On the subject of Islamic violence (see the picture leading the article), Pew says:
While overall public views have shifted only modestly, the partisan gap on this question has steadily widened over the past several years as a growing share of Democrats say Islam does not encourage violence more than other religions, while the share of Republicans who say that it does also has grown.
Islam not more violent than other religions? Then Christians, Shintoists, Buddhists, Hindus, Voodooists and the like, Mormons, and so on are equally violent in the minds of 3 out of 4 Ds, and 3 out of 10 Rs. One wonders how they arrived at that opinion. Not a lot of news of Shintoists in the name of Shinto hiring vans and running down pedestrians.
It’s true that homegrown dislike of Christianity accounts for some of this result. But perhaps some of this is explained by the belief by many that Islam is a race, or rather that Muslims are a race, and there is no crime in this country worse than racism. Those who incorrectly believe Muslims are a race cannot seem to be talked out of it, either. These folks will murmur about “Islamophobia”, but if you ask them to define that term, it comes back to race, or anti-Christian sentiments. That explains, I think, the curious trends: curious for Ds because the observational evidence is in opposition to the opinion.
Well, nobody said politics had to match reality.
What about race?
Overall, 61% say the country needs to continue making changes to give blacks equal rights with whites, compared with 35% who say the country has made the changes needed to give blacks equal rights with whites.
That number was a bit less than half for a long time, rising sharply after 2014. The jumps came in both Ds and Rs: Ds are up to 81%, Rs 36%.
Pew says lots about “equal rights”, but not a word about duties. Except perhaps for this:
Public opinion also has shifted on perceptions of racial discrimination. Overall, 49% say that blacks who can’t get ahead in this country are mostly responsible for their own condition; 41% say racial discrimination is the main reason why many black people can’t get ahead these days.
Negative discrimination is increasing? Self-imposed apartness surely is. The significance of what used to be trivialities is higher. But negative discrimination increasing? Later Pew shows that disagreement with affirmative action has decreased from 30% to 22%. So why the belief that negative discrimination is increasing? Another instance where theory trumps number.
There is not a consensus:
By 59% to 31%, blacks say that racial discrimination is the main reason why many black people can’t get ahead in the country today. By contrast, more whites say that black people who can’t get ahead are mostly responsible for their own condition (54%) than say that racial discrimination is the bigger impediment to black progress (35%).
We discussed homosexuality last time, and here are some other non-surprises.
Women were always more supportive of homosexuality; the effects of this with “equal rights” are plain. Blacks have always been least accepting. From this plot, we might guess black men are the least accepting, though even here there is probability a majority.
Those who have had less indoctrination are also the least accepting; but again, even here is a majority. Same thing in the oldest age group. Now there are always age effects, meaning age is not only a measure of older version younger opinions (not ages), but of those who have survived version a more diverse pool of the young. There is nothing in these numbers, or in any other evidence, to suppose the estimate of about twenty years until the last vestiges of Christian (and Muslim and Torah-Judaism) beliefs are erased in all but a remnant.
Under the Sigh category, the global warming picture. I refuse to discuss it.
Let’s leave on an unhappy note, with the picture above. It’s self-explanatory.
Categories: Culture, Statistics
People who try to draw conclusions from surveys such as these often make the mistake of failing to consider uncertainty. Dr. Briggs seems to commit this common error several times in his summaries here. Interested readers should consult the PDF version of the complete report by PEW, especially the methodology section. This was a telephone survey, which means that what the researchers claim is a “randomized” sample is actually a sample of people who are odd enough to spend a long time on the phone sharing their opinions with a random person who has barged in on their privacy. PEW claims an uncertainty up to 4.4 percentage points in some of their results. This uncertainty is not shown the graphs (something that Dr. Briggs seems to consider important when he is eager to cast doubt on the results, but that seems not to concern him here), but it wipes out some of the purported differences that the author thinks are significant. These are all common layman’s errors, that you will no doubt see repeated in various journalistic outlets.
There is also this:
“Now there are always age effects, meaning age is not only a measure of older version younger opinions (not ages), but of those who have survived version a more diverse pool of the young. There is nothing in these numbers, or in any other evidence, to suppose the estimate of about twenty years until the last vestiges of Christian (and Muslim and Torah-Judaism) beliefs are erased in all but a remnant.”
I can only guess that Dr. Briggs is experimenting with some kind of deep-learning natural language generator, and that it’s not going well.
I had been about to comment on the same sentences as Mr. Phillips. They do rather create an image of Briggs, wild-eyed and hyped on coffee, pounding maniacally on his keyboard. Far from the truth, I’m sure.
The uncertainty figures would apply to each yearly figure (if any: they apply only to sampling error, provided the sample was taken randomly from the population of interest). But the interest here centers on the trend over time; so unless the uncertainty is biased consistently in the same direction year-after-year, it really doesn’t much matter. The magnitude of the differences are not a big deal. Each year may be within the zone of sampling error; but is every single year in the series? The important comparison is initial year to final year and the consistency of the difference betwixt and between across years.
Keep in mind too that between 2006 and 2017, the vast majority of the pool being contacted is likely unchanged. A few younger folk have entered the pool by 2017; a few older folks from 2006 have dropped out by 2017. But these are for most practical purposes repeated samples from the same population, and so might be regarded as representing the cumulative effects of being tutored in the Correct Answers to give pollsters.
–>”might be regarded as representing the cumulative effects of being tutored in the Correct Answers to give pollsters.”
That was roughly my thought. The objective of polling appears to be as indications of where the Media Megaphone needs to direct more volume regarding The Narrative.
Blacks protesting, e.g. BLM–answer with more “white privilege” stories, demands for more Affirmative Action, etc. Gay marriage still resisted in some quarters, pound the table for fairness in love and marriage–blame heteros for rampant divorce stats. Any polling that conflicts with The Narrative, e.g. environment, immigration, et al., merits the response that the beatings will continue until morale improves. Self-fulfilling as we all learn the Correct Answers.