The same way that someone gets satisfaction out of taking a shit in the middle of a restaurant, and smirking at everyone holding their nose.
Irony
Answer: 100 million.
As in, “100 million people are right, and voted for trump, and your [sic] wrong”
That’s what some conservative nutjob had lettered across his pickup truck down at the grocery store the other day. Wish we could have got a picture, for it was truly a work of staggering genius!
It also had, from memory, the following:
B = Bisexual
I = impeachable
D = demon!
E = ex-felon
N = (something)
The guy spent a lot of time and money to letter the whole thing. Usually conservatives are fiscally sound - they generally can look stupid for a lot less money!
Yes, they are supposed to scrimp on personal expenditures so that they can send the money to Trump.
While I don’t think that the people who answered were necessarily “messing” with the pollsters a lot of people view these sorts of polls as a chance to demonstrate their party loyalty, and so will tend to exaggerate their beliefs. For this reason I tend to take polls asking questions that have clearly partisan answers with a grain of salt. This isn’t purely a Liberal phenomenon conservatives do it too. Which is why I also don’t give a lot of credence to “look at all the crazy things the Republicans believe” threads either.
Come on now. The angry lawnmower lady at least counts as five just on her own.
Yeah, say what you will, that is still push-polling. Give respondents a “About _________ unarmed black men were killed by police in 2019”. It is pretty easy to score, given that the answer is a number, and it does not give them ridiculous options to choose from.
This reminds me of a bumper sticker on a truck that would park here in town - back in the Great Before Time of 2009 - which read,
“One Big Ass Mistake, America”. Alas, the saying was much too long, so the type between the capitals was too small to be read at any sort of distance. From a distance, all you could really read was “OBAMA”.
While I agree that it may have been better to ask the question in the format you propose, I don’t agree that the poll, as constructed, constitutes push-polling. My understanding of push-polling is that it’s a form of pseudo-polling designed to “push” recipients towards the pollster’s preferred conclusion. For example, Senator Smith might commission a push-poll asking “If you learned Senator Jones was a crack addict, would you be more likely to vote for him, or less?” That kind of thing.
I don’t see how the mere presence of outlandish answers can induce people to pick them. The idea that the police are killing 1,000 or 10,000 or even more than 10,000 unarmed black men per year is one which, to put it as mildly as possible, doesn’t stand up to scrutiny. You don’t actually need to know anything to avoid picking those answers. You just need a smidgen of common sense.
I’m (mostly) a conservative, but I know plenty of sensible left-leaning liberals. There’s no correlation between liberalism and a lack of common sense. It seems to me that the most reasonable explanation for why so many people on the left chose such wildly incorrect answers is that they have an unhealthy media diet. If a person gets all their information about police brutality from dishonest sources like The Root or Salon, it’s almost inevitable that, over time, they’ll acquire some pretty serious misconceptions about the extent of police brutality. And these misconceptions can override what their gut may otherwise be telling them. I don’t understand why anyone would pick an obviously absurd answer just because it’s there unless they were operating under the influence of misinformation.
I can see your point, but I’m not sure if it truly applies in this specific instance. I can totally believe someone on the far left or the far right being much more willing to exaggerate the failings (real or imagined) of the other side. However, this particular survey’s questions don’t overtly pit respondents on either end of the political spectrum against each other. I think this matters because if the question doesn’t explicitly ask about one party or the other, there’s nothing which might induce a respondent to exaggerate their answer in order to stick it to the other side.
It’s one thing for a staunch Democrat to say that the Republicans (or vice versa) are far worse custodians of the economy (for example) than they actually are, but in that case the Democrat is explicitly being asked about the Republicans, so the incentive to exaggerate is right there in the question.
However, in this particular survey, the question isn’t phrased in that kind of overtly partisan way. For this reason, I’m not sure if the popularity of the more outlandish answers among left liberal respondents can be explained by the phenomenon described in your cite. As such, I think it’s more likely that the poll results are down to the kind of news media these people are consuming.
I agree that poll isn’t a classic “push-poll”, but I do think its design could skew otherwise reasonable people into picking higher numbers than the facts support.
From your first post on the topic,
So, taking the Washington Post and BLM numbers together, “about 20” would be a reasonable answer. Imagine a poll that had the options:
a) Zero
b) 20
c) 20,000,000
Hardly anybody would get that one wrong!
Now of course the actually existing poll you cited wasn’t anything like that blatantly manipulative. But
is going to “pull” people towards the larger numbers. I myself might easily have answered “b” to that one. I don’t actually have a running tally of “unarmed black men killed by the police last year” in my head; I know for damned sure it isn’t “zero” (or even “one”); I like to think I would have enough basic sense (and numeracy) to steer clear of “d” and “e”; and if I thought about it for ten seconds, even of “c”. But “b” could have easily “gotten” me. And someone more under-informed about the topic could pretty easily have been suckered into going for the seemingly “moderate” choice “c”–after all, it’s right there in the middle.
I think your comparison of liberals vs. conservatives here is super-skewed by the fact that the poll’s answers allowed the possibility of ideologically-influenced error in only one direction.
It may well be that conservatives are just as prone to underestimate the problem of racism and police violence as liberals are to overestimate it. But in this poll, the lowest possible answer happened to be the correct one.
So the poll completely fails to differentiate between the rational well-informed conservative who knows that the answer is “about 10”, and the rabid “Blue Lives Matter” (or just plain innumerate) conservative who would have picked “about .001” (i.e., one victim every hundred years, an estimate about a thousand times smaller than the true rate) if that answer had been available.
At the same time, the poll results do differentiate quite sharply between the rational well-informed liberal who knows that the answer is “about 10”, and the rabid “Abolish Police” (or just plain innumerate) liberal who thinks the rate of such killings is a thousand times greater than it really is.
Does that mean that there are really a lot more stupid/ignorant liberals on this subject than stupid/ignorant conservatives? Not necessarily: it just means that the framing of the poll is naturally going to reveal stupidity and ignorance on the liberal side and hide it on the conservative side.
(Of course I’m not claiming that the poll creators deliberately designed it that way to make liberals look stupid. It’s just a basic fact of survey design that every time you put a correct answer at one end of a spectrum of possible choices, you’re going to lose the ability to detect about half of your dumbest or worst-informed respondents.)
I’m in the same boat. In fact, I wonder if the definition of “unarmed” isn’t such that they include small and unbrandished knives. I looked through the cite, and all of its cites, and couldn’t find exactly how they determine if someone is armed. And that’s before getting into who decided how that’s coded, i.e. will they be looking to pin something on them and count it as a weapon?
If we use the high end of the range because some people were “armed” with a closed swiss army knife, then 27 is only 2.7 times the size of 10, and 100 is only 3.7 the size of 27, which is too close to fault people for guessing incorrectly.
I find this dubious for another reason. As someone who is somewhat informed about this, I know that one of the major issues being brought up by BLM and its predecessors was that this data isn’t actually well known, because there’s no central database. And, if you don’t have solid statistics, how can you do a good comparison of perception vs. reality? It would be very easy for those who reported higher numbers to think there are a lot more which are left unreported.
This actually led me to doing what I should have done in the first place, and check out the credentials of this skeptics.com site. The first place I checked was its Rational Wiki article. And even if I discard the parts that say citation needed, I definitely see some problems. It describes some articles they accepted that are rather dubious, and were apparently bamboozled by a fake psychologist for over 3 years. And it’s not good at all to have race realism anywhere near your work, let along having them as your senior editor.
Now, obviously, that’s not enough to say the entire publication is suspect. But that last part, with the racist senior editor, does give me pause due to my misgivings about the survey. A race realist would be the type of person who would want to push the idea that killing unarmed black people wasn’t that severe a problem.
Throw in the comment @Kimstu made about how the answers pull you to the higher numbers, and I am very skeptical of this study.
This is a very important point.
As is the point by Ludovic:
If we use the high end of the range because some people were “armed” with a closed swiss army knife, then 27 is only 2.7 times the size of 10, and 100 is only 3.7 the size of 27, which is too close to fault people for guessing incorrectly.
I won’t call this a push-poll. I’ll call it a misleading poll designed to come up with a specific answer.
It’s also misleading in other ways. For example, Philando Castile was armed. But he wasn’t remotely a threat, and yet he was still gunned down in front of his wife and children.
It’s not just about “armed vs unarmed”. It is even more about whether police are reacting much more violently to black men than to white men under comparable circumstances - that is, whether being black in itself is seen as a multiplier of perceived threat level. And there’s lots and lots of evidence that it is.
On the margins that might be true, but at the extremes I’d say it’s more the total perceived threat level than the racism component of it, while of course the racism component has implications for society at large.
If there were 10 unarmed Black people killed versus zero in other categories, that would be quite obviously better than if the numbers were a more equal 1000 versus 4000. Comparing the former to the latter would be like saying that bigotry by people of color is just as bad as that of White people - while it undeniably exists, it is less of a problem because it doesn’t have as big an effect.
True dat. How many unarmed white men were shot by police last year. I would bet the number is larger but not as large as their percentage of the population.
And how disgusting is it that it’s only twenty? “Why. that’s not even one every two weeks! Suck it up.”
That’s a fair point, but if the poll showed that the vast majority of ‘very liberal’ and ‘liberal’ respondents picked (a) or (b), I wouldn’t have found the results to be especially indicative of anything untoward.
The reason for this is that the Washington Post’s figures and the Mapping Police Violence figures are based on incomplete data. They both draw from a central database (the FBI’s National Use Of Force Data Collection) and only 5,030 of 18,514 federal, state, and local law enforcement agencies have provided their use of force data. Admittedly, many of these agencies are from very small localities with low populations that didn’t record any shootings whatsoever, but it’s not beyond the realm of possibility that the actual total of unarmed black men shot by the police is closer to 100 than 10.
That’s a very good point which I hadn’t considered. While the poll seems to indicate that a significant percentage of liberals are quite misinformed on the subject of police violence, it doesn’t prove that they’re any more or less misinformed than conservatives, for the reason you stated.
I think a better poll question would be to simply ask "How many people of any race and gender were killed by the police in 2019?" Sources vary, but they all seem to coalesce around 1,000. Asking that question with the same options as before might give a clearer idea of whether one side is less informed than the other.