In Question 74 of the New York Times CBS Poll, 69 percent of what were judged usable respondents said that illegals should be prosecuted and deported. This is reported in their lengthy pdf file available here.
74. Should illegal immigrants be prosecuted and deported for being in the U.S. illegally, or shouldn’t they?
- Should be prosecuted 69
- Should not 24
- DK/NA 6
Immigration Bill Provisions Gain Wide Support in Poll
By JULIA PRESTON and MARJORIE CONNELLY
Published: May 25, 2007
A careful analysis of the pdf and its inconsistencies is reported on at Times Watch:
The Poll’s Goals
“The New York Times reports its own poll on immigration, complete with slanted questions and a slanted cherry-picking of the answers.” Posted by: Tim Graham 5/25/2007 3:33:00 PM. Graham points out the inconsistency of the answers for Q74, not mentioned in NYT article, and other questions highlighted in the NYT article.
The NYT and its pdf give no information on how many people were contacted initially to come up with the final list of usable responses. Many of the questions are not reported even in the pdf.
Q74 is almost at the end of the NYT survey. This is after all the positive information on the bill has been given as part of or before other questions. At this point, the pollster can’t drop the person without losing all the time spent on the previous, apparently, 73 questions. A cynic might infer that the NYT poll couldn’t disqualify people for answering yes to Q74 without losing the answers they wanted for the other questions?
In some cases, the NYT CBS poll gave people information and then asked them questions in light of that information. This information was only positive for the bill? So it appeared to this reader.
What about giving information about jobs. Consider the following potential questions based on information provided.
The census, an office of the US government, reports that the median wages of men have been the same since 1973, see graph page 18. The increase in legal and illegal immigration started in 1965.
The Bureau of Labor Statistics (BLS) reports (in an Excel spreadsheet) that since1965 the labor force participation rates of white men fell from 80 percent to 75 percent and black men from 80 percent, about the same as whites to 66 percent.
NYT CBS style Questions following this fact:
Do you think immigration is why 75 percent of white men and 66 percent of black men still have jobs?
Do we need more immigration so that white and black men can keep the jobs they still have?
Do you believe that median wages for men would be substantially lower except for the positive benefits of immigration on wages?
Do you think that the only thing that can keep wages of men from going down is more immigration?
- Republican bigotry
- Secret plan by Trent Lott and Strom Thurmond
- Don’t know
Black wages are still below white wages.
How many immigrants will it take to get black wages to converge to white wages?
- 1 million per year
- 2 million per year
- 3 million per year
- 4 million per year.
- All of the above.
- Men are pigs
- Employers are pigs
- Secret plan by jocks who resent male teams being cut to comply with federal regs
- Not enough immigrants.
Why did Black wages converge towards white wages before the 1965 Immigration Act but stop afterwards?
- Klan was strong before 1965.
- Public interest law firms can use 1964 Civil Rights Act to control principals and teachers.
- Lack of multicultural education curriculum materials before 1965.
- Blacks got poor education before 1964 in schools to keep their wages down.
- Not enough immigrants.
The census graph shows that women’s wages still don’t match men’s, see graph page 18.
What will it take for women’s wages to converge to men’s?
- 1 million immigrants per year.
- 2 million immigrants per year.
- We need employers to determine the answer and tell us.
- Matching every willing immigrant to a willing employer.
A Canadian Harvard study shows that wages in Canada and the US are depressed by immigration. (See Patrick Cleburne at Vdare on this study.)
After hearing about this study do you think:
- Canada needs more immigration to undepress these wages.
- America needs more immigration to undepress wages.
- America and Canada are the problem: Mexico is the solution.
- Don’t know, Don’t care, bigot or otherwise unusable response.
The Bureau of Labor Statistics projects that black men, white men, Hispanic men and Asian men will all have lower labor force participation rates in 2014 compared to 2004. This was before the 2006 and 2007 Bush McCain Kennedy Kyl amnesty plans.
Do we need more immigration to prevent this?
- No, but count it as yes anyhow.
–Information for next question
The top 1 percent got 20 percent of income before income restriction in the 1920’s, 10 percent and heading lower before the 1965 Immigration Act (legal immigration), and are now back up to 20 percent and headed higher.
Would income inequality have been even lower in the 1950’s if there had been immigration at the bottom of the U?
Is the only thing that is keeping income inequality from being higher today the high influx of poor people whose great number averages out with that of the rich to keep income inequality down?
- No, I went to school before they fell apart, but I’m sure if I went to them today I would answer yes.
If there are enough poor people, who are all equally poor, it doesn’t matter if there are a few rich people. Its just envy of the rich to disagree with this outcome, and bigotry of the middle class towards the poor to try to keep them out?
- I think I heard a luncheon speaker say this, it must be true.
7 of the top 8 wealthiest Senators voted for S. 2611, amnesty, affirmative action, non-deportable crime, and a pathway for the top 1 percent of households to continue to enjoy 20 percent of each year’s income, compared to 10 percent before Kennedy’s 1965 Immigration Act. The only 1 of the top 8 who didn’t vote for S. 2611 didn’t vote, Jay Rockefeller. McCain is 7th and Kennedy 8th in wealth.
Is it possible for us to pay our Senators enough for what they do for us?
- Yes, but count it as no anyhow. (We will.)
–Information Next Question
“Numbers Drop for the Married With Children
Institution Becoming The Choice of the Educated, Affluent”
By Blaine Harden
Washington Post Staff Writer
Sunday, March 4, 2007; Page A03
PORTLAND, Ore. — Punctuating a fundamental change in American family life, married couples with children now occupy fewer than one in every four households — a share that has been slashed in half since 1960 and is the lowest ever recorded by the census.
The first census was in 1790. The census is reporting the unhelpful information that men’s median wages are the same since 1973, (graph page 18) and that women’s wages today, and blacks are lower than all men’s wages were in 1973.
How can we teach the census to do better work?
- Get them to drop respondents who have the wrong answers.
- Tell people with the wrong answers they are bigots.
- Exclude blacks, whites, men, women, Hispanics, and Asians, unless they get speaking fees of $50,000 per lunch.
- Find one lesbian Hispanic immigrant female who makes more than the median wage of men in 1973 and put her picture on census.gov in place of all these pdf’s full of wrong graphs (page 18).
- If we had called Pat Buchanan a bigot one more time, we wouldn’t have these problems.
- Restrict census data to the S & P 500.
What do you think of this poll?
- More immigrants are the answer.
- Invalid response.
- Don’t know.
- Survey respondent disqualified.
- Spoiled survey ballot.
- My chad is hanging.
- Do pollsters get speaking fees?
Media pundits are paid $50,000 or more to speak at corporate lunches and ride corporate jets to and from them.
- More immigrants are the answer.
- Is this a question?
- A prayer?
- More immigrants are the answer.
- Final Jeapardy, What is the reason the pundits support immigration?
==Appendix: others commenting on this poll.
Posted by: Michael Medved at 10:29 PMMedved slams Rasmussen for not push polling for immigration. Instead Rasmussen just asks if they support the bill without giving them any information.
Here are two data that blow out of the water the idea that most Muslims are “moderates,” i.e., that most Muslims believe in Islam only as an individual religion, not as a political religion. Andrew Bostom writes:
Polling data just released (April 24, 2007) in a rigorously conducted face-to-face University of Maryland/ WorldPublicOpinion.org interview survey of 4384 Muslims conducted between December 9, 2006 and February 15, 2007–1000 Moroccans, 1000 Egyptians, 1243 Pakistanis, and 1141 Indonesians–reveal that 65.2% of those interviewed–almost 2/3, hardly a “fringe minority”–desired this outcome (i.e., “To unify all Islamic countries into a single Islamic state or Caliphate”), including 49% of “moderate” Indonesian Muslims. The internal validity of these data about the present longing for a Caliphate is strongly suggested by a concordant result: 65.5% of this Muslim sample approved the proposition “To require a strict [emphasis added] application of Shari’a law in every Islamic country.”
Steve Sailer on “pollaganda”
Stein Report on NYT CBS poll:
CIS executive director Mark Krikorian critiques poll at NRO. This is particularly informative and short.
Poll: Public Wants Illegals to Go Home
Public Prefers Enforcement, Not Senate Legalization Approach
The poll that counts: Call your Senators. Scroll down for lists of numbers to call, stop immigration free fax services, etc from Frosty Wooldridge:
Insightful comments on current and past polls related to immigration at Vanishing American.
Worse, the more information the pollsters give those questioned — about current levels of immigration, alternative policies, fiscal costs — the more strongly they oppose the legislation. So you need to get hurry this bill through the entire legislative process before Americans learn what’s in it.
May 25, 2007 12:00 AM
Double Agent Karl
Machiavelli takes a look at the immigration bill.
By John O’Sullivan
National Review seems to be tacking back towards being a conservative magazine from being an arm of BOG (Bush occupied government). If National Review had still been on the American side before the 2000 and 2004 elections, it might have mattered. But better to have them late than never. This is the Return of the Apostate Son.
= Postscript and unresolved issue:
Because Q74 (prosecute and deport 69% yes) comes at the end, its late for the pollster to discover that the respondent is not responsive and should be considered an unusable response without losing their work? If the person answers against immigration early in the poll, the pollster can try to get them to drop out, e.g. by suggesting the person responding is a bigot. Or the pollster can resort to using more subtle means, like marking the response invalid or unusable.
How many people did they contact for the NYT CBS poll? How many dropped out? At what stages? At what questions? Why were the results of Q74 at variance with other answers as pointed out by Tim Graham? Aren’t pollsters supposed to be steeped in statistics and scientific method? Why didn’t the New York Times have an article by the pollster scientists who did the poll and designed it discussing this issue?
If this poll was submitted to a peer reviewed scientific journal, wouldn’t they make the authors discuss the internal inconsistencies that Tim Graham and others have pointed out? Doesn’t the NYT claim to be scientifically rigorous? Now that critics have pointed these issues out, why doesn’t the NYT have the pollsters discuss these in an article published in the print edition and on the web page? Isn’t that what real scientists would do in a controversy over results?
=Compare Scientific Article and Poll
A scientific article might have a readership of 1000 people or even less. If there are inconsistencies or problems, at least 100 of those 1000 and likely even more will learn of it.
For an NYT poll at least 10 million hear of it. However, only 10,000 or less hear of the controversies and maybe only 1000 or even the same 100 the scientific details of polling.
So if the NYT poll was published in the Journal of Polling Science it would end up with the same 100 people knowing the methodological problems as when published in the NYT.
In addition, a science article with these flaws, especially if concealed and then exposed, would be remembered by the 100 for decades. So it would be permanently linked to the authors as dubious and sloppy work at best. NYT polls don’t survive that long in memory.
This is why they published it in the NYT even though they knew the scientific problems before publication. If the same poll was submitted to the Journal of Polling Science, they would not have published it without all these issues being resolved. That would have taken months at least, and the poll would not be fresh.
The NYT knows or believes that after a couple news cycles, the only people still concerned are a few people, and the NYT thinks they don’t matter. They went through this calculation in advance of publication, knowing the polls inconsistencies. They chose to hide the poll inconsistencies by not discussing them in the article and even withholding some questions and answers from the pdf. They said these were for future use. But no one is interested in poll data that is very old, except the 10,000 down to 100 poll wonks.
The NYT and CBS have adopted as a business practice to present poll findings when the poll results are inconsistent and don’t support the statements in their main news articles or stories. This is intentional scientific fraud. This is a business policy of engaging in fraud for profit.
== Some final questions
What percentage think that Senators who support immigration should be prosecuted?
At least retired?
What about Senators who vote their stock portfolio instead of the median wages of their constituents?
Do we have a Senate of stock portfolios?
Do we have a nation of 1973 median wage earners?
This article represents speculation, hypotheses or opinion. All statements should be restated as questions. All other disclaimers apply.