Well, T.G.G. didn't say "US population" he just said "population." As to the ballpark, there is this from
YouGov that has 33% of Republicans either affirming Chauvin should not be arrested or otherwise "don't know" (which is the same as affirming he should not be arrested). Among Independents, there is a total of 25% saying he should not be arrested and/or "don't know," which can be pretty much split down the middle between left-leaning and right-leaning (as most independents are split), so that's roughly an additional 12% added onto the Republican side for around 45% right-leaning people affirming that Chauvin should not be arrested?
So, yeah, ballpark.
What kind of math is that?
Ballpark.
You can't mix and match different samples like that.
Of course you can. We know from multiple sources (e.g., PEW) that Independents are essentially split down the middle--right-leaning and left-leaning--and almost always vote accordingly. So in a loose, ballpark estimation, you can most certainly add the right-leaning portion to the Republican portion to give a ballpark on how the right-leaning population thinks.
As to equating "don't know" with "should not be arrested," wrong metric. T.G.G. was estimating the percentage of people who thought Chauvin had done nothing wrong. Had you bothered to click on the source, this is the question the
YouGov poll asked (emphasis mine):
Based on what you have read, seen, or heard about the death of George Floyd in a police-involved incident, do you think the officer responsible for his death should or should not be arrested?
So the respondents are specifically being asked a question
based on their existing knowledge of the situation that includes affirmation language (
the officer responsible for his death), not just some question in a vacuum that they know nothing about. Which, again, in a ballpark estimation, would argue that anyone choosing "don't know" is at least tacitly affirming that he should not be arrested. It certainly isn't affirming (tacitly or otherwise) that he
should be arrested.
Again, we're talking about a cop keeping his knee on a man's neck for almost ten minutes resulting in that man's death, which is affirmed by the question being asked. This cop IS responsible for Floyd's death. Should he be arrested? Choosing, "don't know" to that question is certainly akin--i.e., in the same ballpark--to affirming that the cop did nothing wrong. After all, logic dictates that if they did, in fact, think that Chauvin had done something
wrong--after being told he IS responsible for Floyd's death in the question--then they would affirm "should be arrested."
Here, just break it down:
He's responsible for his death, so he should be arrested.
He's responsible for his death, but he shouldn't be arrested.
He's responsible for his death, but I don't know if should be arrested.
Well, then, somehow, in spite of the fact that you're agreeing that he's responsible for his death, you must think he did nothing wrong for him not to
also be arrested for that responsible act.
So, yeah, in a ballpark estimation of what people are
thinking about a given situation, that is sufficient, especially when you give it the haircut I gave it.
Regardless, all one would have to do is add up the percentages of just those who positively affirmed that they thought Chauvin should not be arrested and you get 29% of respondents, which is only 11% off of T.G.G.'s 40% and is therefore, likewise,
in the ballpark.