from the Washington Post, along with the piggybacking stories all over the internet, phrase the result as I have just done, namely that Trump and the GOP are "to blame†for the shutdown. But that is not what the poll asked. The poll question asks, "Who is mainly responsible?†Responsibility and blame are two entirely different things. I am responsible for the dinner I just cooked. I can only said to be "to blame†for it if you have judged it to be a bad dinner. By framing the question as a matter of "responsibility,†the poll invites respondents who are supportive of the GOP’s position to accept their party’s responsibility, without equating this with blame, which therefore pads the results on the side of "Republicans to blame,†for the purposes of reportage, when that is not what the result actually shows.
Ah, how right he is!Q: As you may know, the federal government has been partially shut down because Donald Trump and Republicans in Congress and Democrats in Congress cannot agree on laws about border security. Who do you think is mainly responsible for this situation?
The results of the poll, unsurprisingly, showed Trump and the Republicans as 53% responsible, the congressional Democrats as 29% responsible, both equally responsible at 13%, "neither†at 2%, and "no opinion†at 4%.
First of all, if you asked this exact question in a vacuum, without the respondents having been primed in any way, or provided any context, I guarantee the result would show "Donald Trump and the Republicans†as mainly responsible. Why? Because Donald Trump is the only individual person named in the question, and of course he is an extraordinarily famous person.
Furthermore, by lumping Trump together with "Republicans in Congress†as a single option, Trump’s name recognition, as well as his being the only actual person named, forces respondents to provide the optically convenient and obviously desired poll result: Trump and Republicans to blame; Democrats not to blame.
Notice, further, that the accompanying story from the Washington Post, along with the piggybacking stories all over the internet, phrase the result as I have just done, namely that Trump and the GOP are "to blame†for the shutdown. But that is not what the poll asked. The poll question asks, "Who is mainly responsible?†Responsibility and blame are two entirely different things. I am responsible for the dinner I just cooked. I can only said to be "to blame†for it if you have judged it to be a bad dinner. By framing the question as a matter of "responsibility,†the poll invites respondents who are supportive of the GOP’s position to accept their party’s responsibility, without equating this with blame, which therefore pads the results on the side of "Republicans to blame,†for the purposes of reportage, when that is not what the result actually shows.
I haven't looked at the internals of this poll, but it mirrors something I found about the Trump border wall address to the Nation.
The Left-wing website Mediaite is crowing about a recent Quinnipiac Poll trying to determine how the American People saw Donald Trump's speech on the Border wall:
When asked "Did President Trump’s recent televised address to the nation change your mind about building a wall along the border with Mexico, or not?†only 2 percent responded "yes,†while 89 percent said no, and the remaining 9 percent hadn’t heard enough or didn’t answer.
The poll did not ask how respondents had changed their minds, leaving open the possibility that Trump actually talked a few Americans out of supporting a wall. The poll also had a margin of error of plus or minus 3.3 percent, which means Trump could have changed as few as zero minds, or as many as 5.3 percent.
Respondents also believed Senate Minority Leader Chuck Schumer’s (D-NY) and House Speaker Nancy Pelosi’s (D-CA) rebuttal over Trump’s address by a margin of 46 to 36 percent.
According to Quinnipiac’s press release, Americans were very clear in their views of the wall, and those views were overwhelmingly negative:
American voters are negative in every question about the wall, saying:
59 – 40 percent that it is not a good use of taxpayer dollars;
55 – 43 percent that the wall would not make the U.S. safer;
59 – 40 percent that the wall is not necessary to protect the border;
52 percent say the wall is against American values as 41 percent say the wall is consistent with American values.
Finally, the poll found that Americans oppose shutting down the government over funding for the wall by a two-to-one margin, and 56 percent blame Trump and the Republicans for the shutdown.
End excerpt.
Sounds pretty bad, si? But, as always, this is messaged data and who was asked is as important as what.
But here is the methodology employed:
This RDD telephone survey was conducted from January 9 – 13, 2019 throughout the nation.
Responses are reported for 1,209 self-identified registered voters with a margin of sampling error of +/- 3.3 percentage points, including the design effect. Margins of sampling error for subgroups are available upon request.
Surveys are conducted in English or Spanish dependent on respondent preference with live interviewers calling landlines and cell phones.
All data was collected and tabulated by the Quinnipiac University Poll.
PARTY IDENTIFICATION QUESTION WORDING - Generally speaking, do you consider yourself a Republican, a Democrat, an Independent, or what?
REGISTERED VOTERS
PARTY IDENTIFICATION
Republican 27%
Democrat 31
Independent 35
Other/DK/NA 8
End excerpt.
Huh? four percent more Democrats than Republicans and a whopping thirty five percent "independents (which means either people strongly supportive of Democrats or people farther left wing.) This thing was clearly weighted strongly in favor of opponents of the Wall. Also, notice they conducted many of these surveys in Spanish; hardly a way of finding people who want border security.
Also, please note that it took them the better part of a week to conduct this survey. Why? The longer it went the more likely the average, uninformed person would accept the media narrative on this.
This wouldn't be the first time Quinnipiac was caught with biased polling.
It should also be pointed out that as late as the midterm elections :
The vast majority of Republican voters and more than 53 percent of swing voters in the recent midterm elections support President Donald Trump’s most central promise: The construction of a border wall along the U.S.-Mexico border.
Exit polling conducted by Zogby Analytics for the Federation for American Immigration Reform (FAIR) reveals the vast majority of Republican voters — more than 80 percent — support the construction of a border wall along the southern border. Less than five percent of Republicans oppose a border wall.
Meanwhile, a majority of about 53.3 percent of swing voters in the midterm elections support Trump’s proposed border wall to stop illegal immigration into the country. A plurality of more than 47 percent of all midterm election voters said they support the construction of a border wall.
End excerpt.
and as of December:
1 Dec 2018
Nearly 2-in-3 supporters of President Trump and Republican voters say they support shutting down the federal government in order to secure full funding for a border wall on the United States-Mexico border.
The latest Marist Poll reveals that 65 percent of Trump supporters and 65 percent of GOP voters say the president should not compromise on his proposed border wall, even if it means a government shutdown.
Among those who call themselves "strong Republicans,†nearly eight-in-ten say Trump should shut down the government to secure the border wall funding.
A plurality of about 47 percent of Trump’s base, the white working class, say the president should stay strong on the border wall and not compromise, even with the possibility of a government shutdown.
Non-college educated white men, 50 percent; white evangelical Christians, 53 percent; and Americans living in rural communities, 51 percent, are the most supportive demographic groups of shutting down the government to secure the $25 billion needed to build the border wall.
When asked if building the border wall to stop soaring illegal immigration at the southern border was an "immediate priority,†about 67 percent of Trump supporters and 63 percent of Republican voters said yes, the border wall is an immediate priority for the nation.
A plurality of the white working class, 42 percent, said the border wall is an immediate priority, as well as 45 percent of non-college educated white men and 42 percent of Americans in rural communities, who said the same.
As Breitbart News reported, Trump said on Tuesday he is fully committed to shutting down the federal government if Democrats do not help fund his border wall, although Sen. Chuck Schumer (D-NY) said he and fellow Democrats would not waver in refusing to fund the wall.
End excerpt.
But now we are to believe that the public hates the wall, hates Trump and the Republicans for demanding a scant five billion for it, and everyone is happy with gangsters and terrorists coming into the country.
Interestingly enough, the Quinnipiac poll showed:
There is a security crisis along the Mexican border, American voters say 54 - 43 percent, and voters say 68 - 26 percent there is a humanitarian crisis.
End excerpt.
How do you square that with this lack of support for Trump and the Wall? You can't.
In other words, recent polls are worthless on this issue.
As Rassmussen pointed out on January 11:
In the midst of a government shutdown over disagreements about building a border wall, two-out-of-three voters still think illegal immigration is a serious issue, but nearly half of voters think the government isn’t working hard enough to stop it.
A new Rasmussen Reports national telephone and online survey finds that 48% of Likely U.S. Voters think the government is doing too little to stop illegal immigration, up five points from 43% who felt the same way in August. Twenty-eight percent (28%) believe the government is doing too much to stop illegal immigration, down from 34%, while 17% think the government’s level of action is about right. (To see survey question wording, click here.)
The survey of 1,000 Likely Voters was conducted on January 8-9, 2019 by Rasmussen Reports. The margin of sampling error is +/- 3 percentage points with a 95% level of confidence.
End excerpt.
How do you do more? Clearly, even if people think the Wall wouldn't be all that effective it would be an attempt to do something. Why would the public NOT support that?
A recent Marist poll has Trump's numbers up 19% with Latino voters as of January 17. And while Trump's overall approval is down from forty two to thirty nine percent, one wonders if that is not a statistical rounding error. Most polls are plus or minus three percent.
Considering the endless drumbeat of negativism by the media, it is amazing Trump is where he is. The fact is, Trump is winning just by not being down near zero support. These polls are designed to push people, not to inform. I wouldn't trust them in the least.
Posted by: Timothy Birdnow at
09:59 AM
| No Comments
| Add Comment
Post contains 1776 words, total size 16 kb.
34kb generated in CPU 0.172, elapsed 0.7155 seconds.
35 queries taking 0.7104 seconds, 157 records returned.
Powered by Minx 1.1.6c-pink.
35 queries taking 0.7104 seconds, 157 records returned.
Powered by Minx 1.1.6c-pink.
