many sources for this development, and the reasons people are increasingly likely to hold negative feelings about the opposite party. this growth in negativity means there is this growing reserve of fear, anger, and distrust among the public that the ted cruzes and alan graysons around the world can mine for political profit. i was talking about the paper about partisan warriors in congress. i think at the mass level, the people these partisan warriors may be responding to are the folks who dislike the opposite party the most. there are political opportunities here. the downside is if politicians are taking their signals from that segment of the public, which says, we do not like the opposite party. we do not want to to compromise with them at all. that contributes to the crisis over the government shutdown,

and these type of government crises are likely to occur in the near future as long as these high rates of negativity toward the opposite party persist. >> thank you very much. should i take the handheld mic? i am going to launch the questions -- one of my own. i do not want to hog the microphone. you have this conference every four years, right after the election. i am going to guess that new jersey and virginia get a lot of analysis every four years. i hope this is not too far off topic. as a political reporter, i am very interested in your take on what happened two days ago. in new jersey, you had governor christie shattering the gop gender gap in a solid blue state. he had 57% of the women vote, a third of democrats, a majority of latinos, and a third of union

voters. i wonder if his victory shows republicans a way out of the trend toward partisan polarization. or is his model tied to him personally and not widely duplicatable, nationally or in other states? would anyone like to try their hand at that one? >> i will jump into that, and i will make a bold prediction right now, which is that chris christie will not be the republican nominee in 2016. i would not say it is impossible, but i think it would take an unusual set of circumstances, where you have a very divided and weak field of candidates to his right that was splitting the vote, although we have seen that play out in republican primaries. in 2016 -- and to be fair, christie is not another

giuliani. i see him as somewhat like rudy giuliani in terms of his appeal, and he was the front-runner for the republican presidential nomination in 2007. given where the republican party is now, or having to run the gauntlet of the republican primaries in 2016, first of all, he is going to have to reposition himself, just as john mccain and mitt romney did. secondly, i do not think he is going to be able to do it sufficiently or successfully enough. it is hard to predict what is going to happen. >> i would say chris christie has more drive than giuliani did. giuliani ran, but did not really give it his all. >> he was sort of boxed in. he chose to skip iowa, which sort of made sense. what he also skipped new hampshire, which did not.

mccain could not run a much worse campaign then giuliani did. on the other hand, the republican party, if anything -- the base is even more conservative now. the tea party was not there yet, and now it is. and the base has gotten more conservative. i think he is going to have a lot of trouble duplicating that success. >> does anybody else want to chime in? >> i will try. i think there are two problems with giuliani -- excuse me -- with christie as candidate. ideologically, i do not think he will be acceptable to the conservatives in the republican party who dominate the primary. the second problem he faces is that he does not seem to epitomize, symbolically, what americans are looking for, like barack obama or michelle obama. both look physically fit. i would say that is one thing he should work on.

>> he is working on it. >> anybody else? >> to be a slight devil's advocate, the way he is different than giuliani is, he is pro-life. he matches -- if i said to you an antiunion, pro-life, economic conservative is running for president and he won a blue state, you would say, that is chris christie. his issue preferences are largely there to appeal to blue voters. it is kind of the public being kind to the president when the president comes after a hurricane that he is getting trouble from people on the far right. i am not sure that stuff is durably negative toward him. in terms of his fitness, there is research that suggests people are more positive about larger people than bitter people, so i am not sure whether he should be eating more and dieting less.

>> david, did you want to comment? >> the -- i would say that christie -- not to take too much from his reelection. the democrats sat this out. the democratic donors, operatives. he has not had serious partisan fire directed at him yet. when he becomes a serious nominee, we will see. >> one follow-up, and then open it up to the audience. cuccinelli lost. did that signal to the republican party that you cannot put up a tea party-oriented campaign? >> not to the tea party people. reading their comments the last couple of days, they are almost all along the lines of, we were sold out by the establishment. they thought he could have won, given the close margin. >> there is no learning from that? >> i do not think so.

>> i think the republicans have to learn how to woo and recruit the libertarians. the libertarian candidate was the deciding difference. there was only about a 5000 vote margin. >> actually, i do not think that is true. you look at the pre-election polls and the exit polls -- it was a protest vote, largely. those votes would have split almost evenly. i do not think, in this case, that that was so much an ideological vote so much as -- it was young people and independents who did not like either candidate, which was a large group. you had a large potential pool of people to appeal to. >> now, i would like to -- a hand is already up. i would like you to give us your name and your affiliation before your question. thanks. >> i am bill connoly.

thank you for an excellent panel. his first question is directed toward david and alan, because you both alluded to the issue. you talked about the growing intensity in the affective polarization, and the negative image of the opposite party. i am wondering to what extent -- and i welcome the entire panel addressing the issue -- this is a function of the changing media environment, the greater fragmentation of the media and the capacity for us to cocoon or cul-de-sac in our own comfort zone with particular media outlets. david, you began to touch on that. i am interested in the panel addressing that in a greater light. >> i think you raised a good

point. i do not think our measures quite get at you are talking about, which is more social networks and maybe who people discuss politics -- which i know some other folks here studied. i think it is worth analyzing. i do not think we have the evidence to answer your question very well. >> ok. over here? right here. >> dave mackelbie from byu. you looked at negative movements toward the parties. i wonder if you see the same negative feelings toward the candidates. if so, why would you pick her keys rather than candidates as the major marker? for david, the one slide where you look at negative feelings over time, it looked like the negativity was more pronounced for republicans toward the

republican party then toward the democratic party, in terms of change. i want to make sure i did not misread that slide. if so, why have negative feelings grown much greater toward the republicans? >> the answer is that it shows the same trend, if you look at the candidate affect measures over time. they are a little more variable. because this is the state of the parties now. [laughter] it is not the state of the candidates conference. more seriously, the parties are the longer-term -- they continue from election to election. that is us looking for long-term trends. you do get more fluctuation when you look at candidate affect. some of that is candidate- specific. although of course it is strongly correlated with partisan affect. and the overall trend is

definitely in the same direction. >> we looked at the presidential thermometer ratings and found similar -- the same trends alan mentioned. there are questions that asked how angry does a democratic candidate make you, and those trended up over time, with 2012 being the record anger and fear cycle. we found the same pattern of declining ratings for both democrats rating the republican party and republicans rating the democratic party. on average, maybe, democratic ratings of the republican party are a point or so lower on average than republican ratings of the democrats, but differences are pretty small. >> up front.

>> thank you very much. this one is for joel. >> could you identify yourself? >> i am a graduate student with a masters in political science at the university of akron. you talked about -- this is about polarization, after all. and you showed your information on the asian population, they seemed to trend 42-43 the republicans to the democrats. that was amazing to me. and your study of that, what drove the asian population to be republicans 42%, and what drove the 43% to the democrats? how is that so different when it seems like they live in areas that are quite homogenous? >> excellent question. i have a colleague that is asian-american. he lives in an upper-class suburb on the eastside. if you look at census statistics, you find that asian- americans are the most educated

racial group in america, and they have the highest levels of education. they also have the highest levels of income. about $15,000 or more above the income for white families. i think that is a major factor. where you stand depends upon where you sit. i think asian-americans sit very comfortably, because they work hard and they are successful. >> was your data -- as i recall, when you presented it, you had combined data over a long time in that table. and i think i could be misleading, in the sense that i think if you look at more recent data -- we know what happened in the 2012 election, according to the exit polls, and i think this is true in 2008 as well. the asian vote went heavily democratic. and some of those other groups have shifted over time. the americans -- there has been research recently.

people looked at a group of people who say "i am just american." they are increasingly republican and in places like appalachia in large concentrations. that is a group that i think is trending in the opposite direction with this growing ethnic and racial heterogeneity. >> i think that is a very valid point. doing the first quick and dirty cut, we did not have significant differences, and i could not present this in just one slide. i did not have time. i think that is something we need to explore, to look at changes over time. we did not see significant ones initially, but as alan points out, everything is in flux. >> over here. >> i am from the university of texas, and i have a comment and

question for the ignored paper up until now in the comments section. i thought it was ingenious of you to show that one chart, the dimensional chart. i thought that really compellingly presents who is participating in the partisan debate and who is not. i wonder if you can speak a little bit to my question now, a bit of the cause and effect of that. we know they are voting less, but are they voting less because the parties are not speaking to them in a consistent way that they can engage the debate? could you talk about the cause that affects participation in the process? >> i think they are responding to the choices that they are given. i think you particularly see this among the group we labeled as populist. i do not think they are seeing candidates that represent them consistently. of all the groups we have looked at, they seem to be the ones that participate the least, not only with respect to voting, but other campaign activities.

i think they look at the choices they are presented with and do not get excited about it, and are less likely to turn out as a result. libertarians are a little bit different. they tend to turn out. the populists and moderates stay away. i think they are being turned off by the choices. >> i am a grassroots democrat activist. you all study the past. i am interested in your gut feelings about the future. is some event going to happen to break this up? i know the demographics say the democrats are going to win more votes unless something else happens. i am not directing this to any to anyone who wants to comment on that. >> within the context of our paper, one we looked at these different dimensions, we had

some evidence looking at hispanics, even the attention they have received lately. what we find is, they look like the rest of the american public with respect to these dimensions. it is not like the democrats can say, we are going to grab the hispanics. the hispanics have diversity with respect to many different issues. there are our core issues like immigration that matter to that group probably more than a lot of other issues. but it is not something the democratic party can do to grab the diversity within those groups. >> i have to partially disagree with that. if you look at hispanics, they are predominantly lower income as a group. there is certainly diversity among them. but in terms of their attitudes toward the role of government questions, government activism, health care, things like that, they are predominantly liberal. they are more diverse on social issues, but they do not vote primarily on social issues.

there is no question they tend to be on the progressive side on the economic issues, and that is the way they are voting. especially with the republican party moving further to the right -- the problem the republican party has is not a single, how can we appeal to hispanic voters? immigration is only one of their problems. and the party has been moving in exactly the opposite direction, if it really wants to appeal to these nonwhite voters. not just the african-americans, hispanics who are definitely on the liberal side. hispanics not as much as african-americans, but still, they are definitely on that progressive side on those economic issues. >> we ended up not including the hispanic data in this paper, but i have it in the slideshow. if it is possible to get it back up. alan's intuition is right. or at least compares two hours very similarly. the hispanic voters have been

oversampled in 2012 to the left economically, and slightly conservative socially. as mike said, they are quite heterogeneous, but there is a little bit of a density in this popular area. as we look at the 2008 and 2012 study, and ask how many hispanics, or which percentage, the liberals were the largest group in 2008, at the moderates were in 2012. the liberal policy preferences dropped, i think, 11%. but most of that went to moderates, and not to conservatives or libertarians. i think it is a complicated picture that perhaps slightly advantages the democrats on our dimensions. immigration probably advantages them a bit more. >> one other comment. i agree with the other

panelists. i do not think hispanic voters are on the horizon for the republicans, name it for the reasons stated. the was a book written by thomas and larry edsall, "chain reaction." they argue basically that democratic groups are democratic because they depend more upon government services, and therefore tend to be liberals who want more government. the republicans basically are those who are less dependent upon government, and therefore see government as a vehicle of taking away their hard-earned money. mitt romney made the unfortunate comparison of the 53% versus 47%. but i think there are many republicans who see themselves as cash cows for the democrats. i do not see them going along. >> that is definitely true in

terms of the way they think of themselves, that the interesting thing is that income itself is not a very good predictor of the vote. it is not that you have these affluent republicans who are concerned that the low income democrats are going to take from them. once you control for race, the relationship between family income and presidential vote in 2012 is about 0.05. among whites, it is almost nonexistent. you get a slightly higher democratic vote. at the very, very high end -- you can only go to $250,000 and higher in the data. they are a little republican, but not much. religiosity, frequent church attendance, outweighs income. low income religious whites vote heavily republican. high income secular whites vote

much more democratic. >> i am dan o'shea of colby college. this is also about the future. if polarization is being driven by attitudes towards the other party, a growing disdain for the other side, do you think that will eventually have an impact on housing patterns? that is to say, i would like you to weigh in on the sorting argument. i do not want to live with them. i know this is a bit off-topic. but the impact this could have, disdain for the other side, and sorting. >> bill bishop says -- >> his findings suggest that is already happening. to some extent, at least. people are sorting geographically for a variety of reasons.

some of that probably has to do with values and lifestyle choices, and other things that are correlated with party and ideology. but some of it may be exclusively political. i do not want to live in an area that is predominantly republican if i am a democrat. and vice versa if i am a republican. i think some of that is happening, certainly. >> i think i would be -- take not quite as negative a view as alan. i think there is some evidence of this geographic sorting, but i think people are choosing where to live based on other lifestyle choices that happened to coincide with partisanship, not partisanship being the main driving force. i think it is also easier -- this gets to bill's question, social media and what not. people may have high disdain for the other party, but they can

certainly keep it to themselves, or keep it to their own network of friends or fellow partisans if they so choose. >> again, it is an excellent question. i see partisan sorting as an extension, as you know, of the kinds of sorting that bishop talks about in his book. and i think it has to do with the translation of cultural identifications into political identifications. cultural scholars suggest three causal links. one is how we feel toward different groups in american society, whether we feel positively or negatively toward them. i did an article some years ago, showing how positive and negative reference groups influence the presidential elections. another factor is religious belief systems.

and finally, from an evolutionary or sociological or anthropological standpoint, there is a struggle for cultural dominance in america between those groups who are contending for the dominance of their preferred values and ways of life. >> i would quickly echo david's point about being hesitant about a correlation meaning causation in terms of this relationship. i think it is an interesting theoretical question. is it like how people think about congress? they hate democrats, but the democrat neighbor is ok? i think there is a difference between how people see parties and see their neighbors. that is wrapped up in where your kids go to school, where your job is, whether you like old or new homes. >> there is something to that. at the same time, it is probably more of a factor in terms of

shuffle. that is having a bigger impact on how people interact, and who they talk to about politics, especially. if you have friends, neighbors, and coworkers who support the other party, you may still be friends, and you are kind of stuck with being neighbors and coworkers unless you move or change your job, you do not talk about politics with them anymore. i think people increasingly talk about politics and use social media to interact with other people who share -- especially people who care about politics. the more you care about politics, the more that is true. >> franklin is sitting over there, and the surveys he did in wisconsin before the recall included the question, did you stop talking with somebody because of their views on the recall? i want to say 20% -- >> a third. >> a third stopped talking to folks about politics in that highly contentious episode where i work.

>> i wanted to say something for the bishop book. ron rappaport, william and mary. bishop had a really adjusting point on this. the oecd countries, americans were the most likely to talk about politics. there were the least likely to talk about politics with people with whom they disagreed. that follows up with what you guys were saying. that is bill bishop, eight years ago, or whenever he wrote the book. >> ron, as usual, jumped ahead of me while still sitting down. i am paul beck, from ohio state university. the comment is, we have done national surveys in this country and 20 other countries around the world, asking about discussion networks.

americans are the ones most likely to talk only to people they agree with, and that has increased over time. americans are more polarized than people in any of these other societies, along the lines the panel has, i think very nicely, commented on. my question to joel, your result that whites who live in the most diverse counties are the most likely to vote republican or be republican. i can see that in cities in the south quite well, and some of the most diverse counties in the country are these black belt counties from the rural south and maybe the suburban south. does it hold in the north as well? i think of places -- you think of new york city. a very diverse series of counties. and yet whites were overwhelmingly voting democratic there, and probably did for president as well.

does it extend beyond the south? is it the south driving that? maybe a broader point to make is, one of the things i think we are seeing here is a southernization of the republican party. i think i have read that half the republicans in congress represent southern districts. so we are seeing that as a powerful driver. it extends to the rest of the country, but is the driver of the south. actor the question of most diverse counties. >> that is an outstanding question. we have explored it fairly extensively. initially, we were concerned that this might be a southern phenomenon. that is why we then tiered the racial context variable. initially, we had it less than 60%, and 90% and over. whether you tier it by percentage white or this index of racial and ethnic fragmentation, you have this

middle segment, this 50%, whose racial and ethnic identities are activated by racial context as well. it is not just in the south. there is sort of a continuum from the southern states to the central midwestern states like ohio, going to upper midwestern states like minnesota, where i come from, where my cousin vote solid democratic, and race is not an issue in states like minnesota, vermont, and maine.