The New York Times[/paste:font]
SUBSCRIBE
LOG IN
The Upshot
[/paste:font]
SURVEY SAYS
Why Trump’s Base of Support May Be Smaller Than It Seems
By BRENDAN NYHAN
JULY 19, 2017
Donald J. Trump’s approval ratings are the lowest at this stage of a presidency since modern polling began in the mid-20th century. But his party’s base still appears highly supportive, which has discouraged Republicans in Congress from abandoning the president and his legislative agenda.
Even the most recent Russia revelations seemingly haven’t dented support for Mr. Trump among Republicans, who continue to approve of his job performance at very high rates — 82 percent in an ABC News/Washington Post poll released Sunday, for instance.
But numbers like these may mask a decline in support for Mr. Trump among his original party base.
A new working paper by the Emory University political scientists B. Pablo Montagnes, Zachary Peskowitz and Joshua McCrain argues that people who identify as Republican may stop doing so if they disapprove of Trump, creating a false stability in his partisan approval numbers even as the absolute number of people approving him shrinks. Gallup data supports this idea, showing a four-percentage-point decline in G.O.P. identification since the 2016 election that is mirrored in other polling, though to a lesser extent.
President Trump remains popular among Republicans and Republican leaners, but some erosion of support may be hidden.
ERIC THAYER FOR THE NEW YORK TIMES
One example is the MSNBC host Joe Scarborough, a former Republican member of Congress and an increasingly vocal Trump critic who has announced that he is leaving the party. If most people who have stopped identifying as Republicans are Trump disapprovers like Mr. Scarborough, we could observe stable party approval ratings for the president despite waning popularity with his previous party coalition.
Individual news media polls typically do not have a measure of how respondents previously answered questions about party identification, so Mr. Montagnes and his co-authors had to estimate a range for Mr. Trump’s G.O.P. approval rating among those “missing” respondents — people who previously identified as Republican but no longer do so. If all of them approve of Mr. Trump, the resulting estimate would be the upper bound on his partisan approval rating. If all such respondents disapprove of him, that would be the lower bound.
These lower-bound estimates are likely to be much closer to reality — Pew Research Center data show that disapproval of Mr. Trump was 84 percent among people who switched from identifying or leaning Republican in December 2015 to identifying or leaning Democrat in April 2017.
When the Emory political scientists use the Gallup data to account for Republicans who have stopped identifying with the party since the election, they find that partisan support for Trump could be substantially lower than it appears. The lower bound is often from 70 percent to 80 percent instead of the 80-to-90 range that Gallup polls typically show. Given the decline in Republican identification since last November, they find, “the lower bound on Trump’s partisan approval rate is much lower” than partisan approval at a comparable point in the Obama presidency and is lower than it was even during Mr. Obama’s second term.
This finding contains an important lesson for interpreting presidential approval polls. Mr. Trump remains popular among the people who currently identify as Republican, but there is a group of “missing” former Republicans who most likely disapprove of him and are excluded from that calculation altogether.
Brendan Nyhan is a professor of government at Dartmouth College. Follow him on Twitter at @BrendanNyhan.
The Upshot provides news, analysis and graphics about politics, policy and everyday life. Follow us on Facebookand Twitter. Sign up for our newsletter.
RELATED COVERAGE
Back to top
Home
NYTimes
© 2017 The New York Times Company
SUBSCRIBE
LOG IN
[/paste:font]
SURVEY SAYS
Why Trump’s Base of Support May Be Smaller Than It Seems
By BRENDAN NYHAN
JULY 19, 2017
Donald J. Trump’s approval ratings are the lowest at this stage of a presidency since modern polling began in the mid-20th century. But his party’s base still appears highly supportive, which has discouraged Republicans in Congress from abandoning the president and his legislative agenda.
Even the most recent Russia revelations seemingly haven’t dented support for Mr. Trump among Republicans, who continue to approve of his job performance at very high rates — 82 percent in an ABC News/Washington Post poll released Sunday, for instance.
But numbers like these may mask a decline in support for Mr. Trump among his original party base.
A new working paper by the Emory University political scientists B. Pablo Montagnes, Zachary Peskowitz and Joshua McCrain argues that people who identify as Republican may stop doing so if they disapprove of Trump, creating a false stability in his partisan approval numbers even as the absolute number of people approving him shrinks. Gallup data supports this idea, showing a four-percentage-point decline in G.O.P. identification since the 2016 election that is mirrored in other polling, though to a lesser extent.
President Trump remains popular among Republicans and Republican leaners, but some erosion of support may be hidden.
ERIC THAYER FOR THE NEW YORK TIMES
One example is the MSNBC host Joe Scarborough, a former Republican member of Congress and an increasingly vocal Trump critic who has announced that he is leaving the party. If most people who have stopped identifying as Republicans are Trump disapprovers like Mr. Scarborough, we could observe stable party approval ratings for the president despite waning popularity with his previous party coalition.
Individual news media polls typically do not have a measure of how respondents previously answered questions about party identification, so Mr. Montagnes and his co-authors had to estimate a range for Mr. Trump’s G.O.P. approval rating among those “missing” respondents — people who previously identified as Republican but no longer do so. If all of them approve of Mr. Trump, the resulting estimate would be the upper bound on his partisan approval rating. If all such respondents disapprove of him, that would be the lower bound.
These lower-bound estimates are likely to be much closer to reality — Pew Research Center data show that disapproval of Mr. Trump was 84 percent among people who switched from identifying or leaning Republican in December 2015 to identifying or leaning Democrat in April 2017.
When the Emory political scientists use the Gallup data to account for Republicans who have stopped identifying with the party since the election, they find that partisan support for Trump could be substantially lower than it appears. The lower bound is often from 70 percent to 80 percent instead of the 80-to-90 range that Gallup polls typically show. Given the decline in Republican identification since last November, they find, “the lower bound on Trump’s partisan approval rate is much lower” than partisan approval at a comparable point in the Obama presidency and is lower than it was even during Mr. Obama’s second term.
This finding contains an important lesson for interpreting presidential approval polls. Mr. Trump remains popular among the people who currently identify as Republican, but there is a group of “missing” former Republicans who most likely disapprove of him and are excluded from that calculation altogether.
Brendan Nyhan is a professor of government at Dartmouth College. Follow him on Twitter at @BrendanNyhan.
The Upshot provides news, analysis and graphics about politics, policy and everyday life. Follow us on Facebookand Twitter. Sign up for our newsletter.
RELATED COVERAGE
Political Calculus: A 2016 Review: Why Key State Polls Were Wrong About Trump MAY 31, 2017
Disapproval: Trump’s Approval Ratings Are Down. How Much Does It Mean?FEB 17, 2017
Trump Seems Much Better at Branding Opponents Than Marketing Policies JUL 18, 2017
- PUBLIC HEALTH
Obamacare’s Future Now Depends on an Unhappy White House
President Trump has said he’d like Obamacare to fail, and the Department of Health and Human Services has advertised its failures.
- Trump Seems Much Better at Branding Opponents Than Marketing Policies
Exploring the patterns behind the president’s branding powers.
- THE NEW HEALTH CARE
Can Psychedelics Be Therapy? Allow Research to Find Out
These drugs carry risks that can’t be ignored, but it may be time to reconsider the current classification of controlled substances.
Back to top
Home
NYTimes
© 2017 The New York Times Company
