Presidential Elections, TMBSL History (362 views)

[eric]

Whom did you vote for in the last elections, and for whom do you plan to vote in this election?

[Odin]

I see no option for Gary Johnson

[Lazy Pete]

Johnny mac, mittens and rubby

[kn88]

Jackson, Keyes, Obama, Carson

[kn88]

but for real, none.

[eric]

so far it looks like TMBSL as a whole leans right: got both Bush elections correct, predicted much closer results in the Obama elections

[eric]

There was some talk in shout yesterday of the socially liberal fiscal conservative. Data on this is surprisingly hard to come by but I did find one duo that indicates some interesting underlying phenomena. Here are two results from Gallup based on the same poll. It frustratingly bins in a couple places but we can still work with that. What I want to draw your attention to is this: we know from the first result how many Democrats and Democrat leaning independents self-identify as socially lib/mod/con and fiscally overall, and we know from the second how many self-identify in each combo. If the axes are truly independent, we should be able to multiply across the results in the first to obtain the results in the second. For example, if 45% of this group are fiscally liberal and 53% are socially liberal, then 45% * 53% = 24% should be both fiscally liberal and socially liberal. Here is what actually happens:

fcsmsl fmsl flsl
fmsm
fcsc fmsc flsmsc

12 22 25
18
7 5 7

16 17 24
10
3 5 20
Is that crazy or what? People are dramatically less likely to self-identify as fiscally liberal if they have to at the same time say they are socially moderate or conservative, and that percentage drops mostly into fiscally moderate but even into fiscally conservative. Meanwhile social liberals have no problem saying they're also fiscally liberal, but they're slightly more likely to say fiscally moderate than conservative than they would if you just ask them where they stand fiscally. This effect is much smaller than the first, clearly, but in both cases we see that people want to self-identify towards the coherent axis: liberal and liberal, moderate and moderate, conservative and conservative. Is this a trick of grammar, or are people not thinking it through (for one or both questions)? And either way, what does that tell you about how well you can rely on peoples' self-identification?

[SPL]

Clinton, Nader, Kerry, McCain, Romney and ???

[SugarShaun]

I've voted republican in all the presidential elections since I've been of age but it depends on who win the nominations, I'm not opposed to voting for a democrat even though I'm a registered republican

[Odin]

Mar 2, 2016 10:21:10 GMT -6 eric said:

There was some talk in shout yesterday of the socially liberal fiscal conservative. Data on this is surprisingly hard to come by but I did find one duo that indicates some interesting underlying phenomena. Here are two results from Gallup based on the same poll. It frustratingly bins in a couple places but we can still work with that. What I want to draw your attention to is this: we know from the first result how many Democrats and Democrat leaning independents self-identify as socially lib/mod/con and fiscally overall, and we know from the second how many self-identify in each combo. If the axes are truly independent, we should be able to multiply across the results in the first to obtain the results in the second. For example, if 45% of this group are fiscally liberal and 53% are socially liberal, then 45% * 53% = 24% should be both fiscally liberal and socially liberal. Here is what actually happens:

fcsmsl fmsl flsl
fmsm
fcsc fmsc flsmsc

12 22 25
18
7 5 7

16 17 24
10
3 5 20
Is that crazy or what? People are dramatically less likely to self-identify as fiscally liberal if they have to at the same time say they are socially moderate or conservative, and that percentage drops mostly into fiscally moderate but even into fiscally conservative. Meanwhile social liberals have no problem saying they're also fiscally liberal, but they're slightly more likely to say fiscally moderate than conservative than they would if you just ask them where they stand fiscally. This effect is much smaller than the first, clearly, but in both cases we see that people want to self-identify towards the coherent axis: liberal and liberal, moderate and moderate, conservative and conservative. Is this a trick of grammar, or are people not thinking it through (for one or both questions)? And either way, what does that tell you about how well you can rely on peoples' self-identification?

Sometimes i think you just throw shit in a box and say "wow, look at that!"