their first snowman in new hampshire. they are one of the reasons we fight so hard for what we winston churchill once remarked that success is not final, failure is not is the courage to continue that counts. while this campaign has ended, the mission continues. down here in florida, we will continue to show the country how to lead. thank you and god bless. >> a healthy democracy does t just look like this. it looks like this, where americans can see my chrissy at work because citizens are truly informed. get informed straight from the source on c-sunfiltered, unbiasr word, from the nation's capital to wherever you are. because the opinion that matters the most is your own. this is what democracy looks like. c-span, powered by cable.

rst this morning, two days ahead of the new hampshire primary, with andrew smith. he is the survey center directr at the university of new hampshire. we are going to be talking about new hampshire primary polling and other trends. good morning. guest: good morning. host: let's start off. tell us about the new hampshire survey center. how do you get your funding? is this partisan or nonpartisan research? let's start there. guest: the unh survey center is an academic survey center. it has been around since the 1970'i've been the director sine 1999. i am also a political science professor. we are not funded by the university, we get our fding largely through research that university faculty are doing,

government grants or projects on various■ój forth. we also do media polling during elections. we do not do work for candidates or parties.we are a nonpartisan organization, but we do work for media organizations, and at least going back to 1996've been doing polling during the presidential primary for cnn but we have also done polling for fox news, with newspapersosthat. i am a political science professor. media polling is a small part of what we do, maybe 5% of all the work we do. host: a lot of attention on that polling right now. ld■&tell us before we get into e of the substance of your unh polling, how has the■q methodoly for polling changed? we hear about the trends, we

polling, with cell phones and other things. tell us how you ensure your polling is accurate and reflective of theguest: that iso unpack because frankly the whode stry is going through a paradigm shift in how we do our polls. this is similar to the shift that happened in the 1960's movd from in person surveys to telephone surveys. it took about 10 years for the best practices to be developed, how to reach samples, how to call people. how many times to call them, etc. we a from telephone surveys to web-based surveys. we are moving to web for the same reasons we moved to telephone surveys. they are far less expensive. i will give you an example of

the costs. between 2016 and 2020, the costs for us to do a statewide survey in new hampshire almost doubled from about $45,000 to over $90,000. it was a big increase and most organizations can't afford that. we moved to web-based surveys. what we do at the unh survey center is a probability panel. this is a key concept. for surveys to be generalizable to the general public, yoa(u have to start wita random sample that allows us to use some statistical models. if you don't start with a random sample, you can't do things like how collate margins of sampling errors. you are just violating the assumptions of math. we use a probability sample where we recruit people from across the state at random, by

telephone,ext, and address-based sampling, sending mailers to people's houses. we hava panel in new hampshire of about 8000 people that have been selected at random and we go to do a poll, we have email addresses +hd we send them an email invitation. we interview them that way. it is much quicker. we don't have to ask demographic questions as often because we already have tharms. it is much less expensive. there are drawbacks to web-based surveys as there are two telephone surveys. it is hard to reach people, hard to get people to want to talk about political questions. and the people who do talk to you tend to really want to talk about political questions. the methods used to make sure we

have a representative sample of random sample but a represented of sample, we are still working through those. the 2016-20 20 elections really pointed out those problems and i hope the public understands that the whole industry is not quite flying blind, but the ground underneath our feet is shifting. host: we are going to get into some new hampshire specific questions but i wanted to go ahead and open up the phone lines for our viewers, if they have any questions for andrew smith, the survey center director at the university of new hampshire, or if you have if you are a democrat, call us at (202)-748-8000. republicans, (202)-748-8001. dependents, (202)-748-8002. if you are a new hampshire

resident, new hampshire voters, we want you to call us at (202)-748-8003. as we go into the primary, i want to ask you. new hampshire, tuesday the union leader, which is the big pete -- big paper up there, did have an endorsement. it is endorsed nikki haley. i'm going to show a tweet. this is from governor chris sununu who is so endorsing nikki haley. he writes, the union leader said it best. nikki haley is catching fire in new hampshire and is theext choice -- is theest choice to be the next president of the united states. he then links to that union leader endorsement. i first question for youi( is -- the latest polling, where does it show nikki haley in relation to donald trump? is it close enough that haley1

could win on tuesday? guest: i wish i could tell you the specific answer to your queson but our polling is embargoed until 9:00. i can give you hints about where things are going. we are seeing that nikki haley is within striking distance, but not that close to donald trump. this yearless movement than we n in previous primaries because trump has a quasi-incumbent supporters and he has 40% plus of the support in new hampshire. haley is an unknown commodity and what she is doing is garnering support republicans who really don't like trump and then some other people, independent voters who may be considering voting in the republican party. around -- i don't think things will move around too much. endorsementsargely they are

preaching to the choir. the union leader wasn't anti-trump newspaper since 2016 and the mainstreaof the certains not pro-trump. but there isother states there a significant segment of the republican party that is the trump maga party. the rest of the party, presuming trump will be the nominee will come around to vote for him but they are going to be supporting the team, rather than the captain. host: i know you're polling will be out with cnn and a little bit, less than an hour but this is 5:38, a compilation essentially of polling averages. it -- polling around 49%. nikki haley is around 34%, 35%.

and then ron desantis at 5%. can you tell us about -- ron desantis was second in iowa, but he is barely in the single digits in new hampshire.çg what is the difference? why did ron desantis decide not to compete in new hampshire? i think it is really important to understand that the iowa republican caucus electorate and the new hampshire primary electorate are extremely different. in turnout, iowa has very low turnout, typically 5% to 10% of voters across the state turnout for the caucuses. in new hampshire, turnout is estimated to be -- this cycle 0. you have a different type of voter. iowa voters tend to be very

socially conservative, driven largely by even joe little churches. in new hampshire, republicans here tend to be much more moderate in the sense that they -- that their concerns are more on taxes and the size of government. republican electorate in new hampshire is more highly educated than an iowa. -- than in iowa. i don't meant that they have things like college degrees. iowa is a very religious state. newhe country. the social conservative messages of ron desantis, that work for him in florida and that he hoped top in iowa just don't resonate with voters here. you can see this historically, when more conservative republicans win iowa, think of ted cruz and pat robinson. they don't do well in new

hampshire because the audience is completely different. i think desantis's biggest mistake was the message that worked for him in florida really can't be replicated in new hampshire, and he didn't change his message. he was talking to voters about things that they weren't that interested in and i think that is why we saw his nuers decline from a year ago. he was leading the new hampshire primary at 43% over trump in our polling. he has declined everas soon as d campaigning here, his numbers really went down and i think it is the fact that he is not campaigning here now, he is basically skipping n■ew an indication that he knows he does not have any chance. host: we are speaking this morning with andrew smith. he is the director of the survey center at the university of new hampshire. we are talking about polling and other issues ahead of that state's first in the nation primary on tuesday. the phone lines for you to call in with your questdrew, all yout

politics in general or new hampshire specifically, democrats, (202)-748-8000. . independents, (202)-748-8002. you are a new hampshire voter, we want you to call us at (202)-748-8003. this is open to all now. the democrat line. what is your question or comment? caller: good morning. my question is when polls or surveys are taken, are the same households or individuals called yearly? is it a randomask different peot the country? because i'62k& old and i have voted since i was 18 and i've never been pulled or surveyed about anything.

guest: the short answer -- thank you -- the short answer is yes. we call different people. randomly selected people from across the country, but if you think you have a country like united states with over 300 million people, about 200 million or two and half-million households, the chances of being called are not that great. in new hampshire, your chances are pretty high because there are so many people that we survey, but it is a random process. we you may have been called from -- in the past but maybe you were not home or you didn't recognize the numbe so diit isn't a surpru have not been called or participated in a survey. i compare it to getting called for jury duty. everyone can get randomly called but i have not been called for jury duty since i was 18 years old.

i keep hoping it'll happen but it does not. host: andrew, i wanted to ask you, we had a couple more in. viewers on twitter. j sanrs writes, are there any differences between people who know they are parts of panels as opposed to those who didn't know they were going to be surveyed before being studied? i guess he is ag, or how do you ensure that people are coming at it and you wynne lee -- coming at it genuinely? guest: there are differences in the people who participate in any kind of survey. they be -- they tend to be more willing to give their opinions. there are differences between the people who participate in any surveys, and that is true of panels as well. they do new that they are a part

of a panel, we tell them specifically and then we continually recruit new people as this goes on. ■ it is important to remember that in any survey, response rates are very low, telephone survey results are about -- response rates are about 5% or less and that is well-funded government connected surveys, only -- 100 households, only 5% of them are going to complete the survey. that had not been that big of as because the people who didn't answer surveys tended to be like the people who did. what we have seen increasingly and in particularly was evident in 2016 and 2020 elections is that some people are more likely for example in 2020, i was on a task force and looked at

, and our major conclusion as to why a trump's numbers were low in polls across the country was th republicans were less likely to participate. they just didn't answer the phone, didn't participate in surveys as much as democrats did. there are some theoretical modeling's for that, a spiral■cf silence. if you think about what trump had done since 2016, he talked about fake news and fake polls'j . a lot of republicans took that to heart. they said i'm not going to participate in any of those fake polls and as a result, his numbers and other republican numbers tended to be lower in the 2020 election cycle. host: let's go back to the phone lines. ronald is calling from orford, new hampshire, independent line. go ahead. caller: thank you for taking my call.

ago from morris. i picked up the phone and they up on me. as far as i'm concerned, all of these polls are just made up. you guys are just trying to get the two impound -- incumbent candidates to run against each other in a condition where dissatisfaction is higher than it's been since the civil war. i think it's all corrupt. the pollsters, the media, the politicians, everybody. i support the revolution. host: andrew, how do you convince people like ron that you guys don't have a political agenda, or ither way to convince skeptical people to participate? guest: it is difficult. people believe what they want to believ certainly our political environment today is very polarized. the news media is polarized. polls have become used as a

weapon in elections, used by candidates. if you're winning in the polls, it's a good pole. i think trump said it best a couple weeks ago. he talked about the des moines register polls, saying it was a wonderful pole, a very powerful paul because he was leading. then he said if he wasn't winning, he would say it was a terrible poll. as a we are nonpartisan. we don't care who wins. we want to make sure that what we are doing with our methods is to accuratelre that is tough because part of it, part of the problem we have is most people really don't pay that much attention to politics. voters like ronald, they do, they really focus on these things that are very important to them. to most people, that is not the case. we do have problems in getting representative samples of people, but i can tellorganizatd

democratic media organizations like cnn there polling directors and people on the polling teams really just want to get an accurate reflection of what is going on. /,i can't prove that the people like ronald. i can invite him to show him what we do b if they have that opinion, there is not anything that they see or hear that is likely to change that opinion. their minds are made up. host: let's hear from don, austin, texas, republican line. caller: thank you. you had mentioned that the base in iowa is a little different from the base in new hampshire, that the new hampshire may be do you see that is going to be the case across the nation, that trump is gaining more support in

the educated area of voters? guest: i don't think so. what i was doing was comparing the republican electorate in iowa and the republican electorate here. new hampshire like much of the northeast is a highly educated wealthyish part of the country and frankly the help -- the wealthy and highly educated people tend to vote democrat. the democrat party is sort of the social and economic elite party. the republican party is sorting itself out as the more blue-collar party, attracting those people that the democratic party has kind of abandoned. others have written books about this. academic research looking at this phenomenon but it is up of it has been going on for the stany primary election, the electorates in new hampshire and iowa are very different and any general election, you will see that new hampshire is going to

-- the highly educated people in new hampshire are like going to vote democrat regardless of who the nominees are. guest: staying -- host: staying on the republican line, next is elaine in washington. caller: i have a question about what polls are being run. have you ever run a poll to find out the number of people who have committed voter fraud? have you ever done that? guest: we haven't done that specifically but i have been involved in many studies looking and trying to find voter fraud in new hampshire. yes, there are occasions of voter fraud in new hampshire but they are very very small. a handful of people in any given election. there are myths about all ofc$ these voters supposedly coming up from massachusetts and voting in the state but there is no

evidence to support that that is the case. after the 2016 election, you to say that there was voter fraud in the state and new hampshire was the first date that they came worked with the secretary of state's office and looked at voter fraud and people's trust in the election which i think is more important, and nothing is really vevery few people are going to admit that they commit voter fraud because that would be illegal, and basically tarsayina crime but there are a lot of methods being done, looking at databases across states, to see if voter fraud occurs and it does occur occasionally but it is usually on the part of people with two homes like a home in one state and a vacation home in another state. new hampshire does have a lot of that. our tourism industry is very large. the voter fraud that has occurred tends to be among people wive in florida and

the winter and new hampshire in the summer and they voted in both places thinking no one would find out. there voter fraud the any numbers -- in any numbers that would sway an election. host: let's hear from rick now, in madison, alabama on the democratic line. caller: good morning. thank you for taking my call. your guest, how has the integrity of the surveys in his mind changed sinoi versua traditional phone call? thank you. guest: that is an excellent question. i think that is one of the reasons we are seeing this increasing use of web-based surveys. research scholars have done, looking at the accuracy of surveys done through different methodologies did not show any consistent that one methodology is more accurate than the others.