Iowa W 4 2016: Fill & Download for Free

The Sunday after he said all the churches should reopen, President Trump was playing golf at the Trump National in Potomac, Virginia.Trump himself said he went to church "as much as I can. Always on Christmas. Always on Easter. Always when there's a major occasion. And during the Sundays. I'm a Sunday church person. I'll go when I can.”[1]The Washington Post wrote:“He only occasionally attends a service, including multiple visits to St. John’s Episcopal across from the White House, which past presidents historically have attended, and Christmas and Easter worship at the Florida Episcopal church where he was married or at Washington National Cathedral, another Episcopal church.”Trump has always maintained that he is a member of the Marblegate Collegiate Presbyterian Church, but the church in an official statement said he was “not an active member.”[2]If the reader knows of any specific date when Donald Trump attended a regular church service, I would like to see it in comments. Please provide a source. So far I have the following from news reports and Trump’s official presidential calendar.[3]April 21, 2019, Easter, The Church of Bethesda by the Sea, Palm Beach, FLMarch 17, 2019, St. Patrick’s Day, St. John’s Episcopal, Washington, DCDecember 24, 2019, Christmas Eve, Family Church, West Palm Beach, FloridaDecember 5, 2018, George H. W. Bush Memorial Service, Washington National CathedralDecember 24, 2018, National Cathedral, Washington DC.September 3, 2017, St. John’s Episcopal, Washington DC[4]Prior to becoming president:January 24, 2016, First Presbyterian Church, Muscatine, IowaJanuary 20, 2017, St. John’s Episcopal, Washington DC, Private service. This was on the day of the inauguration, but before Trump took office.References:Trump mocks the faith of others. His own religious practices remain opaque.Update:The intent of this post was to answer the question with as much factual information as I could find, and I hoped that the comments would contain additions and corrections. There are a great many comments, but none that I can find adding or correcting the church attendance list.Some on the Christian side have gotten very upset about the question, and very defensive about the fact that Trump rarely goes to church. They take exception to something that is not in the answer here, an idea that Trump’s Christian credentials are tarnished because he doesn’t go to church. I’ve been a Christian for 60 years and I have heard many sermons attacking the excuse “you don’t have to go to church to be a Christian”; nevertheless, that excuse appears several times in the comments below.The comments have degenerated into name calling and personal attacks. I have gone through my own comments to remove any BNBR violations that might be implied. Since the original purpose of comments, to update the list, didn’t get information and since the discussion in comments really has nothing to do with the question, I am disabling further comments.Footnotes[1] Trump talks about religion[2] https://www.bemidjistate.edu/academics/departments/political-science/wp-content/uploads/sites/40/2017/08/herbert-thesis.pdf[3] President Donald Trump - Public Schedule Calendar | Factbase[4] KSL.com News Photo Viewer

A2AThis is a question many are asking.First, it’s worth assessing is it true? Do polls ‘tend to overestimate Democratic support and underestimate Republicans?’Historically, no. They don’t. A review of the 2012 polling data versus outcomes on RealClearPolitics (a site that track polls among other things) showed that the results of the polls provided a mixed bag in terms of the outcomes achieved in the election and the final poll numbers[1].Looking a their so-called “battelground” states, Mitt Romney actually tended to do slightly better in final polls than his actual performance against incumbent president Barack ObamaIn Ohio, polls favoured Obama by 2.9%; Obama won Ohio by 3.0%, a very, very slight underestimate of his support. (R +0.1)In Florida, Romney actually polled slightly ahead, with a weighted average of 1.5%. Obama won Florida by 0.9% (R+2.4)Virginia leaned to Obama in polls, favoured by 0.3%. Obama won Virginia by 3.9% (R+3.6)Wisconsin, poll: Obama +4.2%, vote Obama +6.9% (R+2.7)Iowa: poll: Obama +2.4%, vote: Obama +5.8% (R+3.4)Going down the list, the slant to Romney in polls vs the final vote was:NC (R+1.0)NH (R+3.6)PA (R+1.6)MI (R+5.5)AZ (D+1.6)You get the point. The 2012 polls leaned Republican on the whole, not Democrat.So the claim that polls lean to Democrats is not supported by historical data. It’s a mixed bag.Now, in 2020, the polls did seem to underestimate Trump’s performance. And more strongly, polls seemed to underestimate Republican support in senate races. In 2016, more to the point, the polls not only underestimated Trump’s performance, they actually got key states wrong. Notably, Pennsylvania, Wisconsin and Michigan were all going to be won by Clinton per the polls. And all of the polls in Florida, with the exception of pro-Republican Trafalgar polling, showed Clinton with a lead of 1–2 points. She lost Florida by more than 1%.In the case of Michigan (Clinton +3.5%) and Wisconsin (Clinton +6.5%), the polls were really wrong.Nate Silver, who leads the influential Five Thirty Eight blog, has been campaigning that, well, the polls were actually pretty good based upon two truths. First, the final “popular vote” results (Biden +3%) are actually pretty close to the national polls, second, the polls got 48 of 50 states right, and third, the gaps between polling and reality are close to the traditional margins of error anyways.Here is what’s wrong with Silver’s defence.First, the fact that the gap nationally in the popular vote being close to the result of national polls is a vanilla washing-out of errors. Setting aside that the presidential election is not a national referendum, the reality in statistical sampling is that the “accuracy” of a poll depends on the size (the truth is, the actual election is a poll whose sample size is 100% of the voting cohort) and the internal variance. One of the properties of sampling from an ostensibly binomial (Bernoulli) distribution [2]is that the confidence interval (‘error’) in the estimate gets smaller as one approaches 0% and 100%, and is at its maximum at 50%.That is to say, the inherent uncertainty in your estimate is at its maximum when the two candidates are at about 50–50. Which fundamentally makes sense - if the election is a true coin toss, my sample had better be damned large because the uncertainty of who will win is at the highest where support is the most evenly distributed. This has the unfortunate effect that your estimate is the least stable when you need it to be most accurate.Put another way, we do not see many polls from places like California, despite its being the largest state. Why? Because we already know who is going to win. It’s just a matter of whether the Democrat will get 60, 65, or 70 per cent of the vote.When the “national” polling data are coalesced, it is doing two things. One, it is making the sample size largest, which reduces the total sampling variance. And two, it is including data from many states where the outcome is already known, which reduces the sampling variance.The national sample just washes out much of the “noise” that is inherent in the state by state samples. In an election where the intra-state samples are important.The second point that Silver makes - that polls got 48 of 50 states “right” - is silly on its face. We all know that the lion’s share of the states are going to be one way or the other. We do not need polls to know that Wyoming is going to go the Republican, or that New York to the Democrats.Put another way, I know nothing about football. But in the early NCAA season, teams like Ohio State routinely match up against teams like Miami of Ohio. That I can “predict” with close to 100% accuracy how Big Ten football teams will do in the first three weeks of the season does not mean I know a darned thing about NCAA football.Polls should get virtually all of the states right. Their utility is in the handful of states where the results will be close. So, let’s look at 2016 and 2020 in how the battleground polls did in the 10 battleground states that in a sense decided the election.2016 (Predicted, Winner)Florida (Trump +0.2, Trump +1.2) CORRECTIowa (Trump +3.0, Trump +9.5) CORRECTMichigan (Clinton +3.4, Trump +0.3) INCORRECTPennsylvania (Clinton +1.9, Trump +0.7) INCORRECTWisconsin (Clinton +6.5, Trump +0.7) INCORRECTOhio (Trump +3.5, Trump +8.1) CORRECTNorth Carolina (Trump +1.0, Trump +3.7) CORRECTNew Hampshire (Clinton+0.6, Clinton +0.3) CORRECTColorado (Clinton +2.9, Clinton +4.9) CORRECTVirginia (Clinton +5.0, Clinton +5.4) CORRECTThe polls got 7 of 10, which at 70% is ok. It’s a passing grade in school, but barely a C. It’s also worth noting that in 8 of the 10 states, Trump did better with the actual vote than the polls. 80% of the polls leaned Clinton (back to this in a second). Two of the eight were wrong by more than five percentage points (Iowa and Wisconsin), and a third (Ohio) was off by 4.6.That’s actually a pretty poor showing.In 2020Florida (Biden +0.9, Trump +3.3) CORRECTIowa (Trump +2.0, Trump +8.2) CORRECTMichigan (Biden +4.2, Biden +2.7) CORRECTPennsylvania (Biden +1.2, Biden +1.0) CORRECTWisconsin (Biden +6.7, Wisconsin +0.7) CORRECTOhio (Trump +1.0, Trump +8.2) CORRECTNorth Carolina (Trump +0.2, Trump +1.3) CORRECTNew Hampshire (Biden +8.0, Biden +7.2) CORRECTColorado (Biden +9.0, Biden +13.1) CORRECTVirginia (Biden +11.0, Biden +9.4) CORRECTThis year, the polls did much better, getting the 10 “battlegrounds” right in all cases. But part of that may be down to the reality that the final polls were just not as close, as the errors in many cases were actually worse. For example, polls underestimated Trump’s final result by 5 points or more in Ohio (-7.2), Iowa (-6.2), and Wisconsin (-7.4). Three out of ten estimates were wrong by more than 5 points. Two, by more than 7.Third, and finally, it is virtually certain that polls that are short of a 100 per cent sample are going to be “wrong” due to sampling vagaries. It’s unlikely that a poll is going to pick the exact support level in the final vote.One would expect then that about half the polls would overestimate one candidate, and half the other.This was not the case in the sample. In 2020 battlegrounds, directionally, 9 of the 10 had errors in the same direction. Colorado alone underestimated Biden’s support. Adding in the 2016 data, polls over-estimated the Democrat in 17 of 20 cases. (Oddly, Colorado in both 2016 and 2020 gave a higher share of its final votes than polls “predicted).17 out of 20 coin tosses came up ‘heads.’If you tossed 20 coins and got 17 heads, you might get suspicious about the guy tossing the coin, or perhaps the coin itself.Pew Research dives into the outcomes in some depth here.All that said, it’s clear that national and many state estimates were not just off, but off in the same direction: They favored the Democratic candidate. To measure by how much, we compared the actual vote margins between Republicans and Democrats – both nationally and at the state level – with the margins in a weighted average of polls from FiveThirtyEight. Looking across the 12 battleground states from the upper Midwest (where many polls missed the mark) to the Sun Belt and Southwest (where many were stronger), polls overestimated the Democratic advantage by an average of about 4 percentage points.Why is that?Polling is a sort of science, and its accuracy (or lack of accuracy) is really a function of sampling. You must get an adequately large sample - the size of which is proportionally related to how close to 50–50 your population (again, if you sample 1000 people, and all 1000 say they will vote Democrat or Republican, chances are, the Democrat or Republican is actually going to win; if you get 505 and 405, you better take a larger sample) - and your sample had better be carefully taken to avoid any obvious biases.I suspect in 2020 (as in 2016), the sample sizes were not the problem.Thus the issue is almost surely a function of sampling frame. Are you systemically sampling in such a way that your sample is not representative of your population? For example, if you took your data exclusively on college campuses, that is probably not going to be representative of the state. If you survey on or around golf courses, that, too is likely to mis-represent your population. Polling companies earn their money by creating sampling frames to take into account demographics.Second, there is the suggestion that Republicans simply ignored polling or gave answers not aligned with the truth. This is the so-called “Shy Trump Voter” phenomenon that is described in the news. Are Trump supporters more likely to give the wrong answers to a pollster? I think it’s more likely that Trump supporters, who have listened to six years of attacks on the media, complaints about biases, and tirades about polls, are likely to just not reply to pollsters. This is different from giving the wrong answer, but nevertheless has the same outcome.If Trump supporters are more likely to refuse to participate in surveys, this almost surely means any polls are going to oversample Democrats and undersample Republicans.This would align with the data presented - that in 2012, polls overestimated Romney’s support, but in 2016 and 2020, they underestimated Trump’s.I suspect that Trump’s attacks on the media have convinced supporters not to participate or cooperate with “the mainstream media,” which has the effect observed.Here is what I would rule out as explanations.First, the impact of mail-in voting. Because so many more people this year opted to use absentee ballots, the pool of “likely voters” is going to look different from the pool of final, actual voters.That at first blush sounds reasonable. But if it were the case, it would make sense that polls would undersample likely Democrativ voters, as Democrats were far more likely to cast absentee votes than Republicans. Turnout this year - fuelled in large part by mail-in votes, was much much higher than 2016. Traditional sampling cannot take into account the vagaries of absentee votes, likely underestimating how many mail-in voters would participate. But then, when they do, these voters likely broke for Biden, not Trump.Unless the polling firms were trying to correct for the impact of mail-ins, and over-corrected, the results would be the opposite of what was witnessed.Second, replies here comment on the “gutting” of the 1965 Voting Rights Act. But the portions of the Act that were tossed out in 2013 by the Supreme Court were specifically oversight rules for a handful of states (and counties) that historically had patterns of discrimination. Prior to 2013, and Shelby v Holder, election boundaries and other rules proposed by jurisdictions had to be approved by court review.Below is a list of constituencies affected.Mostly, these were states in the Jim Crow south, plus Arizona and Alaska.Virigina is covered (one of the two states in 2016 that Trump got fewer votes than his polls indicated), as is Georgia (which polls showed leaned Trump, but ultimately was won by Biden). Arizona as well - polls in Arizona showed Biden as +0.9; he won by +0.3. Polls slightly underestimated Trump’s performance (+0.6).States that you won’t see represented: Ohio, Michigan, Pennsylvania, or Wisconsin.We could argue the propriety of the 2013 decision, but arguments that it is the cause for the non-random way polls were off is simply not supported by data.Footnotes[1] Battleground State Polls | RealClearPolitics[2] Bernoulli Distribution

I think there are four main reasons:First, Hillary Clinton was a poor overall candidate, and a terrible candidate for the particular mood of the country. A few years ago, when President Obama was talking about the 2016 election, he noted that after 8 years, the American people would likely want a “new car smell” in the White House. And he was absolutely right: even the most successful presidents struggle to pass the baton to their party for a third consecutive term. We’ve had numerous two-term presidents since WWII, but only Reagan’s anointed successor, George H.W. Bush, was able to win an election (and of course, he was ultimately a one-term president). So even though Obama has been a reasonably popular president—and has been extremely popular with Democrats—Democrats were always facing an uphill battle in 2016.But that being said, Clinton was a terrible choice; voters (both with Obama in 2008 and 2012, and the Tea Party in 2010 and 2014) have been screaming at the top of their lungs for years that they are dissatisfied with the status quo and perceived corruption of Washington, so it was strange to pick someone who more or less personifies the corrupt Washington establishment. Add to it that Clinton had no core message—“I’ll be Obama’s third term! No, wait! I’ve always been a strong progressive who hates globalist policies!”—and is a poor “retail” politician, and it’s amazing that she came as close to the presidency as she ultimately did. A lot of blame for Clinton's defeat has been placed on the FBI, Russia, and even the American people themselves, but Clinton deserves the most blame, as a better candidate would have beaten Trump and provided coattails to other candidates seeking office.Second, for all the talk over the past few years about the GOP allegedly dying—since about 2008, hopeful lefties have written countless “concern” pieces and obituaries for Republicans—the Democrats’ civil war wreaked havoc on the party’s messaging. Hillary Clinton, in my opinion, has few core principles, but leaked emails regarding paid speeches showed some of them, namely that she is a free trade loving globalist. Again, these were not the right views for 2016, and Bernie Sanders's successful primary campaign forced Clinton (an economic moderate) to rebrand herself as a lifelong progressive, a suit that simply didn't fit and a message that most progressives didn't buy. Instead of staking out a position in the political center, Clinton was forced to take a sharp left turn to try to stave off Sanders and appease his supporters. Again, this was not the campaign Clinton wanted to run, and as a result, Democrats struggled to find a unifying message other than “we've got to make sure that Trump doesn't win.”Third, Democrats hadn't spent much time working on their “bench” of young politicians since George W. Bush was president. After 2004—a year Democrats thought they'd retake the White House—the party worked very hard to up its ground game and recruit better candidates who appealed to a wider array of voters. As a result, Democrats retook Congress and later controlled the government for the first two Obama years. But the party learned the wrong lessons, thinking that because Nancy Pelosi was so good at raising money and Barack Obama was such a great campaigner, developing the next generation of leadership didn’t matter. The last few years have also witnessed so many nonsensical books/articles about how Democrats would enjoy permanent majorities simply as a result of demographic changes. Partisans ignored, however, that (1) party platforms aren’t static (Trump ran far to the left of conventional GOP policies on many issues), (2) voters aren’t static in their allegiances, (3) talk of “never-ending majorities” always seems to peak right before a fall, e.g. Republicans making the similar claims of extended dominance in 2005, and (4) even with a supposed advantage, it’s wise to outwork your opponents and leave nothing to chance. What 2016 shows is that demographics aren't destiny, and picking and supporting good candidates still matters a great deal; as a result of complacency, there are now huge parts of the country where Democrats are not at all competitive.Fourth, as Neeraj Shukla correctly pointed out in a comment below, perhaps due to overconfidence regarding a “demographic advantage,” Democrats ran a poor campaign from a technical perspective. Many liberals have been shouting “Clinton won by 3 million votes” as though that has some meaning in the electoral college system. Another—more accurate—way of looking at 2016 is that Trump put a laser focus on votes he absolutely had to have in order to win the election and got them, whereas Clinton showed no similar discipline. Trump campaigned hard in states that Obama had won in 2012 such as Michigan, Pennsylvania, Wisconsin, Iowa, and Ohio. Due to those efforts, he carried the first three states by razor-thin margins, and the latter two by surprisingly large margins; if a few of these contests went the other way, Trump would not have won. Trump didn’t spend time in states that were lost causes for him, such as California, Massachusetts, or Illinois, which allowed Clinton to run up (meaningless) margins of victory in states she was always going to win. In contrast, Clinton spent precious little time in swing states other than Florida (and when she did, it was celebrity-oriented events like a concert in Philadelphia), instead spending a lot of time in California (mostly fund-raisers) and Texas of all places (another big source of funds and a place liberals have dreamed of turning blue for decades without success). A few more days in Milwaukee, Cleveland, Philadelphia, and Detroit, and things might have been different.And fifth, the backlash against political correctness and perceived liberal smugness hit Democrats hard. Within the liberal echo chamber, things like “safe spaces” and privilege theory may make sense, but in much of the country, these concepts are offensive, to the extent they are even comprehensible. People struggling to make ends meet do not want to be lectured about their supposed privilege by finger painting rich kids. Similarly, progressives experienced a lot of success during the Obama years, but instead of being magnanimous, decided that people who opposed them were worthy of contempt. Take LGBT rights as an example: over the last 10 years, gay marriage went from being a fringe issue to something that the majority of the country supported. This would seem to speak highly of the compassion of the American people as well as their ability to change views when confronted with compelling arguments. But not to progressives: instead, there's always another battle to fight, always new politically correct words to learn, and if people aren't on board 110%, they're scum. I live in Houston, for example, which was the first major US city to elect an openly gay mayor (who was reelected twice). New York has never had a gay mayor, nor has Chicago, nor has San Francisco; nope, it's us hicks down in Texas. But when the city voted down an ordinance over concerns about biological men using women's bathrooms (a concern that was overblown, in my opinion, but nevertheless mattered to a lot of people), we suddenly became the city of hate. Never mind that the ordinance was initially passed in secret to avoid a fair debate, or that the mayor's office was literally subpoenaing churches in an attempt to stifle criticism. Nope, if you weren't on board with the progressive agenda without question, it was only because you are stupid and bigoted.Is it any wonder that Clinton's “basket of deplorables” remark had such staying power? It wasn’t a surprise that a certain class of liberals disdains much of the country; rather, the only surprise was that such a remark was made openly. Again, I think the cause of gay rights illustrates that given time and reasonable argument, the American people can change their minds a great deal. But by 2016, too much liberal “discussion” was along the lines of “we're good people, and here's what we want. If you aren't on board, it's because you're a bad person.” Very few people respond well to having their lives, their culture, and their faiths attacked, and so a backlash was inevitable.