As a self-proclaimed “polling nerd,” it drives me crazy when people either don’t know how to read polls or don’t know how to interpret them properly. This shouldn’t be the case when it comes to CNN’s supposed “Polling Director,” Jennifer Agiesta. The problem is, CNN has either willfully misrepresented poll results for cheap clicks or doesn’t know how to read their own polling results.
This is the headline I woke up to this morning (which has since been changed on the article itself but is still being used by CNN elsewhere):
Poll: Debate Win Doesn’t Help Hillary Clinton
That headline was followed by this:
Clinton stands at 45% in the race for the Democratic nomination, with Vermont Sen. Bernie Sanders behind her at 29%. Vice President Joe Biden, who is considering a run for presidency and did not participate in last week’s debate, follows at 18%.
Compared with pre-debate polling, Sanders’ support is up five points since mid-September, but no other candidate showed significant change. (Source)
Now, statistically speaking, with a margin of error of plus or minus three percent, Clinton didn’t “gain” over her mid-September number of 42 percent. So the headline is technically accurate – but misleading. The fact is, her raw numbers did go up. Not only that, but considering that same margin of error, I’m not exactly sure I would call Sanders gaining two percent “significant.” Especially when you consider the lower your polling numbers initially, the more likely you are to see larger gains. If we really want to get technical about it, Clinton gained anywhere from 0-6% after the debate, while Sanders gained anywhere from 2-8%. Is that enough to single out Sanders as being the only candidate who saw a “significant change”? You tell me.
Then respondents were asked who they would support if Joe Biden isn’t in the race, and Clinton’s lead over Sanders climbed to 23 points.
But it gets even worse.
Only 31 percent of the people polled actually watched the debate. Now, this isn’t CNN’s fault – a random sample is called a “random” sample for a reason – but it’s hard to judge the bump a candidate receives following a debate when almost 7 in 10 respondents to your poll didn’t even watch the debate.
That’s like asking 10 people to critique and review a movie, when 7 of them haven’t seen anything more than the trailer – if even that.
Not only that, but the key part of this poll is the “mid-September” cutoff. While it’s true that Ms. Agiesta was going by the most recent CNN/ORC numbers, I think it’s significant to point out that since early September, Clinton’s numbers have gone up 8 points whereas Sanders’ numbers have climbed 2 points. In other words, statistically (when factoring in the margin of error) Clinton has shown fairly sizable gains since early September while Sanders’ numbers haven’t had the same steam.
Now, to be fair, Ms. Agiesta does mention the aspects of the poll that are extremely positive for Clinton. But anyone who works in any form of journalism knows that most people just read a headline, and many of those who do click the actual article don’t read much more than a paragraph or two – which doesn’t include any of the positive information for Clinton. Besides, the mundane aspect of the details of a poll such as this are basically just white noise to the average reader. All most people want to know is “who’s winning” – many people don’t have spare time to sort out the details.
It’s also significant to point out (as CNN actually did) that in a scientifically done poll (not the online surveys many blogs – and even some supposed legitimate news sites – had been touting) respondents said Clinton overwhelmingly won the debate, 62 percent to Sanders’ 34 percent.
Furthermore, when it comes to who respondents felt would be better at handling:
- The economy
- Health care
- Race relations
- Climate change
- Foreign policy
- Gun control
- Income inequality (yes – income inequality)
Clinton leads in all categories.
Though perhaps the most mind-boggling stat of all shows that in a head-to-head matchup with Ben Carson, both Clinton and Sanders are trailing. I’ve given up trying to understand why people view Carson favorably. I would honestly rather see Trump as president than Ben Carson.
So, when all is said and done, all this poll really says is that statistically speaking, neither Clinton or Sanders saw any real solid boost from the debate – though most Americans who watched it say (overwhelmingly) that Clinton “won.”
That says a lot going forward. Basically, it solidifies the odds that Clinton becomes the party’s nominee unless she falters greatly between now and next summer; that Sanders is going to have to do something significant if he wants to gain ground; and if Joe Biden decides not to run – Clinton becomes an even bigger favorite to win the nomination.
Not only that, but despite the media’s (especially left-wing blogs) incessant harping on how “people don’t trust Clinton,” according to this scientifically sound poll (not worthless online surveys), she’s viewed as the candidate – in every category – who’s best equipped to handle the key issues facing this country.
The bottom line is, no matter if you’re a Clinton supporter, a Sanders supporter, a Biden supporter or someone like me who likes all three – it’s clear our candidates are substantially better than anyone Republicans have. Because when it gets right down to it, those of us on the left are debating over which candidate is best equipped to lead our nation.
Meanwhile, on the Republican side, it’s a race to see which clown performs best in a circus.