cultural rescue staff. they have supported this vision for such a long time. other smithsonian staff supported the training. they worked hard to realize this important program. i want to thank my boss for believing in the aid and supporting us 100%. my heartfelt thanks to my husband, paul wagner. he has supported my dream for this course. he continued to support everything as a volunteer and chief logistics training officer. [applause] -- last but not least, may i present the 2022 class of the officers. [applause] >> good morning everybody. a warm welcome to everybody

today. it's good to see you all, today we're going to take up the long and fitful an easy relationship between the news media and election polling. we will consider a number of specific cases, today. in the context of presidential elections that, for whatever reason, the polls didn't get right. the polls went wrong. we will also consider factors that may have considered affected the easing of poll bashing among american journalists. much of the content into these lectures is drawn from my most recent book lost in a gallop, pulling failure in presidential elections and here are some names that we are going to be mentioning during the next bit of time. edward r. murrow. he was a legendary broadcast journalist in the 50s and 60s, even the 40s. who worked for cbs news.

a lengthy and storied career from cbs news. another name we will hear is that of eric -- he was a commentator for a long period of time from cbs news on television. cbs news evening report. another name we will encounter mike rio. a prominent journalist in chicago a columnist who was regarded as chicago's most witty in ornery observer, news observer. ariana huffington is another name we will encounter. she was a founder of huffington post, half post now. a former syndicated columnist. haynes johnson of the washington post is yet another name we will encounter today as is jimmy brazil and the celebrated writer and columnist for newspapers in new york city. each of these journalists was prominent in some fashion in

print or broadcast media and as we shall see all of them were suspicious about election polling and even hostile and vehemently hostile of election polling. all of them can be called pull bashers vigorous even harsher critics of election surveys in considering pull bashing journalists we will touch on presidential elections of 1948, 1952, 19 82, 008, 2016 and 2020. the polls in each of those elections were off target in some fashion in some way and as we have discussed polling failure doesn't always happen the same way. in any case, these cases that we will touch on today sparked

confusion, surprise, even disillusionment and no small embarrassment for pollsters. so let's get going, let's step back 70 years to the election in 1952. it was the first presidential election after the polling debacle of 1948. pollsters than, in 1948, had confidently predicted the victory of republican thomas dewey, then governor of new york state in an election with harry truman, the incumbent president. truman won in an upset that no pollster anticipated. it opened it anticipated and the news media did not anticipate, either. it was a shock. a shock. so, in 1952 pollsters inevitably were pretty wary. they were pretty cautious. in the projections and

estimates of how this race was going to go. in fact, they were so cautious that one humor columnist for newspaper said a tie would suit them fine. that the pollsters would be satisfied if the election turned out to be a tie they were that cautious. that wary. that race in 1950, to 70 years ago, pitted at american war hero named general dwight eisenhower against the governor of illinois, at least stephenson. at least stephenson. the polling indicated that this race was so close that a could be won by any candidate, could go either way. a narrow election was anticipated by the polls, as this article about the gallup poll suggests. despite the re-quid occasion, the pollsters,

as they often do, set the agenda. set the agenda for the news media and others on for the election in 1952. the expectation was a tight race between eisenhower and stevenson. that's not how it turned, out that's not how it turned out. eisenhower wanted a landslide that no post or anticipated. indeed, almost no pundit anticipated a landslide. he won, eisenhower carried the popular vote by almost 11 percentage points and received more than 440 electoral votes. it was a crushing landslide for the republican candidate, eisenhower. pollsters were confused as to what happened, trying to come up with explanations in the immediate aftermath of the surprise. it seems that undecided voters were a major factor in the polling surprise of 1952. in that, there was a substantial

number of people responding to the polls saying that they did not know we there going to vote for. eisenhower, stevenson. so, that undecided vote was and under some difficult proposition for the pollsters. in the sense that, how do you allocate the undecided vote? do you split it in half? do you give both to one candidate or the other? do you ignore it all together? it's a conundrum. it was a conundrum for pollsters in 1952. gallup, george gallup of the gallup pole, decided to allocate the undecided voters in a way that resembled allocation of undecided voters on the two previous presidential elections. in 1944 and 1948. and one scenario, he said those undecided voters are going to break 2 to 1 for the democrat, adlai stevenson. in another scenario, he said it would be 3 to 1 for the

democrat adlai stevenson. and that latter scenario, you come up with a 50/50 tie. a 50/50 tie. why the undecided voters may have swung to dwight eisenhower late in the campaign was his pledge, in late october 1952, to go to korea. go to korea. then the scene of a long protracted stalemate of a war between the united states, its south korean allies and the north koreans and the chinese. that war had begun in 1950. and by late 52, had settled into a stalemate. eisenhower, a war hero from world war ii, suggested that he was going to go. in, fact stated openly that he would go to korea to seek a way to end the stalemate, and the conflict, and the bloodletting. that announcement may have been enough to tip most of the undecided voters in

his direction. in any, case the undecided vote and how to allocate undecided voters was a difficult situation for the pollsters in 1952. criticism out the outcome was weathering. a newsper and massachusetts, the berkshire county eagle, said the pollsters re unable to tell a tidal wave from a photo finish.

as i read in my book, i lost in a gall, journalist took no small satisfaction in the st or his failure in 1952. edward r. murrow, a broadcast legend, with cbs news, said in the day after the election that the people surprise the pollsters, prophets, and many politicians. they demonstrated, the people demonstrated, he said, as they had in 1948, that they are mysterious they're not meant to be measured by mechanical means. doubting that polling had the ability to define what americans really thought. this sentiment edward r. murrow resonated throughout the news media in 1852 and beyond. a newspaper in denton, texas, picked up on that the even said, the outcome may be discouraging news for pollsters, but we like the independents demonstrated by the voters. the american voter, this newspaper said in its editorial, is an independent critic. it was not 1952 when coal bashing a

rejuvenated, but the outcome of that election intensify the sentiment, this tendency, among prominent journalists. after all, the polls had misfired and back-to-back presidential elections. 1948, 1952. this lens, and there are different reasons for the pole failure in those two elections. this lent a distinct impression that polls were imprecise and unreliable. in addition, the notion was widespread among journalists that polling was just a presumptuous pastime, and posed in unwarranted intrusion in the life and thinking of american that american voters. these arguments maintain that americans were just too complex, idiosyncratic, for their views to be accurately captured by opinion polls, by survey research. so in a way, polling failure was seen perversely,

perhaps, as a triumph of the individual and, individual. critics of election polling, included morrow and others. and others. and 1964, eric sever i, a commentator for cbs news, acknowledged and a newspaper article that he had a secret plea and really when the polls go wrong. he said the reasons for his feelings were obvious. we hate, he said, to have the mystery, and suspense, of human behavior eliminated by clinical deception. in other words, the polls presume they were able to divide what american thought. he thought that this was an unwarranted intrusion. eric several. in time, pull bashing among journalist became more

vigorous, more vehement, and more hostile. micro cow, who were chicago's best known journalist, he was a columnist for the chicago tribune after many years as a columnist for the chicago sometimes. we're go, in the mid 1980s, melted a noisy campaign, encouraging his readers to live a pollster should ty er be under viewed by exit pollsters. an exit polls are conducted on election day, our polling places, as voters leave after casting ballots.heare approached by exit pollsters. this ichanged in recent years, and sex to posters you a little

more than just that, they try to reach voters who have already voted, and early voting. now, or back, than they were still being interviewed, voters were interviewed, are they left the polling place, after having voted. so rock is urging his readers to light pollsters. he wrote this in 1984. that urging people to lie has been what appears you constructive civic endeavors he's probably already fine and she kara. he says the encouragemenbend voters to lie to pollsters is to try t mass a polling results to have the announcement of the exit polls on television projected incorrectly of course, all of

this was somewhat naive, because not even work out and the reach that he had amongst chicagoans back in his day, then ladies, could influence enough people, for interviewed by ex a pollsters, which is only a small number of people, to mess up the exit polls so it wasn't really going to happen that way. nonetheless, his, his encouragement, his columns, his pull busting columns, were remembered and cited years after the fact at a national convention of a four, which is the american association for public opinion research it's the leading organization of survey research, the president, then president of poor, invoked the column a said how can we expecpublic to some of our opinion leaders, like royko are trying to make a mockery of that? so his lying

to pollsters advice resonated and lived on 30 years later, 30 years after. even more improbable and adamant campaign led by ariana hopping ten, a founder of huffington post, formerly a syndicated columnist as well. she lied what she called a campaign for a potpourri america. a poll for aerial air gonna. by that she meant that she wanted people to take and no poll pledge, and to hang up on pollster should they call. or, if people were not able to hang up, they should lie to them. anything to contaminate e sample and demonstrate how unreliable polls e. lighter pollsters. thnopal pledge. very colourlly, she described polina's having become a stevens, and rarded with the same sort of reverence as ancient roman's gave to chicken on tils. eventually, her pull bashing campaign faded away. and fact, huffington post, in 20, can acquired appalling

aggregator called pollster. com. this caused a shift in her views about pulling a little bit, saying at that time, polling, whether we like it or, not is a big part of how we communicate about politics. another persistent and perhaps even more prominent poll bashir was keynes johnson. and national correspondent for the washington post. james johnson. he says in an interview with c-span that he wish that we could disband all polls. let me

play a brief clip. from that interview. >> i wish we were to spend all polls. i hate the balls. i mean, we rely on them. i've used, them cited them, i've gone out, got healthy, don't pulling myself. knocked on doors and so forth. i think we make too much of polls. poll. i think we make too much of polls. on another occasion, he said that polls no substitute for hard reporting, for street reporting by journalists to get

out and talk to people. haynes was both a practitioner and an of what is called shoe leather journalism? shoe leather journalism journalism, which essentially means getting out of the office, talking to people, doing interviews, getting a sense trends through these qualitative measures. so intense, supposedly, that you begin to wear the soles of your in shoe leather journalism, shoe leather reporting. for example, in the fall of 1980. haynes crisscrossed the country, going to places as diverse as boston and youngstown, ohio, interviewing and in writing long really long articles for the washington that the post called portrait of america that

discussed concerns disillusionment, apprehensions, troubles. in 1980, more 40 years ago. in his travels, johnson sometimes sneered polls and pollsters saying he disliked polls because they were unable to pick up the complexities. the complexities of people echoing the sentiments expressed by r murrow a generation earlier failed to pick up the complexities of people and opinions. in 1980, somewhat 1952, the polls projected anticipated estimated a close race between ronald reagan, the republican, and jimmy carter, the incumbent president. and near the end of the campaign. although he did not in his

articles offer an estimate as to who was likely to winhe election in 1980. haynes johnson was asked on a television program for his appraisal of the race, and he gave this answer. i don't know, he said. i think all my bones tell me that reagan is going to win. but i think somehow that carter is going to slip through and win reeltion based on his intensive interviews, crisscrossing the countrduring the fall campaign, carter going to slip through. in an outcome that resembled some respects 1952, the outcome of 1952. ronald reagan won in a near landslide. he defeated jimmy carter by nearly ten percentage points. an outcome pollsters did not anticipate. granted, haynes johnson, he did not dwelmu onisrediction in the aftermath of the 10

election. instead, he in a news article that reporters would he been served by relying on own stincts, their own legwork, which said in turn produces their own lical itit rather than the presumably scientific sample offered by pollsters. the surprise outcome of 1980 left pollsters bickering among themselves quite openly, which is quite unusual. disputes among pollsters election pollsters tends to be more private, more quiet, more low level. not so in 1980s. aftermath. they bickered about how things got so wrong. and one of the disputes was whether there was this late in the campaign shift of support to reagan from carter or not.

some pollsters said, yep, that explains it. others said, that's nonsense. we did not see such a shift late in the campaign. and really, it's still kind of 42 years later. somewhat unresolved. i think the standard explanation is that, yes, there was this shift to ronald reagan late in the campaign, but it was a disputed analysis. it was not embraced unanimously unanimously. it inspired itorial cartoonists to the polling failure of 1980, led tony off the philadelphia inquirer to put together thifairly imaginative and a bit insulting. editorial cartoon for the philadelphia inquirer. the notion that pollsters somewhat akin to sorcerer is making all kinds of a witch's brew, if you will, resonated and

found many years later. in 2004, a commentator on the fox news program newswatch named jim pinkerton said in that program that polls are as accurate and precise as human nature, which is to say they are not accurate and they are not precise. this practice, he said, is witchcraft. the 24 presidential election problem really represented the zenith or if you will the nadir of pop among prominent journalists that year, jimmy breslin, a well-known, well-read, celebrated columnist writer, playwright in new york city. took on a campaign late in the late the race to chastise polls on one occasion, calling them cheap, meaningless blatant lies.

and. jimmy breslin's complaint was that pollsters were not interviewing voters with cell phones who had gen up landlines and were relying primarily on cell phones that. number was fairly small a cell phone only. voters back in 2004. it has since become overwhelming. many people have given up their landline for cell phone only use not so much in 2000 for pollsters. he was right did not generally call cell phone numbers. there were complications. federal regulations seem to impede that practice. pollsters nowadays do call cell phones. in fact, most people pollsters who are using the phone in any fashion for their polling are more cell phone numbers than they are landline. in any case, this breslin's principal complaint, he said

they're failing to up and failing to interview people who were probably younger and more likely to vote for the democratic candidate in thousand for john kerry. on election day 2000 for. breslin writes this that i'm not even going to bother to watch the results on television. i'm going to go to bed early. besides, he wrote with typical bluster, if i was up so many people upon seeing every word i said of this election coming true would be kissing my hands and embarrassing me with outlandish praise. jimmy breslin, an election day column. 24. john kerry lost to george bush. george w bush by 2.5 percentage points was a close election pivot on the outcome of ohio that year, which narrowly, to

george bush amid some dispute, amid some dispute about that outcome. so why all this periodic poll bashing among journalists? what explains this tendency that we've seen in the years past? edward murrow is a legendary was a legendary broadcast journalist, forties, fifties and sixties sixties. haynes johnson, mike royko, jimmy breslin. were pulitzer prize winners at one time in their career. they were all respected journalists. so what factors explain cross generation noel paul bashing. i'll offer three three factors. one is that pollings record of unreliability and inaccuracy see made it an easy target.

journalists prominent journalists picking the polls, if you will, calling them out for their mistakes, their misfires, their errors. another factor, which we've seen from murrow to haynes johnson, is a dislike of the presumption is presumptive of polling. by that i mean that polling seemed to say and does say in many respects we can figure out what americans are thinking. we can figure out what their preferences are. this is a procedure that can do that. that content can detect what americans are thinking. a lot of journalists over the years had with that. they resented the presumptions presumptively as a. and also polling represented a direct challenge to shoe leather journalism to shoe reporting, which we discussed a moment ago.

this is journalistic direct out of the office, talking to people, doing what haynes johnson did, for example, for election cycles, crisscrossing the country talking to people about their inclinations, their preferences in terms of politics and how this was likely to influence the outcome of the election. getting out of the office, talking to people was a fundamental element of shoe leather journalism, which still exists. still exists and still celebrated in media. so why has pope abated among journalists? it's not as if journalists have become suddenly more polite and courteous. that's not an explanation. what.

possible explanations might you offer? what do you think? what might account for the abatement you will? the decline of poll bashing? gentlemen, is it because, especially with trump's rise and him bashing the polls fake or they're underestimating his support and not to look at the polls. is it because if journalists were to also bash the polls, they would kind of be aligning themselves someone who is also attacking the press itself? that's an excellent. one that i hadn't thought of. does that mean then if if that's the case, it might be a short duration? because trump is not going to be on the political scene forever. i mean, he's been there for a while, but it's not going to be forever. what do you think? is this a short term phenomenon? shortish, it's definitely short term in terms of trump himself, but i think the practice of attacking the press and also polls is going to be around in the republican scene for at least a little bit longer than trump with like ron desantis and other republican politicians kind of picking up that mantle.

so i think that it's not necessarily to trump itself, although he's definitely one who's like pioneered it on the republican side. very interesting point out william, because the public already such a distrust for the polls that the media like they don't really need to add more to bashing on it as in the media. so why pile on in other words, why should the media pile on when when the distrust, as you put it in the polls is pretty intense. yeah, that's good. gabriel and i'm wondering if paul bashing has abated because arab, i think i'm thinking about the shift from news like more entertainment based in the past 10 to 20 years and maybe it's similar to sports that's in which they're good for reading things not more of like there's something easily digestible the public can easily digest and interest in. and also, it's a lot easier to look at a poll than it is to like listen to like a ten minute report on the complex inner

workings of american minds when it comes to political decisions. that's a good point, right? it's related to maybe the entertainment factor of the news media. but do you think this like ten, 20 years in the making or, is it probably than that? um, this phenomenon you described i'm thinking that. i'm thinking that paul bashing has slowed down probably in the past 20 years. and that's because i'm like 22 and i haven't really heard of it in the really in the news media since i've been a watcher. the news media. i think also when it came to 2016, i feel like american public was a lot more like okay now. trump's president then oh, what did the polls do wrong? i don't know. i felt like if there were to be if was there might have probably was back to 2016. i just was in high school so you didn't notice it but i feel like

the news i didn't really like have reports on like what polling went wrong or it's it's, it's, it was not as vehement or intense or hostile among the media after 2004. it's kind of puzzling as why that seems to be a demarcation line. and maybe you're right, maybe you're right. the public's respect for polling has dropped off to a point where the media just feels that they don't have to, to, to to weigh in. it could kind of also because you said that it didn't change a lot, but someone jimmy breslin that you mentioned prior also made it incorrect like assumption about the election so i think that it could just like people are wrong like i don't know like it's something of sort where like the journalists also got some journalists did get it wrong prominent journalists and so then there's obviously something going on that people are not like picking up on in terms of like human behavior or from journalists who are doing the like those shoe leather like

method. yeah, the shoe leather method or people that are doing the polls. it's great. i that i like that explanation. the journalists essentially were chastened by their own mis predictions, such as jimmy breslin in thousand four and therefore kind of to tamp down the poll bashing. interesting. either. i also wonder if it could be just the development of like technol and data and itself. i feel like that our perception of data, that type of stuff, specifically in terms like media news has become a lot more potentially like welcoming than it has been in previous just from other areas using it so much more. i think gabrielle's point about, um, about sports is an excellent understanding of that just seeing that so many people are like very much focused on stats and even nate silver the like, the person who runs

fivethirtyeight is a sports statistician at and i think that that might be something like has made the entire environment a little bit more welcoming to the data pull bash or the data of polling. why would that be, though? why would people suddenly become, if not enamored then more respectful of the data of polling i mean, what would because because statistics do put a lot of people and a lot of journalists. yeah. i mean i think that that's definitely i think however in the past i would say even ten years to data and technology has just become so entrenched in everyone's lives. i think that it makes it so people be a little bit less so worried about it than before because i mean, can remember back ten years ago the idea of like someone having a phone when i was like 11 is like absolutely. but now that's that's something that's normal. we need to make sure that people are to contact like their children and stuff that.

so i think that the greater societal shift towards technology and big and all that stuff might one of the reasons why it's more it's more okay that's a good point that's a good point as well. yeah. this is a point to that. i think that like you mentioned, nate silver, i think that the rise of the polling aggregators has played some in this partially because it's taken some of the it's taken some of the celebrity and some of the attention away from pollsters themselves, which i think was something that, um, i think journalists that they were competing with. whereas now you see very much a fusion between journalism and the polls, you often see folks like nate silver being brought on to election night coverage and think that the, um, the especially with the rise of online journalism, it has made it a lot easier for journalists to get to make polling part of

their coverage as opposed to having to compete polling. and i it's also made it if it hasn't made polling more accurate, it's created a more there's there's more of a respect, if not from the public from journalists, that it's a scientific process as opposed to something that's guesswork. so it's a little more sophisticated than paul bashing was and a little more thoughtful than the reflexive. the polls are wrong let's insult them some more. dan, you've been very patient. yeah, i think thinking like specifically to 2016 where i think the people were extremely with the result. so the people and i guess by extension the media were more i guess inclined to be forgiving of the polls, even though they were completely wrong as well. so it's just sort of that forgiveness factor of if the people are surprised in the polls, also have the right to be surprised. interesting point. i how many pollsters that embrace that notion people were forgiving because i think they

remember 16 and then 2020 which was another embarrassment in some respects a different kind of embarrassment. but the polls collect were off in a way that was unmatched since the reagan carter election. the point that i was going to get at, i was confident answer your question or ask into the question with the question, whereas is it poll bashing has stopped or do people just not really care polling anymore as a mass is it kind of goes towards isaac's of the advancement of technology. i think we've even discussed in the class before that people have started closing their doors more and syrian calls last especially with the invention of caller i.d. and being able to see who's calling your phone. most people don't want be bothered anymore, and i think it has a lot do with our culture as a society, as individualistic. so i don't think it's that paul bashing has really like dropped

per se. i just don't think people rely on polling again going towards what we've been. people don't think that polling captures, the complexity of the human mind and what people are really thinking or what if people change their minds. again, going back to the in time. so it's really it's, really complicated. yeah, i so. these are great responses, folks. wish i had thought about half of them or more. cassie yeah, kind of off of priscilla's point about if people care or not. i was thinking about this question and then trying to think of other trends that could kind of go along parallel to the decline in poll bashing. and i thought with the decline in poll bashing since 2004, we've also the increase of polarization in our politics. so i'm kind of thinking that maybe this could be part of an explanation. so after a surprising election

result, people are not as interested in what went in this past instance, but focusing now, okay, we have this polarized government now what? the next steps? how are the branches going to be interacting with each other? and it's much more focused on that partizan polarization. now, with presidential elections. so i think that future looking minds that focused on polarization has sort of taken over looking in the back on polling mistakes in a way. so that doesn't seem to really matter that much anymore because the the electorate's interest perhaps lies elsewhere. it's interesting point as well. lily, did i see your hand. yeah i was going to respond to priscilla's with another question. if people don't trust the polls and the media doesn't like to use the polls then why do we still have polls during elections season? because if the public already doesn't think that they're accurate, the media already doesn't like that. their presumption was like, why is polling still?

and i think it's because like humans want to know what's going to happen. they want that sense of certainty. and i think, like cassie's point is very accurate, that polarization is a very big effect because i don't think it's about the individual candidates anymore necessarily, but i think it's about the parties more. and so i think that's why people like the polls now. it's because it's more the parties and it's more broad than just projecting a specific, which i think is what people got wrong. and people were frustrated with because they were expecting a specific candidate to win. but now like when you talk about candidates and it's always like the democratic presidential candidate, it's not like you're saying joe biden's name and the polls are very interesting. but to that, i would add that polling is probably not going to go away, in part because it's pedigree so long researchers have found evidence of sort of protocol polls, straw polls of the sort as far back as 1824 and

the election of 1824, newspapers in pennsylvania and north carolina, a reporting on the sort of sentiment detection that that some were. okay, who who who's for john quincy adams and who's for andrew jackson and so going back that far to 1824 is is really a long long, long period. so it's it's hard see polling be completely uprooted. it might be it might be i kind of doubt it am i my argument. no that it will not be completely uprooted and we'll still have election polls every election cycle midterms or presidential or even off year elections there's going to be polling a couple points, angela, than i yeah. sort of two point about people really closing their doors, not answering. i think that that is related. the trend of declining institution trust right across

american society. and i think part of why poll bashing abated is because journalists are concerned about that. and similar to this idea that, poll bashing has become political. i journalists now looking at this poll bashing don't necessarily see just this sort of maybe friendly, maybe harsh between polls and journalists, but they see a broader, more concerning trend about institutional trust and they're less likely to engage in it because feel that it's actually a societal problem. they should be investigating and not something that they should be participating in. very good, thoughtful points. i think responding more to lili's point on who are the for i would completely agree that i don't think that they're for the people anymore i think that and we've had this discussion before in class that think that it's much more for the candidates because of the fact that polling provides such a good indicator

for potential donors to like get money to candidates that they think are going be successful in the campaigns. and i think that that's definitely something that as we have seen a decline line in the actual trust and interest of polls, maybe from people in general, i think that that might be the alternative of where it innovates to, not for the people anymore. it's for institutions and others such as that. right. an election polling does tend to be focused on candidates and the public polls are out there perhaps in in greater numbers ever before. we'll i mean, people like nate silver, who as i mentioned, do kind of keep track of how many polls are being conducted and by whom there are internal polls. the candidates run themselves and don't necessarily make public as to the trends that they're picking up. so you're right. but there are these other.

this other sort of subterranean aspect of of opinion research, of polling of election polling that we don't often see that we don't often pick on, that are not often disclosed, is great points, folks. these are great points. let me offer my three observations. john's one of them for explain the decline of poll bashing is that it's some of the more prominent poll bashers have passed away. and are no longer on the scene scene. secondly, major news organizations, even local some local news organizations have become polling entities themselves. this is not necessarily a new trend. we've seen the new york times cbs news get in an association that began in 1975, but they were the pioneers nowadays. we see many national and even

local polling organizations or news organizations conduct polls or commission polls, cnn, the new york times, the washington post, the wall street journal are among the news organizations, prominent news organizations have become polling pollsters. so it makes it obviously a little awkward to be bashing your own work in a way, this may be an explanation. angela sort of hinted at this. in fact, you kind of nailed it, saying that the rise of the data journalist is another explanation for the decline, the abatement, if you will, of poll bashing among prominent journalists, the rise of the data journalists now that all journalists in this country are steep and are comfortable with. big data, but a growing are and nate silver, who launched the

polling analysis site fivethirtyeight.com. in 2008 began became almost an instant celebrity when he predicted or forecast the outcome of. 49 of the 50 states elections that year's presidential election between obama and john mccain, he got it right, 49 out of 50 in his poll based modeling, his poll based data analysis and even top. in 2012. in the race between obama and mitt romney, republican a race that some organizations such as the gallup organization thought was going to go romney's way narrowly. silver predicted outcome in all 50 states and an obama victory. really becoming something of a celebrity. now, he's not infallible. he's not infallible. 2016, he gave donald trump than

a 30% chance of. and he said he forecast that hillary clinton would carry key states such as wisconsin, florida, michigan, north carolina and pennsylvania, all of which trump won in his narrow election victory. he an electoral college victory of some substance but and lost the popular vote. but the forecast fivethirtyeight.com offered in 19 in 2016 about those states was incorrect. and he also figured that trump's chances of winning a split election a split election in which he loses the popular vote but wins the electoral vote fivethirtyeight.com, that likelihood was 10.5%, very all, not impossible, but

pretty small. and that's what happened. a split decision in 2016 and in 2020, fivethirtyeight aecom gives trump a 10% chance of victory, saying biden is likely to win by eight percentage points in the popular vote. but he says and wrote two days before the election i'm here to tell you that trump can still win, even though there's a 10% chance that fivethirtyeight's model was trump in that election. all hedging aside, and i would argue this is a bit of a hedge. silver and other data journalists have shown a comfort with and indeed a sophistication and understanding of polling data and. this project's a cogent sophistication that wasn't matched by poll bashers of

earlier days of earlier times. so to recap, why has vehement poll bashing abated the passing of poll bashers, journalists who are bashing polls, major news organizations have really become and have confirmed their presence on the polling scene. they are polling as well as news organization and then the rise the rise of the data journalists so what one might ask have poll bashing journalists accomplished in their criticism, in their denunciations? what effect effect did this cross. poll bashing have overall? what would you suspect are some of the results? upshot consequences of cross

generation poll bashing? yes. well, i think one of the columnists that was criticizing pulling of only using telephone and landlines and not including cell phones, that probably prompted the industry to make that move towards cell phones that we see today. and so hopefully in theory, that is making the pollsters available reach a more wide audience. so i think that was a really thing that happened for the industry and then also i was thinking that some of the that the poll bashers were offering were saying how polling does not capture a realistic representation because. it's just a number and it doesn't really gauge how the public is actually feeling. so maybe pushing journalists and pollsters take that more of a shoe leather approach has happened and made polling

hopefully a bit more so i feel like if that movement happened and that was a good thing so think those two criticisms were pretty prominent from what i gathered from the readings and this lecture, it's very good taking the first one for a moment. cassie so it was something of a reminder that there are a large number of people, a growing number people who are using cell primarily that jimmy was calling attention to perhaps insulting and an aggressive way. but nonetheless, this was underscoring a trend that became even more of a powerful and decisive trend because as i say, most, you know, have given up landlines, many people have. and our cell phone only cell phone only households. it's great knowing. i agree with cassie. it's like saying like how paul bashing has of forced pollsters to change their methodologies. and from chapter of the book that we read this week, like

going back to like the 1936 election and like the poll bashing that there, it says like the digest failure, like the literary alec introduced, like a sense of like nagging doubt about the polls that's a continued so like for so long and like in election polling in general and it also kind of presented like a competitiveness in the polling and how like pull bashing you can bash like one poll and then like praise another poll for getting it more accurate. so kind of made a sense of like a competitor nature and this industry. those are good points. those are good points. do you think that that's a direct ramification or consequence of of poll bashing or is it a sort of just a sidelight of of this phenomenon? i like it's a little bit of both. i think it depends on the election year. and like what the polls have said, because of all of the polls, it wrong. obviously, there's not really going to be a competition because everybody it wrong. but in some places like some years like we call like you know different poll on going against

organized nations have different numbers so like years might be more prominent of like look who got it right and then look who got it wrong right you know what comes to mind when were saying that is the 2020 election and a lot of pollsters got it wrong thinking that biden had a 1012 percentage point lead in the popular but as we've seen there some pollsters emerson college polling for one it was pretty close their final poll biden ahead by five percentage points and he won by 4.5. so yeah you're right are some clearly some differences in in the polling results but i think it's made just everyday americans not care about the polls and like trump has obviously the process for republicans but i think republicans democrats and independents they like their tone about the polls and they go, oh, i don't believe the polls. and i think it's because that everyday americans are kind of engaged in their own form of shoe leather journalism because they're the ones talking to like friends and neighbors, like other people out and about

during their normal lives, that they just talk to people about these and they develop a better sense than, like maybe a poll could. so i think that in that respect it's made people more cautious of polls or even outright just dismissive of them because they're feeling something different. do you think that that may be to and not sort of scientific enough? yeah, especially talking to your neighbors doing your own shoe leather reporting, especially with republicans tending to live with more republicans and democrats living with democrats. it definitely produces more of a bias but i agree is too qualitative but i think it's still is something that's more i don't know maybe real i guess in a sense interesting. yeah. going directly onto chandler's point i think that poll has definitely brought a lot of disillusion to the institution of to these major institutions that we all used to hold dear. and i think that one of the reasons why is because of the

formulation polling at its beginning as a very scientific quick way of doing things well, at least was characterized as that yes. exact more quasi scientific than well in that exact point though. dr. gallop was always called dr. gallop, and it was always seen as this very sophisticated measure of always trying to do things a very pretty pristine and scientific way. but even as you say in your book, all that stuff, it's sometimes gets taken to the side when you're actually down in the field and there's a whole bunch of issues with that. and i think that with the resulting disillusion and when they wrong and all the poll bashing that right of it, i think that tacitly also like hurt the characteristic of hard as well in that and i think that and the idea that these are for civic duty and and related to these hard characteristics that then made it so people are then a lot less like a lot less

interested or welcome to the thoughts of these hard data sciences because we see that this thing that was characterized as a data science was wrong, like a lot of the time, or often enough to, to, to raise people's skeptics. ism polls are not always wrong, but they have been wrong often enough. we've seen for people to be of wary about them, kind of skeptical and not certain about whether they should put much trust in in them. although i think we've talked it before. but could that also be the lesson and of itself, like failure maybe i'm like a proponent proponent of failure, but think that in election polling in particular when there's like a fail, like they get it wrong or something like that, it makes everyone question like what happened in our member like back in 2016. and it may be because like people live with one another like in terms of like political ideology, but that made me question like what in the world happened to like the election polling results or just in general because when talking to

other people you i don't think anyone around me would have guessed that donald trump would have won the election. and so then i think there's to it to failing and even the 1948 election or even in 1952, i think that also other people question in those elections like what happened in that situation. and in 1948, we talked about this before and even it's in some of the that people like some the pollsters stopped polling at before like after they didn't poll after labor day and things of that nature. and so that was like one of the examples of like this where we went wrong in polling. also like it also tells you the shift in attitude amongst society like where does shifts happen? does it happen right before the elections the day before or is there some kind of announcement that happens that changes people's minds like in the case of eisenhower? so i just think that there's value in failure. interesting. in election polling failure in

particular, that sounds a great research project to take on the value of failure in election polling. interesting. final thought, folks, and i would like to offer a final thought. go ahead. that yeah, i think the idea that this that the election, just as cathy said, are just number that they just focused sort of on the horse race has led to some changes especially in media of polls because now any exit poll you always see with not just where are the candidates standing but what are the issues, the people saying or most important to them. and so i that poll bashing has had that accomplishment of leading journalists to put some context and not just focus on the immediate context, but where's the horse race at? i'm sorry, what would you say about this point, though? major news organizations have themselves polling organizations. i think that's definitely.

a fact. and i think it's probably led to some it's led to changes. it's led to journalists looking at what kind of polls are going to be the most to their coverage and that means more polling that can sort of serve as a substitute for leather journalism. well said. on that note, folks, we are going to wrap up today. i want to thank you very much. your observations, comments, suggestions and insights. they were great. that list i showed a moment ago could expanded, what, four times as long? in any event, i appreciate your attention and your