Jump to content

U. S. Politics: A noun, a verb and no collusion.


LongRider

Recommended Posts

Mind the Gap

A new paper debunks the persistent myths about the racial wealth gap.

https://slate.com/news-and-politics/2018/04/debunking-the-persistent-myths-about-the-racial-wealth-gap.html

 

 

Quote

A new paper from William Darity of Duke University, Darrick Hamilton of the New School, and a team of researchers seeks to debunk these narratives and establish that the racial wealth gap was not produced by supposed black pathology and cannot be closed by individual effort. Darity and Hamilton focus on 10 myths about the racial wealth gap.

 

Link to comment
Share on other sites

@Yukle

Quote

Here's another failure of America: according to 538, looking ahead there's more pain to be had for the Democrats. Changing demographics will see their overall vote share increase, as it has trended since the year 1996, in federal elections. However, due to the utterly stupid Electoral College, they will find it harder to win the White House as the vote increases are unlikely to translate into state wins any time soon.

The study that article is talking about is really interesting and useful stuff.  However, I didn't link it here because I figured its findings would be misinterpreted, even with 538's able summary, and I'd feel compelled to spend a considerable amount of time putting it into context.  Well, screw it, here we go:

1.)  The study is on demographic shifts, and as such is not predictive.  Hell, their findings are not even projections the way that term is known colloquially in political journalism.  They are simulations:

Quote

First and foremost, these are simulations—not predictions. When talking about results under a given set of assumptions, the authors are not expressing the belief that this is what will happen in a given presidential election. At heart, these are thought experiments—revealing tomorrow’s contours under a certain set of assumptions.

Additionally, those assumptions will almost certainly be wrong in ways both big and small. While the expectations the authors have about the underlying race, age, and education makeup of the electorate are probably the soundest of their assumptions, those pertaining to voter behavior are more likely to stray from reality [9]

Basically, what they did is provide estimates for demographic shifts on a state-by-state (even county-by-county) basis.  Then, they essentially mapped the demographic margins in the 2012 or 2016 elections (actually, their data integration process is much more rigorous than this - see "Methodology" starting on pg. 31 - but I'm summarizing) onto these estimates of demographic turnout for a given future election.  

As the authors say, these are really cool thought experiments, but the demographic margins can and will shift from the current electoral makeup - hell, this should be abundantly clear considering they use both the 2016 and 2012 elections as baselines (and actually a third in which the large amount of third party voters in 2016 are estimated to "go home" to their most likely mainstream Party) since there were shifts between those two results!

This is emphasized by both the authors and the 538 writeup repeating the mantra "demography is not destiny."  Take a comparison of the 2012 and 2016 exit polls as an example.  In 2012, Obama won Latina women by 53 points.  In 2016, Clinton won them 44 points.  If you can think of a plausible or even rational explanation for why Trump gained 9 points among Latina women, you're a better analyst than I.  The point is, it is very difficult to predict or even project both future turnout and margins of future elections.  And even as the election approaches and polling makes such predictions more reliable, there will be unpredictable demographic shifts during every election cycle.

2.)  The findings are, as expected, generally very encouraging for the Democrats based on the country becoming less and less white.  The study does not, in the slightest, find Dems "will find it harder to win the White House."  Just look at the top-line results for mapping the three baselines onto the projected demographic shifts in 2020:

Quote

The results of each simulation are discussed in detail below, but the top-line results for 2020 are displayed in Figure 2. These 2020 results may be summarized as follows:

• 2012 baseline: Democrats would win the popular vote by 6 points, and the electoral vote would go to the Democrats 332-206—same as the actual 2012 election.

• 2016 baseline: Democrats would win the popular vote by 3 points, and they would take back Michigan, Pennsylvania, and Wisconsin to carry the Electoral College 279-259.

• 2016 baseline, coming home: Democrats would take the popular vote by a little under 3 points, but there would be a 269-269 tie in the Electoral College. [11] 

So, if the (32) demographic groups voted by the exact same margin as in 2016, but only taking into account the estimated turnout shifts of the study for 2020, the Dems would win the electoral college in 2020.  This demonstrates how narrow Trump's EC victory was (particularly in the aforementioned states).  It's only when third party voters "come home" that estimates an EC tie - this reflects the fact that more third party voters in 2016 tend to vote GOP rather than Dem, which makes sense as more intraparty "leaners" were turned off by Trump than Clinton.

In fact, when you look at all 16 of their different scenarios (see Figure 2, 13), in only 4 of their simulations does the GOP win the EC:  when minorities shift to GOP; when uneducated whites shift to GOP; when educated whites shift to GOP; and when the GOP trades off gains in uneducated whites for losses in educated whites (more on this last one later). 

Overall, these scenarios actually show that if the Dems can simply hold steady, or make any type of marginal gains in the various demographic groups, the expected shift in demographic turnout should ensure their victory in 2020.

3.)  What these findings do show is how integral uneducated whites are as a demographic group, and how they've been underestimated in exit polling:

Quote

The authors’ synthetic data analysis indicates that this group made up about 44 percent of those who showed up at the polls in 2016, compared with 34 percent in the national exit polls.

This finding underscores a great truth: Whites without a college degree are still a massive demographic group and even small shifts in their voting behavior can have dramatic effects on American politics. [23]

44 percent of the electorate is huge.  You don't need a study to know if Dems continue to lost that group 66-29 like they did in 2016, they're gonna be in trouble - even as uneducated whites decreases as a percentage of the electorate.  This is why it's crucial to try to maintain competitiveness with this group, rather than write them off as hopeless racists.  In fact, this study estimates if Dems can simply get the uneducated white margins back to 2012 levels, that would flip 7 swing states (WI, MI, PA, IA, FL, NC, OH) back to the blue column giving them 347 EC votes in 2020 (see Figure 8, 25).

What this Figure does show, however, is if uneducated whites continue to shift towards the GOP, then the Dems would be screwed in the EC.  However, what that simulation doesn't take into account is the "tradeoffs" part of their scenarios.  If the GOP continues to double down on being the party solely of white racial resentment, then it would be a rather safe assumption we'd see corresponding shifts towards Dems among both minorities and educated whites.  When those tradeoffs are simulated, the Dems regain their EC advantage (see Figure 9, 28).

4.)  The other key finding of this study, in my book, is the importance of black turnout (and margins).  If the Dems can generate the same type of enthusiasm black voters had for Obama, this can help secure the EC.  Reverting back to black turnout and support in 2012 would give the Dems 339 EC votes in 2020 - flipping NC, GA, PA, MI, FL, WI - and an estimated 350 votes by 2036 (see Figure 6, 20).

5.)  The final key thing to consider is the aging population:

Quote

Second, the population is aging. Those 65 years old and older will make up a larger share of EVs—going from 21 percent in 2016 to 22 percent in 2020 and to 27 percent in 2036—while those ages 18 through 64 will shrink. While seniors will continue to be less racially diverse than younger age groups over this period, white senior EVs as a share of all EVs will rise by only a single percentage point from 2020 to 2036. The overwhelming majority of growth in the 65 and older age category is going to come from people of color. [6]

This study doesn't (and can't, really) take into account how these individuals will be demographically cross-pressured in terms of partisanship trends.  Nobody can figure out right now how these dynamics will interact with each other and how each party reacts to a population that is at once older, more diverse, and more educated.

6.)  No matter what the demographic trends suggest, it's rather silly to refer to demography as a "failure" on behalf of America.  The electoral college is a complete abject disgusting failure, but this is hardly news.  The trends among uneducated whites is independent of the EC, certainly more subject to change, and yes, it's incumbent on Dems to either recoup their losses or at least stop the bleeding there.  But again, this is known.

Link to comment
Share on other sites

31 minutes ago, dmc515 said:

This is emphasized by both the authors and the 538 writeup repeating the mantra "demography is not destiny."  Take a comparison of the 2012 and 2016 exit polls as an example.  In 2012, Obama won Latina women by 53 points.  In 2016, Clinton won them 44 points.  If you can think of a plausible or even rational explanation for why Trump gained 9 points among Latina women, you're a better analyst than I.  The point is, it is very difficult to predict or even project both future turnout and margins of future elections.  And even as the election approaches and polling makes such predictions more reliable, there will be unpredictable demographic shifts during every election cycle.

The obvious answer is that the ones that voted for Obama stayed home, especially in states that added significant voter suppression techniques. 

31 minutes ago, dmc515 said:

In fact, when you look at all 16 of their different scenarios (see Figure 2, 13), in only 4 of their simulations does the GOP win the EC:  when minorities shift to GOP; when uneducated whites shift to GOP; when educated whites shift to GOP; and when the GOP trades off gains in uneducated whites for losses in educated whites (more on this last one later). 

But...three of those four scenarios are the current existing balance of the GOP, and while educated whites appear to be shifting a bit towards dems, uneducated whites are super for Trump's GOP. 

31 minutes ago, dmc515 said:

5.)  The final key thing to consider is the aging population:

This study doesn't (and can't, really) take into account how these individuals will be demographically cross-pressured in terms of partisanship trends.  Nobody can figure out right now how these dynamics will interact with each other and how each party reacts to a population that is at once older, more diverse, and more educated.

The problem I mostly have with this is ignoring a lot of the undemocratic processes that are being encouraged and put in place - or removing existing democratic processes - and ignoring that. It's not a surprise that in many of the states that instituted voter suppression we saw a larger percentage of minorities voting for Trump and a smaller overall turnout of minorities - but that's just the tip of the iceberg. We still haven't figured out how to stop another nation-state from influencing the election rather overtly, we still have a POTUS and a party that is singularly uninterested in doing anything about that issue, and every report we get shows that it was far worse than anyone actually thought. 

But yeah, demographically things are looking vaguely okay-ish for the Democrat party, provided that they also work at being something for the white people. It's all the other things that suck. 

Link to comment
Share on other sites

28 minutes ago, dmc515 said:

3.)  What these findings do show is how integral uneducated whites are as a demographic group, and how they've been underestimated in exit polling:

White, noncollege educated voters making up 44% of the electorate instead of 34% (as exit polls estimated) is a huge error.  And really that group is Trumps entire base.  He did significantly worse than Romney with educated white voters (they voted R+4 in 2016, ten points worse than Romney's R+14)  and any improvement Trump saw in his performance with minorities is minimal and likely the result of either exit polling error or voter disengagement without Obama on the ballot. 

The big question that noone really knows is if Trump can repeat that trick of both a +35 margin and very strong turnout amongst white working class voters.  Because the numbers for him in essentially every other group have gotten worse since Nov 2016.  Since Trump won, minorities and educated white voters have both been voting at higher rates in special elections and in VA/NJ.  In contrast, Trump has been able to generally maintain the margin with white working class voters, but not the high turnout. 

If in 2020, Trump wins white working class voters by +35, but turnout in that group is a bit lower than we saw in 2016, then there's little chance of Trump winning reelection.  Because he has been doing a very poor job of building bridges with any new constituencies.  He needs to keep that group both very loyal and very motivated.  We'll see if he can do it. 

Link to comment
Share on other sites

It's so deeply unfeminist that the very walls of my office are bearing down on me in anger, but Pamela Brown's lip gloss was just awful for TV. Her lips were reflecting so much light that I think my left eye is blinded.

No CoLLUSION!

Link to comment
Share on other sites

[mod] Quick moderator note.

We tend to give this thread a bit more latitude than others because people often get understandably passionate, because generally the standard of discussion is strong, and because people don't abuse the privilege. Recently people have been abusing the privilege.

I'm not going to single out people because that is the first thing that has to go, for everyone. Talk about the issues, not about each other. No more singling people out, and no personal attacks. Stick to the topic. Which is US politics, by the way. Digressions should be short and justified, if they happen at all.

Be polite and be on point. Thank you. [/mod]

Link to comment
Share on other sites

24 minutes ago, Kalbear said:

The obvious answer is that the ones that voted for Obama stayed home, especially in states that added significant voter suppression techniques.

First, they're both national polls, so states really don't have anything to do with it.  Second, the latina female turnout was 6% in both elections, so there's nothing there to suggest turnout shifts.  Without corresponding evidence - or even rationale for why - simply writing it off a nine point shift as "GOP supporters turned out more and Dem supporters stayed home" is an unfounded (and rather lazy) assumption, not an explanation.

27 minutes ago, Kalbear said:

But...three of those four scenarios are the current existing balance of the GOP, and while educated whites appear to be shifting a bit towards dems, uneducated whites are super for Trump's GOP. 

Um, those are all proposed shifts from the existing 2016 margins, so it's impossible to say any of the scenarios are the current balance of the GOP.  Perhaps you're misunderstanding.  What you could say is the GOP enjoyed a shift in more support for uneducated whites - and even in 2016 compared to 2012 saw gains among minorities - but there's no reason to assume these shifts in margins will continue in the same direction.  That's why I said the Dems simply need to "hold steady" or "stop the bleeding" in terms of margins based on the study's turnout estimates for 2020.

31 minutes ago, Kalbear said:

The problem I mostly have with this is ignoring a lot of the undemocratic processes that are being encouraged and put in place - or removing existing democratic processes - and ignoring that.

I'm not trying to ignore the undemocratic processes.  Rather, I'm trying to clarify what that study actually found, which clearly can be considerably misinterpreted.

22 minutes ago, Maithanet said:

White, noncollege educated voters making up 44% of the electorate instead of 34% (as exit polls estimated) is a huge error.

Yeah it is.  These firms really to work on that problem.

23 minutes ago, Maithanet said:

If in 2020, Trump wins white working class voters by +35, but turnout in that group is a bit lower than we saw in 2016, then there's little chance of Trump winning reelection.

Right, it doesn't have to have a huge shift because of the size of that demographic.  If Dems can shift it from 66-29 to 63-32 - whether that's generated from a turnout shift or a shift in preferences (probably a little of both), that could mean a huge difference (even controlling for gains among minorities/educated whites).

Link to comment
Share on other sites

22 minutes ago, Maithanet said:

White, noncollege educated voters making up 44% of the electorate instead of 34% (as exit polls estimated) is a huge error. 

The authors of the original report seem to think this is because educated voters are simply more likely to answer pollsters' questions. But I still wonder if many of the exit polls aren't asking their questions about education too vaguely, so that they are actually including many people who do not have four year college degrees but who instead have associate's degrees from community colleges (or are otherwise in the "some college" group) in their "college educated" category. If some of these exit polls are simply asking people "Do you have a college degree?", I bet many of the millions with AA degrees will say "Yes."

Link to comment
Share on other sites

7 minutes ago, dmc515 said:

Right, it doesn't have to have a huge shift because of the size of that demographic.  If Dems can shift it from 66-29 to 63-32 - whether that's generated from a turnout shift or a shift in preferences (probably a little of both), that could mean a huge difference (even controlling for gains among minorities/educated whites).

That's one way, but I think that what we're seeing with special elections (and is likely to continue in 2018 at least) is that Trump is still putting up incredible margins in Trump country, he just isn't getting the turnout he needs.  For example, in both AL Senate and PA-18, Moore and Saccone both hit or exceeded their predetermined "county margin benchmarks" that 538 put out.  The problem was that turnout was lower in those districts than it was in the more democratic leaning areas.  The 538 Alabama Senate Live Blog discussed that extensively.  The 10:40 post shows Jones hit his "county target" in terms of margin in only 13 of 67 counties.  But the 10:57 post (after Jones was declared the winner) shows that turnout was always way higher in Pro-Jones areas, and the very rural, very white areas had the lowest turnout. 

Link to comment
Share on other sites

14 minutes ago, Maithanet said:

That's one way, but I think that what we're seeing with special elections (and is likely to continue in 2018 at least) is that Trump is still putting up incredible margins in Trump country, he just isn't getting the turnout he needs.

It's possible, but I refer you (again, think we've discussed this already) to this article that shows Dems are (albeit very slightly to a null result) overperforming more in poorer and less educated districts/states.  Could that be primarily due to a shift in turnout the way you're describing it?  Sure, but it could also be because even some uneducated whites are shifting preferences on Trump.  Special elections can only tell us so much.

30 minutes ago, Ormond said:

The authors of the original report seem to think this is because educated voters are simply more likely to answer pollsters' questions.

I think another thing to consider is the study above uses ANES and CCES data.  I would suspect that with exit polls in which the respondent often is right outside the precinct, there may be increased social desirability bias to (falsely) claim you have a college degree.

Link to comment
Share on other sites

1 minute ago, Tywin et al. said:

And in a moment of cowardice, 73 years of history making goes out the window...

History is written by the survivors.

Have fun going to war, fuckboi :kiss: I'm sure you'll die a hero or whatever. Feel free to send me letters, but don't kid yourself expecting a response.

Link to comment
Share on other sites

15 minutes ago, Pony Empress Jace said:

History is written by the survivors.

Have fun going to war, fuckboi :kiss: I'm sure you'll die a hero or whatever. Feel free to send me letters, but don't kid yourself expecting a response.

Fun fact, if not for a clerical error I would have joined the Navy after college. I guess the life of an intelligence officer was not meant to be.

That said, I doubt this will have any major impact on whether or not we start any new wars. Pompeo is a hawk, no doubt, and an aggressive one at that, but ultimately Trump’s just going to do whatever he wants.

Link to comment
Share on other sites

1 hour ago, dmc515 said:

First, they're both national polls, so states really don't have anything to do with it.  Second, the latina female turnout was 6% in both elections, so there's nothing there to suggest turnout shifts.  Without corresponding evidence - or even rationale for why - simply writing it off a nine point shift as "GOP supporters turned out more and Dem supporters stayed home" is an unfounded (and rather lazy) assumption, not an explanation.

It's a hypothesis, not an assumption or an explanation. Without knowing other data such as actual turnout numbers it's a reasonable one as well. And given that the overall hispanic voting-eligible bloc grew by 37% during that time, it's still a reasonable one to consider; as a percentage of total voters it was unchanged, but as a percentage of eligible population it went down. In particular, it was expected that it would increase tremendously and did not

Essentially the weirdest thing about the election is that in spite of all demographic trends increasing as a share of population in hispanic and AA voting blocs, the actual percentages were almost identical to 2012. When balancing for demographic trends that suggests a major surge in white voters and a downturn in everyone else. 

The other possibility that this article suggests is that while Clinton didn't get as many Latina voters, Trump wasn't the lucky recipient either - he got basically the same number as Romney did, and even fewer than McCain did. 

1 hour ago, dmc515 said:

Um, those are all proposed shifts from the existing 2016 margins, so it's impossible to say any of the scenarios are the current balance of the GOP.  Perhaps you're misunderstanding.  What you could say is the GOP enjoyed a shift in more support for uneducated whites - and even in 2016 compared to 2012 saw gains among minorities - but there's no reason to assume these shifts in margins will continue in the same direction.  That's why I said the Dems simply need to "hold steady" or "stop the bleeding" in terms of margins based on the study's turnout estimates for 2020.

No, I understood; my point is that thinking that 2016 is the blip instead of the trend is wishful thinking. Again, you're basing methodologies on the same firms that missed the estimate of the actual uneducated white voter turnout by 33%. That's kind of insane, and it's also insane to think that this is something that will happen just once. Populists once they gain power don't tend to lose it from that base unless they turn truly vile towards that base, and so far that's not happened - nor has polling indicated that it's happened among that base at all. 

 

Link to comment
Share on other sites

2 hours ago, dmc515 said:

@Yukle

The study that article is talking about is really interesting and useful stuff.  However, I didn't link it here because I figured its findings would be misinterpreted, even with 538's able summary, and I'd feel compelled to spend a considerable amount of time putting it into context.  Well, screw it, here we go:

1.)  The study is on demographic shifts, and as such is not predictive.  Hell, their findings are not even projections the way that term is known colloquially in political journalism.  They are simulations:

Basically, what they did is provide estimates for demographic shifts on a state-by-state (even county-by-county) basis.  Then, they essentially mapped the demographic margins in the 2012 or 2016 elections (actually, their data integration process is much more rigorous than this - see "Methodology" starting on pg. 31 - but I'm summarizing) onto these estimates of demographic turnout for a given future election.  

As the authors say, these are really cool thought experiments, but the demographic margins can and will shift from the current electoral makeup - hell, this should be abundantly clear considering they use both the 2016 and 2012 elections as baselines (and actually a third in which the large amount of third party voters in 2016 are estimated to "go home" to their most likely mainstream Party) since there were shifts between those two results!

This is emphasized by both the authors and the 538 writeup repeating the mantra "demography is not destiny."  Take a comparison of the 2012 and 2016 exit polls as an example.  In 2012, Obama won Latina women by 53 points.  In 2016, Clinton won them 44 points.  If you can think of a plausible or even rational explanation for why Trump gained 9 points among Latina women, you're a better analyst than I.  The point is, it is very difficult to predict or even project both future turnout and margins of future elections.  And even as the election approaches and polling makes such predictions more reliable, there will be unpredictable demographic shifts during every election cycle.

2.)  The findings are, as expected, generally very encouraging for the Democrats based on the country becoming less and less white.  The study does not, in the slightest, find Dems "will find it harder to win the White House."  Just look at the top-line results for mapping the three baselines onto the projected demographic shifts in 2020:

So, if the (32) demographic groups voted by the exact same margin as in 2016, but only taking into account the estimated turnout shifts of the study for 2020, the Dems would win the electoral college in 2020.  This demonstrates how narrow Trump's EC victory was (particularly in the aforementioned states).  It's only when third party voters "come home" that estimates an EC tie - this reflects the fact that more third party voters in 2016 tend to vote GOP rather than Dem, which makes sense as more intraparty "leaners" were turned off by Trump than Clinton.

In fact, when you look at all 16 of their different scenarios (see Figure 2, 13), in only 4 of their simulations does the GOP win the EC:  when minorities shift to GOP; when uneducated whites shift to GOP; when educated whites shift to GOP; and when the GOP trades off gains in uneducated whites for losses in educated whites (more on this last one later). 

Overall, these scenarios actually show that if the Dems can simply hold steady, or make any type of marginal gains in the various demographic groups, the expected shift in demographic turnout should ensure their victory in 2020.

3.)  What these findings do show is how integral uneducated whites are as a demographic group, and how they've been underestimated in exit polling:

44 percent of the electorate is huge.  You don't need a study to know if Dems continue to lost that group 66-29 like they did in 2016, they're gonna be in trouble - even as uneducated whites decreases as a percentage of the electorate.  This is why it's crucial to try to maintain competitiveness with this group, rather than write them off as hopeless racists.  In fact, this study estimates if Dems can simply get the uneducated white margins back to 2012 levels, that would flip 7 swing states (WI, MI, PA, IA, FL, NC, OH) back to the blue column giving them 347 EC votes in 2020 (see Figure 8, 25).

What this Figure does show, however, is if uneducated whites continue to shift towards the GOP, then the Dems would be screwed in the EC.  However, what that simulation doesn't take into account is the "tradeoffs" part of their scenarios.  If the GOP continues to double down on being the party solely of white racial resentment, then it would be a rather safe assumption we'd see corresponding shifts towards Dems among both minorities and educated whites.  When those tradeoffs are simulated, the Dems regain their EC advantage (see Figure 9, 28).

4.)  The other key finding of this study, in my book, is the importance of black turnout (and margins).  If the Dems can generate the same type of enthusiasm black voters had for Obama, this can help secure the EC.  Reverting back to black turnout and support in 2012 would give the Dems 339 EC votes in 2020 - flipping NC, GA, PA, MI, FL, WI - and an estimated 350 votes by 2036 (see Figure 6, 20).

5.)  The final key thing to consider is the aging population:

This study doesn't (and can't, really) take into account how these individuals will be demographically cross-pressured in terms of partisanship trends.  Nobody can figure out right now how these dynamics will interact with each other and how each party reacts to a population that is at once older, more diverse, and more educated.

6.)  No matter what the demographic trends suggest, it's rather silly to refer to demography as a "failure" on behalf of America.  The electoral college is a complete abject disgusting failure, but this is hardly news.  The trends among uneducated whites is independent of the EC, certainly more subject to change, and yes, it's incumbent on Dems to either recoup their losses or at least stop the bleeding there.  But again, this is known.

I wrote with some ambiguity; for me the failure isn't demographics, it's the EC and nothing else. There is no reason that America's presidential elections should be decided by a group that doesn't represent the electorate at large, but disproportionately rewards only key parts of it. Whoever wins the election should actually win the election. I don't care if a particular demographic votes one way, another way, or another way, just that whatever the outcome is should be the mandated actual result. Instead, what's happening is that states with large populations but huge Republican majorities are seeing their vote-shares become more competitive. But it doesn't translate into any actual changes to the outcome.

I was also musing on the fact that the lower-educated white vote was about 44% of the electorate. It was interesting as I think that most pundits assumed that, while huge, the Republicans had more or less hit the ceiling with how much support they would ever get among that group. Therefore it was a surprise in some respects when Trump won it by a much bigger margin than expected; he pretty much didn't need any other core group to vote for him, just a little more here and there.

The study has a bit of an information void which often happens with any demographics study - and it's one that isn't easy to fix - which is that it's hard to know the overlap between any group. For instance, to use your example of female Latina voters, how many of those who voted for Trump also voted for Obama? Was it that one group who voted in 2012 didn't vote in 2016, while another who didn't in 2012 did in 2016? It's really hard to account for this.

I can't remember if it was the same group or not, but a few years ago there was a similar effort: using projections based on demographic shifts (and you're right, it's projections, not predictions) and it said that states like Pennsylvania and Maine were trending away from being solid blue to becoming competitive. Similarly, it said that Texas should be similarly purple, or at least much more so than now, but there is a noticeable discrepancy between its demographics as a state and its voting population (my assumption is strict voter suppression laws have some impact).

For the most part, these demographic shifts should be having more impact than they actually are. The idea that candidates need to flip "states" in a federal election is ridiculous. There should be nothing other than the popular vote underlying the election. The changes that America is witnessing aren't having enough of an impact on their federal government. I don't just blame the EC, of course, there's also the fact that electoral boundaries are drawn by the parties and voting isn't universally accessible, but at the moment the major parties don't need to pay as much attention as they should to demographics, because for the most part they're not aiming to win demographics. They're aiming to win only sub-sections of them within particular states, at the expense of paying any attention to the country at large.

In summary, I don't care about shifting demographics, I care that these changes don't have as much impact as they should. Abolish the EC, gerrymandering and voter suppression!

(And when you've done those really easy tasks, let's move onto gun violence, and the day after solving that piece of cake, there's some sort of kerfuffle going on between Iranians and Saudis).

Link to comment
Share on other sites

4 minutes ago, Fragile Bird said:

I just saw a story that not one Democrat has been invited to tomorrow's state dinner for French president Macron.

Isn't that just...weird?

"Ill" igetimate DEMS obstructionust whiners! Their ATTACK on our institutions will not be at "state" dinner.

NO COLLUSION

WALL

Link to comment
Share on other sites

15 minutes ago, Kalbear said:

It's a hypothesis, not an assumption or an explanation. Without knowing other data such as actual turnout numbers it's a reasonable one as well. And given that the overall hispanic voting-eligible bloc grew by 37% during that time, it's still a reasonable one to consider; as a percentage of total voters it was unchanged, but as a percentage of eligible population it went down. In particular, it was expected that it would increase tremendously and did not.

No, it's still an assumption, although at least it's a founded one now.  The increase in eligible voters without a corresponding increase in turnout should lead one to assume latinos are being disproportionately disenfranchised due to voter id laws/voter suppression.  That would lead to a hypothesis (that I'd agree with) that the gap between eligible voters and turnout among latinos is significantly higher within states with high levels of voter suppression.  Even then, that doesn't explain why the remaining voters tend to support Trump at higher rates - and it especially doesn't explain why Latina women saw a 9 point shift while Latino men only saw a 1 point shift during the same time period in the same polling.

21 minutes ago, Kalbear said:

No, I understood; my point is that thinking that 2016 is the blip instead of the trend is wishful thinking.

I really (really) don't think assuming the marginal gains Trump made among minorities are not going to be exacerbated is wishful thinking.  In fact I think assuming he's going to continue to make gains among minorities is rather ridiculous.  As for his gains among uneducated whites, I certainly think it's possible for the Dems to stem the tide.  66-29 is a pretty big number, one it's plausible to think is a ceiling and had as much to do with the economy as racial resentment.

24 minutes ago, Kalbear said:

Again, you're basing methodologies on the same firms that missed the estimate of the actual uneducated white voter turnout by 33%.

See, this continues to demonstrate you are misunderstanding.  The study in question were the one's that estimated the actual turnout at 44%, and it was based on firms I trust much more than exit polls - namely the ANES and CCES.  It's the exit polls that got it wrong, and that's what the study details.

27 minutes ago, Yukle said:

I wrote with some ambiguity; for me the failure isn't demographics, it's the EC and nothing else.

[...]

 

In summary, I don't care about shifting demographics, I care that these changes don't have as much impact as they should. Abolish the EC, gerrymandering and voter suppression!

(And when you've done those really easy tasks, let's move onto gun violence, and the day after solving that piece of cake, there's some sort of kerfuffle going on between Iranians and Saudis).

Ok, but that's not really what that study was about, so I wanted to put it into context - and also point out that the findings in actuality (generally) favored Dems.  As to the bolded, sounds like so much fun...

1 hour ago, Pony Empress Jace said:

Rand Paul once again proves he's the hero Libertarians deserve.

He's gonna vote Yes on Pompeo

Meh, once Heitkamp, Donnelly, and Manchin announced they were voting for him it didn't matter anyway.  And I don't blame any of those three for supporting him - hard to explain in an election year why you voted for Pompeo for CIA but not for State.  The flip-flop ads write themselves.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Guest
This topic is now closed to further replies.
×
×
  • Create New...