Let's see if we can make this last, or if it's just a one-off. Somehow, while examining the Quinnipiac Poll results, I missed that Gary Johnson and Jill Stein were also included in the Monmouth University poll this week. Even better, Monmouth puts Stein's aggregate at 4% (after rounding).
The unfortunate qualifier is that a) Monmouth's sample size is half as large, and b) all the respondents are registered voters in New Jersey, Monmouth's own state, rather than a nationwide sample.
The unfortunate qualifier is that a) Monmouth's sample size is half as large, and b) all the respondents are registered voters in New Jersey, Monmouth's own state, rather than a nationwide sample.
This poll also highlighted questions about how a New Jersey–based running mate might affect voters' choices: specifically, Chris Christie running with Donald Trump, or Cory Booker running with Hillary Clinton.
Christie draws some heavy negatives among voters in his own state, even among Republicans: 14% of Republicans would be more likely to vote for Trump with Christie on the ticket, but 23% would be less likely. For Millennial voters of whatever affiliation, those figures are 2% (more likely) and 49% (less). The largest percentages of those polled answered "No Effect." Adding Booker to the Clinton ticket does much better among Democrats and the youth contingent.
One Reason I Love/Hate These Polls: False Hope
Yes, false hope. For just one example: According to at least one poll In 2012, as late as September, the Stein-Honkala ticket was supposed to receive close to 2% of the vote in Texas and about the same nationwide. Her final tally was 0.31% in Texas, 0.34% nationwide. (Sorry, can't find a link.)
A gap of 1.7 percentage points is well within the margin of error, and it doesn't make much difference if you're expecting 60% or even 40%. Obviously, the closer you get to 50%—or 5% to maintain state ballot access or gain federal funds—the more significant small numbers can look.
After all these years of polling, we still don't have a solid answer as to whether substantial differences between opinion polls and actual results for independent and minor-party candidates come down to:
UPDATE: a paragraph from P-Diddie for which I am particularly grateful today.
Christie draws some heavy negatives among voters in his own state, even among Republicans: 14% of Republicans would be more likely to vote for Trump with Christie on the ticket, but 23% would be less likely. For Millennial voters of whatever affiliation, those figures are 2% (more likely) and 49% (less). The largest percentages of those polled answered "No Effect." Adding Booker to the Clinton ticket does much better among Democrats and the youth contingent.
One Reason I Love/Hate These Polls: False Hope
Yes, false hope. For just one example: According to at least one poll In 2012, as late as September, the Stein-Honkala ticket was supposed to receive close to 2% of the vote in Texas and about the same nationwide. Her final tally was 0.31% in Texas, 0.34% nationwide. (Sorry, can't find a link.)
A gap of 1.7 percentage points is well within the margin of error, and it doesn't make much difference if you're expecting 60% or even 40%. Obviously, the closer you get to 50%—or 5% to maintain state ballot access or gain federal funds—the more significant small numbers can look.
After all these years of polling, we still don't have a solid answer as to whether substantial differences between opinion polls and actual results for independent and minor-party candidates come down to:
- polling samples not being truly reflective of the voting public;
- those third-party sympathies not lasting all the way to the voting booth, with voters chickening out at the last minute and voting for their preferred major party;
- election officials cooking the numbers;
- some combination of the above; or
- some other explanation.
UPDATE: a paragraph from P-Diddie for which I am particularly grateful today.
And pay no heed to Debbie Downers who say that the Greens and Libs always poll higher than they actually wind up with on Election Day -- this isn't going to be a normal year -- or that dreaming big on the part of alternate candidates and parties is something to be mocked. That kind of discouragement is not how we raise our children and grandchildren; it's not even how our mayor manages the city.
RSS Feed