Margin of Σrror

Margin of Σrror -

You can trust Gallup’s numbers … for now | Harry J Enten

Method changes last year have substantially improved Gallup’s polling accuracy. That’s good news for early trend spotting

Gallup screwed up in 2010 and 2012. They called for a much wider Republican House victory in the former than what actually occurred, and they polled a Mitt Romney victory in the latter. That put them on a blacklist of sorts with some polling analysts. And while I am just one person, I want to say that I trust Gallup’s presidential approval numbers as much as any mainstream pollster at this point.

Why? The main question when looking at a poll is whether or not the numbers accurately reflect the current state of the country. Methodology is important, but without accuracy, methodology is rather useless. In Gallup’s case, solid technique has led to Gallup’s approval track being reliable since last October.

While Gallup was losing a lot of creditability in the presidential horse race, they were making adjustments to their presidential approval methodology. They added a higher percentage of cell phones, changed weights for the geographic distribution of Americans, and many other minor changes starting on 1 October 2012. These alterations made all the difference in the world (or, at least, the country).

From the beginning of August through the end of September 2012, Gallup pegged President Obama’s approval at only 47.1%. All other pollsters who also used live interviews and called all adults found an average presidential approval of 49.7%. Given undecideds, the difference between the two percentages is that of a president who is in major re-election trouble and one who is probably going to win a tight race.

From 1 October 2012 through the election in early November, Gallup all of a sudden was projecting Obama’s approval at 50.8%. We’re talking over 17,000 interviews in October and nearly 30,000 in the eight weeks prior, so that movement in Gallup’s Obama’s approval rating was outside any margin of error.

The change was not seen by other polling outfits. Although most pollsters switched to a “likely voter electorate”, the ones that continued to poll all adults discovered an average Obama approval of 49.7% – exactly the same as they had produced in the two months prior.

Thus, the movement seen in Gallup’s weekly numbers can only be ascribed to its change in methodology. A change that produced an approval rating indicating President Obama’s re-relection, unlike horserace numbers. The anti-Obama house effect in approval rating essentially disappeared overnight.

Since the election, Gallup has continued to keep the reliable work up. HuffPollster aggregates approval ratings from all pollsters and allows the ability to sort by population (adults, registered, and likely voters) and mode (live telephone, automated telephone, and internet). I have selected adults and live telephone, like Gallup, from all non-Gallup pollsters and compared this plot of local regression to Gallup’s trend* since October 2012.

What we see is what we’d want to see from a trusted pollster. Gallup’s numbers have tracked very well with the aggregate of the other pollsters. Both groups gave Obama post-re-election boosts. Both have shown Obama’s approval rating dropping since January. Both have Obama’s approval fall accelerating since the NSA story broke in early June.

The only difference you’ll notice is that Obama’s approval and disapproval ratings are slightly higher in the overall group. The disparity is only about a point for both. All it means is that other pollsters have designed questionnaires that end up pushing undecideds a little harder. The net approval of both pollster groups is the same.

So what’s this all mean? It means we have the ability to catch trends more quickly thanks to Gallup’s daily and weekly tracking data. Gallup also has very good crosstabs, which thanks to high sample sizes, have relatively small margins of error.

Of course, the major question going forward is whether or not Gallup can translate this success with all adults to registered and likely voters for elections. If they can’t, Gallup will continue to earn a bad rap. If they can, then Gallup’s reputation may be restored.

*Note: An average may produce slightly different results from the averaging technique spoken about earlier.

theguardian.com © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Another clue to Obama’s 2012 victory: Americans weren’t doing so badly | Harry J Enten

Revised income data shows that Americans have had more disposable income than we thought

What happens if Americans got wealthier when nobody was looking? There are a number of ways to measure the health of the economy: gross domestic product (GDP) growth, job growth, and how much disposable personal income (DPI) Americans have. It turns out Americans have had more of the latter, if we are to believe a re-calculation from the Bureau of Economic Analysis (BEA).

In its August report, the BEA revised a bunch of a measures including DPI, which was “revised up for 1929-2007, down for 2008, and up for 2009-2012″. That means essentially that Americans have had more money in their pockets for most of the last 84 years than originally thought.

Why upward? Mostly because of “revisions to supplements (specifically, employer contributions for employee pension and insurance funds) for 1929-1975, for 1989-2002, and for 2004-2011. A number of other definitional and statistical changes affected the revisions to personal income”. Put another way, employers were giving more to benefit packages for its employees than originally thought.

While it’s interesting from an economic angle, it has political ramifications as well. No, I don’t mean that President Obama is fudging the numbers or anything like that. It goes a long way in explaining in why President Obama won last year.

In 2012, most, though not all, economic indicators predicted a small Obama victory. As Nate Silver illustrated, however, personal income presented an economy that was far weaker than other measures such as job growth. That’s why the Douglas Hibbs’ bread and peace model that takes into account election data from 1952 to 2008 failed miserably.

The model, which is based off real disposable personal income growth and military fatalities in wars instigated by the United States*, predicted Obama would only win about 47% of the two-party vote. That, of course, ended up very far away from his actual 52% of the two party vote.

Had Hibbs had the data we have now for income growth, he would have projected that Obama win about 49% of the vote with a weighted income growth of 0.76%. The higher estimate is because Americans had more money than we thought, and growth was better coming off of a more depressed 2008 than originally believed.

Now, this is still an error of 3pt, but it’s well within the model’s margin of error of greater than +/- 4.5pt. In other words, it’s the type of error that we would expect to come along every once in a while. It just so happened that in a close election, the error went the wrong way.

My own amended Hibbs model would have had Obama winning 51% of the two-party vote–quite accurate. My model was aimed at trying to account for institutional factors that shape presidential elections.

My model added the number of terms a party has been in power with a penalty for those which had been in power for more than one term. It’s the same variable featured in the very accurate Drew Linzen model. In the real world, Obama benefitted because some of the blame for a bad economy was laid at the feet of the prior president.

Obama also may have benefitted from the House of Representatives being in Republican control. My model says the national vote would have been decided by less than a point (instead of four) had Democrats had complete control of Congress. Voters were more likely to apply any blame for the current woes of this country to both parties equally.

Of course, all of this is in hindsight. Both my model and Hibbs’ original were less accurate with data available at the time of the election. It’s not that the models were over-fitted like those from the University of Colorado or that the models needed to be refitted; it’s that the data needed simply wasn’t available. For forecasting purposes, DPI perhaps isn’t the best tool.

For explaining elections, however, disposable income growth over the course of a president’s term continues to do a pretty good job. If people have more money in their pockets, they’re more likely to reward the party in power. This reward is heightened when a president’s party has only had one term in the White House. My model explains over 92% of differences in past election results since 1952. The 2012 election didn’t change that.

*Per Hibbs, a president gets a term to end a conflict before fatalities from a war started under another party’s watch affect his party’s re-election chances.

theguardian.com © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Another clue to Obama’s 2012 victory: Americans weren’t doing so badly | Harry J Enten

Revised income data shows that Americans have had more disposable income than we thought

What happens if Americans got wealthier when nobody was looking? There are a number of ways to measure the health of the economy: gross domestic product (GDP) growth, job growth, and how much disposable personal income (DPI) Americans have. It turns out Americans have had more of the latter, if we are to believe a re-calculation from the Bureau of Economic Analysis (BEA).

In its August report, the BEA revised a bunch of a measures including DPI, which was “revised up for 1929-2007, down for 2008, and up for 2009-2012″. That means essentially that Americans have had more money in their pockets for most of the last 84 years than originally thought.

Why upward? Mostly because of “revisions to supplements (specifically, employer contributions for employee pension and insurance funds) for 1929-1975, for 1989-2002, and for 2004-2011. A number of other definitional and statistical changes affected the revisions to personal income”. Put another way, employers were giving more to benefit packages for its employees than originally thought.

While it’s interesting from an economic angle, it has political ramifications as well. No, I don’t mean that President Obama is fudging the numbers or anything like that. It goes a long way in explaining in why President Obama won last year.

In 2012, most, though not all, economic indicators predicted a small Obama victory. As Nate Silver illustrated, however, personal income presented an economy that was far weaker than other measures such as job growth. That’s why the Douglas Hibbs’ bread and peace model that takes into account election data from 1952 to 2008 failed miserably.

The model, which is based off real disposable personal income growth and military fatalities in wars instigated by the United States*, predicted Obama would only win about 47% of the two-party vote. That, of course, ended up very far away from his actual 52% of the two party vote.

Had Hibbs had the data we have now for income growth, he would have projected that Obama win about 49% of the vote with a weighted income growth of 0.76%. The higher estimate is because Americans had more money than we thought, and growth was better coming off of a more depressed 2008 than originally believed.

Now, this is still an error of 3pt, but it’s well within the model’s margin of error of greater than +/- 4.5pt. In other words, it’s the type of error that we would expect to come along every once in a while. It just so happened that in a close election, the error went the wrong way.

My own amended Hibbs model would have had Obama winning 51% of the two-party vote–quite accurate. My model was aimed at trying to account for institutional factors that shape presidential elections.

My model added the number of terms a party has been in power with a penalty for those which had been in power for more than one term. It’s the same variable featured in the very accurate Drew Linzen model. In the real world, Obama benefitted because some of the blame for a bad economy was laid at the feet of the prior president.

Obama also may have benefitted from the House of Representatives being in Republican control. My model says the national vote would have been decided by less than a point (instead of four) had Democrats had complete control of Congress. Voters were more likely to apply any blame for the current woes of this country to both parties equally.

Of course, all of this is in hindsight. Both my model and Hibbs’ original were less accurate with data available at the time of the election. It’s not that the models were over-fitted like those from the University of Colorado or that the models needed to be refitted; it’s that the data needed simply wasn’t available. For forecasting purposes, DPI perhaps isn’t the best tool.

For explaining elections, however, disposable income growth over the course of a president’s term continues to do a pretty good job. If people have more money in their pockets, they’re more likely to reward the party in power. This reward is heightened when a president’s party has only had one term in the White House. My model explains over 92% of differences in past election results since 1952. The 2012 election didn’t change that.

*Per Hibbs, a president gets a term to end a conflict before fatalities from a war started under another party’s watch affect his party’s re-election chances.

theguardian.com © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Why the Republican coalition will still work in 2016 | Harry J Enten

Predictions of demographic doom for the GOP are wishful: polling shows that winning big with white voters can deliver

The faultline in the GOP revealed by the party’s internal debate on immigration reform – over whether a future Republican coalition should rely more heavily on whites than it already does, or should try and bring more Latinos into the fold to win the presidency – remains unresolved. What we can say is that the last election and current polling suggest that the Republicans’ path of least resistance is to win even more non-college-educated whites and to try to win somewhat more of the minority vote.

Start with the fact that in 2012 Obama lost a little more than 3pt off his margin of victory in 2008. That swing was not uniform. It’s fairly clear that Obama’s share fell significantly with white voters without a college education, stayed about level with whites with a college degree, gained a few points with Latinos, and may have lost a point or two with black voters.

I don’t view the incongruity between those shifts as a sign that Obama should have done worse or better. In any election, you’re given a certain amount of leverage by the state of the economy, and you need to use it where you can. In 2012, the parties found the coalitions that worked for them. It just so happened that Obama had more room for growth electorally because the economy was “good enough”.

Indeed, there is little to no sign that Mitt Romney did worse than he should have, given the state of the economy. President Obama won by a little less than 4pt, when the two best fundamental models (based solely on numerous different economic factors) had him winning by 3pt and 5pt respectively. Taking into account Obama’s approval rating and the economy, as the original Alan Abramowitz model does, shows Obama should have won by a little more than 4pt.

That’s the reason why I don’t buy the argument that the shrinking white population in this country necessarily spells doom for the Republicans. This is a two-party system where the economy almost always dictates who wins and loses elections. No one has yet proved that the 2012 election indicates that the Republican party needs to change fundamentally in order to win, despite hundreds of column inches expended on the subject.

For Republicans to win, they’d need economic conditions slightly more favorable to the out-party (that would have been, in 2012, a worse economy and less confidence). Following the 2012 pattern, this would allow them in 2016 to continue to do exponentially better among white working-class voters. Sean Trende, who believes that the GOP could win with a mostly white coalition, anticipates Republicans also gaining a few more points among minorities.

Of course, many doubt this steady-state strategy could work for Republicans. Karl Rove said a few months ago that Republicans would have a hard time regularly winning the white vote by 25pt or more. But that’s the funny thing about electoral rules: they’re made to be broken. For example, the aforementioned Alan Abramowitz said that Republicans would have a very hard time getting above the 58% of the white vote in 2010 that they had in 1994. In fact, they won 62% of the white vote in the last midterms.

That’s why Trende has vigorously argued that the demographic wall facing the GOP doesn’t really exist. The worsening Democratic performance among white voters we have seen recently is part of a longstanding trend. If the pattern continues, then white support for Democrats will continue to drop below its current historic low.

So, now we have a test case of sorts in 2014. The economy really hasn’t gone south. Economic confidence is far higher than it was at the beginning of 2013, although it has stalled slightly. The percentage of those who view economic recovery as imminent has fallen slightly over the past few months, but it’s only slightly lower than where it was at the time of the 2012 election. Put another way, there hasn’t been that level of decline in the economy which many thought would need to happen for Republicans to win with the coalition they have.

Yet, Obama’s net approval rating with white voters is no better than -25pt right now. His approval in Gallup’s polling among white voters since the NSA leaks is now only 34.5%. That’s a 4.5pt drop since the election. Pew has it slightly lower, at 33%, which is a 6pt drop off what Obama was showing in their final pre-election poll last year. It seems as though that wall keeps moving.

Most, if not all, of this drop for Obama is among whites without a college degree. Pew found support from that segment of the electorate dropping by a little less than 10pt. In other words, there is a continuity here with the November election result, in which Obama’s support fell the furthest among whites without a college degree. If you think this might be pegged to public reaction to the Trayvon Martin case, the Zimmerman verdict and the president’s response, it’s not. Pew found college-educated whites reacted in the same way as non-college-educated whites to that issue – yet Obama has seen no decline in his standing among college-educated white voters.

Indeed, the swing looks much as expected: we have that drop among the non-college-educated whites, and among minorities to a smaller degree. We can see this by pooling Gallup’s data since the Edward Snowden/NSA affair; this gives us a very large sample size. That’s important because minority sample size in polls is often very small, which is why Pew and NBC News/Wall Street Journal differed so much in their minority findings this week. Gallup splits the difference and discovers that non-white approval of Obama has fallen since the election by 4pt, or about half of the drop we’ve seen with non-college-educated whites.

Now, things could change and this analysis may end up way out-of-date come 2016. I should also add that the eventual party coalitions may look somewhat different then, with Republicans winning back some African-American voters once Obama is out of office, but maybe also Democrats making a slight recovery with white voters.

But the most salient fact is that Obama’s approval among registered voters is a weak 45% – and that’s without a bad dip in the economy or any grave scandal. This soft approval rating gives credence to what the polling suggests: that the Republican party is most likely to win the White House in 2016 with a coalition that includes even more non-college-educated whites and a slight increase among minority voters.

theguardian.com © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Why the Republican coalition will still work in 2016 | Harry J Enten

Predictions of demographic doom for the GOP are wishful: polling shows that winning big with white voters can deliver

The faultline in the GOP revealed by the party’s internal debate on immigration reform – over whether a future Republican coalition should rely more heavily on whites than it already does, or should try and bring more Latinos into the fold to win the presidency – remains unresolved. What we can say is that the last election and current polling suggest that the Republicans’ path of least resistance is to win even more non-college-educated whites and to try to win somewhat more of the minority vote.

Start with the fact that in 2012 Obama lost a little more than 3pt off his margin of victory in 2008. That swing was not uniform. It’s fairly clear that Obama’s share fell significantly with white voters without a college education, stayed about level with whites with a college degree, gained a few points with Latinos, and may have lost a point or two with black voters.

I don’t view the incongruity between those shifts as a sign that Obama should have done worse or better. In any election, you’re given a certain amount of leverage by the state of the economy, and you need to use it where you can. In 2012, the parties found the coalitions that worked for them. It just so happened that Obama had more room for growth electorally because the economy was “good enough”.

Indeed, there is little to no sign that Mitt Romney did worse than he should have, given the state of the economy. President Obama won by a little less than 4pt, when the two best fundamental models (based solely on numerous different economic factors) had him winning by 3pt and 5pt respectively. Taking into account Obama’s approval rating and the economy, as the original Alan Abramowitz model does, shows Obama should have won by a little more than 4pt.

That’s the reason why I don’t buy the argument that the shrinking white population in this country necessarily spells doom for the Republicans. This is a two-party system where the economy almost always dictates who wins and loses elections. No one has yet proved that the 2012 election indicates that the Republican party needs to change fundamentally in order to win, despite hundreds of column inches expended on the subject.

For Republicans to win, they’d need a slightly more favorable economic conditions for the out-party. Following the 2012 pattern, this would allow them to continue to do exponentially better among the white working class. Sean Trende, who believes that Republicans could win with a mostly white coalition, gives Republicans a few more points among minorities.

Of course, many doubt this steady-state strategy could work for Republicans. Karl Rove said a few months ago that Republicans would have a hard time regularly winning the white vote by 25pt or more. But that’s the funny thing about electoral rules: they’re made to be broken. For example, the aforementioned Alan Abramowitz said that Republicans would have a very hard time getting above the 58% of the white vote in 2010 that they had in 1994. In fact, they won 62% of the white vote in the last midterms.

That’s why Trende has vigorously argued that the demographic wall facing the GOP doesn’t really exist. The worsening Democratic performance among white voters we have seen recently among is part of a longstanding trend. If the pattern continues, then white support for Democrats will continue to drop below its current historic low.

So, now we have a test case of sorts in 2014. The economy really hasn’t gone south. Economic confidence is far higher than it was at the beginning of 2013, although it has stalled slightly. The percentage of those who view economic recovery as imminent has fallen slightly over the past few months, but it’s only slightly lower than where it was at the time of the 2012 election. Put another way, there hasn’t been that level of decline in the economy which many thought would need to happen for Republicans to win with the coalition they have.

Yet, Obama’s net approval rating with white voters is no better than -25pt right now. His approval in Gallup’s polling among white voters since the NSA leaks is now only 34.5%. That’s a 4.5pt drop since the election. Pew has it slightly lower, at 33%, which is a 6pt drop off what Obama was showing in their final pre-election poll last year. It seems as though that wall keeps moving.

Most, if not all, of this drop for Obama is among whites without a college degree. Pew found support from that segment of the electorate dropping by a little less than 10pt. In other words, there is a continuity here with the November election result, in which Obama’s support fell the furthest among whites without a college degree. If you think this might be pegged to public reaction to the Trayvon Martin case, the Zimmerman verdict and the president’s response, it’s not. Pew found college-educated whites reacted in the same way as non-college-educated whites to that issue – yet Obama has seen no decline in his standing among college-educated white voters.

Indeed, the swing looks much as expected: we have that drop among the non-college-educated whites, and among minorities to a smaller degree. We can see this by pooling Gallup’s data since the Edward Snowden/NSA affair; this gives us a very large sample size. That’s important because minority sample size in polls is often very small, which is why Pew and NBC News/Wall Street Journal differed so much in their minority findings this week. Gallup splits the difference and discovers that non-white approval of Obama has fallen since the election by 4pt, or about half of the drop we’ve seen with non-college-educated whites.

Now, things could change and this analysis may end up way out-of-date come 2016. I should also add that the eventual party coalitions may look somewhat different then, with Republicans winning back some African-American voters once Obama is out of office, but maybe also Democrats making a slight recovery with white voters.

But the most salient fact is that Obama’s approval among registered voters is a weak 45% – and that’s without a bad dip in the economy or any grave scandal. This soft approval rating gives credence to what the polling suggests: that Republican party is most likely to win the White House in 2016 with a coalition that includes even more non-college-educated whites and a slight increase among minority voters.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

How the Republicans could win big in the 2014 Senate elections | Harry J Enten

The urban diversity that helps Democrats in presidential elections will hurt them in the race to control the Senate

Republicans’ chance to win back the United States Senate in 2014 are up. Analysts from Nate Silver to Sean Trende put the chances at just south to about 50% for Republicans to take the six seats necessary to gain 51 seats. Why are Republicans doing so well?

The immediate answer is Democrats can’t seem to find strong Democrats to run in states more Republican than the nation as a whole, as measured by the 2012 presidential vote. Democrats have yet to field any candidates in either Montana such as Brian Schweitzer or West Virginia, even though Democrats currently represent those states. Indeed, the majority candidates Democrats are finding to run in red states are lackluster. Even the better than average candidates such as incumbent Mary Landrieu of Louisiana are facing mediocre polling numbers.

Democrats’ red state blues are a big deal considering Democrats have to defend seven seats more Republican than the nation as a whole, while Republicans only have to beat back a Democratic challenge in one state won by President Obama. Seven minus one equals the six seats the Republicans need for control.

Yet, I would argue that 2014 recruitment failures are as much about the current state of each party’s coalitions than anything particular to this election cycle. There are now more states that lean Republican than lean Democratic. Twenty-six states are more Republican than the nation as a whole. Only 23 states are more Democratic. Virginia votes with the nation. Translating that to Senate seats, we’d expect something like a 53 to 47 Republican advantage on presidential vote alone.

Just twenty years ago, this Republican state edge might not have been so big of a deal. In 1993, 49% of the Senate Democratic caucus came from red states, and 28% of the Republican caucus came from blue states.

Now though, straight-ticket voting is becoming the rule for senate elections. Only 25% of the Democratic Senate caucus comes from red states, while 16% of Republicans come from blue states. The 25% and 16% listed above should fall even further as Republicans bump up their numbers with red staters in 2014, while the Democratic caucus will become more limited to blue staters.

These statistics explain why Democrats are having such recruitment problems in red states in 2014, and why they should for years going forward. People don’t want to run in races they’ll probably lose.

What demographics are behind the Republican state edge? Ironically, it’s the same factor that many are citing as the possible reason for their demise in presidential elections: Republicans are increasingly the party of white and rural voters. Mitt Romney won white voters by 20pt, if you believe the exit polls. Romney took rural areas by 24pt. President Obama won non-whites by 62pt and took the largest cities by 40pt and moderate sized cities by 18pt.

The relationship between racial and geographic voting patterns is no accident. Only about 26% of the country as a whole live in the top 10 metropolitan areas. For minorities, those percentages climb significantly: 37% of blacks, who voted for Obama by a 10:1 ratio, live in the top ten metropolitan areas. This includes Washington, DC, which doesn’t have any representation in Congress. Further, 45% of Latinos, who voted for Obama by greater than a 2.5:1 ratio, live in the top ten metropolitan areas.

That’s why it should be no surprise that 36 of the 50 states in the union have a higher percentage of non-Hispanic whites than the nation as a whole. That puts the Democratic party behind the eight ball in winning more states than Republicans. If racial polarization in terms of voting patterns continue to exist or get worse, then it will only go downhill for Democrats in winning more states.

This isn’t that big of a deal in winning presidential contests. The electoral college may be a lot of things, but it does take into account population in assigning electoral votes. Indeed, at this point, the Democrats seem to be enjoying an advantage in the electoral college. That is, they are in a better position to win the electoral college and lose the national vote than are Republicans.

On the Senate level, it is, however, a very big deal. Each state gets two senators, regardless of population. The deeply Republican and white smallest state Wyoming has a population of about 600,000 people. The deeply Democratic and diverse largest state California has a population of 38 million – 37.4 million people more than Wyoming.

But then why haven’t Republicans already won the senate? Republicans have had recruitment problems of their own. Think of some of the Republicans put up over the past few years, including Todd Akin, Sharron Angle, Ken Buck, Richard Mourdock, and Christine O’Donnell.

If Republicans get some decent candidates like they have in North Dakota with Mike Rounds and West Virginia with Shelley Moore Capito this year, then the Senate should be a gold mine for them. Not only should Republicans have an easier shot at the majority, but they should have a better chance to run up the score in years in which they do well. Had Mitt Romney had won by the same percentage nationwide as President Obama did, he would have won 34 states compared to the 26 Obama took. That’s eight Senate seats’ worth.

Now, none of this means that Republicans will win back the Senate in 2014. Republicans need a slightly more favorable national environment than seen in 2012. Then, they still need candidates to take advantage of the math – candidates who don’t make women run away when they speak.

What the math does mean, however, is that keeping all other things equal, the combination of the new Republican coalition and straight-ticket voting gives Republicans a big step-up in winning the Senate.

• This article on 19 July to correctly reflect the red state/blue state divide in the Senate in 1993.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

How the Republicans could win big in the 2014 Senate elections | Harry J Enten

The urban diversity that helps Democrats in presidential elections will hurt them in the race to control the Senate

Republicans’ chance to win back the United States Senate in 2014 are up. Analysts from Nate Silver to Sean Trende put the chances at just south to about 50% for Republicans to take the six seats necessary to gain 51 seats. Why are Republicans doing so well?

The immediate answer is Democrats can’t seem to find strong Democrats to run in states more Republican than the nation as a whole, as measured by the 2012 presidential vote. Democrats have yet to field any candidates in either Montana such as Brian Schweitzer or West Virginia, even though Democrats currently represent those states. Indeed, the majority candidates Democrats are finding to run in red states are lackluster. Even the better than average candidates such as incumbent Mary Landrieu of Louisiana are facing mediocre polling numbers.

Democrats’ red state blues are a big deal considering Democrats have to defend seven seats more Republican than the nation as a whole, while Republicans only have to beat back a Democratic challenge in one state won by President Obama. Seven minus one equals the six seats the Republicans need for control.

Yet, I would argue that 2014 recruitment failures are as much about the current state of each party’s coalitions than anything particular to this election cycle. There are now more states that lean Republican than lean Democratic. Twenty-six states are more Republican than the nation as a whole. Only 23 states are more Democratic. Virginia votes with the nation. Translating that to Senate seats, we’d expect something like a 53 to 47 Republican advantage on presidential vote alone.

Just twenty years ago, this Republican state edge might not have been so big of a deal. In 1993, 49% of the Senate Democratic caucus came from red states, and 28% of the Republican caucus came from blue states.

Now though, straight-ticket voting is becoming the rule for senate elections. Only 25% of the Democratic Senate caucus comes from red states, while 16% of Republicans come from blue states. The 25% and 16% listed above should fall even further as Republicans bump up their numbers with red staters in 2014, while the Democratic caucus will become more limited to blue staters.

These statistics explain why Democrats are having such recruitment problems in red states in 2014, and why they should for years going forward. People don’t want to run in races they’ll probably lose.

What demographics are behind the Republican state edge? Ironically, it’s the same factor that many are citing as the possible reason for their demise in presidential elections: Republicans are increasingly the party of white and rural voters. Mitt Romney won white voters by 20pt, if you believe the exit polls. Romney took rural areas by 24pt. President Obama won non-whites by 62pt and took the largest cities by 40pt and moderate sized cities by 18pt.

The relationship between racial and geographic voting patterns is no accident. Only about 26% of the country as a whole live in the top 10 metropolitan areas. For minorities, those percentages climb significantly: 37% of blacks, who voted for Obama by a 10:1 ratio, live in the top ten metropolitan areas. This includes Washington, DC, which doesn’t have any representation in Congress. Further, 45% of Latinos, who voted for Obama by greater than a 2.5:1 ratio, live in the top ten metropolitan areas.

That’s why it should be no surprise that 36 of the 50 states in the union have a higher percentage of non-Hispanic whites than the nation as a whole. That puts the Democratic party behind the eight ball in winning more states than Republicans. If racial polarization in terms of voting patterns continue to exist or get worse, then it will only go downhill for Democrats in winning more states.

This isn’t that big of a deal in winning presidential contests. The electoral college may be a lot of things, but it does take into account population in assigning electoral votes. Indeed, at this point, the Democrats seem to be enjoying an advantage in the electoral college. That is, they are in a better position to win the electoral college and lose the national vote than are Republicans.

On the Senate level, it is, however, a very big deal. Each state gets two senators, regardless of population. The deeply Republican and white smallest state Wyoming has a population of about 600,000 people. The deeply Democratic and diverse largest state California has a population of 38 million – 37.4 million people more than Wyoming.

But then why haven’t Republicans already won the senate? Republicans have had recruitment problems of their own. Think of some of the Republicans put up over the past few years, including Todd Akin, Sharron Angle, Ken Buck, Richard Mourdock, and Christine O’Donnell.

If Republicans get some decent candidates like they have in North Dakota with Mike Rounds and West Virginia with Shelley Moore Capito this year, then the Senate should be a gold mine for them. Not only should Republicans have an easier shot at the majority, but they should have a better chance to run up the score in years in which they do well. Had Mitt Romney had won by the same percentage nationwide as President Obama did, he would have won 34 states compared to the 26 Obama took. That’s eight Senate seats’ worth.

Now, none of this means that Republicans will win back the Senate in 2014. Republicans need a slightly more favorable national environment than seen in 2012. Then, they still need candidates to take advantage of the math – candidates who don’t make women run away when they speak.

What the math does mean, however, is that keeping all other things equal, the combination of the new Republican coalition and straight-ticket voting gives Republicans a big step-up in winning the Senate.

• This article on 19 July to correctly reflect the red state/blue state divide in the Senate in 1993.

theguardian.com © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Gallup’s 2012 election polling errors were only part of the problem | Henry Enten

Gallup was caught out badly, but other national pollsters were off, too. It’s time to look at different methods and new technologies

We all know that Gallup screwed the pooch in the 2012 presidential election. It had Mitt Romney leading through most of October and in its final poll by a point – a 5pt error. Gallup sought to prove to the polling world that it was seriously investigating its 2012 polling errors by issuing a report on Tuesday. In the write-up, Gallup noted that although there was no single cause, a likely faulty voter screen and too few Hispanics were among the problems. This comes as no surprise to others including Mark Blumenthal and myself.

It’s worth the time, though, to point out, as I have and Gallup did on Tuesday, that the Gallup effect was only half the problem.

The average of polls done in the final week, excluding Gallup and Rasmussen, had Obama’s lead over Romney more than 2pt too low. I might be willing to look the other way, except the polling average in 2000 had George W Bush winning and had a margin error of again more than 2pt. The error in margin in 1996 was off by 3pt. The 1980 average saw an error of more than 5pt. The years in between 1980 and 1996 were not much better. In other words, the “high” error in national polling even when taking an average isn’t new; in fact, it seems to be rather consistent over the years.

Worse than the error in the final polls was how the national polls took the consumer for a ride in October 2012 before finally settling in the final week. Anyone remember when Pew Research published a poll after the first debate in 2012 that had Mitt Romney up by 4pt among likely voters? I don’t mean to single out Pew, but because of Pew’s sterling reputation, this poll got an outsized amount of attention even as most of us suspected that it probably didn’t reflect the truth. Other pollsters, too, showed a bounce for Romney that propelled him into the lead after the first debate, though not all to the same extent.

The state polling, meanwhile, did not show an analogous large bounce. It consistently had Obama leading in the states he needed to be leading in. Moreover, it showed Obama holding very similar positions to those he did prior to the first debate in the non-battleground states.

Look at YouGov, for example, which polled before and after the first debate. In Florida, Obama was ahead by 2pt before the debate and 1pt after it. In blue New York, Obama was ahead by 22pt before the first debate and 24pt after. In red Georgia, Romney was up by 7pt before the debate and 8pt afterward. Pollsters like ABC/Washington Post, CNN/ORC,and Public Policy Polling (PPP) all did better on the state level throughout October than they did at the national level.

It’s not the first time the state polling beat the national surveys. Back in 2000, for instance, state poll followers knew that Al Gore had a really good shot at winning. National survey followers, though, were surprised when Gore won the national vote. That’s why smart poll aggregators like Drew Linzer, Nate Silver and Sam Wang barely looked at national polling in 2012 when trying to project the winner. It’s also why the Obama campaign didn’t conduct national surveys.

I asked Gallup about state polling on Tuesday, and why it didn’t try to do individual state polls and/or then sum up, as Silver did, to calculate the national vote. After all, besides polling accuracy, the ball game of presidential elections for pollsters is state elections. Gallup’s response was telling. First, it said that polling 50 individual states via live interviewer to come up with a national estimate would be too time-consuming and cost too much money. That’s fair. Second, Gallup said that it didn’t just poll the swing states because it was interested in knowing what all Americans thought, not just swing-state voters. (I agree and made the same point in an earlier column.)

But for those of us who are interested in knowing who is going to win, Gallup’s answer is not satisfying. Other live pollsters like CNN/ORC, Marist, Quinnipiac and the Washington Post did very good statewide polling in 2012. Gallup hasn’t conducted a statewide general election poll since 2006 and hasn’t done so in a general presidential election since 2004. (Those 2004 polls, by the way, weren’t very good.)

Moreover, the option now exists for pollsters to use other technologies to poll most states, if not all 50. We have interactive voice response (IVR) or robo-polls that are relatively cheap and can survey many people quickly. As long as you properly weight in younger voters, as does PPP and SurveyUSA, these polls work quite well in predicting who is going to win the national election. We also have the somewhat less expensive, randomly selected internet surveys such as Knowledge Networks, and the cheaper volunteer internet polling, which YouGov and Ipsos have implemented successfully. These volunteer surveys hold a lot of promise in the future as more and more people get rid of landlines and have computers.

The point is that there are proven ways to poll that produce more consistently accurate portrayals of the election than doing a single live telephone interviews of a randomly selected population in a national poll. In fact, it’s already being done. That’s not to say that good-quality probabilistic national surveys don’t have a place. No one has proven to me that IVR or non-random internet surveys are as good as probabilistic telephones surveys on issue questions beyond the ballot test. The problem, again, is that I’d look to other sources in preference to a survey that interviews some number of respondents in one survey, a different set of respondents in the next survey, and so on.

One of the biggest takeaways from the American Association for Public Opinion Research (AAPOR) conference is on the usefulness of panel research. That is, you have a set number of respondents, weighted to the correct population parameters, who get interviewed over and over again. This leads to less volatility, and you can actually see how different respondents are reacting to the campaign. Panels can be difficult to do by phone, yet are rather easily obtained in randomly-selected internet samples that pretty much everyone, including those against volunteer internet samples, agrees do just as good a job at finding the true public opinion on issue questions. You can actually see how well panels worked with the Rand American Life Panel. It was the only national tracking survey in 2012 that had both convention bounces and Obama leading throughout the month of October.

None of this is to say that live telephone surveying is bad or useless, by any stretch. Most of the national telephone polls in 2012 were better than Gallup’s. It just seems to me that we shouldn’t only be examining Gallup for 2012′s polling failings. It might be time for even the most ardent defenders of live telephone national interviews to look at other methods in greater depth. Whether it be for the presidential horserace or more in-depth issue questions, different and (in some cases) less expensive survey styles have shown a trend to do better or at least as well.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

It’ll take more than a presidential library to make George W Bush popular | Harry J Enten

Don’t be fooled by a new poll showing Americans think fondly of ‘Dubya’. Of recent presidents, only Richard Nixon was less liked

The George W Bush presidential library is opening this week. Not surprisingly, fans of the former president are out in full force trying to help with the rehabilitation’s of Bush’s image.

Bush, of course, ended his presidency with an approval rating around 30%. This previous low, combined with the library opening, has helped give an impression of elevation to a new ABC/Washington Post poll that put his retrospective job approval rating at 47%.

The fact that President Obama’s approval rating in the current HuffPollster aggregate is a similar 48% only helps with the intrigue. The Washington Free Beacon wrote an article titled “Dubya’s Approval Matches Obama’s”. Add on the fact that Obama’s approval rating and Bush’s retrospective approval on the economy approval rating are about the same, and you got the makings of a great press narrative.

Let me be the one who tries to nip this story in the bud. I have no clue what type of person George W Bush is, or how history will view him in the future. What I do know is that the 47% retrospective approval rating should be put in context.

First, retrospective approval ratings should almost never be compared to current job approval ratings. Humans have a tendency to remember their elected officials more fondly than they did when they left office. Back in 2010, Gallup asked Americans what their retrospective approval rating was for Presidents John F Kennedy through George W Bush. In every instance except for one, the retrospective approval was higher than the final approval was when they left office.

Most Republicans, for instance, love to make fun of Jimmy Carter. Carter was the only president of the 20th century to lose re-election after replacing a president of a different party. He left office with a 34% job approval rating. His retrospective job approval rating in the 2010 Gallup poll jumped by 18pt.

Second, Bush’s retrospective approval is the second worst among presidents in the last 50 years ago. To save you doing the math, Carter’s 52% approval rating is higher than Bush’s 47%. Only the Watergate-tainted Richard Nixon recorded a lower retrospective approval than Bush.

Most presidents have retrospective approval ratings above 60%. All but Nixon and Bush have +10 or better retrospective net approval ratings. Thus, not only did Bush tie for the second worst final approval rating while in office, but he is also has the second worst retrospective job approval rating.

Third, the Gallup data should make clear that George W Bush hasn’t seen much recovery over the past three years in his retrospective approval rating. The 47% in the latest Washington Post poll is the same as the 47% that Gallup found in 2010. The 50% disapproval now is nearly identical to the 51% three years ago. Yes, people view him more rosily now than they did during his presidency, but that effect has been baked in for a while now. He’s not getting more loved as time goes by.

Finally, the retrospective approval rating probably gives a false sense of how Americans view Bush now. Remember that a retrospective approval is exactly that – retrospective. The better way to view how Bush stands with the American public is his current favorable ratings. In the past two years, there have been five polls conducted that have asked about Bush’s favorable rating with the American public.

Bush currently holds an average -5pt net favorable rating with the American public. President Obama’s favorable rating is almost exactly the reverse of that, at +7pt in the HuffPollster aggregate. Indeed, even the losing Republican nominee Mitt Romney ended the 2012 campaign with a higher net favorable than Bush, at -3pt.

It’s no wonder that President Bush continues to hurt the Republican party. Mitt Romney tried his hardest to tie what many saw as a lackluster economy in 2012 to President Obama. The problem was that most Americans still blamed Bush over Obama. In the network exit polls, 53% said the economic problems were more Bush’s than Obama’s fault. Only 38% of Americans disagreed.

So, George W Bush may be more fondly thought of now than he was when he left office, yet this is to be expected – and discounted. Almost all presidents see a boost after they leave office. Bush is still quite unpopular compared with other former presidents, and his current favorable rating is far worse than President Obama’s. No library opening is changing that. Republicans would be wise to stay away from embracing George W Bush for the time being.

• This article has been amended to reflect updated polling data as of 25 April 2013.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Why Richard Cohen is mistaken about the Republican primary process | Harry J Enten

The thesis that the Iowa caucus and New Hampshire primary skew the GOP to pick a conservative candidate is simply wrong

My high school teachers and college professors always told me to look at the evidence and come up with a thesis that is supported by it. The completely wrong thing to do was to think up a thesis and then fit the evidence around it. But that, in my view, is exactly what the Washington Post’s Richard Cohen did on Monday.

Cohen believes that Iowa and New Hampshire should not be the first nominating contests during the primary season. He decries that the recent Republican “Growth and Opportunity Project” didn’t dare to make the suggestion that Iowa and New Hampshire lose their special “first in the nation” status. There’s nothing wrong with this belief, and many others support that position.

The problem I have is Cohen’s reasoning for why Iowa and New Hampshire shouldn’t be first. He thinks that their Republican electorates are too conservative compared to the Republican party as a whole. If this were true, then they would be likely to select a candidate who is very conservative. This is an issue because, as we know, the first contests help to winnow the primary field. If you don’t win either the Iowa caucus or New Hampshire primary, you probably won’t last too long as a serious candidate. Thus, he proposes, the Republican party ends up with its presidential candidates being too conservative.

On Iowa, at least, Cohen’s case can be made pretty strongly. After all, it’s where Pat Robertson came in second in 1988, and Mike Huckabee won in 2008. We know caucuses themselves are lower turnout affairs that require effort to attend and thus tend to be attended proportionally by more conservative activists than vote primaries, but let’s get more specific.

Using Ohio as a control case because it was right at about the center of where the primary electorate was at, we can see Iowa’s tilt strongly: 56% of Iowa’s 2012 voters self-identified as white, born-again or evangelical Christian. That’s far above Ohio’s level, at 47%.

Whereas 64% of Iowa’s 2012 voters supported the Tea Party, just 59% of Ohio’s voters did. And 47% of Iowa’s 2012 voters self-identified as very conservative. That’s not even close to Ohio’s 32%.

Not surprisingly, this led to Rick Santorum winning the Iowa caucus over Mitt Romney, even as Romney would go on to win Ohio and the nomination. Romney took only 24.5% of the vote in Iowa, while Santorum, Michele Bachmann, Newt Gingrich, and Rick Perry combined for an astronomical 53.2% – or double Romney’s percentage. When we combine all the nation’s caucuses and primaries, it was actually Romney who got 52.1%.

Here’s the rub: none of this same argument can be applied to New Hampshire. New Hampshire is actually pretty well to the left of the national Republican primary electorate. Again, let’s use Ohio as our control case.

Only 21% of New Hampshire’s 2012 voters self-identified as white born-again or evangelical Christian. That’s 26 points below Ohio. Only 51% of New Hampshire’s 2012 voters supported the Tea Party. That’s 8pt below Ohio.

Only 34% of New Hampshire’s 2012 voters viewed themselves as very conservative on spending issues such as taxes and spending. That’s not even close to Ohio’s 47%. Only 24% of New Hampshire’s 2012 voters, less than a quarter, self-identified as very conservative on social issues such as abortion. That’s 18pt below Ohio’s 42%. And only 21% of New Hampshire’s 2012 voters self-identified as very conservative overall. That’s 11pt down on Ohio.

As you can see, New Hampshire’s Republican electorate isn’t anywhere near being more conservative than the median Republican electorate. In fact, it’s a lot more liberal.

That’s why you see strict social conservatives hit a wall when they come to New Hampshire. After his Iowa victory, Santorum was uncompetitive in New Hampshire receiving just 9.4% of the vote. He, of course, got 37% of the vote in Ohio and nearly won that primary. The Bachmann, Gingrich, Perry, Santorum foursome, combined, got less than 20% of the vote, compared to their 53.2% back in Iowa.

It’s no wonder that the Republican mainstream candidate, Mitt Romney, won the primary by 16.4pt – far greater than his 1pt squeak win in Ohio.

Meanwhile, more moderate candidates thrive in New Hampshire as they do in few other places. Jon Huntsman, whose campaign was marked by speaking Mandarin Chinese during debates, saying he believed in global warming and evolution, and voicing support for civil unions, actually was competitive in New Hampshire. He still lost, with 17% of the vote, but he finished ahead of the eventual runner-up Santorum. Of course, Huntsman never finished much above 2-3% in national polls and had to drop out after losing New Hampshire.

Cohen could argue that New Hampshire did allow the very conservative Ron Paul to place second, with nearly 23% of the vote. Most of us would agree, however, that that is because Paul is a libertarian. He wanted government out of people’s lives, and so he fit New Hampshire’s “live free or die” motto to a tee. That’s not exactly the type of strong social conservatism that Cohen would argue is apparent in New Hampshire.

Finally, some might note that New Hampshire has an open primary (that is, independents can vote) – and that this skewed the 2012 results because there was no competitive Democratic contest in the state. That belief doesn’t hold water, either. For example, in 2008, the same percentage of New Hampshire Republican primary voters self-identified as very conservative, despite a primary between Barack Obama and Hillary Clinton occurring on the Democratic side.

The truth is that far from New Hampshire’s Republican electorate obliging the party to select more conservative nominees, it’s actually making them more liberal and mainstream. New Hampshire helped propel John McCain’s Straight Talk Express to victory in 2000; it stopped social conservative Mike Huckabee after his Iowa win in 2008; and it put Mitt Romney on the right track in 2012.

Any view to the contrary just isn’t looking at the facts.

guardian.co.uk © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds