2021 Polling Is Telling Us The Industry Is Broken

By most approval polls Joe Biden is sailing high. Despite having a record low approval at 100 days (minus Trump who nobody else may ever reach), it is explained away as a result of partisanship and most voters being inelastic. After – all, most studies have shown voters who identify as Independent truly lean heavily to one party or the other (sometimes even moreso than their registered partisan brothers and sisters). Hence, most Republicans disapprove of Biden and most Democrats approve while Independents disapprove. Wait, what?

No. You heard that right. Most Republicans and Independents disapprove of the President and he is still above 50%. So, again, what gives? Well, if anything, 2016 and 2020 showed Republicans and conservative Independents, particularly in rural areas, are not likely to report their feelings to pollsters. In other words, they don’t answer surveys. And while online and cell line surveys help cancel this effect out somewhat, the GOP is older than the Democratic coalition and thus many of the GOP’s voters won’t respond to these options either.

This year, the clear split in Biden’s approval along these lines can be seen in his approval ratings. Take for example the difference between a recent CNN survey of all adults and a Politico/Morning Consult Survey. The Morning Consult survey found President Biden with a whopping 60/37 approval to disapproval rating. The CN survey had a more modest 52/45 split. Both surveys sampled general adults (different than surveys of actual registered voters) and were conducted around the same time period.

Want to know the biggest difference? The samples themselves. Whereas the CNN survey sampled a pool of adults identifying as +5 D the Politico survey sampled a population which was 49% Democratic. As in for Biden to be underwater he would need to be at about 20 percent approval of Independents. And this is after they weighted the survey respondents.

For the uninitiated weighting is what pollsters do after a raw number of opinions have been collected from a survey to predict what the general population thinks. They generally tie this to past electorates or what they think the partisan and demographic make – up of the same is in their targeted group (state, nation, etc.). Well, I am not sure what went into Politico’s weighting decisions but it certainly was not the 2020 electorate.

The 2020 electorate, according to national exit polls of in – person and by mail voters saw 36 percent of voters ID as Republican, 37 percent as Democrats and 27% as Independents. Compare this to a survey with a 49/35/16 spread with leaners included. If this sample is truly what America now is, conservatives and Republicans and their ideas are endangered species.

Politico/Morning Consult is not the only pollster to find absurdly high approval ratings for the President or highly biased samples. But it is a departure from Obama approval polls where the President maintained his high approval ratings through the first year of his term on the back of majority Independent support. Now, it is possible more Independents are identifying as Democratic in surveys as they are asked questions on issues which makes them identify more with a party as they think about where they stand on the issues, but I would not bet on it. If that were the case, the same thing would see increased GOP identification in surveys. We are not.

So what does this tell us? Well, in terms of Biden’s true approval, nothing. The actual composition of the country is hardly D + 14. Pretty sure we just had an election with the most unpopular incumbent running in history and he lost by 4.5 points. So, we don’t really know what the public actually thinks of Biden (unless you want to auto peg his actual ratings as a few points lower than any given survey.

There is the option of using strong approval/disapproval as a proxy. These tend to be partisans and most likely to turn out in an election, donate money, campaign or voice their opinion on issues. But, even so, that is imperfect at best.

More broadly, the polling industry’s seeming inability or maybe refusal to fix their issues despite knowing about it for more than half a decade now is disconcerting. It is likely an entire swathe of the country and their views are being missed in surveys and only noted in electoral results (see GOP overperformance at multiple levels in 2014, 2016 and 2020).

For all the sources telling us how the modern GOP and conservatives are a threat to democracy (too many to list) this is a far graver and often ignored threat. The media, politicians and other talking heads often do not understand or ignore learning the complicated caveats of polls and thus weave a narrative of absolutes in a poll’s results. I forgot, margins of errors, sample sizes. etc. are not things anymore.

Rather, what this tends to create is the impression of popular support on major issues without it actually being there. Heck, even politicians recognize the problem. Take surveys on major progressive issues in this country and their poll results correlated with Biden’s agenda.

Per surveys, large majorities of voters agree climate change is an issue, infrastructure needs to be repaired, support taxing the rich and believe inequality is an issue. But, if so, why are so many Democrats baking at supporting Biden’s agenda? After – all, they are popular. Why have Republicans not suffered a backlash over opposing Stimulus Checks with a majority of Republicans even supported in surveys? I think we know the answer.

In truth, 2021 is reinforcing the notion polling is broken and unable to bridge a chasm in Americans opinions, ideologies and politics. As a result. Americans trust in the laws written, policies implemented, and electoral results are further undermined. Contrary to the belief Republican voting laws are restricting voting, this one worldview mindset is likely to lead to much less turnout. The polling industry is broken and 2021 has shown it.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s