Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[MrRoboto75]

I'm concerned. What can Trump do that would cause voters to lose confidence?

Publicly shit his pants.

[mainiac]

Someone covered that earlier in this thread.  The media would fawn over how he goes to any lengths to help the american farmer.

[smjjames]

Someone covered that earlier in this thread.  The media would fawn over how he goes to any lengths to help the american farmer.

That makes me think of Jimmy Carter though somehow, not Trump.

[Reelya]

Yes and evidence shows that conducting random site inspections of ground water at suburban tampa homes will knock a couple of (statistically significant) percentage points points off the sale price of the homes with good water.  Humans are extremely weird with their valuations.  I just dont see what your specific examples have to do with the graph.

Because when O'Malley left the race people dropped from considering three choices to two. Going by the research, even a choice which has 0% favorability in itself can cause wild swings in the popularity of the other two. If O'Malley was considered to be more similar to Clinton than Sanders, then his mere presence in the race might have boosted her perceptions of favorability. After he left the race, then people re-evaluated the two candidates without the third choice. This would have taken some time though.

We tend to assume that someone with very low polling numbers must be having an equally small impact on the primaries/elections etc. But the behavioral psychology research I'm citing tells otherwise. In one cited research case (which was a real-world economic example of a company accidentally adding the useless third choice), a "useless third choice" switched things from 66%/34% to 16%/84%, without choice 3 ever actually being selected - so the mere presence of a choice nobody ever takes can literally invert people's decision-making process.

[smjjames]

Yes and evidence shows that conducting random site inspections of ground water at suburban tampa homes will knock a couple of (statistically significant) percentage points points off the sale price of the homes with good water.  Humans are extremely weird with their valuations.  I just dont see what your specific examples have to do with the graph.

Because when O'Malley left the race people dropped from considering three choices to two. Going by the research, even a choice which has 0% favorability in itself can cause wild swings in the popularity of the other two. If O'Malley was considered to be more similar to Clinton than Sanders, then his mere presence in the race might have boosted her perceptions of favorability. After he left the race, then people re-evaluated the two candidates without the third choice. This would have taken some time though.

We tend to assume that someone with very low polling numbers must be having an equally small impact on the primaries/elections etc. But the behavioral psychology research I'm citing tells otherwise. In one cited research cases, a "useless third choice" switched things from 66%/34% to 16%/84%, without choice 3 ever actually being selected - so the mere presence of a choice nobody ever takes can literally invert people's decision-making process.

So, you're saying we should never include third parties in polls? There IS a reason that polls randomize the order of the choices, which should weed out that effect to an extent.

[Reelya]

Maybe or maybe not. By randomizing the order, you make sure the voter is taking all the choices into account, which might even strengthen the effect. Because the effect comes from the need to process all the irrelevant choices. Sure, the gut feeling is "randomizing it makes it fair/objective/true" but that assumes that bias works in a specific way, and it's not really taking things like this psych research into account at all. Polls where you had to rate pairs of candidates directly would be the type that would circumvent this particular "psychological spoiler effect".

The sad thing is that this research really puts a crimp on things like IRV and multi-party elections, because it implies that having many choices can cause people to make irrational sub-optimal choices, rather than choosing better. The modern mantra is that more choices is always better, but the research suggests that people don't really deal with selecting from a number of alternates very well.

Although it suggests that political parties might learn from this, and back spurious candidates that make their main candidate look better, sort of a reverse of how we normally view the Spoiler Effect. e.g. having the choice "Republican, Democrat, Communist" might change how people select between Republican and Democrat even though only 1% of people would ever choose Communist directly.

[Frumple]

Someone covered that earlier in this thread.  The media would fawn over how he goes to any lengths to help the american farmer.

That makes me think of Jimmy Carter though somehow, not Trump.

I'unno, carter would probably have actually tried to help the american farmer, followed by the media taking a dump on him. Trump's more like a reversal of that.

[smjjames]

Maybe or maybe not. By randomizing the order, you make sure the voter is taking all the choices into account, which might even strengthen the effect. Because the effect comes from the need to process all the irrelevant choices. Sure, the gut feeling is "randomizing it makes it fair/objective/true" but that assumes that bias works in a specific way, and it's not really taking things like this psych research into account at all. Polls where you had to rate pairs of candidates directly would be the type that would circumvent this particular "psychological spoiler effect".

The sad thing is that this research really puts a crimp on things like IRV and multi-party elections, because it implies that having many choices can cause people to make irrational sub-optimal choices, rather than choosing better. The modern mantra is that more choices is always better, but the research suggests that people don't really deal with selecting from a number of alternates very well.

I wonder if it also has an effect on those multiple choice questions that you'd take for a SAT test or a regular school exam.

[penguinofhonor]

This effect definitely seems present when people are presented with unfamiliar options, but I'm not convinced that this applies in politics or areas where people have firm, established opinions.

If you poll people on whether they like creamy or crunchy peanut butter, adding "peanut butter with gravel in it" as an option isn't going to make more people like crunchy peanut butter. There's a reason those people don't like crunchy peanut butter.

[Reelya]

We should test that. You assume that it wouldn't, as I would. But if you look at the range of examples in the video then they get rather big swings from small stuff. So I definitely wouldn't say that as a guaranteed belief.

But I'd argue that the case under question - election primaries specifically, is probably the area where it would apply the most in politics. Since people have pretty strong tribal lines along party lines, but much less strong lines for internal party politics, and primaries are often where people are exposed to "unfamiliar options", and comparing a number of options which are superficially similar yet wildly varying in popularity - which all sounds like it would be a ripe area for this sort of effect to have a big sway.

[smjjames]

Yeah, it sounds like the kind of thing that needs further study on.

[Criptfeind]

This effect definitely seems present when people are presented with unfamiliar options, but I'm not convinced that this applies in politics or areas where people have firm, established opinions.

If you poll people on whether they like creamy or crunchy peanut butter, adding "peanut butter with gravel in it" as an option isn't going to make more people like crunchy peanut butter. There's a reason those people don't like crunchy peanut butter.

I mean, I liked smooth peanut butter before, but now that you've brought it up like this, I'm sorta thinking I want crunchy now.

[smjjames]

We should test that. You assume that it wouldn't, as I would. But if you look at the range of examples in the video then they get rather big swings from small stuff. So I definitely wouldn't say that as a guaranteed belief.

But I'd argue that the case under question - election primaries specifically, is probably the area where it would apply the most in politics. Since people have pretty strong tribal lines along party lines, but much less strong lines for internal party politics, and primaries are often where people are exposed to "unfamiliar options", and comparing a number of options which are superficially similar yet wildly varying in popularity - which all sounds like it would be a ripe area for this sort of effect to have a big sway.

And often times, a lot or maybe even all of the candidates are unknown (or at least not well known) outside of their homestates. At least this time around for the republican side as Clinton is well known nationally. There have been candidates during primaries who were well known nationally in the past, not counting incumbent VPs.

[Max™]

I'm concerned. What can Trump do that would cause voters to lose confidence? I'm worried Clinton isn't up to the task of taming Trump on national television. That's not a mark against her personally: I think it would be a daunting task for anyone. But if Trump goes in and doesn't insult someone's dead relatives, it's a win for him. Clinton's got to be on her game, Trump could virtually phone it in. In fact the odds of that aren't too bad: Trump IS known for making a mess of something in the bag, but I hope she is up to it.

It's worth noting that Hillary can crib from all sorts of shit that Bill did, and all the experience therein.

I honestly hadn't noticed even though I saw parts of the debate, but the Clinton/Bush/Perot one in 92 I think?

Turns out Bill sent people in beforehand and switched out the stools to ones like he had been using while practicing, they were tall and unusual compared to the ones that Bush/Perot had been using, so Perot kinda stood next to it, while Bush perched on it unsure how to sit, as Bill lounged casually. Then he makes use of knowing what the camera cuts and such were going to be so he could keep one or both of the other candidates visible in the background, so you catch Perot fiddling around, you catch Bush Sr. looking like he's about to doze off, while Clinton is maintaining his stage persona the whole time, and thus looking professional and presidential by comparison.

The Cuban/Flowers shit is a similar type of game, who knows what else they have planned, heh, apparently Quayle pushed to bring a prop out on stage, he wanted to read from Gores book and pick passages to make him look silly, Gore and his people finally agreed to the "one prop" condition, but said they would be bringing a potatoe.

[smjjames]

@Max, you spelled Potato the same way as Quayle did which he got laughed for.

Anyhow, on the debates, sounds like Clinton could very well be running into the overprepared zone while Trump is apparently doing minimal preparation other than reviewing Clintons past debates (of which there are A LOT) and going over her positions.

Hopefully Clinton doesn't run into the same error that Rubio did and repeatedly use the same canned lines. I can't even imagine her blundering into it the way Rubio did. However, it seems like she has done enough debates to know where she feels comfortable and where she would be overprepared where Rubio was pretty much a rookie. At least Rubio seemed like a rookie and he literally walked into Trumps trap in a later debate.