Susan elliott morris, poster and data journalist, your new book is entitled, strength in numbers how polls work and why we need them. This title reflects your philosophy about polls. Why do you think polls really play an important part in our society . Elliott that is the question of the book. When i set out to write it, i started from a statistical perspective. I am a reporter, not a pollster. When you consider a lot of polling information, you started to notice that they get things wrong frequently. The average in polls is what we like to focus on when forecasting elections but more importantly is what pollsters ask of people, what they want from the government and what leaders to Pay Attention to, the problems they want them to solve. So we have a plethora of clinical polling information before elections and they sometimes get things wrong but the issue of polls is accurate. I wanted to ask what we can make of the issue polling data, not to forecast an election but to increase representation for the average american, give the people and extra fighting chance in their democracy. Susan so your title is data journalist for the economist. Elliott the Data Journalism team was created in 2018 in 2000 14. Before i got there in 2018. The aggregation is political polling information on trying to figure out what people think and want in a statistical way that has not assisted for more than two decades. Some of the first polling aggregates came out 20 years ago. So that is a pretty limited time span. You notice when you cover polls that there is a lot of uncertainty about what we know about the world and the statistical aggregation of polls speaks to that uncertainty. The question i ask day today is are these polls showing real movements in the electorate, can we show what people want or is it statistical noise . Susan what is that training for a data journalist . Elliott i went to school for Political Science, not statistics at all although i have some statistical training. I graduated pretty recently. I think you can take any road you want to get there. Some of my colleagues were phds in Political Science, others were in the Health Industry in the u. K. Before becoming data journalists and others are just regular conventional journalists who became really interested in this. Susan what was it like writing your first book . Elliott very rewarding. At the economist most of our articles are short, it hundred words. This book is about 800 words. This book is 70 times that long. You can ask more questions about the subject of the book. It was great and there are Great Stories in the book that would require more articles to tell and i was able to tell those and bring the information to the people. Susan lets start on your subject matter with the two last president ial elections because they are fresh in peoples minds and how pollsters got it so wrong. Lets watch a compilation of pundits after the facts were talking about the polling. How could polls have missed it so badly . There will be a rigorous autopsy done on the polling that pervaded this campaign. How did the polls miss this . Because people are voting in much larger numbers for donald trump than anyone expected. The polls got the popular vote nearly right but missed badly on how the states were going to fall in terms of electoral votes. The group is predicting a Trump Victory even though other reputable posters predict a biden victory. You have to consider that we have been right so many times, [indiscernible] tune into american tv news recently and you might be left with the impression that joe biden was on course for a landslide. The National Polls were shoved down our throats. When i took over as Campaign Manager in 2016 we did zero National Polls. Susan how did the industry get it wrong twice . Elliott i understand the frustration or anger at pollsters. If you are a Hillary Clinton voter in 2016 and the polls say she is going to win and then she doesnt, i would blame that. Similarly, if you are a trump supporter and the polls say there are more biden voters, i would feel angry it is rigged against me personally. One thing to remember is when we have a whole, its only a survey of 1000 people on average. They do some math to make sure they represent the population as a whole but in those statistical tricks on the science of sampling where pollsters arrive at an estimate of what the population begins. Those estimates rely on some assumptions pollsters make about statistics about people and those assumptions can go wrong over time. In the 2016 election, pollsters field a survey and assume democrats and republicans will respond equally. Its hard to correct for. They can do some statistical tricks to make sure there are the right number of black or white voters, high income or low income but it is almost impossible to have a correct estimate of the number of democrats and republicans so if those groups are not answering your polls, if one is not responding, its almost impossible to correct for that all the time. Thats what you saw in 2016, republicans not really picking up the phone. Susan was it intentional . Was it because of a concerted effort that posters now understand . Elliott its hard to answer because we would answer that by using a poll but if theyre not taking the pole, they cant tell you if they are lying to you. Evidence on people lying to posters is pretty thin. What i call shy trump voters. They tried to have a balanced sample based on if people trust the government were neighbors and they do not find differences in support for a trump but i still say we do not have a final answer. Susan is the challenge trying to do this on a National Level when looking at a president ial race, or should people learn to look at key state and see the poles of a smaller population . Elliott one of my friends in that clip would tell you you should disregard the National Polls, Pay Attention to the aggregate polls and take a National Poll and do some statistical tricks to get an estimate in every state, as long as the u. S. Has an Electorate College, you want to aggregate the numbers in each state. There are a lot of polls and if there is shifting wind in the public, the National Polls will pick it up first but in terms of inferring where states lie, its missing the mark. Susan you have a message for pollsters in the book. Elliott yes. To be more honest about what they are doing. If you bring pollsters onto cspan and give them an extended interview, they will tell you they make estimates and they can get things wrong and you should not trust polls 100 and there is a margin of error in the numbers. But those messages dont really come through in the reports. When you read a report on the pole and there is an interview, its just usually about results. So suggestions are to do their own due diligence. But if they know from the past two election the polls can be wrong, they should remind the public, there is a margin of error. 5 , 10 , whatever it is and then give the story that it is a scientific estimate, not a true fact about the public. Susan explain margin of error. Elliott a statistical term for how wrong an estimate in a pole can be. 95 of the time. It goes back to sampling. If you pick up 1000 americans out of a bag of millions, you will get weird americans experts on the time in the margin of error will tell you there is a 5 chance that a candidate will win if they have a 53 support in the polls, a 3 margin of error. They base it off of sampling error. Assembling out of the back of americans. But other things can go wrong. There is waiting adjustments, figuring out if they have the right share of democrats and republicans, if theyre using recent estimates from the census for white and black americans, the tendency of some groups not to respond, and measurement error in the polls, whether they are asking the questions. So one thing to note is it does not cover 100 of the probability. Susan when people report on polls, you scroll to the bottom and there is always small print, the margin of error. What is an acceptable margin of error . Elliott they should put that at the top. An acceptable origin of error would be whatever the sample tells you. When there are 1000 people in the pole, typically does 3 . The numbers could be off 3 . You do not want to listen to an estimate that has a 20 margin of error but more polls are accurate than that. You want to be careful with the population of the sample size. Polls can change wildly between surveys. Susan you have a message for media. Elliott highlight the uncertainty. If we know the pole can be wrong, as we do, they can be more honest with the public when reporting the numbers. Abc news has a website with election forecasters. That is a more honest discussion. The errors inherent in polling and asking people what they want. Other posters at other pollsters at other news outlets, its the other journalists we have to be careful with the who might not be as trained in surveys as those employed for major media organizations. They should be honest when discussing numbers and give a better indication to consumers about how there is a margin of error. The message to the public is that no matter what the polls say, take it with a grain of salt. In 2020 the election forecasters were saying biden is up 8 but the average sized error means he could win by 1 or 2 , which might not be enough to carry Electorate College states. If reporters report honestly, the info gets to the public but they should not have to rely on the press to get this right. Susan today we had a university of virginia reporter same the ultimate pole is in the voting booth. What is it about the irresistible ability of preelection polling that the media continues returning to it year after year . Elliott there is an innate human desire to want to know the future and who will run the country. The stake in politics are really high. Straw polls were used in the 1824 election. Its nothing new. We have had elections before. In rome and greece, also stories in the book. More egalitarian than we have now. There is a desire to know who will run the country and preelection polls will give you that information and can serve as benchmarks on questions. The support of the percent of americans who support gun rights or abortion, they are that barometer of how accurate polls are. There is other information. You can ask the percent of americans who go on the internet or who own a refrigerator. These are key government statistics. There are a few reasons why you might want to take a preelection poll. Susan you said most of the work is not in preelection polls but in issues related. What is the value of issues . You talk about pollsters working issues. They reveal the will of the people, they are instrumental in shaping Supreme Court decisions, you have a whole paragraph on that, and informs legislators sometimes constraining them about their view of the public they are legislating. Elliott those are some examples of [indiscernible] of how the data gets used by the government. We do not just have anecdotal answers. In the book i try to all rely on Political Science evidence. You find that leaders are pretty responsive when you give them a poll of what people in their district want. There are a couple examples. One story in new mexico after a Budget Surplus in 2008 that finds that after letters were sent to lawmakers about Public Opinions in their district they were more likely to vote with the Public Opinion and lots of antidotes, one congressman in the 1960s says i am here to represent the people and give me as much information as possible on what they want and i will represent them. So the public can steer the ship of government and why not use polls in those areas to push the government in the right direction . That is one of the ideas of one of the founders of polls and it was a valuable instrument in taking the pulse of democracy and the book tries to elevate the idea beyond that susan do you want to say more about the Supreme Court . Elliott there is evidence from the Supreme Court decision that legalized gay marriage and then the over and then the legalization of abortion in roe v. Wade [indiscernible] she wants to be careful legalizing gay marriage because she thinks there is a backlash in Public Opinion polls after so she wants to make sure the public is in the right place before legalizing gay marriage. Justice roberts makes reference to Public Opinion polls in deciding to take up the case in the hodges case and the direction of the case in voting for more rights for samesex couples. Its anecdotal evidence but you will never get a quantifiable study at a Supreme Court that says this is moving the needle. Susan much of your book is on the history of polling that i noted you described the story of history of polling as enthralling, which only a data journalist would. Lets start with the 1824 election. The first recorded occasion of straw polls. How did they come into american politics . Elliott thats the earliest known reported straw poll in a newspaper. The reason they are called straw polls is because they are not statistical. They were voice votes taken at fourth of july parades and both leaders from the organization could know what people thought but also so newspaper reporters could write about what people honestly think. Essentially they were used as propaganda by the newspapers but they are the first real polls [indiscernible] you would not want to go back to this method of sampling the people. We dont have a great idea of the share of americans but we have some Great Stories about the first polls and what they might have told reporters that the time about people. Susan the iowa party still uses straw polls and widely reported. Why . Knowing how invalid they are statistically . Elliott i think they are fun. You see you go to a fair and maybe there are some statistical properties. Maybe we did them originally because they were fun. I do not advise people to listen to the straw polls from the iowa county fair if they are trying to figure out who could win iowa but maybe if the reporter puts the caveat in the piece they can still write about it. Susan you said straw polls began a dramatic change when new pavers got involved. Newspapers got involved. Elliott 1936 election predict the roosevelt would be defeated, a bad prediction famously. But there were strong polls before that conducted by newspapers owned by the Hearst Family and in partnership between newspapers in ohio and chicago and l. A. They were more accurate, at least in the records we have of the surveys they took. They were more systematic. Instead of sampling a fourth of july parade, you had employees of the newspaper or magazines go door to door or walk around town and ask people what they thought. Maybe they would stand at a light post with a clipboard. They were more systematic and Data Collection and there were more of them so you could average so they seemed more accurate. But it is still not a scientific poll but it is the rise of more systematic Data Collection. Susan 1936, literary digests story. Elliott the literary digest magazine conducted a nationwide poll of americans. They would send out their magazine with a square ballot you could cut out and mail that or consult lists of phone and car owners they had addresses for and send them a postcard and say check here if you are going to vote for roosevelt. And get back to us. So you had millions of ballots across the country at their headquarters and they count them up and report that roosevelt will lose by 12 Percentage Points and they were dramatically wrong. They were 38 off in the end on roosevelts margin of victory. It speaks to the weakness of the straw polling method. What goes wrong is the types of people who own automobiles and phones are different than the population of americans at the time. Those who do not own those are poor workers are poor and will likely vote for new deal politics and roosevelt. If you are voting for lent and you are probably also more enthusiastic to answer the poll. Susan they went out of business a few years later. Do you think this might have something to do with it . Elliott yes. It is a monumentally bad prediction. They were doing pretty well before that. Their revenue was fine. I would say they went out of business because of the failed prediction. Susan 1936, 1 of the people whose company is still the biggest name in polling entered the scene, George Gallup senior. On youtube we found a clip of him and i am going to play it so we can see. In the summer of 1922, i was a jr. At university of iowa and hired as an interviewer by an agency in st. Louis that was working on a project on a great survey of st. Louis for the st. Louis globe democrats. I was an interviewer and my assignment was to find out by going house to house what readers, what newspapers people were taking and what they read in them, what they liked. I went back to the university of iowa with a resolve to devise a better system and eventually, that became my doctoral thesis, a more objective way of measuring reader interest. Susan what was George Gallups contribution . Elliott that is the whole introduction for gallup in the book. He learned the first example of sampling, how to figure out what people want from the agency and he develops a method for figuring out what newspaper articles people like, partly by asking them, partly by watching them and marking down what they read, partly by picking up discarded newspapers and seeing where they stopped reading. He brings the tips to advertising agencies after graduating with the doctoral thesis on how to figure out what people want from their newspaper in order to sell more newspapers to an Advertising Agency where as the director of research, he spends some time and in 1930 one his motherinlaw runs for secretary of state in iowa and he wonders if it works for toothpaste, why wont it work for politics . So he goes home and runs primitive scientific holes polling to figure out what policy she might emphasize to get ahead in the race. And she wins. Thats our first example of electoral horse race scientific polling and i think its important to note it came from the same instincts as the advertising man to figure out what people want and to sell it to them. George gallup founds the American Institute of Public Opinion research in 1935 and does the polling in 1930 six and predicts the literary digest poll will miss the election, as they do and the gallup poll was more accurate so he becomes the electoral savant of the day and discovers along the way some incredible writing by james bryce who writes treatises about Public Opinion in america. They were used in textbooks in history class. In which he says Public Opinion, if we had a device to measure it, could lead america into a forced stage of democracy where Public Opinion not only informed leaders but world itself. From that point, George Gallup is the steward of polls has information for democracy as something again to tell leaders what people want and hopefully they will do it. His family is involved in progressive politics in the 1930s and 1940s to try to have Electoral Systems that are more reflective of the will of the people. Hes a big supporter of the new england town hall in politics. Ever since polling has changed as a tool for democracy, not just snapshots, George Gallup gives it a democratic [indiscernible] susan who is the first president to have their own poster . Elliott franklin roosevelt. He does not hire George Gallup. Gallup wants to conduct his own surveys and is doing a fortune. George gallup hires a littleknown former reporter from crystal falls, michigan, which still today only has 1000 people in the upper peninsula. He was from finish immigrants. He is George Gallup before there was George Gallup. Roosevelt hires him in 1932 campaign for president to run early polls of the Democratic Party standing in congressional races, of counties in america and states that needed more resources from the Democratic Party to push him over the finish line and he has to build this from scratch. He has never done a poll before doesnt know how but he is a miner so he knows about sampling. This is his speech to the director of the dnc at the time, who brings him aboard and he conducts early political polls and t and takes polls from George Gallup and suggests or corrects them and creates the first trend line of what americans want and he leads Congressional Democrats to victory in the 1938 midterms and continues to be roosevelts poster pollster through the remainder of his first three terms in office. He in 1937 because he doesnt like roosevelts Court Packing scheme but before that he is the political savant, the socalled wizard of washington. Susan to understand how big a deal he was, 1936 he was on the cover of time magazine, as we can see here. A name that is littleknown in washington today but hugely important and wellknown at the time during the new deal. Lets move to the 1950s. The landline telephone became almost ubiquitous in the 1950s and 1960s. How did it change polling . Elliott random sampling over the phone was invented in the 1970s. Until then the way you pulled was by going door to door and asking for someone at the house to tell you something about politics. You record their demographic information. George gallup could no if there were too many of one type in their sample and take make the appropriate adjustments. In 1948 sample it was predicted do we would defeat truman. Gallup and a group of political scientists figured out how to make doorknocking more systematic but it is expensive. You needed a lot of people to go to the doors. And a White Reporter does not want to go to a black part of town or neighborhood they feel is dangerous. There is examples of those people not being interviewed in the polls and people feeling uncomfortable doing it. So random telephone calling is cheaper and invented in the 1970s and more accurate and less susceptible to the biases of the interviewer. It is the obvious successor to the more scientific area sampling polls of the 1950s and 1960s. Susan 1960, john kennedy was the first to game out how a decision might impact voter behavior and we learn about the similar medics corporation and the calculating machine they use. Elliott that is the first microtarget are in america. They collected a lot information from the polls. The actual interview were rescuers records from gallup and roper and krauss lee and put all of the information in a computer. Not the type of laptop we have today, a computer that would fill the whole room and they gave it a personality of its own and ask to questions. They were figuring out if the vote share increased among some group, how does it impact the whole election. If he makes a play for catholics, either enough catholics to carry him across the finish line . You can ask the computer to run calculations instead of doing it by hand. It was a success. They pitch the idea to kennedys campaign and the dnc had given them money to run initial studies of his prospects and they were eventually acquired by the New York Times to run analysis and Election Night results in the 1960s to see if computers could lead this revolution in electoral handcrafting. Electoral handicapping. Kennedy uses it whether or not speech in support of civil rights will help him win. They published the report. I guess they were right because they debuted on the stock market as one of the richest corporations in america and a number of weeks and they went on to lose that money and declare bankruptcy but the intervening story of how they shaped politics is the first example of mathematical political analysis. Susan you said after losing, nixon became a voracious consumer of Public Opinion polls. Elliott i dont think you need polls to tell you that Richard Nixon was constantly concerned with his appearance. A chronic narcissist, at least in how we commission done absorbed polls. In the white house he has this constant apparatus in the west wing by his chief of staff to get polls from other posters, to do their own pulling and constantly assess his Approval Rating. So there are in in the white house and studies from interviews of people there at the time to see how the white house used the pulling information. The striking stories are if you gave nixon bad news, he would shut the door and shut down. So constantly the white house was searching for an area where nixon was doing well or a policy area they could focus on that was more popular than the president. There are some questions we can ask about this. If polls are used systematically in the white house to elevate the relevance of a certain policy, does it affect whether it passes in congress . On the other hand, if the president is continually figuring out where the public likes him and only focusing on those areas, are we missing the portrait of the man . In 1969, i guess it is that yes we did not know what nixon would turn out to be. Susan and then reflecting or leading in a democracy. Lets move on because we have less than 20 minutes left. In the 1970s and 1980s, Computer Technology and the phone pole people responded to was greater participation levels than they do today, lets go to early to thousands. You reference the technique earlier, pole aggregation. Why was it important . Elliott 1996 was the first instance in the American Press of someone averaging the polls together. Its important because if you have one single pole it can be off for any number of reasons. Sampling error, nonresponse. But if you have two or three, your chance that they are all biased in the same way is lower and if you go to five or 10, the chance you are having a quality reading of what americans want in terms of leader or policy is very high. At least thats the theory behind polling averages. In practice we know it is not entirely true. The last few elections, averages have been wrong. If people are systematically less likely in one party or one group of people to answer the polls, they are probably also less likely to answer the other polls so aggregation is not perfect to forecast and election were figuring out what people want but it is a revolutionary step in the right direction for controlling. Susan how important was political scientist Charles Franklin to this . Elliott he is the first public poll aggregator. He created a website in 2004 called political arithmetic. Com and contributed later to pollster. Com, a project between him and another polling pundit card called Mike Blumenthal and then dan rivers who funds their enterprise to put up the first website of polling aggregates and they are particularly popular in tracking president bush and then biden obama was oh obamas Approval Rating and there are some of the first averages for election forecasting. They had competition from nate silver, now famous for pulling aggregation and forecasting. So this conversation we are having could be quantified statistically. I it takes the steps of aggregation and forecasting to a new level. On one hand it is more systematic. I think it is more honest about what data says. Its just a reading, not a truth about what people think. George gallup believed in his polls. When they were wrong, they said we did something wrong. We have to figure out what it was so we can fix it the next time. For him the steaks were huge. It was this it was the future of democracy. But if you are trying to forecast and election, you do not want just one pole, you want hundreds. That is the environment we are in today. It also might give people a false impression of how accurate they can be. If you are telling people after the election you were right in 49 at 50 states, they might disregard the margin of error and use your model as the truth rather than the estimates. Susan what was nate silvers day job before politics . Elliott baseball. He came up with an algorithm that forecast Baseball Players success and that aggregation of baseball stats and forms what you can do with political pulling data. At the time he had been blogging on progressive websites to come up with forecast of whether or not clinton or obama would win the primary and what polls said about chances in a general election and that comes from the same place. Aggregating information together and studying the patterns had passed data to predict the future, you can come up with pretty reliable predictions. Susan how much did the media rely on nate silvers predictions in 2016 . Elliott not enough. There were other forecasters who gave clinton a 98 probability of victory. People in the press wrote about forecast as putting their thumb on the scale. At the time he predicted that trump had a 30 chance of victory in the 2016 election not because National Polls will be wrong but because the race is close. I wonder if everyone in the media had understood the 30 as meaning these polls, if you have three elections, they will be wrong one time. If your sample sizes is 18 elections, the polls are going to miss enough that the losing candidate could actually win. I wonder how the tone of the campaign wouldve been different and what people wouldve thought would happen. Susan heres nate silver in 2016 the importance of models. The good thing about system models is it commits you to rule. Instead of saying that early pools arent predict live and probably not, it says, ok, early pools arent predictive but at some point they become more so. To have an answer set up [indiscernible] assigned ahead of time is more helpful than people would think. One way of saying it is, i am not sure i am any better than the average pundit unless i have a model and the disciplining effect of a model doing thinking in advance and setting up rules of evidence is important. Susan you credit nate silver with changing political reporting forever in your book. Elliott and for the better. There are some downstream effects to presenting forecasts as confident readings of the public. And there are also effects of creating Data Journalism where there shouldnt be or creating a class of News Reporters who are only relying on data journals. These questions are hard to answer. But if we focus on if people are consuming polls with the appropriate sized grains of salt and margins of errors at the front of their mind, not at the back, i think the forecasts recently are a step in the right direction. Susan the next advance in technology is big data. How has that changed things . Elliott its counterintuitive with pulling. Polls are data and are big. So the laws of statistics say if you have enough data you will have an unbiased reading of the population. But it doesnt work for election forecasts. With the rise of the internet, posters take the study of public online and think, maybe we do not have to call 10,000 phone numbers. Maybe we could compile lists of americans and their information and ask them questions repeatedly over time. Those are called nonprobability polling because they are not based on the laws of probability sampling like previous polls. And they are incredibly powerful. They are even cheaper. They can be done almost instantly. The first nonprobability panel pull was conducted on settop boxes on tops of tvs. They had little red lights that went on when the Polling Company wanted you to answer a question. You would change her to handle two the settop box and get out a Remote Control and answer it and it would to send it back through the tv to the headquarters and they would pull that way. So you could conduct polling almost instantly and scale the practice to millions of people. Doug rivers is now the head of research and now you can go on your phone and conduct an answer polls and now they have 5 Million People who can answer a poll in a day. A departure from theories of how to poll but it is pretty accurate and thats another way of how polls are conducted that make them more accurate over time and solves some tech problems that have led them to be more susceptible to bias. Susan but you have to have internet. Elliott you have to have the internet, the time, to know enough about politics to want to take a survey voluntarily. So they have to do more modeling and math than they used to to get the right estimates. On one hand it is a jumps forward. But asking people what they want is getting harder and posters have to keep coming up with math techniques to come up with good predictions. Susan both methodology and communication continues having to change. In the state of things right now, to have confidence in the pool and journalists feel they are putting out accurate information, what needs to happen . Elliott five suggestions in the book. First, you should not conduct a poll via one method only. We have moved beyond only serving people by mail or by the phone. Now in the last few years, it has become cheap and easy to combine polls from different methods so you do not have the various problems of polling. That suggestion rise arises from posters. Another is for reporters to be more clear when writing about polls that these are estimates, they are not god telling you what americans want. We have discussed science and art in this hour and i think pollsters, reporters should reflect that in their work. There are some other suggestions put forward in the book. I think the official society of posters should be more clear about the polls they think are good or biased or bad and that serves as a useful signal to reporters and what they should produce. Susan your fourth suggestion is more political Interest Groups should devote themselves to more opinion. Does that not bring bias . Elliott it does. When you only have one or two Interest Groups elevating the voice of the people, if you have a lot of them from all corners, the bias theoretically cancels itself out but i would say in response that we should not listen to information from politically biased Interest Groups at all. In this case that recommendations from Interest Groups or companies that have Public Opinion polls, elevating the voice of the people. Connecting pools and saying what people want and relating that information to those in charge without the bias. Susan is pew Research Example of that . Elliott they are a public poster, not necessarily an interest group. In the latter chapters in the book i talk about having some commitment to what the numerical majority of americans are saying and relentlessly championing that. That is what gallup did but we have Interest Groups today who do it. That is what the mission should be rather than just conducting the poll. Really hammering that lawmakers should listen to it. It couldnt hurt to try. Susan the book is called strength in numbers how polls work and why we need them. Elliott morris is a data journalist for the economist on the author of the book. Thank you for spending the hour with us. Elliott thank you so much. All q a programs are available on our website or as a podcast on cspan now at app. Cspan is your unfiltered view of government. We are funded by these Television Companies and more, including cox. Hummer can be hard. Squatting in a diner for internetwork is even harder. That is why we are providing lower income students access to affordable internet. So homework can just be homework. Cox connect to compete. Cox along with these other television providers, giving you a front row seat to democracy. This week on the cspan networks, thursday night, the january 6 committee holds their eighth hearing investigating the attack on the capital. Also on capitol hill, tuesday, Pete Buttigieg testifies before the House Transportation Committee on the implementation of the 1. 9 trillion infrastructure bill passed last year. Wednesday morning, two hearings on guns and gun violence. The Senate JudiciaryCommittee Examines the Highland Park attack and civilian access to military assault weapons. And ceo from gun manufacturers will testify before the House Oversight committee on the practices and profits of gun makers. The house and senate are both in session with the house taking up the First Federal spending bill for 2023 and also work on legislation to protect up to protect a persons abilityyour f government. British Prime Minister burson johnson fielded questions from members of the house of commons almost a week after another he would step down as soon as every easement was in place. Before the start, the speaker threw out two members of parliament for disruptive behavior. Speaker before we come to Prime Ministers questions, i should point out that a british sign language interpretation of of proceedings is available to watch on parliamentlive. Tv. I start with robin millar. [interruption]